From victor.aleo at estudiant.upc.es Thu Jan 2 13:37:28 2003 From: victor.aleo at estudiant.upc.es (Victor Aleo) Date: Thu, 02 Jan 2003 19:37:28 +0100 Subject: Resetting card problem Message-ID: <3E1486E8.9060308@estudiant.upc.es> Hi all, I have the next problem using 2 APs and 1 station configured as wlan1 (all of them running the latest release hostap-2002-10-12.tar.gz): when the station (which is associated with one AP and sending data) is disassociated (upon a "Disassociation" message from the AP) most of the times produces a reset of the station card (wlan1). Has been this problem solved in the last CVS version? Thanks, Victor wlan1: LinkStatus=2 (Disconnected) wlan1: LinkStatus: BSSID=xx:xx:xx:xx:xx:xx wlan1: hfa384x_setup_bap - timeout after wlan1: TXEXC - fid=0x0258 - could not read txdesc wlan1: scheduled card reset hostap_pci: wlan1: resetting card wlan1: removed pending cmd_queue entry (type=1, cmd=0x010b, param0=0x0138) prism2_hw_init() prism2_hw_config: initialized in 17874 iterations prism2_tx_timeout: wake up queue wlan0: xx:xx:xx:xx:xx:xx auth (alg=0 trans#=1 stat=0 len=6fc=00b0) ==> 1 (authentication denied) wlan1: LinkStatus=1 (Connected) wlan1: LinkStatus: BSSID=xx:xx:xx:xx:xx:xx From jasonb at edseek.com Thu Jan 2 15:40:39 2003 From: jasonb at edseek.com (Jason Boxman) Date: Thu, 2 Jan 2003 15:40:39 -0500 (EST) Subject: HostAP and D-Link 614+ Message-ID: <32804.192.168.0.12.1041540039.squirrel@rebecca.edseek.com> I tired to use a WPC11v3 Prism2 card to speak to a D-Link 614+ to no avail. My syslog has the following entries. I am using 2002-10-12. It associates, but when I send a DHCP discover things stop working. Anyone have any thoughts on this encounter? Jan 1 00:56:08 rachael kernel: hostap_cs: CS_EVENT_CARD_INSERTION Jan 1 00:56:08 rachael kernel: prism2_config() Jan 1 00:56:08 rachael kernel: hostap_cs: setting Vcc=50 (from config) Jan 1 00:56:08 rachael kernel: Checking CFTABLE_ENTRY 0x01 (default 0x01) Jan 1 00:56:08 rachael kernel: IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 Jan 1 00:56:08 rachael kernel: io->flags = 0x0046, io.base=0x0000, len=64 Jan 1 00:56:08 rachael kernel: hostap_cs: index 0x01: Vcc 5.0, irq 3, io 0x0100-0x013f Jan 1 00:56:09 rachael kernel: hostap_cs: Registered netdevice wlan0 Jan 1 00:56:09 rachael kernel: prism2_hw_init() Jan 1 00:56:09 rachael kernel: prism2_hw_config: initialized in 17300 iterations Jan 1 00:56:09 rachael kernel: wlan0: NIC: id=0x801a v1.0.0 Jan 1 00:56:09 rachael kernel: wlan0: PRI: id=0x15 v1.0.5 Jan 1 00:56:09 rachael kernel: wlan0: STA: id=0x1f v1.3.4 Jan 1 00:56:09 rachael kernel: wlan0: defaulting to host-based encryption as a workaround for firmware bug in Host AP mode WEP Jan 1 01:01:26 rachael kernel: wlan0: LinkStatus=2 (Disconnected) Jan 1 01:01:26 rachael kernel: wlan0: LinkStatus: BSSID=00:06:25:a9:f7:da Jan 1 01:01:30 rachael kernel: wlan0: LinkStatus=2 (Disconnected) Jan 1 01:01:30 rachael kernel: wlan0: LinkStatus: BSSID=44:44:44:44:44:44 Jan 1 01:01:34 rachael pumpd[7862]: starting at (uptime 0 days, 3:19:10) Wed Jan 1 01:01:34 2003 Jan 1 01:01:34 rachael pumpd[7862]: PUMP: sending discover Jan 1 01:01:34 rachael pumpd[7862]: breq: opcode: 1 Jan 1 01:01:34 rachael pumpd[7862]: breq: hw: 1 Jan 1 01:01:34 rachael pumpd[7862]: breq: hwlength: 6 Jan 1 01:01:34 rachael pumpd[7862]: breq: hopcount: 0 Jan 1 01:01:34 rachael pumpd[7862]: breq: xid: 0xa9250d59 Jan 1 01:01:34 rachael pumpd[7862]: breq: secs: 0 Jan 1 01:01:34 rachael pumpd[7862]: breq: flags: 0x0000 Jan 1 01:01:34 rachael pumpd[7862]: breq: ciaddr: 0.0.0.0 Jan 1 01:01:34 rachael pumpd[7862]: breq: yiaddr: 0.0.0.0 Jan 1 01:01:34 rachael pumpd[7862]: breq: server_ip: 0.0.0.0 Jan 1 01:01:34 rachael pumpd[7862]: breq: bootp_gw_ip: 0.0.0.0 Jan 1 01:01:34 rachael pumpd[7862]: breq: hwaddr: Jan 1 01:01:34 rachael pumpd[7862]: breq: servername: Jan 1 01:01:34 rachael pumpd[7862]: breq: bootfile: Jan 1 01:01:34 rachael pumpd[7862]: breq: vendor: 0x63 0x82 0x53 0x63 Jan 1 01:01:34 rachael pumpd[7862]: breq: vendor: 53 1 0x01 Jan 1 01:01:34 rachael pumpd[7862]: breq: vendor: 0xff Jan 1 01:01:34 rachael kernel: wlan0: TXEXC - fid=0x0120 - status=0x0004 ([Discon]) tx_control=000c Jan 1 01:01:34 rachael kernel: retry_count=0 tx_rate=0 fc=0x0108 (Data::0 ToDS) Jan 1 01:01:34 rachael kernel: A1=00:00:00:00:00:00 A2=00:06:25:a9:f7:da A3=ff:ff:ff:ff:ff:ff A4=00:00:00:00:00:00 Jan 1 01:01:38 rachael kernel: wlan0: TXEXC - fid=0x0137 - status=0x0004 ([Discon]) tx_control=000c Jan 1 01:01:38 rachael kernel: retry_count=0 tx_rate=0 fc=0x0108 (Data::0 ToDS) Jan 1 01:01:38 rachael kernel: A1=00:00:00:00:00:00 A2=00:06:25:a9:f7:da A3=ff:ff:ff:ff:ff:ff A4=00:00:00:00:00:00 Jan 1 01:01:45 rachael kernel: wlan0: TXEXC - fid=0x014c - status=0x0004 ([Discon]) tx_control=000c Jan 1 01:01:45 rachael kernel: retry_count=0 tx_rate=0 fc=0x0108 (Data::0 ToDS) Jan 1 01:01:45 rachael kernel: A1=00:00:00:00:00:00 A2=00:06:25:a9:f7:da A3=ff:ff:ff:ff:ff:ff A4=00:00:00:00:00:00 Jan 1 01:01:58 rachael kernel: wlan0: TXEXC - fid=0x0161 - status=0x0004 ([Discon]) tx_control=000c Jan 1 01:01:58 rachael kernel: retry_count=0 tx_rate=0 fc=0x0108 (Data::0 ToDS) Jan 1 01:01:58 rachael kernel: A1=00:00:00:00:00:00 A2=00:06:25:a9:f7:da A3=ff:ff:ff:ff:ff:ff A4=00:00:00:00:00:00 Jan 1 01:02:04 rachael kernel: wlan0: TXEXC - fid=0x0176 - status=0x0004 ([Discon]) tx_control=000c Jan 1 01:02:04 rachael kernel: retry_count=0 tx_rate=0 fc=0x0108 (Data::0 ToDS) Jan 1 01:02:04 rachael kernel: A1=00:00:00:00:00:00 A2=00:06:25:a9:f7:da A3=ff:ff:ff:ff:ff:ff A4=00:00:00:00:00:00 Jan 1 01:02:04 rachael kernel: wlan0: TXEXC - fid=0x018b - status=0x0004 ([Discon]) tx_control=000c Jan 1 01:02:04 rachael kernel: retry_count=0 tx_rate=0 fc=0x0108 (Data::0 ToDS) Jan 1 01:02:04 rachael kernel: A1=00:00:00:00:00:00 A2=00:06:25:a9:f7:da A3=ff:ff:ff:ff:ff:ff A4=00:00:00:00:00:00 From perw at vismaprovider.no Thu Jan 2 16:36:53 2003 From: perw at vismaprovider.no (Per Weisteen) Date: Thu, 2 Jan 2003 22:36:53 +0100 (CET) Subject: Use Linksys WMP11 (PCI) with hostap ? Message-ID: I feel there is some confusion re. the use of Linksys PCI WMP11 card as an Access Point. I've seen some info indicating that the card is not compatible with the hostap driver (http://trekweb.com/~jasonb/articles/hostap_20021012.shtml) while AVS (http://www.linux-wlan.org/) in their compatibility list states that the card is a PRISM2 compatible card. Could anyone enlighten me on this subject ? If the WMP11 Card is not supported could anyone suggest a PCI card (preferrably with external antenna) that is ? Regards, -- Per W. From andre.correa at pobox.com Thu Jan 2 18:33:12 2003 From: andre.correa at pobox.com (=?iso-8859-1?Q?Andr=E9_Corr=EAa?=) Date: Thu, 2 Jan 2003 18:33:12 -0500 (EST) Subject: Association and Disassociation info Message-ID: <44813.200.213.210.132.1041550392.squirrel@webmail.hiperlinks.com.br> Hi list. I need to do something unusual and maybe somebody in this list can help me. I need to know, as quickly as possible, when a station disassociates from the AP. Looking at /var/log/debug I've found that it takes 5 minutes for the AP to realize that the station disassociated. Is there a way for me to know quickly that a STA disassociated? I've thought that I may be able to change hostap's code to decrease association timeout to 1 minute or less. Are there any implications in doing this? Is there a better way to know an station disassociated then looking at /var/log/debug or to files created at the /proc/net/hostap/wlan0 directory? Any help will be apreciated. Tks in advance. Andre From Peter.DeCleyn at ua.ac.be Thu Jan 2 18:48:21 2003 From: Peter.DeCleyn at ua.ac.be (Peter De Cleyn) Date: Fri, 3 Jan 2003 00:48:21 +0100 (CET) Subject: Use Linksys WMP11 (PCI) with hostap ? In-Reply-To: Message-ID: On Thu, 2 Jan 2003, Per Weisteen wrote: > > I feel there is some confusion re. the use of Linksys PCI WMP11 card as an Access Point. > I've seen some info indicating that the card is not compatible with the hostap driver > (http://trekweb.com/~jasonb/articles/hostap_20021012.shtml) while > AVS (http://www.linux-wlan.org/) in their compatibility list states that the card is a PRISM2 compatible card. Per, Only the firmware of your card is not compatible with the HostAP mode, so other modes should work!. This can easily be changed as is also described in the article ( or one of the other 2 about wlans) on the site you mention. The trick is to load another secondary firmware in RAM. I tried this with the WPC 11 (pcmcia version) and this works pretty well. Peter > > Could anyone enlighten me on this subject ? > > If the WMP11 Card is not supported could anyone suggest a PCI card (preferrably with external antenna) that is ? > > Regards, > > -- > Per W. > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From jasonb at edseek.com Thu Jan 2 18:55:56 2003 From: jasonb at edseek.com (Jason Boxman) Date: Thu, 2 Jan 2003 18:55:56 -0500 (EST) Subject: Use Linksys WMP11 (PCI) with hostap ? In-Reply-To: References: Message-ID: <3950.209.208.68.56.1041551756.squirrel@rebecca.edseek.com> Per Weisteen said: The WMP11 PCI card is no longer Prism2 chipset based. It is now Broadcom based. This happened some months ago, so it's possible AVS hasn't updated their compatibility list. > > I feel there is some confusion re. the use of Linksys PCI WMP11 card as an > Access Point. > I've seen some info indicating that the card is not compatible with the > hostap driver > (http://trekweb.com/~jasonb/articles/hostap_20021012.shtml) while > AVS (http://www.linux-wlan.org/) in their compatibility list states that > the card is a PRISM2 compatible card. > > Could anyone enlighten me on this subject ? > > If the WMP11 Card is not supported could anyone suggest a PCI card > (preferrably with external antenna) that is ? > > Regards, > > -- > Per W. > From jasonb at edseek.com Thu Jan 2 18:59:47 2003 From: jasonb at edseek.com (Jason Boxman) Date: Thu, 2 Jan 2003 18:59:47 -0500 (EST) Subject: Use Linksys WMP11 (PCI) with hostap ? In-Reply-To: References: Message-ID: <3956.209.208.68.56.1041551987.squirrel@rebecca.edseek.com> Peter De Cleyn said: > On Thu, 2 Jan 2003, Per Weisteen wrote: > >> >> I feel there is some confusion re. the use of Linksys PCI WMP11 card as >> an Access Point. >> I've seen some info indicating that the card is not compatible with the >> hostap driver >> (http://trekweb.com/~jasonb/articles/hostap_20021012.shtml) while >> AVS (http://www.linux-wlan.org/) in their compatibility list states that >> the card is a PRISM2 compatible card. > > Per, > > Only the firmware of your card is not compatible with the HostAP mode, so > other modes should work!. I think you're thinking of cards with STA v1.4.2. That fails to function with HostAP mode. New the WMP11 (labeled v2.7 on the chipset itself, but you cannot tell from the labeling) cards use the Broadcom chipset and fail to work at all. HostAP does not support them. > This can easily be changed as is also described in the article ( or one of > the other 2 about wlans) on the site you mention. The trick is to load > another secondary firmware in RAM. I tried this with the WPC 11 (pcmcia > version) > and this works pretty well. Indeed you can, as long as it is a Prism2 card. (I think you're limited to STA <= 1.4.9 unless your PRI is >= 1.1.0, though.) > Peter > >> >> Could anyone enlighten me on this subject ? >> >> If the WMP11 Card is not supported could anyone suggest a PCI card >> (preferrably with external antenna) that is ? >> >> Regards, >> >> -- >> Per W. >> From jkmaline at cc.hut.fi Thu Jan 2 21:27:21 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 2 Jan 2003 18:27:21 -0800 Subject: Association and Disassociation info In-Reply-To: <44813.200.213.210.132.1041550392.squirrel@webmail.hiperlinks.com.br> References: <44813.200.213.210.132.1041550392.squirrel@webmail.hiperlinks.com.br> Message-ID: <20030103022721.GA1338@jm.kir.nu> On Thu, Jan 02, 2003 at 06:33:12PM -0500, Andr? Corr?a wrote: > I need to know, as quickly as possible, when a station disassociates from > the AP. Looking at /var/log/debug I've found that it takes 5 minutes for > the AP to realize that the station disassociated. If the station indeed disassociates (i.e., sends a disassociation frame), you should get information about this immediately. However, in many cases, station does not disassociate (e.g., it re-associates with another AP or it just goes out of range). In these cases, the timeout in the AP is used to disassociate the station eventually. > Is there a way for me to know quickly that a STA disassociated? I've > thought that I may be able to change hostap's code to decrease association > timeout to 1 minute or less. Are there any implications in doing this? Station-initiated disassociation ends up in hostap_disassoc() which will print a log message about this and will send IWEVEXPIRED wireless event. If you need to also know quickly that a station has moved away (without disassociating), you could decrease timeout limit. Host AP driver uses an empty data frame to poll the station when this limit is reached, so you should only get few more frames and connection should be maintained. Disassociation is completed only if this data frame is not ACKed. > Is there a better way to know an station disassociated then looking at > /var/log/debug or to files created at the /proc/net/hostap/wlan0 > directory? Yes, if you are using Linux wireless extensions version 15 or newer. Host AP code will send IWEVEXPIRED wireless event when a station is disassociated (check, e.g., iwevent for how to receive these events). -- Jouni Malinen PGP id EFC895FA From raghu at sraamerica.com Fri Jan 3 12:26:57 2003 From: raghu at sraamerica.com (Raghu) Date: Fri, 3 Jan 2003 12:26:57 -0500 Subject: 802.1X Unicast Keys? Message-ID: <20030103172657.YPHD5165.imta02a2.registeredsite.com@smtp.registeredsite.com> Hi Happy New Year to everyone. A quick update, I have limited success with unicast keys, and need your help to make it to work. Once again my setup is XSupplicant, HostAP(client), HostAP(AP), Hostapd (Authenticator), FreeRADIUS. STA configuration, iwpriv wlan0 host_encrypt 1 iwpriv wlan0 host_decrypt 1 iwpriv wlan0 ieee_802_1x 1 iwpriv wlan0 host_roaming 1 I am not sure if I can use, iwpriv wlan0 bcrx_sta_key 1 ? AP command, $ ./hostapd -x -b13 -i13 -o192.168.1.1 -a192.168.1.1 -stest -SSSID wlan0 Though I am using the same hostap cvs snapshot for both STA and AP, Wireless extensions version differs at AP(v12) and STA(v15). After hostapd generates both broadcast and unicast keys, iwconfig and iwlist shows only the broadcast key. # iwlist wlan0 key wlan0 2 key sizes : 40, 104bits 4 keys available : [1]: off [2]: 64A7-4D5C-D6C6-3653-505E-FB52-85 (104 bits) [3]: off [4]: off Current Transmit Key: [2] Encryption mode:open How can I see the unicast key for each STA at AP ? Xsupplicant receives the unicast key, occasionally, I am not sure when it receives the unicast key and when it doesnot. I was under the impression that whenever Xsupplicant sends EAPOL start, a new authentication takes place and AP should generate a new unicast key and STA should always receive it, but this doesnot seem to be happening. After Xsupplicant receives both the broadcast key and unicast key, iwconfig lists the broadcast key and iwlist shows both the keys, ie # iwlist wlan0 key wlan0 2 key sizes : 40, 104bits 4 keys available : [1]: 056D-031D-E10E-5780-4E2A-07A2-3F (104 bits) [2]: 64A7-4D5C-D6C6-3653-505E-FB52-85 (104 bits) [3]: off [4]: off Current Transmit Key: [2] Encryption mode:open Because of this Incompatibility STA is still not able to ping the AP. Can you please let me know what are the other configurations/commands that I need to set at STA/AP to make unicast keys working ? -Raghu From wxdai at utstar.com Fri Jan 3 22:55:10 2003 From: wxdai at utstar.com (Wenxiang Dai) Date: Fri, 3 Jan 2003 22:55:10 -0500 Subject: how to check out the source code Message-ID: Hello, everybody, I just found out this project, can anybody kindly tell me how to check out the latest code? I am currently working on the intersil chipset driver stuff, this mail list will definitely help me lot. Thanks From wxdai at utstar.com Fri Jan 3 23:08:34 2003 From: wxdai at utstar.com (Wenxiang Dai) Date: Fri, 3 Jan 2003 23:08:34 -0500 Subject: how to check out the source code In-Reply-To: Message-ID: please forgive my blindness! I already figured out the homepage, and got the latest code, cool stuff! I like to raise my dumb questions: Do you know which version of intersil tertiary F/W which can support the host authentication while it's running tertiary mode? and how to handle the "authenticationRequest" info frame? many thanks! -----Original Message----- From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com]On Behalf Of Wenxiang Dai Sent: Friday, January 03, 2003 10:55 PM To: hostap at shmoo.com Subject: how to check out the source code Hello, everybody, I just found out this project, can anybody kindly tell me how to check out the latest code? I am currently working on the intersil chipset driver stuff, this mail list will definitely help me lot. Thanks _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap From wireless at douglasheld.net Sat Jan 4 00:26:24 2003 From: wireless at douglasheld.net (wireless at douglasheld.net) Date: Fri, 3 Jan 2003 21:26:24 -0800 Subject: Fw: deauthentication: reason_code=3 Message-ID: <005401c2b3b1$d5db7d70$0200a8c0@buttrump> Hello, I am running hostap-2002-10-12 (the .plx driver) on a Red Hat 6.2 machine, and I am investigating roaming with a laptop, where I come in and out of range of the AP. My AP consists of the plx card in the RedHat box, and a Dlink DWL 900-AP+ configured as a repeater of the plx radio. What I find is that the card software on the laptop will happily state that is is still associated (or RE-associated), and the LEDs show that it associated. However, no packets get through. I find the following (or similar) in my messages file: My question, I guess, is where can I decode "reason_code=3", and can anyone guess a setting I can use to instruct the AP to be agreeable when clients with bad connections move to a better location? Thanks, Jouni. --Doug Note: MAC 0090D1078103 is the radio in the RedHat box. MAC 00:40:05:D0:4A:1A is the Dlink repeater AP MAC 00:30:AB:14:0F:1C is the moving laptop. wlan0: TXEXC - fid=0x041a - status=0x0001 ([RetryErr]) tx_control=000c retry_count=0 tx_rate=0 fc=0x0a08 (Data::0) addr1=00:30:ab:14:0f:1c addr2=00:90:d1:07:81:03 addr3=00:90:d1:07:81:03 wlan0: STA 00:30:ab:14:0f:1c TX rate lowered to 10 wlan0: sending disassociation info to STA 00:30:ab:14:0f:1c(last=390603, jiffies=391703) wlan0: TXEXC - fid=0x0228 - status=0x0001 ([RetryErr]) tx_control=000c retry_count=0 tx_rate=0 fc=0x08a0 (Mgmt::10) addr1=00:30:ab:14:0f:1c addr2=00:90:d1:07:81:03 addr3=00:90:d1:07:81:03 wlan0: sending deauthentication info to STA 00:30:ab:14:0f:1c(last=390603, jiffies=391803) wlan0: TXEXC - fid=0x0431 - status=0x0001 ([RetryErr]) tx_control=000c retry_count=0 tx_rate=0 fc=0x08c0 (Mgmt::12) addr1=00:30:ab:14:0f:1c addr2=00:90:d1:07:81:03 addr3=00:90:d1:07:81:03 Could not find STA for this TX error wlan0: dropped received packet from non-associated STA 00:30:ab:14:0f:1c (type=0x02, subtype=0x00) handle_ap_item - data frame wlan0: authentication: 00:30:ab:14:0f:1c len=6, auth_alg=0, auth_transaction=1, status_code=0, fc=0x00b0 wlan0: new STA 00:30:ab:14:0f:1c wlan0: association request: from 00:30:ab:14:0f:1c to 00:90:d1:07:81:03 len=30 capability=0x0001, listen_interval=1 - new AID 2 wlan0: deauthentication: 00:30:ab:14:0f:1c len=2, reason_code=3 wlan0: deauthentication: 00:30:ab:14:0f:1c len=2, reason_code=3 wlan0: deauthentication: 00:30:ab:14:0f:1c len=2, reason_code=3 wlan0: deauthentication: 00:30:ab:14:0f:1c len=2, reason_code=3 From jkmaline at cc.hut.fi Sat Jan 4 02:49:16 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Fri, 3 Jan 2003 23:49:16 -0800 Subject: Fw: deauthentication: reason_code=3 In-Reply-To: <005401c2b3b1$d5db7d70$0200a8c0@buttrump> References: <005401c2b3b1$d5db7d70$0200a8c0@buttrump> Message-ID: <20030104074916.GA26262@jm.kir.nu> On Fri, Jan 03, 2003 at 09:26:24PM -0800, wireless at douglasheld.net wrote: > My AP consists of the plx card in the RedHat box, and a Dlink DWL 900-AP+ > configured as a repeater of the plx radio. Do you know exactly what that repeater configuration is doing? Is it supposed to extend range of AP (using Host AP driver)? Are you sure it is compatible with Host AP driver? > What I find is that the card software on the laptop will happily state that > is is still associated (or RE-associated), and the LEDs show that it > associated. However, no packets get through. Associated with which AP (Host AP/D-Link DWL 900-AP+)? > I find the following (or similar) in my messages file: My question, I > guess, is where can I decode "reason_code=3", and can anyone guess a setting > I can use to instruct the AP to be agreeable when clients with bad > connections move to a better location? Reason codes are defined in IEEE Std 802.11 (3 = deauthentication because sending station is leaving or has left). > wlan0: TXEXC - fid=0x041a - status=0x0001 ([RetryErr]) tx_control=000c > retry_count=0 tx_rate=0 fc=0x0a08 (Data::0) > addr1=00:30:ab:14:0f:1c addr2=00:90:d1:07:81:03 addr3=00:90:d1:07:81:03 > wlan0: STA 00:30:ab:14:0f:1c TX rate lowered to 10 > wlan0: sending disassociation info to STA 00:30:ab:14:0f:1c(last=390603, > jiffies=391703) I would assume that the station re-associated with the D-Link AP and Host AP is removing it after timeout. > wlan0: association request: from 00:30:ab:14:0f:1c to 00:90:d1:07:81:03 > len=30 > capability=0x0001, listen_interval=1 - new AID 2 Here the station re-associated with Host AP.. > wlan0: deauthentication: 00:30:ab:14:0f:1c len=2, reason_code=3 .. but, for some reason decided to deauthenticate. -- Jouni Malinen PGP id EFC895FA From nsubtil at vega.dhis.co.it.pt Sat Jan 4 14:39:05 2003 From: nsubtil at vega.dhis.co.it.pt (Nuno Subtil) Date: Sat, 4 Jan 2003 19:39:05 +0000 Subject: Linksys WPC11 ver. 3 hangs on insert Message-ID: <20030104193905.GA1324@vega.dhis.co.it.pt> I am using a Linksys WPC11 version 3 card with hostap-2002-10-12. Sometimes, shortly after inserting the card and loading hostap_cs, my machine hangs. Everything stops (even the console cursor stops blinking), and I have to hit the reset button. The last few messages I get on dmesg are: wlan0: LinkStatus: BSSID=44:44:44:44:44:44 wlan0: LinkStatus=2 (Disconnected) wlan0: LinkStatus: BSSID=44:44:44:44:44:44 wlan0: LinkStatus=6 (Association failed) When it doesn't cause the machine to hang, however, the card works just fine. Is there any workaround for this ? The machine is an Omnibook XE3-GF, and I'm running pcmcia-cs-3.1.33. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030104/e70e8cd1/attachment.pgp From imp at airlinktek.com Mon Jan 6 04:23:14 2003 From: imp at airlinktek.com (=?utf-7?B?K3ZCWFhaY1gwLQ==?=) Date: Mon, 6 Jan 2003 18:23:14 +0900 Subject: hostapd help.. Message-ID: Dear all.. When I run hostapd as #hostapd wlan0 or #hostapd -xm wlan0 the hostapd message always show bellow message.. and the client wireless PC connect to AP then disconnect again and again.. Could you tell me what is my mistake ? The debug message is Using interface wlan0ap with hwaddr 00:30:00:12:01:1b and ssid 'test' Flushing old station entries Station 00:30:00:40:a4:40 authentication OK (open system) Station 00:30:00:40:a4:40 authenticated Station 00:30:00:40:a4:40 association OK (aid 1) handle_assoc_cb(reassoc=1) - too short payload (len=30) Data frame from not associated STA 00:30:00:40:a4:40 Data frame from not associated STA 00:30:00:40:a4:40 Station 00:30:00:40:a4:40 authentication OK (open system) Station 00:30:00:40:a4:40 authenticated Station 00:30:00:40:a4:40 association OK (aid 1) handle_assoc_cb(reassoc=1) - too short payload (len=30) Data frame from not associated STA 00:30:00:40:a4:40 Station 00:30:00:40:a4:40 authentication OK (open system) Station 00:30:00:40:a4:40 authenticated Station 00:30:00:40:a4:40 association OK (aid 1) handle_assoc_cb(reassoc=1) - too short payload (len=30) Data frame from not associated STA 00:30:00:40:a4:40 Station 00:30:00:40:a4:40 authentication OK (open system) Station 00:30:00:40:a4:40 authenticated Station 00:30:00:40:a4:40 association OK (aid 1) handle_assoc_cb(reassoc=1) - too short payload (len=30) Thanks for your attantion... From ftpmaster at deltanet.ro Mon Jan 6 06:02:32 2003 From: ftpmaster at deltanet.ro (ftpMaster) Date: Mon, 6 Jan 2003 13:02:32 +0200 Subject: FW 1.4.9 Message-ID: <004901c2b573$9659a8c0$a923fea9@p4> Hi, just f***ed up a Prism2.5 PCMCIA card... First I updated a Prism2 (0x8002) PCMCIA 0.3.0/0.8.0 to 1.3.4, later to 1.4.9 with the file posted by Pedro Estrela. Everything worked OK. I didn't found any other Pri F/W to D/L for it. Then (tried to) update(d) the Prism2.5 (0x800c) PCMCIA 1.0.5/1.3.4 to 1.4.9. The winupdate program reported successfull update, after wich windblows begun to slow down and locked temporarily, until the card was ejected. After reinsertion, same simptoms. The card wouldn't be recognized again. It seems that it's necessary to update simultaneously the Pri F/W to 1.1.0 (wich comes in the same archive with Sta F/W 1.4.9), wich worked on another identical card. Is there any utility wich permits flashing 1.3.4 again, without the card beeing recognized by the O/S? Best regards, Bogdan From Phillip.Watts at nlynx.com Mon Jan 6 11:30:51 2003 From: Phillip.Watts at nlynx.com (Phillip.Watts at nlynx.com) Date: Mon, 6 Jan 2003 10:30:51 -0600 Subject: Function of hostapd? Message-ID: <86256CA6.005AB7A6.00@nlynx.com> I have built a linux router with a DLink DWL 520 pci card as my internal interface using the hostap_pci.o as the driver and the mode set to MASTER. This router is now functioning as an Access Point, perfectly as far as I can tell.(GREAT software, way to go.) I do NOT have hostapd loaded. I cannot tell from the README what its function is. Anyone enlighten me? From ammirata at econointl.com Mon Jan 6 14:19:23 2003 From: ammirata at econointl.com (Sergio M. Ammirata) Date: Mon, 6 Jan 2003 14:19:23 -0500 Subject: hostapd forking and enhancements Message-ID: <001501c2b5b8$865718a0$017ba8c0@ammirata.net> I have two suggestions: 1) hostapd should support forking. I have tried to use "hostapd &" on my init scripts in order to get it loaded on startup but as soon as the init script dies the hostapd dies with it. 2) The station authentication currently supports open access and mac address list. It would very helpful if hostapd would add an option to support the execution of an external bin for authentication. Perhaps it can pass two arguments, the mac of wlan0 and the mac of the station and based on the response from the bin it lets the station in . Any comments? Sergio -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030106/c546b065/attachment.htm From havanna_moon at gmx.net Mon Jan 6 14:34:23 2003 From: havanna_moon at gmx.net (Yon Uriarte) Date: Mon, 6 Jan 2003 20:34:23 +0100 (CET) Subject: Dead card Message-ID: Hi, I'm having problems with an SMC2632W (prism2) card. After forgetting to plug the AC-cord on the laptop, it run until its battery drained and it automatically shut down. Afterward, the card wont associate with the AP anymore. Monitor mode works (!). This was under linux. I've tested it under windows, and it doesnt work either. Updating to sta 1.4.9 didnt help either (linux or windows). I'm not susbscribed, so please, CC me.. I'll happily provide any information required. BTW, I cant get the last version of hostap to work on my kernel (fails when assigning an IRQ), but I suspect a pcmcia-cs/kernel version mismatch. TIA, regards, yon From BichotG at tce.com Mon Jan 6 17:08:46 2003 From: BichotG at tce.com (Bichot Guillaume (Princeton)) Date: Mon, 6 Jan 2003 17:08:46 -0500 Subject: Wireless on IPAQ Message-ID: Referring some emails exchanges on linux-arm-kernel list in last Novemeber about Wireless on IPAQ; I understood that (and I confirm) only the couple wvlan_cs driver,Lucent/orinoco card works fine whatever the Familiar release and the ipaq series. However I need to work with the hostap_cs driver for different reasons. What I've tested is that the behavior is different depending on the card but also depending on the ipaq version. Any way the best I could obtain was something running but with these so famous timeouts (watch-dog). I've read that this should come from the PCMCIA access time. My question is then how can we change this PCMCIA access timeout ? Thanks in advance for the help. Guillaume Bichot -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030106/8345d2d9/attachment.htm From dave at cc0.net Mon Jan 6 18:24:47 2003 From: dave at cc0.net (Dave) Date: Mon, 6 Jan 2003 15:24:47 -0800 Subject: Compile problem on 2.4.20 Message-ID: <000c01c2b5da$f0e51c40$0507ee3f@ws1> Hi All, I was wondering if anyone might know why PRISM2_HOSTAPD would not be defined in hostap_wlan.h? To build the hostap drivers I used instructions located in kernel-patches/README and copied the .c and .h files to the 2.4.20 tree as well as applied the patch. When compiling the modules hostap_cs.c hits assoc_ap_addr being undefined and cannot continue. I assume that that define should only be used when building the hostapd app and something should change when building hostap as part of the kernel...yes? Anyway, I just defined it and got it built just fine...just a heads up.. Thanks! -Dave From jasonb at edseek.com Mon Jan 6 18:41:18 2003 From: jasonb at edseek.com (Jason Boxman) Date: Mon, 6 Jan 2003 18:41:18 -0500 (EST) Subject: Function of hostapd? In-Reply-To: <86256CA6.005AB7A6.00@nlynx.com> References: <86256CA6.005AB7A6.00@nlynx.com> Message-ID: <37736.65.33.153.223.1041896478.squirrel@rebecca.edseek.com> Phillip.Watts at nlynx.com said: > > > I have built a linux router with a DLink DWL 520 pci card > as my internal interface using the hostap_pci.o as the driver > and the mode set to MASTER. > > This router is now functioning as an Access Point, perfectly as > far as I can tell.(GREAT software, way to go.) > > I do NOT have hostapd loaded. I cannot tell from the > README what its function is. Anyone enlighten me? hostapd is a userspace daemon that deals with 802.1x related things. If you compile with hostapd support enabled, you have to run hostapd for anything useful to be possible with your card in master mode. From jkmaline at cc.hut.fi Mon Jan 6 21:23:28 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 6 Jan 2003 18:23:28 -0800 Subject: Compile problem on 2.4.20 In-Reply-To: <000c01c2b5da$f0e51c40$0507ee3f@ws1> References: <000c01c2b5da$f0e51c40$0507ee3f@ws1> Message-ID: <20030107022327.GA1155@jm.kir.nu> On Mon, Jan 06, 2003 at 03:24:47PM -0800, Dave wrote: > I was wondering if anyone might know why PRISM2_HOSTAPD would not be defined > in hostap_wlan.h? It is supposed to be defined only from hostap_config.h or Makefile.. > When compiling the modules hostap_cs.c hits assoc_ap_addr being undefined > and cannot continue. That's a bug I introduced on Saturday.. It is now fixed in CVS. > I assume that that define should only be used when building the hostapd app > and something should change when building hostap as part of the > kernel...yes? Yes, PRISM2_HOSTAPD should be defined when compiling driver for hostapd-use. assoc_ap_addr is only supported with hostapd, so it should not have been refered at all without this definition. > Anyway, I just defined it and got it built just fine...just a heads up.. That would also mean that you would need to use hostapd for management functionality.. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Mon Jan 6 22:57:40 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 6 Jan 2003 19:57:40 -0800 Subject: hostapd forking and enhancements In-Reply-To: <001501c2b5b8$865718a0$017ba8c0@ammirata.net> References: <001501c2b5b8$865718a0$017ba8c0@ammirata.net> Message-ID: <20030107035740.GB1812@jm.kir.nu> On Mon, Jan 06, 2003 at 02:19:23PM -0500, Sergio M. Ammirata wrote: > 1) hostapd should support forking. I have tried to use "hostapd &" on my > init scripts in order to get it loaded on startup but as soon as the > init script dies the hostapd dies with it. Yes, I agree. In addition, hostapd would benefit of an option to direct some logging to syslog instead of stdout. These are on my todo list.. > 2) The station authentication currently supports open access and mac > address list. It would very helpful if hostapd would add an option to > support the execution of an external bin for authentication. Perhaps it > can pass two arguments, the mac of wlan0 and the mac of the station and > based on the response from the bin it lets the station in . I would pass interface name and MAC address of the station, but yes, something like this has been requested couple of times. Instead of external program, hostapd could also use RADIUS server directly since there is already routines for generating and parsing RADIUS messages. However, there is a small issue with external authentication. IEEE 802.11 station implementations seem to use quite small timeout value for authentication frames. In other words, often there is not enough time to execute query to RADIUS server or external program. hostapd would need to cache the external reply for some time and rely on station trying again. Cached reply could then be used to generate authentication reply quickly enough. If someone can point me to a specification of which RADIUS attributes are used in this kind of authentication query (i.e., how to encode the MAC address of the station into Access-Request so that the RADIUS server understands it), I'll try to add support for it. I can also accept sniffed RADIUS exchange as an alternative for this specification ;-). -- Jouni Malinen PGP id EFC895FA From joshk at mspencer.net Mon Jan 6 23:02:39 2003 From: joshk at mspencer.net (Joshua Kwan) Date: Mon, 6 Jan 2003 20:02:39 -0800 Subject: Function of hostapd? In-Reply-To: <86256CA6.005AB7A6.00@nlynx.com> References: <86256CA6.005AB7A6.00@nlynx.com> Message-ID: <20030106200239.481d5f13.joshk@mspencer.net> Cool, I'm using the 520 as well.. :) I'd like to know too. But it seems to me like a wifi authentication gateway. Enlighten me as well ... Regards Josh Rabid cheeseburgers forced Phillip.Watts at nlynx.com to write this on Mon, 6 Jan 2003 10:30:51 -0600: > > > I have built a linux router with a DLink DWL 520 pci card > as my internal interface using the hostap_pci.o as the driver > and the mode set to MASTER. > > This router is now functioning as an Access Point, perfectly as > far as I can tell.(GREAT software, way to go.) > > I do NOT have hostapd loaded. I cannot tell from the > README what its function is. Anyone enlighten me? > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap -- Joshua Kwan joshk at mspencer.net pgp public key at http://joshk.mspencer.net/pubkey_gpg.asc It's hard to keep your shirt on when you're getting something off your chest. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030106/33d346a0/attachment.pgp From jkmaline at cc.hut.fi Tue Jan 7 01:02:01 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 6 Jan 2003 22:02:01 -0800 Subject: Function of hostapd? In-Reply-To: <86256CA6.005AB7A6.00@nlynx.com> References: <86256CA6.005AB7A6.00@nlynx.com> Message-ID: <20030107060201.GA25433@jm.kir.nu> On Mon, Jan 06, 2003 at 10:30:51AM -0600, Phillip.Watts at nlynx.com wrote: > I do NOT have hostapd loaded. I cannot tell from the > README what its function is. Anyone enlighten me? hostapd is a user space alternative for handling IEEE 802.11 management functions (like authentication and association) that was originally performed in the kernel driver. The default configuration is still to use kernel driver for this; this does not require hostapd at all. However, hostapd is the future :-). Nowadays, hostapd includes quite a bit more than only 802.11 authentication and association. It supports IEEE 802.1x authentication and authorization, RADIUS accounting, parts of IAPP, etc. In addition, most of the new higher level functionality will be added only to the user space daemon and using hostapd (i.e., define PRISM2_HOSTAPD for the driver) will eventually become the default option and I might even remove the option to cleanup the driver code (i.e., remove all the authentication and association related functionality from the kernel driver). -- Jouni Malinen PGP id EFC895FA From dave at cc0.net Tue Jan 7 01:14:50 2003 From: dave at cc0.net (Dave) Date: Mon, 6 Jan 2003 22:14:50 -0800 Subject: Function of hostapd? References: <86256CA6.005AB7A6.00@nlynx.com> <20030107060201.GA25433@jm.kir.nu> Message-ID: <000401c2b614$16477c60$0200a8c0@ws7> If there is a vote going on regarding removing all association/authentication/management from the driver code and putting it in userspace....then I am +1 on it...;) -Dave > hostapd is a user space alternative for handling IEEE 802.11 management > functions (like authentication and association) that was originally > performed in the kernel driver. The default configuration is still to > use kernel driver for this; this does not require hostapd at all. > However, hostapd is the future :-). > > Nowadays, hostapd includes quite a bit more than only 802.11 > authentication and association. It supports IEEE 802.1x authentication > and authorization, RADIUS accounting, parts of IAPP, etc. In addition, > most of the new higher level functionality will be added only to the > user space daemon and using hostapd (i.e., define PRISM2_HOSTAPD for the > driver) will eventually become the default option and I might even > remove the option to cleanup the driver code (i.e., remove all the > authentication and association related functionality from the kernel > driver). > > -- > Jouni Malinen PGP id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From ammirata at econointl.com Tue Jan 7 02:03:09 2003 From: ammirata at econointl.com (Sergio Ammirata) Date: Tue, 7 Jan 2003 02:03:09 -0500 Subject: hostapd forking and enhancements In-Reply-To: <20030107035740.GB1812@jm.kir.nu> Message-ID: <000601c2b61a$d83bae30$350aa8c0@sergioxp> > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] > On Behalf Of Jouni Malinen > Sent: Monday, January 06, 2003 10:58 PM > To: hostap at shmoo.com > Subject: Re: hostapd forking and enhancements > > > On Mon, Jan 06, 2003 at 02:19:23PM -0500, Sergio M. Ammirata wrote: > > > 1) hostapd should support forking. I have tried to use > "hostapd &" on > > my init scripts in order to get it loaded on startup but as soon as > > the init script dies the hostapd dies with it. > > Yes, I agree. In addition, hostapd would benefit of an option > to direct some logging to syslog instead of stdout. These are > on my todo list.. I think the forking is rather critical since it prevents people from deploying the software in a production environment. Specially if the are using it in a custom distro with limited tools. > > > 2) The station authentication currently supports open > access and mac > > address list. It would very helpful if hostapd would add an > option to > > support the execution of an external bin for > authentication. Perhaps > > it can pass two arguments, the mac of wlan0 and the mac of > the station > > and based on the response from the bin it lets the station in . > > I would pass interface name and MAC address of the station, > but yes, something like this has been requested couple of > times. Instead of external program, hostapd could also use > RADIUS server directly since there is already routines for > generating and parsing RADIUS messages. > > However, there is a small issue with external authentication. > IEEE 802.11 station implementations seem to use quite small > timeout value for authentication frames. In other words, > often there is not enough time to execute query to RADIUS > server or external program. hostapd would need to cache the > external reply for some time and rely on station trying > again. Cached reply could then be used to generate > authentication reply quickly enough. There would have to be a timeout on the cache responses though. Just in case a MAC is deactivated on the radius server. > If someone can point me to a specification of which RADIUS > attributes are used in this kind of authentication query > (i.e., how to encode the MAC address of the station into > Access-Request so that the RADIUS server understands it), > I'll try to add support for it. I can also accept sniffed > RADIUS exchange as an alternative for this specification ;-). I like the idea of RADIUS. Perhaps we can use PAP authentication and send the station MAC in both the username and password attributes. I have attached a tcpdump capture of a PAP radius communication between a pppd daemon and a Microsoft IAS Radius Server. It includes connect and disconnect (use tcpdump -r filename to read). I used "testuser" and "testpassword" in this case. > > -- > Jouni Malinen PGP > id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap > Hope this helps, Sergio Ammirata -------------- next part -------------- A non-text attachment was scrubbed... Name: tcpdump.log Type: application/octet-stream Size: 650 bytes Desc: not available Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030107/b7a6514c/attachment.obj From ian.cass at mblox.com Tue Jan 7 04:08:57 2003 From: ian.cass at mblox.com (Ian Cass) Date: Tue, 7 Jan 2003 09:08:57 -0000 Subject: hostapd forking and enhancements References: <001501c2b5b8$865718a0$017ba8c0@ammirata.net> <20030107035740.GB1812@jm.kir.nu> Message-ID: <013201c2b62c$6935e070$0200a8c0@defender> > > 1) hostapd should support forking. I have tried to use "hostapd &" on my > > init scripts in order to get it loaded on startup but as soon as the > > init script dies the hostapd dies with it. I've used this http://www.ncftpd.com/unixstuff/q2getty.html for years now. It was originally designed for Quake game servers, but it will work for just about anything you throw at it. These days I use it to control my perl code. Basically, it spawns the process of your choice on a virtual terminal instead of 'login'. All config is done in /etc/inittab. > I would pass interface name and MAC address of the station, but yes, > something like this has been requested couple of times. Instead of > external program, hostapd could also use RADIUS server directly since > there is already routines for generating and parsing RADIUS messages. Would also be cool if there was a way to tag packets from each user differently as they fly past so we can maintain the concept of sessions all the way to the firewall. This then allows tradition logon processes, such as a web page to stand a chance of working even halfway securely. Vlan tagging perhaps? or TOS at a stretch. -- Ian Cass From reyk at vantronix.net Tue Jan 7 04:30:17 2003 From: reyk at vantronix.net (Reyk Floeter) Date: Tue, 7 Jan 2003 10:30:17 +0100 Subject: Function of hostapd? In-Reply-To: <20030107060201.GA25433@jm.kir.nu> References: <86256CA6.005AB7A6.00@nlynx.com> <20030107060201.GA25433@jm.kir.nu> Message-ID: <20030107093017.GA5846@localhost> On Mon, Jan 06, 2003 at 10:02:01PM -0800, Jouni Malinen wrote: hi! > Nowadays, hostapd includes quite a bit more than only 802.11 > authentication and association. It supports IEEE 802.1x authentication > and authorization, RADIUS accounting, parts of IAPP, etc. In addition, > most of the new higher level functionality will be added only to the > user space daemon and using hostapd (i.e., define PRISM2_HOSTAPD for the > driver) will eventually become the default option and I might even > remove the option to cleanup the driver code (i.e., remove all the > authentication and association related functionality from the kernel > driver). > i'm currently writing a linux- driver for the atheros ar5000 802.11a chipset. the latest driver only supports rx in rfmon mode, but tx and all the other stuff is work in progress (http://team.vantronix.net/ar5k/). the ar5k chipset handles most of it's 802.11- related stuff host- based and it's possible to implement an accesspoint with 802.1x support, host- based crypto (by disabling the on-card WEP engine) and so on... before re- inventing the wheel and implementing all the stuff it would be a fine solution to use use the existing hostapd (with some driver- independent interfaces) with this driver. how do you think about it? reyk -- /* .vantronix|secure systems - (research & development) * reyk floeter - friendly known free software engineer * reyk at vantronix.net - http://www.vantronix.net/~reyk/ */ From mike at bwsys.net Tue Jan 7 12:02:48 2003 From: mike at bwsys.net (Michael Shuler) Date: Tue, 7 Jan 2003 11:02:48 -0600 Subject: Klogd PS Mode buffer error Message-ID: <002101c2b66e$9e8354a0$9600a8c0@oemcomputer> Any idea what this means? ....ignore the date, my clock is wrong on my AP. Apr 28 16:59:01 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:04:02 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:09:03 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:14:04 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:19:05 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:24:06 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:29:07 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:34:08 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:39:09 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:44:10 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Mike From mike at bwsys.net Tue Jan 7 12:09:18 2003 From: mike at bwsys.net (Michael Shuler) Date: Tue, 7 Jan 2003 11:09:18 -0600 Subject: Klogd PS Mode buffer error Message-ID: <000001c2b66f$86b26ae0$9600a8c0@oemcomputer> Any idea what this means? ....ignore the date, my clock is wrong on my AP. Apr 28 16:59:01 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:04:02 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:09:03 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:14:04 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:19:05 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:24:06 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:29:07 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:34:08 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:39:09 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Apr 28 17:44:10 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer Mike From martin.polak at liwest.at Tue Jan 7 12:28:55 2003 From: martin.polak at liwest.at (Martin Polak) Date: Tue, 07 Jan 2003 18:28:55 +0100 Subject: OT: Bad Performance with Senao NL-2511CD ? Message-ID: <3E1B0E57.7010502@liwest.at> Hi! I recently bought a Senao NL-2511CD PLUST EXT2 as a replacement for my netgear MA401 (which i had fitted an external antenna connector before) to get better radio performance on my HostAP-Box. Kernel: 2.4.19, hostap-2002-10-12 the new card is now attached to the same antenna as the old card before. And instead of getting better performance from the 23dbm card, I get less SNR and weaker connections than from the old 15dbm card. I even tried to pull-up txpower, using iwconfig but nothing would help. The card inits fine, using STA 1.4.9 and even loads 1.5.6 perfectly. Does any of you have a suggestion, what I am doing wrong? Martin p.s.: I am using only 1 of the two antenna connectors and the only other difference to the previous setup is, that there now is an N-female-Nfemale coupling in between the pigtail and the antenna-cable. From mike at bwsys.net Tue Jan 7 14:10:44 2003 From: mike at bwsys.net (Michael Shuler) Date: Tue, 7 Jan 2003 13:10:44 -0600 Subject: AP Lockups Message-ID: <000d01c2b680$7d7899c0$9600a8c0@oemcomputer> Ironically a few min. after I sent my last message to the list my AP locked up again. It happens to us randomly from what I can tell. Some times it takes only 30 min. and other times it takes a day to a week. Here is what we are using/know about the system before it locks up: Hardware: Soekris net4511 or net4521 (32MB or 64MB RAM) 32MB SanDisk Compact Flash card Zcomax XI-325HI (23db) (1.4.9 or 1.5.6) Software: Kernel: 2.4.20 and 2.4.20-pre2 HostAP: almost the snapshot or 10-12-2002 Clients are a mixture of Orinoco and Zcomax cards. WEP is disabled on all. Minuets before a lockup: -> The systems have plenty of free RAM and disk space. -> Interestingly enough the watchdog timer does not seem to reset the system like it should which would make you think the hardware is locking up, which is possible or the watchdog timer software/hardware does not work properly. -> It seems to happen only during high traffic. We have another AP that hardly get used that has been running for over a month without a problem. Based off my log file I would guess that there is a possible bug when the HostAP driver lowers the rate to 55 since it never got to log it to the file...here is the tail of my log file: May 5 13:03:49 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:04:04 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer May 5 13:04:31 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:04:35 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:04:46 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:04:50 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:09:05 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer May 5 13:09:50 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:09:54 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:10:31 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:10:37 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:12:26 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:12:31 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:14:06 (none) user.debug klogd: wlan0: No more space in STA (00:02:2d:3f:fe:ea)'s PS mode buffer May 5 13:14:56 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:15:00 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:13 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:13 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:15 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:15 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:17 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:17 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:18 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:18 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:20 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:20 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:20 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:21 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:21 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:21 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:21 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:22 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:23 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:24 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:24 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:24 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:25 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:25 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 May 5 13:16:26 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate lowered to 55 May 5 13:16:27 (none) user.debug klogd: wlan0: STA 00:02:2d:3a:09:88 TX rate raised to 110 And then it locked up. Any help would be appreciated. Mike From ariel at fismat.umich.mx Tue Jan 7 09:36:48 2003 From: ariel at fismat.umich.mx (ariel at fismat.umich.mx) Date: Tue, 7 Jan 2003 14:36:48 -0000 (UTC) Subject: What about 5Ghz cards? Message-ID: <35232.148.216.6.188.1041950208.squirrel@fismat.umich.mx> Greetings The thing is that if 5Ghz cards offer 100Mbps (in theory), it would be great to build an AP using them. Is there any driver like hostap for them?? From dsen at homemail.com Tue Jan 7 15:19:59 2003 From: dsen at homemail.com (D. Sen) Date: Tue, 07 Jan 2003 15:19:59 -0500 Subject: "NETDEV WATCHDOG" problems on IBM Thinkpad T30 Message-ID: <3E1B366F.602@homemail.com> Hi, I am using the hostap drivers on my T30 which has a prism2 based pci card. I am running linux 2.4.20 and the cvs hostap snapshot from Jan 7, 2003. The driver works fine except when I attempt to copy files to/from a NFS filesystem accross the network. Then I get these errors (sometimes hanging the machine, requiring a cold-reboot). Jan 7 00:53:00 localhost kernel: NETDEV WATCHDOG: wlan0: transmit timed out Jan 7 00:53:00 localhost kernel: wlan0 Tx timed out! Resetting card Jan 7 00:53:00 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 00:53:07 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 00:53:25 localhost kernel: Already released txfid found at idx 7 Jan 7 00:53:25 localhost last message repeated 23 times Jan 7 00:53:25 localhost kernel: Already reserved txfid found at idx 7 Jan 7 00:53:25 localhost kernel: Already released txfid found at idx 7 Jan 7 00:53:25 localhost last message repeated 142 times Jan 7 01:46:32 localhost kernel: NETDEV WATCHDOG: wlan0: transmit timed out Jan 7 01:46:32 localhost kernel: wlan0 Tx timed out! Resetting card Jan 7 01:46:32 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:07:34 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:07:38 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:07:47 localhost kernel: NETDEV WATCHDOG: wlan0: transmit timed out Jan 7 02:07:47 localhost kernel: wlan0 Tx timed out! Resetting card Jan 7 02:07:47 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:17:38 localhost kernel: hostap_pci: Netdevice wlan0 unregistered Jan 7 02:17:38 localhost kernel: hostap_pci: Driver unloaded Any help will be appreciated. Here is the lspci entry for the wireless device: 02:02.0 Network controller: Harris Semiconductor: Unknown device 3873 (rev 01) Subsystem: Intel Corp.: Unknown device 2513 Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR+ FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- Hi, I am using the hostap drivers on my T30 which has a prism2 based pci card. I am running linux 2.4.20 and the cvs hostap snapshot from Jan 7, 2003. The driver works fine except when I attempt to copy files to/from a NFS filesystem accross the network. Then I get these errors (sometimes hanging the machine, requiring a cold-reboot). Jan 7 00:53:00 localhost kernel: NETDEV WATCHDOG: wlan0: transmit timed out Jan 7 00:53:00 localhost kernel: wlan0 Tx timed out! Resetting card Jan 7 00:53:00 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 00:53:07 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 00:53:25 localhost kernel: Already released txfid found at idx 7 Jan 7 00:53:25 localhost last message repeated 23 times Jan 7 00:53:25 localhost kernel: Already reserved txfid found at idx 7 Jan 7 00:53:25 localhost kernel: Already released txfid found at idx 7 Jan 7 00:53:25 localhost last message repeated 142 times Jan 7 01:46:32 localhost kernel: NETDEV WATCHDOG: wlan0: transmit timed out Jan 7 01:46:32 localhost kernel: wlan0 Tx timed out! Resetting card Jan 7 01:46:32 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:07:34 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:07:38 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:07:47 localhost kernel: NETDEV WATCHDOG: wlan0: transmit timed out Jan 7 02:07:47 localhost kernel: wlan0 Tx timed out! Resetting card Jan 7 02:07:47 localhost kernel: hostap_pci: wlan0: resetting card Jan 7 02:17:38 localhost kernel: hostap_pci: Netdevice wlan0 unregistered Jan 7 02:17:38 localhost kernel: hostap_pci: Driver unloaded Any help will be appreciated. Here is the lspci entry for the wireless device: 02:02.0 Network controller: Harris Semiconductor: Unknown device 3873 (rev 01) Subsystem: Intel Corp.: Unknown device 2513 Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR+ FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- References: <20030107035740.GB1812@jm.kir.nu> <000601c2b61a$d83bae30$350aa8c0@sergioxp> Message-ID: <20030108030233.GA1134@jm.kir.nu> On Tue, Jan 07, 2003 at 02:03:09AM -0500, Sergio Ammirata wrote: > I think the forking is rather critical since it prevents people from > deploying the software in a production environment. Specially if the are > using it in a custom distro with limited tools. I added it yesterday (-B argument for hostapd). > I like the idea of RADIUS. Perhaps we can use PAP authentication and > send the station MAC in both the username and password attributes. I would like to an encoding that is used by other 802.11 APs (I would assume that there exists one supporting this), but yes, it could be something like that. > I have attached a tcpdump capture of a PAP radius communication between > a pppd daemon and a Microsoft IAS Radius Server. It includes connect and > disconnect (use tcpdump -r filename to read). The snaplength (-s option for tcpdump) seemed to be too short and the packets were truncated so that the interesting part of the attributes was not included. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Tue Jan 7 22:19:52 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Tue, 7 Jan 2003 19:19:52 -0800 Subject: Klogd PS Mode buffer error In-Reply-To: <002101c2b66e$9e8354a0$9600a8c0@oemcomputer> References: <002101c2b66e$9e8354a0$9600a8c0@oemcomputer> Message-ID: <20030108031952.GB1134@jm.kir.nu> On Tue, Jan 07, 2003 at 11:02:48AM -0600, Michael Shuler wrote: > Any idea what this means? ....ignore the date, my clock is wrong on my > AP. > > Apr 28 16:59:01 (none) user.debug klogd: wlan0: No more space in STA > (00:02:2d:3f:fe:ea)'s PS mode buffer It means that the maximum number of buffered frames was reached for the indicated station. The default value for this is 32. That station was using power saving mode and AP started to buffer all TX frames (to be delivered when the station polls for them). The station may have left the range of the AP and there are still frames sent to it from previous connections. That could also happen if there is a burst of more than 32 frames for the station. In addition, there could be a bug in the driver that would leave traffic indication flag off in the beacons and the station would not poll for the frames due to this. -- Jouni Malinen PGP id EFC895FA From jim at terwee.com Tue Jan 7 23:57:27 2003 From: jim at terwee.com (Jim TerWee) Date: Tue, 7 Jan 2003 22:57:27 -0600 (CST) Subject: OT: Bad Performance with Senao NL-2511CD ? In-Reply-To: <3E1B0E57.7010502@liwest.at> References: <3E1B0E57.7010502@liwest.at> Message-ID: <32843.207.41.65.20.1042001847.squirrel@207.41.64.133> Make sure that the external connector is connected to the primary connector. This is the top connector on the card with the label facing you and the connectors on the right side. Mikrotik has a picture here http://www.mikrotik.com/Documentation/200mw.pdf Jim > Hi! > > I recently bought a > Senao NL-2511CD PLUST EXT2 as a replacement for my > > netgear MA401 (which i had fitted an external antenna connector before) > to get better radio performance on my HostAP-Box. > Kernel: 2.4.19, hostap-2002-10-12 > > the new card is now attached to the same antenna as the old card before. > And instead of getting better performance from the 23dbm card, I get > less SNR and weaker connections than from the old 15dbm card. I even > tried to pull-up txpower, using iwconfig but nothing would help. > The card inits fine, using STA 1.4.9 and even loads 1.5.6 perfectly. > > Does any of you have a suggestion, what I am doing wrong? > > Martin > > p.s.: > I am using only 1 of the two antenna connectors and the only other > difference to the previous setup is, that there now is an > N-female-Nfemale coupling in between the pigtail and the antenna-cable. > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap From brianc at palaver.net Wed Jan 8 00:12:27 2003 From: brianc at palaver.net (Brian Capouch) Date: Wed, 08 Jan 2003 00:12:27 -0500 Subject: OT: Bad Performance with Senao NL-2511CD ? In-Reply-To: <32843.207.41.65.20.1042001847.squirrel@207.41.64.133> References: <3E1B0E57.7010502@liwest.at> <32843.207.41.65.20.1042001847.squirrel@207.41.64.133> Message-ID: <3E1BB33B.4020304@palaver.net> Jim TerWee wrote:> Make sure that the external connector is connected to the primary > connector. This is the top connector on the card with the label facing you > and the connectors on the right side. > Mikrotik has a picture here > http://www.mikrotik.com/Documentation/200mw.pdf > Check the firmware version, too. My Mikrotik card was useless until I flashed it up. Even now, after flashing, it works in certain machines and not in others. And as I recall they don't want to help you get it flashed up, and the firmware that came with mine was much much older than another Senao card I bought the same week from Netgate. But with the Mikrotik card as it came, I could associate but had horrible performance. Now I have great performance, but I have to remember which of my machines it is it works in and which it doesn't. FWIW. B. From earl at stanfordalumni.org Wed Jan 8 05:13:25 2003 From: earl at stanfordalumni.org (earl) Date: Wed, 08 Jan 2003 02:13:25 -0800 Subject: Tx timed out! Resetting card (with OpenAP) Message-ID: <3E1BF9C4.4143A1F9@stanfordalumni.org> Hi, I'm doing some development with the OpenAP (Eumitcom WL-11000) hardware: http://tools.collegeterrace.net/openap-ct This is basically a AMD Elan SC400 (486 class system-on-a-chip) with a Prism2-based card plugged into the PCMCIA. Currently i'm using CVS hostap (as of Jan 3), pcmcia 3.2.3, and kernel 2.4.20. I have two of these units set up with bridging between their wireless and ethernet interfaces, with a WDS link between the two units. (wds_type=2) I am testing the WDS link by transferring data from a PC connected to the ethernet interface of one unit to a PC connected to the interface of the other. Generally, it works for a while but eventually one unit will die with the message: NETDEV WATCHDOG: wlan0: transmit timed out wlan0 Tx timed out! Resetting card wlan0: CMD=010b EVSTAT=8099 OFFSET0=0000 OFFSET1=0000 SWSUPPORT0=8a32 which repeats forever. It's easy to reproduce this by ping-flooding, which will trigger this failure within a minute or two at the most. Transferring large files also causes this to kick in reliably after a few mintues. The failure behavior is slightly different if i use use standard WDS by first using prism2_srec to do a nonvolatile update of the firmware then use wds_type=4. In this case when it fails i get the message: NETDEV WATCHDOG: wlan0: transmit timed out wlan0 Tx timed out! Resetting card wlan0: CMD=010b EVSTAT=0019 OFFSET0=0000 OFFSET1=0000 SWSUPPORT0=8a32 which is only printed out once. After that, the link is completely dead so apparently the card reset failed to make things work again. I did verify that, before the failure occurs and while data is still flowing normally, the interrupt count in /proc/interrupts is incrementing for hostap_cs. Does this indicate that this isn't something that would be fixed with an irq_mode setting? Just for fun i tried changing the value of TX_TIMEOUT in hostap.c from 2 * HZ to 200 * HZ to see whether the timeout was in fact a brief glitch that could be safely ignored, i.e. an empirical timing parameter that is being set too strictly. In this case no timeout was reported but the link would die anyway, which i assume means things were broken already regardless of whether the card reset is attempted. That is, the fact that prism2_tx_timeout() is getting called is a symptom of a bug that's already occurred. The fact that the reset fails to put things right may or may not be related to the bug that already occurred. Any ideas what's happening? Can i put in more printk's to give some better tracing info? Here are some possibly relevant startup messages (for the first case): Linux Kernel Card Services 3.1.22 options: none Intel PCIC probe: Intel i82365sl A step ISA-to-PCMCIA at port 0x3e0 ofs 0x00, 1 socket host opts [0]: none ISA irqs (default) = 3,4,5,7,9,10,11,12,15 polling interval = 1000 ms Starting wireless: cardmgr[23]: watcs: IO port probe 0x0c00-0x0cff:ching 1 socke clean. cs: IO port probe 0x0800-0x08ff: clean. cs: IO port probe 0x0100-0x04ff: clean. Jan 1 00:00:06 cs: memory probe 0x0d0000-0x0dffff:cardmgr[24]: starting, version i clean. s 3.2.3 Jan 1 00:00:06 cardmgr[24]: socket 0: Longshine LCR-8531 11Mbps WLAN PCMCIA CARD Jan 1 00:00:06 cardmgr[24]: executing: 'insmod /lib/modules/preferred/pcmcia/ho stap_crypt.o' Jan 1 00:00:07 cardmgr[24]: + Using /lib/modules/preferred/pcmcia/hostap_crypt.o Jan 1 00:00:07 cardmgr[24]: executing: 'insmod /lib/modules/preferred/pcmcia/hostap.o' Jan 1 00:00:08 cardmgr[24]: + Using /lib/modules/preferred/pcmcia/hostap.o Jan 1 00:00:08 cardmgr[24]: executing: 'insmod /lib/modules/preferred/pcmcia/hostap_cs.o ignore_cis_vcc=1' hostap_cs: 0.0.0 CVS (Jouni Malinen ) Jan 1 00:00:08 cardmgr[24]: + Using /lib/modules/preferred/pcmchostap_cs: index , irq 3Vcc 5.0ia/hostap_cs.o , io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 wlan0: NIC: id=0x8002 v1.0.0 wlan0: PRI: id=0x15 v0.3.0 wlan0: STA: id=0x1f v0.7.5 Jan 1 00:00:09 cardmgr[24]: executing: './network start wlan0' Jan 1 00:00:10 cardmgr[24]: + wlan0 Link encap:Ethernet HWaddr 00:90:D1:00:C4:88 Jan 1 00:00:10 cardmgr[24]: + BROADCAST MULTICAST MTU:1500 Metric:1 Jan 1 00:00:10 cardmgr[24]: + RX packets:0 errors:0 dropped:0 overruns:0 frame:0 Jan 1 00:00:10 cardmgr[24]: + TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 Jan 1 00:00:10 cardmgr[24]: + collisions:0 txqueuelen:100 Jan 1 00:00:10 cardmgr[24]: + RX bytes:0 (0.0 iB) TX bytes:0 (0.0 iB) Jan 1 00:00:10 cardmgr[24]: + Interrupt:3 Base address:0x100 thanks earl From antonio.vilei at unile.it Wed Jan 8 05:58:24 2003 From: antonio.vilei at unile.it (Antonio Vilei) Date: Wed, 8 Jan 2003 11:58:24 +0100 Subject: crash with kernel 2.4.17 on ST40 processor Message-ID: <002501c2b704$ddf75000$151e0a0a@stmantoniov> Hi, I've compiled and installed the driver hostap_plx on a system platform based on an ST40 processor with Linux kernel 2.4.17. When I do "modprobe hostap_plx", I get the following messages from printk, then the system crashes: ---------------------------------------------------- hostap_plx: (c) Jouni Malinen PCI INTERRUPT! PCI INT -> 0x4 PCI AIR -> 0x3480 PCI CIR -> 0x0 ---------------------------------------------------- I have an SMC2602W card, that works OK with my PC with kernel 2.4.18. On my ST40 platform, I load the kernel by means of a loader connected to the ST40 platform and to a PC. The filesystem for my ST40 platform is exported via NFS by the same PC. To compile I use a cross-compiler that runs on the PC and puts the bynaries for ST40 on the exported filesystem. I have modified the HostAP Makefile on the PC to include the correct kernel path, cross-compiler and modules path (within the exported filesystem) for my target ST40 system. Before trying "modprobe hostap_plx", I've done "depmod -a" on the ST40. May the problem be related to genksyms? Indeed I am obliged to use the PC genksyms and not the genksyms for my ST40 linux distribution (that doesn't include gcc because it relies on the cross-compiler on the PC). Any idea or suggestions? Thanks in advance, Antonio From alessandro.bondoli at bondoli-campese.com Wed Jan 8 06:11:30 2003 From: alessandro.bondoli at bondoli-campese.com (Alessandro Bondoli) Date: Wed, 8 Jan 2003 12:11:30 +0100 Subject: wds encryption Message-ID: <003401c2b706$b262ee70$af9fa8c0@admi001> Did anybody encrypt (with cipe) connection via wds interface ? From earl at stanfordalumni.org Wed Jan 8 06:14:26 2003 From: earl at stanfordalumni.org (earl) Date: Wed, 08 Jan 2003 03:14:26 -0800 Subject: Tx timed out! Resetting card (with OpenAP) References: <3E1BF9C4.4143A1F9@stanfordalumni.org> <007901c2b702$e52e9c90$53153b0a@nednet.co.za> Message-ID: <3E1C0812.36DF29D3@stanfordalumni.org> Robert Botha wrote: > It seems from the Messages you pasted that you're using the builtin kernel > pcmcia, and not the external package. > Have you tried using the external package seperately and disabling the > kernel drivers? > I've had endless hassles with the kernel pcmcia and most of them went away > when I switched to the pcmcia-cs package. Thanks for the suggestion, i can give that a shot. It's possible that the reason this was never done by the original OpenAP developers, with whose work i'm starting, is that the pcmcia driver modules are too large (the boot media for this platform is a mere 1MB of flash, and there's not much free space left on it) in which case i'd be stuck with the kernel pcmcia. What kind of problems did you have with kernel pcmcia and hostap? Any way to get a better idea whether this is a pcmcia issue or a hostap issue? thanks earl From mpolak at gup.jku.at Wed Jan 8 02:41:07 2003 From: mpolak at gup.jku.at (Martin Polak) Date: Wed, 08 Jan 2003 08:41:07 +0100 Subject: OT: Bad Performance with Senao NL-2511CD ? References: <3E1B0E57.7010502@liwest.at> <1041986029.14785.8.camel@sporran> Message-ID: <3E1BD613.30006@gup.jku.at> alexd wrote: > On Tue, 2003-01-07 at 17:28, Martin Polak wrote: > > >>the new card is now attached to the same antenna as the old card before. >>And instead of getting better performance from the 23dbm card, I get >>less SNR and weaker connections than from the old 15dbm card. I even >>tried to pull-up txpower, using iwconfig but nothing would help. >>The card inits fine, using STA 1.4.9 and even loads 1.5.6 perfectly. > > > Do you *know* for a fact that the Senao card has more output power? > Don't just rely on the manufacturers data. Have you tested both cards > with equipment in a lab? [I'm guessing not]. Perhaps you should do a > little research into the Senao card and see what other WLAN enthusiasts > are saying about it. Right, lacking this lab equipment, i couldn't do this tests.But my sense of logic tells me, that I card stated to have more than 6 times the output power and according to a chart found in the web (dont remeber the link)a better rf sensitivity, it should at least perform equal to the old one, shouldn't it. > Try swapping the card > with a different example of the same model. I think i always did this. > Well it's probably not you ;-) Have you checked the electrical polarity > of the RF connector? Maybe it is idiosyncratic? > > Every coupling causes losses. And remember, if you change more than one > variable at once, It Ain't Scientific ;-) Yes I know, but remeber the thing about 'lacking the right equipment' :-) ? Thanks for your thoughts anyway! Martin From mpolak at gup.jku.at Wed Jan 8 02:42:27 2003 From: mpolak at gup.jku.at (Martin Polak) Date: Wed, 08 Jan 2003 08:42:27 +0100 Subject: OT: Bad Performance with Senao NL-2511CD ? References: <3E1B0E57.7010502@liwest.at> <1041987645.2128.3.camel@localhost.localdomain> Message-ID: <3E1BD663.3040406@gup.jku.at> wlan wrote: >>p.s.: >>I am using only 1 of the two antenna connectors and ... > > > Maybe that could be the reason, is there any option to enable which > connector is transmiting? > > maybe you are transmiting using both connnectors and that is the reason > that make you loose power. I ve already tried the antsel_rx and antsel_tx option of the hostap-driver, without any success. Martin From flash at flash.za.org Wed Jan 8 05:44:13 2003 From: flash at flash.za.org (Robert Botha) Date: Wed, 8 Jan 2003 12:44:13 +0200 Subject: Tx timed out! Resetting card (with OpenAP) References: <3E1BF9C4.4143A1F9@stanfordalumni.org> Message-ID: <007901c2b702$e52e9c90$53153b0a@nednet.co.za> Hi, It seems from the Messages you pasted that you're using the builtin kernel pcmcia, and not the external package. Have you tried using the external package seperately and disabling the kernel drivers? I've had endless hassles with the kernel pcmcia and most of them went away when I switched to the pcmcia-cs package. -Robert. ----- Original Message ----- From: "earl" To: Sent: Wednesday, January 08, 2003 12:13 PM Subject: Tx timed out! Resetting card (with OpenAP) > Hi, > > I'm doing some development with the OpenAP (Eumitcom WL-11000) hardware: > > http://tools.collegeterrace.net/openap-ct > This is basically a AMD Elan SC400 (486 class system-on-a-chip) with a > Prism2-based card plugged into the PCMCIA. > > Currently i'm using CVS hostap (as of Jan 3), pcmcia 3.2.3, and kernel > 2.4.20. > > I have two of these units set up with bridging between their wireless > and ethernet interfaces, with a WDS link between the two units. > (wds_type=2) I am testing the WDS link by transferring data from a PC > connected to the ethernet interface of one unit to a PC connected to the > interface of the other. Generally, it works for a while but eventually > one unit will die with the message: > NETDEV WATCHDOG: wlan0: transmit timed out > wlan0 Tx timed out! Resetting card > wlan0: CMD=010b EVSTAT=8099 OFFSET0=0000 OFFSET1=0000 SWSUPPORT0=8a32 > which repeats forever. It's easy to reproduce this by ping-flooding, > which will trigger this failure within a minute or two at the most. > Transferring large files also causes this to kick in reliably after a > few mintues. > > The failure behavior is slightly different if i use use standard WDS by > first using prism2_srec to do a nonvolatile update of the firmware then > use wds_type=4. In this case when it fails i get the message: > NETDEV WATCHDOG: wlan0: transmit timed out > wlan0 Tx timed out! Resetting card > wlan0: CMD=010b EVSTAT=0019 OFFSET0=0000 OFFSET1=0000 SWSUPPORT0=8a32 > which is only printed out once. After that, the link is completely dead > so apparently the card reset failed to make things work again. > > I did verify that, before the failure occurs and while data is still > flowing normally, the interrupt count in /proc/interrupts is > incrementing for hostap_cs. Does this indicate that this isn't > something that would be fixed with an irq_mode setting? > > Just for fun i tried changing the value of TX_TIMEOUT in hostap.c from 2 > * HZ to 200 * HZ to see whether the timeout was in fact a brief glitch > that could be safely ignored, i.e. an empirical timing parameter that is > being set too strictly. In this case no timeout was reported but the > link would die anyway, which i assume means things were broken already > regardless of whether the card reset is attempted. That is, the fact > that prism2_tx_timeout() is getting called is a symptom of a bug that's > already occurred. The fact that the reset fails to put things right may > or may not be related to the bug that already occurred. > > Any ideas what's happening? Can i put in more printk's to give some > better tracing info? > > Here are some possibly relevant startup messages (for the first case): > > Linux Kernel Card Services 3.1.22 > options: none > Intel PCIC probe: > Intel i82365sl A step ISA-to-PCMCIA at port 0x3e0 ofs 0x00, 1 socket > host opts [0]: none > ISA irqs (default) = 3,4,5,7,9,10,11,12,15 polling interval = 1000 > ms > Starting wireless: cardmgr[23]: watcs: IO port probe 0x0c00-0x0cff:ching > 1 socke > clean. > cs: IO port probe 0x0800-0x08ff: clean. > cs: IO port probe 0x0100-0x04ff: clean. > Jan 1 00:00:06 cs: memory probe 0x0d0000-0x0dffff:cardmgr[24]: > starting, version i clean. > s 3.2.3 > Jan 1 00:00:06 cardmgr[24]: socket 0: Longshine LCR-8531 11Mbps WLAN > PCMCIA CARD > Jan 1 00:00:06 cardmgr[24]: executing: 'insmod > /lib/modules/preferred/pcmcia/ho > stap_crypt.o' > Jan 1 00:00:07 cardmgr[24]: + Using > /lib/modules/preferred/pcmcia/hostap_crypt.o > Jan 1 00:00:07 cardmgr[24]: executing: 'insmod > /lib/modules/preferred/pcmcia/hostap.o' > Jan 1 00:00:08 cardmgr[24]: + Using > /lib/modules/preferred/pcmcia/hostap.o > Jan 1 00:00:08 cardmgr[24]: executing: 'insmod > /lib/modules/preferred/pcmcia/hostap_cs.o ignore_cis_vcc=1' > hostap_cs: 0.0.0 CVS (Jouni Malinen ) > Jan 1 00:00:08 cardmgr[24]: + Using > /lib/modules/preferred/pcmchostap_cs: index > , irq 3Vcc 5.0ia/hostap_cs.o > , io 0x0100-0x013f > hostap_cs: Registered netdevice wlan0 > wlan0: NIC: id=0x8002 v1.0.0 > wlan0: PRI: id=0x15 v0.3.0 > wlan0: STA: id=0x1f v0.7.5 > Jan 1 00:00:09 cardmgr[24]: executing: './network start wlan0' > Jan 1 00:00:10 cardmgr[24]: + wlan0 Link encap:Ethernet HWaddr > 00:90:D1:00:C4:88 > Jan 1 00:00:10 cardmgr[24]: + BROADCAST MULTICAST MTU:1500 > Metric:1 > Jan 1 00:00:10 cardmgr[24]: + RX packets:0 errors:0 dropped:0 > overruns:0 frame:0 > Jan 1 00:00:10 cardmgr[24]: + TX packets:0 errors:0 dropped:0 > overruns:0 carrier:0 > Jan 1 00:00:10 cardmgr[24]: + collisions:0 txqueuelen:100 > Jan 1 00:00:10 cardmgr[24]: + RX bytes:0 (0.0 iB) TX bytes:0 > (0.0 iB) > Jan 1 00:00:10 cardmgr[24]: + Interrupt:3 Base address:0x100 > > thanks > earl > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > > > From flash at flash.za.org Wed Jan 8 06:59:15 2003 From: flash at flash.za.org (Robert Botha) Date: Wed, 8 Jan 2003 13:59:15 +0200 Subject: Tx timed out! Resetting card (with OpenAP) References: <3E1BF9C4.4143A1F9@stanfordalumni.org> <007901c2b702$e52e9c90$53153b0a@nednet.co.za> <3E1C0812.36DF29D3@stanfordalumni.org> Message-ID: <008401c2b70d$61255690$53153b0a@nednet.co.za> > > It seems from the Messages you pasted that you're using the builtin kernel > > pcmcia, and not the external package. > > Have you tried using the external package seperately and disabling the > > kernel drivers? > > I've had endless hassles with the kernel pcmcia and most of them went away > > when I switched to the pcmcia-cs package. > > Thanks for the suggestion, i can give that a shot. It's possible that the > reason this was never done by the original OpenAP developers, with whose work > i'm starting, is that the pcmcia driver modules are too large (the boot media > for this platform is a mere 1MB of flash, and there's not much free space left > on it) in which case i'd be stuck with the kernel pcmcia. In theory the size difference should be minimal. After all, the kernel pcmcia comes from the same tree..it's just a much older version. In the kernel you're using it's version 3.1.22 > What kind of problems did you have with kernel pcmcia and hostap? Any way to > get a better idea whether this is a pcmcia issue or a hostap issue? Well, my issues were before I even knew about hostap :) I use a PCMCIA-PCI bridge. I'd get alot of Resource Unavailable or IRQ 0 type errors. And if it worked (normally by restarting the pcmcia subsystem) I'd get low throughput and Tx Timeouts, this was using an Orinoco Silver card with orinoco_cs. It might be different on a Prism2 card with the hostap driver. I never tried kernel pcmcia again since I changed to pcmcia-cs. Most of the IRQ issues went away as pcmcia-cs matured. I think the best version I've used so far is 3.1.34 I find that 3.2.x doesn't work well for me, it keeps crashing the machine after a day or two. Regards, -Robert. From doug at aircomwireless.net Wed Jan 8 08:01:29 2003 From: doug at aircomwireless.net (Doug Yeager) Date: Wed, 8 Jan 2003 08:01:29 -0500 Subject: OT: Bad Performance with Senao NL-2511CD ? In-Reply-To: <3E1BD613.30006@gup.jku.at> Message-ID: <000001c2b716$17639180$6e00a8c0@DOUG1> I have this card and I am using it w/ an 32 slot omni-directional antenna. I was using the wmp11. I can notice an improvement over that card definitely. I did notice that only one of the jacks worked at all for tx. Only the primary jack transmits. I thought something was messed up bad and then I just switched the jack it was plugged into and I did an iwconfig and watched the driver adjust the tx power. That was really cool. It climbed up to 20db and then stopped. I then took my laptop w/ a crappy pcmcia antenna and drove in my car up to about 1/4 miles away and was able to surf the web...that was pretty cool. Line of site is important though. I'll try the yagi antenna next week when I get it. Fun geeky stuff! :) -doug -----Original Message----- From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] On Behalf Of Martin Polak Sent: Wednesday, January 08, 2003 2:41 AM To: alexd Cc: hostap at shmoo.com Subject: Re: OT: Bad Performance with Senao NL-2511CD ? alexd wrote: > On Tue, 2003-01-07 at 17:28, Martin Polak wrote: > > >>the new card is now attached to the same antenna as the old card before. >>And instead of getting better performance from the 23dbm card, I get >>less SNR and weaker connections than from the old 15dbm card. I even >>tried to pull-up txpower, using iwconfig but nothing would help. >>The card inits fine, using STA 1.4.9 and even loads 1.5.6 perfectly. > > > Do you *know* for a fact that the Senao card has more output power? > Don't just rely on the manufacturers data. Have you tested both cards > with equipment in a lab? [I'm guessing not]. Perhaps you should do a > little research into the Senao card and see what other WLAN enthusiasts > are saying about it. Right, lacking this lab equipment, i couldn't do this tests.But my sense of logic tells me, that I card stated to have more than 6 times the output power and according to a chart found in the web (dont remeber the link)a better rf sensitivity, it should at least perform equal to the old one, shouldn't it. > Try swapping the card > with a different example of the same model. I think i always did this. > Well it's probably not you ;-) Have you checked the electrical polarity > of the RF connector? Maybe it is idiosyncratic? > > Every coupling causes losses. And remember, if you change more than one > variable at once, It Ain't Scientific ;-) Yes I know, but remeber the thing about 'lacking the right equipment' :-) ? Thanks for your thoughts anyway! Martin _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap From dave at cc0.net Wed Jan 8 14:18:22 2003 From: dave at cc0.net (Dave) Date: Wed, 8 Jan 2003 11:18:22 -0800 Subject: hostapd startup Message-ID: <000901c2b74a$b6548310$0507ee3f@ws1> Hi all, When starting hostapd and providing the wlan0 iface as an argument, I get the following: [root at w hostapd]# ./hostapd wlan0 ioctl(SIOCGIFINDEX): No such device ioctl[SIOCGIFFLAGS]: No such device 2.4.20 hostap from CVS snapshot hostap built in kernel (patch/copy drivers to 2.4.20 tree) [root at w hostapd]# iwconfig wlan0 wlan0 IEEE 802.11b ESSID:"CICAP" Nickname:"w.calinet.com" Mode:Master Frequency:2.422GHz Access Point: 00:02:6F:03:60:50 Bit Rate:11Mb/s Tx-Power:-8 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:0 Signal level:0 Noise level:0 Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 [root at w hostapd]# ifconfig wlan0 wlan0 Link encap:Ethernet HWaddr 00:02:6F:03:60:50 inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:8 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:0 (0.0 b) TX bytes:576 (576.0 b) Interrupt:5 Base address:0x100 Any ideas? -Dave From hueiping at yahoo.com Wed Jan 8 14:23:30 2003 From: hueiping at yahoo.com (Huei-Ping Chen) Date: Wed, 8 Jan 2003 11:23:30 -0800 (PST) Subject: hostapd startup Message-ID: <20030108192330.77269.qmail@web40711.mail.yahoo.com> did you compile hostap with "-DPRISM2_HOSTAPD" ? > [root at w hostapd]# ./hostapd wlan0 > ioctl(SIOCGIFINDEX): No such device > ioctl[SIOCGIFFLAGS]: No such device __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com From cijoml at volny.cz Wed Jan 8 14:34:04 2003 From: cijoml at volny.cz (Michal Semler) Date: Wed, 8 Jan 2003 20:34:04 +0100 Subject: connect, disconnect in dmesg Message-ID: hello, i have link via very long distanc. i had in dmesg so many connect, disconnect and other messages, which could not be in dmesg - nor wireless driver in kerne writes these into dmesg if it's not in debug mode compiled. So here is little patch removing them. Please apply or make hostap to write these only in debug mode... Thanks Michal -------------- next part -------------- A non-text attachment was scrubbed... Name: hostap.patch Type: text/x-diff Size: 3810 bytes Desc: hostap.patch Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030108/1b3e4815/attachment.patch From softlib at turbolinux.com.cn Wed Jan 8 20:43:38 2003 From: softlib at turbolinux.com.cn (softlib) Date: Thu, 9 Jan 2003 09:43:38 +0800 Subject: Where can I download the newest firmware for Prism2x In-Reply-To: <20030108193700.29329.7306.Mailman@sisyphus.iocaine.com> References: <20030108193700.29329.7306.Mailman@sisyphus.iocaine.com> Message-ID: <133796795.20030109094338@turbolinux.com.cn> Hello, I have a linksys WMP11 PCI card and I have already upgraded firmware to 1.4.9, In order to use all functions in hostap driver, I have to upgrade it to at least 1.5.x. Do you know where can I download the newer prism2x firmware for my card? Thanks! Merlin From jkmaline at cc.hut.fi Wed Jan 8 22:00:22 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 8 Jan 2003 19:00:22 -0800 Subject: connect, disconnect in dmesg In-Reply-To: References: Message-ID: <20030109030022.GB1199@jm.kir.nu> On Wed, Jan 08, 2003 at 08:34:04PM +0100, Michal Semler wrote: > i have link via very long distanc. i had in dmesg so many connect, disconnect > and other messages, which could not be in dmesg - nor wireless driver in > kerne writes these into dmesg if it's not in debug mode compiled. So here is > little patch removing them. > > Please apply or make hostap to write these only in debug mode... They are already logged only in debug mode, but the debug mode is enabled by default.. If you want to get rid of this, you can change DEBUG_MASK definition (remove DEBUG_EXTRA) in hostap_wlan.h. If you want to get rid of all PDEBUG() logging, compile driver with PRISM2_NO_DEBUG defined. Your patch removed lot more than only the LinkStatus debug information, e.g., couple of error messages. I'm not going to remove those from the driver. -- Jouni Malinen PGP id EFC895FA From imp at airlinktek.com Thu Jan 9 02:42:30 2003 From: imp at airlinktek.com (=?utf-7?B?K3ZCWFhaY1gwLQ==?=) Date: Thu, 9 Jan 2003 16:42:30 +0900 Subject: The IAPP Message-ID: Dear all.. Could I ask when the IAPP protocol fully implemented ? As I know iapp.c only implement about the associate request action.. isn't it ? Thanks for your attention.. From pedro.estrela at inesc.pt Thu Jan 9 03:10:02 2003 From: pedro.estrela at inesc.pt (Pedro Estrela) Date: Thu, 9 Jan 2003 08:10:02 -0000 Subject: management frames are lost during high load situations (REPOST) References: <86256CA6.005AB7A6.00@nlynx.com> <20030107060201.GA25433@jm.kir.nu> <000401c2b614$16477c60$0200a8c0@ws7> Message-ID: <008901c2b7b6$84609c10$94e016d5@pedropc> Hi! I've got a problem related to management frames in high load situations. This was already sent by me some months ago (http://lists.shmoo.com/pipermail/hostap/2002-October/000053.html) and i'm hoping that now someone has any ideia of how i could give priority to the transmission of management frames: "Simply put, when the card is put in stress with a major downstream UDP flow (greater than a flood ping), the AP is unable to process further associations/reassociations. In this scenario, a station moving from one unloaded AP to a loaded AP will keep on sending associations until the load stops." Thanks, Pedro Estrela From BichotG at tce.com Thu Jan 9 08:45:47 2003 From: BichotG at tce.com (Bichot Guillaume (Princeton)) Date: Thu, 9 Jan 2003 08:45:47 -0500 Subject: management frames are lost during high load situations (REPOS T) Message-ID: <0BCAE6FBFB16D511922A0002A507DACE030C8E15@INDYEXCH24> I've got the same problem. Jouni answer was this. > With the laptop: > The average association time (i.e. the client detects and associates > with the AP)is 10 seconds between 256 Kbit/s and 1 Mbit/s. With a > 2Mbit/s throughput it becomes impossible to associate with the AP. Lot of date traffic in the network will result in association problems since most stations seem to use quite strict limits on how long time receiving of the auth/assoc reply can take. If you upgrade your AP to use latest CVS version, you should be able to see whether the station is acknowledging auth/assoc replies (if not, the delay was probably too long). At the moment, the driver does not prioritize outgoing frames at all, but I might need to do this at some point to help association under heavy load. It should help, if the AP would simply by-pass any outgoing queue of data frames when sending auth/assoc replies. There can still be some frames buffered in the wlan card, but at least the host-side queue would not be adding latency. In addition, number of buffered frames in the wlan card could be reduced, if needed. Guillaume Bichot > -----Original Message----- > From: Pedro Estrela [mailto:pedro.estrela at inesc.pt] > Sent: Thursday, January 09, 2003 3:10 AM > To: HostAP at shmoo.com > Subject: management frames are lost during high load > situations (REPOST) > > > Hi! I've got a problem related to management frames in high > load situations. > > This was already sent by me some months ago > (http://lists.shmoo.com/pipermail/hostap/2002-October/000053.h tml) and i'm hoping that now someone has any ideia of how i could give priority to the transmission of management frames: "Simply put, when the card is put in stress with a major downstream UDP flow (greater than a flood ping), the AP is unable to process further associations/reassociations. In this scenario, a station moving from one unloaded AP to a loaded AP will keep on sending associations until the load stops." Thanks, Pedro Estrela _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030109/571d52be/attachment.htm From mpolak at gup.jku.at Thu Jan 9 02:09:35 2003 From: mpolak at gup.jku.at (Martin Polak) Date: Thu, 09 Jan 2003 08:09:35 +0100 Subject: OT: Bad Performance with Senao NL-2511CD ? RESOLVED References: <3E1B0E57.7010502@liwest.at> <1041987645.2128.3.camel@localhost.localdomain> <3E1BD663.3040406@gup.jku.at> <1042053553.3e1c79b1b731e@webmail.cbnco.com> Message-ID: <3E1D202F.5040506@gup.jku.at> Thanks for all the advice! It seems to work out fine now, here is what I did: -I first changed the antenna pin, but it had already been the right one. (like some folks suggested on this list) changing to the other one, brought txpower to 20 dbm, but made the signal worse -I swapped the two cards I own (again), but this time it worked out, so the clue seems to be, that my prior attempt in ram uploading STA 1.5.6 instead of of using 1.4.9 supplied with the card, somehow messed it up. After "hard resetting" it by removing it now works fine Martin From a2587c at motorola.com Thu Jan 9 03:11:51 2003 From: a2587c at motorola.com (Hu Tao-a2587c) Date: Thu, 9 Jan 2003 16:11:51 +0800 Subject: Crash problem of HostAP driver-2002-10-12 on MPC857 platform Message-ID: <01139FC052A0D411900B00508B9535FC09DE32D1@ZCH07EXM04.corp.mot.com> Kernel Version: 2.4.2 PCMCIA Version: 3.1.31 HostAP Version: 2002-10-12 WLAN NIC: Prism II, manId: 0x0156, 0x0002 HW Platform: IAD857 (Motorola PowerPC857) All of the modules(pcmcia_core, m8xx_pcmcia, ds, hostap_crypt, hostap, hostap_cs)had been loaded successfully, and the cardmgr had been started. When "ifconfig wlan0 up" is used to bring up the NIC, crash occured. Following is the oops: Oops: kernel access of bad area, sig: 11 NIP: C010B83C XER: E000017F LR: C001BBAC SP: C1EF3D70 REGS: c1ef3cc0 TRAP: 0300 MSR: 00009032 EE: 1 PR: 0 FP: 0 ME: 1 IR/DR: 11 DAR: 64756C69, DSISR: 0000013F TASK = c1ef2000[79] 'ifconfig' Last syscall: 54 last math 00000000 last altivec 00000000 GPR00: 00000000 C1EF3D70 C1EF2000 C0159DE8 00000001 C1F73000 C02523F8 00000000 GPR08: 00000000 C0141BB0 C010B83C C0141BB0 91003139 1005A5C0 01FAC000 FF780000 GPR16: 00000000 00000000 02000000 00000000 00009032 FFFF8914 C1EF3DD8 00008914 GPR24: 00000000 00000000 C1F73000 00000002 00000000 00000001 C1F73000 64756C65 Call backtrace: 00000001 C001BBAC C00C0510 C00C1BDC C00EFDBC C00F216C C00BA674 C00454CC C000245C 100028B4 1001AC4C 1002E128 1002DD78 0FF0A3A4 00000000 When "kill" cardmgr, a another carsh occured: unbind_request(0, 'hostap_cs') prism2_detach hostap_cs: detach postponed, 'wlan0' still locked prism2_release cs: CardServices(ReleaseConfiguration, 0xc0253860, 0x00000000) m8xx_pcmcia: Service(0, 4, 0xc02f780c) m8xx_set_socket: Enter m8xx_pcmcia: SetSocket(0, flags 0x200, Vcc 50, Vpp 0, io_irq 0, csc_mask 0x80) voltage_set: Enter voltage_set: Exit m8xx_set_socket: Exit m8xx_pcmcia: Service(0, 6, 0xc1f0faf8) m8xx_pcmcia: SetIOMap(0, 0, 0x00, 0 ns, 0x0000-0x0001) m8xx_get_graycode: Enter m8xx_get_graycode: Exit cs: CardServices(ReleaseIO, 0xc0253860, 0xc0253b90) cs: CardServices(ReleaseIRQ, 0xc0253860, 0xc0253ba4) release - done cs: CardServices(DeregisterClient, 0xc0253860, 0x00000001) cs: deregister_client(c0253860) Oops: kernel access of bad area, sig: 11 NIP: C010B860 XER: E000017F LR: C001BBAC SP: C1F0FA90 REGS: c1f0f9e0 TRAP: 0300 MSR: 00009032 EE: 1 PR: 0 FP: 0 ME: 1 IR/DR: 11 DAR: 64756C69, DSISR: 00000069 TASK = c1f0e000[29] 'cardmgr' Last syscall: 54 last math 00000000 last altivec 00000000 GPR00: 00000014 C1F0FA90 C1F0E000 C0159DE8 C1F73000 C1F73000 C02523F8 00000000 GPR08: 00000000 64756C65 C010B858 C0141BB0 91BB5135 10021E54 01FAC000 FF780000 GPR16: 00004000 00000014 0200000C 20040000 00009032 01F0FE80 00000000 C02C6260 GPR24: C0150000 C014F170 00000000 C1F0FB18 C02C61A4 00000006 C1F73000 64756C65 Call backtrace: 00000006 C001BBAC C00C2948 C0099C34 C3436E90 C34374A4 C3418B98 C3419898 C00454CC C000245C 00000000 10003664 10003E64 100045E8 0FF0A3A4 00000000 Is there any one can help me? Thanks in advance! From koma at lysator.liu.se Thu Jan 9 09:56:18 2003 From: koma at lysator.liu.se (Magnus Ekhall) Date: Thu, 09 Jan 2003 15:56:18 +0100 Subject: Host based encryption needed with 1X? Message-ID: <5.1.1.6.0.20030109155113.02da2318@imap.lysator.liu.se> Is host based encryption really needed when running 802.1X in hostap? If so, why? Is the implemented 802.1X code compliant to the upcoming 802.11i standard or is it some kind of ad-hoc solution? /Magnus From Jacques.Caron at IPsector.com Thu Jan 9 10:08:05 2003 From: Jacques.Caron at IPsector.com (Jacques Caron) Date: Thu, 09 Jan 2003 16:08:05 +0100 Subject: Host based encryption needed with 1X? In-Reply-To: <5.1.1.6.0.20030109155113.02da2318@imap.lysator.liu.se> Message-ID: <5.2.0.9.0.20030109160409.02158828@pop.mail.yahoo.com> At 15:56 09/01/2003, Magnus Ekhall wrote: >Is host based encryption really needed when running 802.1X in hostap? >If so, why? If you are using encryption (which you should, with 802.1X), then host based encryption is needed because there's one key per station, and there is no known interface in the hostap firmware code to tell the card to use different keys for each station. >Is the implemented 802.1X code compliant to the upcoming 802.11i standard >or is it some kind of ad-hoc solution? 802.11i contains a lot more than just 802.1X, and even 802.1X is changing (with 802.1aa), so I think the hostap 802.1X code is more "de facto standard" (mainly Cisco and XP) compliant than anything (the exact way to use 802.1X with 802.11 was/is not defined yet). Jacques. -- Jacques Caron, IP Sector Technologies Join the discussion on public WLAN open global roaming: http://lists.ipsector.com/listinfo/openroaming From venky at tataelxsi.co.in Thu Jan 9 11:19:12 2003 From: venky at tataelxsi.co.in (Venkatesh N) Date: Thu, 09 Jan 2003 21:49:12 +0530 Subject: Hostap-2002-10-12, Installation problem on Linux Message-ID: <3E1DA100.1FA7B6E6@tataelxsi.co.in> > Hello All, > > I have few problems while compiling the "hostap-2002-10-12" code > My configuration are: > --------------------- > Driver used --> hostap-2002-10-12. > Redhat Linux Kernel version --> 2.4.20. > PCMCIA adapter with Prism version2.5 chipset. > > The Driver compiled successfully but > I get the following errors when executing "make install_pccard" > ------- > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_ap.o > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o > > depmod: *** Unresolved symbols in > /lib/modules/2.4.20/net/hostap_crypt.o > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_cs.o > A help here is much appreciated. Thanks for your valuable time. regards, Venkatesh N Tata Elxsi Ltd Bangalore India From alessandro.bondoli at bondoli-campese.com Thu Jan 9 12:22:09 2003 From: alessandro.bondoli at bondoli-campese.com (Alessandro Bondoli) Date: Thu, 9 Jan 2003 18:22:09 +0100 Subject: wds encryption II Message-ID: <001901c2b803$a44df3e0$af9fa8c0@admi001> Recently I had need to build a point-to-point encrypted link using another couple of aps (LinuxAP) and cipe. No success, and this mainy due to impossibility of routing packets along the network space. A few months ago I used hostap's wds connections to make a similar job (on OpenSTA) but unencrypted and all was fine. On hostap's readme, I read that wds should be encrypted in order to avoid problems of eavesdropping and intromissions from outer world: this is what I need to do, but don't know how to reach the goal. No googles on this matter out there... did anybody encrypted traffic on wds and tell what are the trick(s) to do that ? regards, Alex From JFulmer at hrblock.com Thu Jan 9 13:22:44 2003 From: JFulmer at hrblock.com (Fulmer, John) Date: 09 Jan 2003 12:22:44 -0600 Subject: wds encryption II In-Reply-To: <001901c2b803$a44df3e0$af9fa8c0@admi001> References: <001901c2b803$a44df3e0$af9fa8c0@admi001> Message-ID: <1042136569.13853.8.camel@muir.hrblock.net> I do encrypted tunnels using Free S/Wan IPSEC tunnels between workstations and my AP. Works well, although re-establishing the tunnels can be tricky sometimes... jf On Thu, 2003-01-09 at 11:22, Alessandro Bondoli wrote: > Recently I had need to build a point-to-point encrypted link using another > couple of aps (LinuxAP) and cipe. No success, and this mainy due to > impossibility of routing packets along the network space. > A few months ago I used hostap's wds connections to make a similar job (on > OpenSTA) but unencrypted and all was fine. > On hostap's readme, I read that wds should be encrypted in order to avoid > problems of eavesdropping and intromissions from outer world: this is what I > need to do, but don't know how to reach the goal. > No googles on this matter out there... did anybody encrypted traffic on wds > and tell what are the trick(s) to do that ? > regards, Alex > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap From raghu at sraamerica.com Thu Jan 9 15:26:07 2003 From: raghu at sraamerica.com (Raghu) Date: Thu, 9 Jan 2003 15:26:07 -0500 Subject: 802.1X Unicast Keys? Message-ID: <20030109202607.ZLET5165.imta02a2.registeredsite.com@smtp.registeredsite.com> Hi, Since my last few mails did not get any response and I am still struggling to figure out a way to make the unicast keys working, I thought to try again and request you all for any kind of help/feedback in this regard. Jouni, Since you made the unicast keys working, can I request you to provide your setup so that I can replicate your setup and make it work. If possible, can you please provide a brief description of how unicast keys work in hostAP? Any hooks(in the code)/hints/tips are also greatly appreciated. -Raghu > > From: Raghu > Date: 2003/01/03 Fri PM 12:26:57 EST > To: hostap at shmoo.com > CC: raghu at sraamerica.com > Subject: Re: 802.1X Unicast Keys? > > Hi > > Happy New Year to everyone. > > A quick update, I have limited success with unicast keys, > and need your help to make it to work. > Once again my setup is XSupplicant, HostAP(client), > HostAP(AP), Hostapd (Authenticator), FreeRADIUS. > > STA configuration, > iwpriv wlan0 host_encrypt 1 > iwpriv wlan0 host_decrypt 1 > iwpriv wlan0 ieee_802_1x 1 > iwpriv wlan0 host_roaming 1 > > I am not sure if I can use, iwpriv wlan0 bcrx_sta_key 1 ? > > AP command, > $ ./hostapd -x -b13 -i13 -o192.168.1.1 -a192.168.1.1 > -stest -SSSID wlan0 > > Though I am using the same hostap cvs snapshot for both > STA and AP, Wireless extensions version differs at > AP(v12) and STA(v15). > > > After hostapd generates both broadcast and unicast keys, > iwconfig and iwlist shows only the broadcast key. > > # iwlist wlan0 key > wlan0 2 key sizes : 40, 104bits > 4 keys available : > [1]: off > [2]: 64A7-4D5C-D6C6-3653-505E-FB52-85 (104 > bits) > [3]: off > [4]: off > Current Transmit Key: [2] > Encryption mode:open > > How can I see the unicast key for each STA at AP ? > > Xsupplicant receives the unicast key, occasionally, > I am not sure when it receives the unicast key > and when it doesnot. I was under the impression that > whenever Xsupplicant sends EAPOL start, a new > authentication takes place and AP should generate a > new unicast key and STA should always receive it, > but this doesnot seem to be happening. > > After Xsupplicant receives both the broadcast key > and unicast key, iwconfig lists the broadcast key > and iwlist shows both the keys, > ie > # iwlist wlan0 key > wlan0 2 key sizes : 40, 104bits > 4 keys available : > [1]: 056D-031D-E10E-5780-4E2A-07A2-3F (104 > bits) > [2]: 64A7-4D5C-D6C6-3653-505E-FB52-85 (104 > bits) > [3]: off > [4]: off > Current Transmit Key: [2] > Encryption mode:open > > > Because of this Incompatibility STA is still not able > to ping the AP. > > Can you please let me know what are the other > configurations/commands that I need to set at STA/AP > to make unicast keys working ? > > > -Raghu > > From alessandro.bondoli at bondoli-campese.com Thu Jan 9 15:26:17 2003 From: alessandro.bondoli at bondoli-campese.com (Alessandro Bondoli) Date: Thu, 9 Jan 2003 21:26:17 +0100 Subject: wds encryption II References: <001901c2b803$a44df3e0$af9fa8c0@admi001> <1042136569.13853.8.camel@muir.hrblock.net> Message-ID: <000f01c2b81d$619c22a0$34d385d9@TOFIPC> Thank you John for your reply: exactly what I need to do is to join two separates segments of the same network, because on this network there are wireless devices that are working on it with fixed address and no dhcp. First segment is from 192.168.159.1 to 192.168.159.128, netmask 255.255.255.0; second segment is from 192.168.159.128 to 192.168.159.255 same netmask as above. Connection between those two trunks have to be realized with two aps with encrypted point-to-point wireless tunnel. Is this possible? What are needed steps? Thanks in advance for enlighting me. regards, Alex ----- Original Message ----- From: "Fulmer, John" To: "Alessandro Bondoli" Cc: Sent: Thursday, January 09, 2003 7:22 PM Subject: Re: wds encryption II > I do encrypted tunnels using Free S/Wan IPSEC tunnels between > workstations and my AP. Works well, although re-establishing the tunnels > can be tricky sometimes... > > jf > > On Thu, 2003-01-09 at 11:22, Alessandro Bondoli wrote: > > Recently I had need to build a point-to-point encrypted link using another > > couple of aps (LinuxAP) and cipe. No success, and this mainy due to > > impossibility of routing packets along the network space. > > A few months ago I used hostap's wds connections to make a similar job (on > > OpenSTA) but unencrypted and all was fine. > > On hostap's readme, I read that wds should be encrypted in order to avoid > > problems of eavesdropping and intromissions from outer world: this is what I > > need to do, but don't know how to reach the goal. > > No googles on this matter out there... did anybody encrypted traffic on wds > > and tell what are the trick(s) to do that ? > > regards, Alex > > > > > > _______________________________________________ > > HostAP mailing list > > HostAP at shmoo.com > > http://lists.shmoo.com/mailman/listinfo/hostap > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From jfulmer at hrblock.com Thu Jan 9 15:52:22 2003 From: jfulmer at hrblock.com (John Fulmer) Date: 09 Jan 2003 14:52:22 -0600 Subject: wds encryption II In-Reply-To: <000f01c2b81d$619c22a0$34d385d9@TOFIPC> References: <001901c2b803$a44df3e0$af9fa8c0@admi001> <1042136569.13853.8.camel@muir.hrblock.net> <000f01c2b81d$619c22a0$34d385d9@TOFIPC> Message-ID: <1042145547.13853.23.camel@muir.hrblock.net> On Thu, 2003-01-09 at 14:26, Alessandro Bondoli wrote: > Thank you John for your reply: exactly what I need to do is to join two > separates segments of the same network, because on this network there are > wireless devices that are working on it with fixed address and no dhcp. > First segment is from 192.168.159.1 to 192.168.159.128, netmask > 255.255.255.0; > second segment is from 192.168.159.128 to 192.168.159.255 same netmask as > above. > Connection between those two trunks have to be realized with two aps with > encrypted point-to-point wireless tunnel. Is this possible? What are needed > steps? What you want is an encrypted bridge. You CAN create a point to point IPSEC tunnel, and then (if they are two linux boxes) use the bridge code to create a VLAN that joins the two networks, but that is way beyond me. The IPSEC is easy enough to set up, setting up the VLAN config is tricky, and I don't remember how to do it anymore. I would find a Linux bridge mailing list and ask. Sorry I can't be more help... jf > Thanks in advance for enlighting me. > regards, Alex > > ----- Original Message ----- > From: "Fulmer, John" > To: "Alessandro Bondoli" > Cc: > Sent: Thursday, January 09, 2003 7:22 PM > Subject: Re: wds encryption II > > > > I do encrypted tunnels using Free S/Wan IPSEC tunnels between > > workstations and my AP. Works well, although re-establishing the tunnels > > can be tricky sometimes... > > > > jf > > > > On Thu, 2003-01-09 at 11:22, Alessandro Bondoli wrote: > > > Recently I had need to build a point-to-point encrypted link using > another > > > couple of aps (LinuxAP) and cipe. No success, and this mainy due to > > > impossibility of routing packets along the network space. > > > A few months ago I used hostap's wds connections to make a similar job > (on > > > OpenSTA) but unencrypted and all was fine. > > > On hostap's readme, I read that wds should be encrypted in order to > avoid > > > problems of eavesdropping and intromissions from outer world: this is > what I > > > need to do, but don't know how to reach the goal. > > > No googles on this matter out there... did anybody encrypted traffic on > wds > > > and tell what are the trick(s) to do that ? > > > regards, Alex > > > > > > > > > _______________________________________________ > > > HostAP mailing list > > > HostAP at shmoo.com > > > http://lists.shmoo.com/mailman/listinfo/hostap > > > > _______________________________________________ > > HostAP mailing list > > HostAP at shmoo.com > > http://lists.shmoo.com/mailman/listinfo/hostap > > From petergun at coffeehaus.com Thu Jan 9 22:21:51 2003 From: petergun at coffeehaus.com (Stephane Matis) Date: Thu, 9 Jan 2003 22:21:51 -0500 Subject: Compaq WL100 & 0.2.1 PRI Message-ID: Is there a way to upgrade a Compaq WL100's primary 0.2.1 ? I would like to get to 0.3.0 or higher, to remove one more potential pitfall in debugging inter-machine weirdness. I have an OpenBSD 3.2 box running AP mode with Linksys 0.3.0/0.8.3 connecting fine with a Mac G4 airport card, hex based WEP. A1 The WL100 in Win98 sees the AP, indicates excellent signal strength, but doesn't seem to send/received. I suspect some sort of WEP related bug. Anyone ? Thanks, Stephane From jkmaline at cc.hut.fi Thu Jan 9 23:29:15 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 20:29:15 -0800 Subject: hostapd forking and enhancements In-Reply-To: <013201c2b62c$6935e070$0200a8c0@defender> References: <001501c2b5b8$865718a0$017ba8c0@ammirata.net> <20030107035740.GB1812@jm.kir.nu> <013201c2b62c$6935e070$0200a8c0@defender> Message-ID: <20030110042915.GC2055@jm.kir.nu> On Tue, Jan 07, 2003 at 09:08:57AM -0000, Ian Cass wrote: > Would also be cool if there was a way to tag packets from each user > differently as they fly past so we can maintain the concept of sessions all > the way to the firewall. This then allows tradition logon processes, such as > a web page to stand a chance of working even halfway securely. Vlan tagging > perhaps? or TOS at a stretch. There is already a "tag" in each packet: station MAC address. If you want to have a unique tag for each user, I would assume this is the easiest one since it does not require any new features.. Station MAC address will be the source address of the Ethernet headers in data frames passed to Linux net stack. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Thu Jan 9 23:35:34 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 20:35:34 -0800 Subject: Crash problem of HostAP driver-2002-10-12 on MPC857 platform In-Reply-To: <01139FC052A0D411900B00508B9535FC09DE32D1@ZCH07EXM04.corp.mot.com> References: <01139FC052A0D411900B00508B9535FC09DE32D1@ZCH07EXM04.corp.mot.com> Message-ID: <20030110043534.GD2055@jm.kir.nu> On Thu, Jan 09, 2003 at 04:11:51PM +0800, Hu Tao-a2587c wrote: > HW Platform: IAD857 (Motorola PowerPC857) I'm using Host AP driver with MPC8260-based board, but I have not tested it with MPC8xx boards. > All of the modules(pcmcia_core, m8xx_pcmcia, ds, hostap_crypt, hostap, hostap_cs)had been loaded successfully, > and the cardmgr had been started. So, the card was initialized properly? Was there anything strange in the kernel log or did it look okay (e.g., Host AP driver reported firmware versions, etc.)? > When "ifconfig wlan0 up" is used to bring up the NIC, crash occured. Following is the oops: > > NIP: C010B83C XER: E000017F LR: C001BBAC SP: C1EF3D70 REGS: c1ef3cc0 TRAP: 0300 > Call backtrace: > 00000001 C001BBAC C00C0510 C00C1BDC C00EFDBC C00F216C C00BA674 > C00454CC C000245C 100028B4 1001AC4C 1002E128 1002DD78 0FF0A3A4 > 00000000 Could you please run this through ksymoops or manually resolve function names for NIP and the backtrace? Without System.map (and kernel module load points) this report is not very useful. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Thu Jan 9 23:49:14 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 20:49:14 -0800 Subject: Host based encryption needed with 1X? In-Reply-To: <5.2.0.9.0.20030109160409.02158828@pop.mail.yahoo.com> References: <5.1.1.6.0.20030109155113.02da2318@imap.lysator.liu.se> <5.2.0.9.0.20030109160409.02158828@pop.mail.yahoo.com> Message-ID: <20030110044914.GE2055@jm.kir.nu> On Thu, Jan 09, 2003 at 04:08:05PM +0100, Jacques Caron wrote: > If you are using encryption (which you should, with 802.1X), then host > based encryption is needed because there's one key per station, and there > is no known interface in the hostap firmware code to tell the card to use > different keys for each station. In addition, IEEE 802.1X requires that EAPOL packets are passed through unencrypted, which requires host-based encryption with current firmware versions. > 802.11i contains a lot more than just 802.1X, and even 802.1X is changing > (with 802.1aa), so I think the hostap 802.1X code is more "de facto > standard" (mainly Cisco and XP) compliant than anything (the exact way to > use 802.1X with 802.11 was/is not defined yet). I have updated hostapd code to latest 802.1aa draft. However, couple of changes are commented out since they seem to break key distribution. As far as I know, it would be fair to say that 802.1X code in hostapd is mostly standard compliant. As far as 802.1X use with 802.11 is concerned, yes, it is currently, for many parts, de facto standard. Anyway, some parts are specified in non-normative Appendix of 802.1X (and also as an informational Internet-Draft, draft-congdon-radius-8021x-20.txt). -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Thu Jan 9 23:53:26 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 20:53:26 -0800 Subject: Hostap-2002-10-12, Installation problem on Linux In-Reply-To: <3E1DA100.1FA7B6E6@tataelxsi.co.in> References: <3E1DA100.1FA7B6E6@tataelxsi.co.in> Message-ID: <20030110045326.GF2055@jm.kir.nu> On Thu, Jan 09, 2003 at 09:49:12PM +0530, Venkatesh N wrote: > > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_ap.o What is this? There is no hostap_ap.o module.. How did you compile the driver? > > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o > > depmod: *** Unresolved symbols in > > /lib/modules/2.4.20/net/hostap_crypt.o > > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_cs.o Could you run 'depmod -ae' to list all the unresolved symbols? I would guess that this is due to nonmatching version symbols, but getting the list would make it easier to confirm this. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Fri Jan 10 00:06:25 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 21:06:25 -0800 Subject: wds encryption II In-Reply-To: <1042145547.13853.23.camel@muir.hrblock.net> References: <001901c2b803$a44df3e0$af9fa8c0@admi001> <1042136569.13853.8.camel@muir.hrblock.net> <000f01c2b81d$619c22a0$34d385d9@TOFIPC> <1042145547.13853.23.camel@muir.hrblock.net> Message-ID: <20030110050625.GG2055@jm.kir.nu> On Thu, Jan 09, 2003 at 02:52:22PM -0600, John Fulmer wrote: > What you want is an encrypted bridge. You CAN create a point to point > IPSEC tunnel, and then (if they are two linux boxes) use the bridge code > to create a VLAN that joins the two networks, but that is way beyond me. Yes, one way to do this is to use IPSec protected IP connection with any virtual tunneling of bridged networks. OpenVPN (http://openvpn.sourceforge.net/) is another alternative for this. It uses non-IPSec tunnel and TUN/TAP device (TAP part, i.e., virtual Ethernet devide, bridged with the real Ethernet device on both tunnel end-points). -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Fri Jan 10 00:31:46 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 21:31:46 -0800 Subject: 802.1X Unicast Keys? In-Reply-To: <20030103172657.YPHD5165.imta02a2.registeredsite.com@smtp.registeredsite.com> References: <20030103172657.YPHD5165.imta02a2.registeredsite.com@smtp.registeredsite.com> Message-ID: <20030110053146.GH2055@jm.kir.nu> On Fri, Jan 03, 2003 at 12:26:57PM -0500, Raghu wrote: > I am not sure if I can use, iwpriv wlan0 bcrx_sta_key 1 ? Not in the station, but in the AP you could try using it and then forcing station to use the individual key for all frames. > After hostapd generates both broadcast and unicast keys, > iwconfig and iwlist shows only the broadcast key. Linux wireless extensions does not yet support setting or showing unicast keys. You will need to use utils/hostap_crypt_conf to view the unicast keys on the AP. Station use only the default four keys (first one is the unicast key, rest are used as broadcast keys), so this can be shown with wireless tools. > After Xsupplicant receives both the broadcast key > and unicast key, iwconfig lists the broadcast key > and iwlist shows both the keys, > [1]: 056D-031D-E10E-5780-4E2A-07A2-3F (104 > [2]: 64A7-4D5C-D6C6-3653-505E-FB52-85 (104 > Current Transmit Key: [2] Hmm.. If I remember correctly, transmit key should be 1 so that the individual key would be used. Otherwise this looks OK. You should see matching key1 in the AP with hostap_crypt_conf. I think I should go through the station case again at some point.. I do not remember anymore, whether I implemented support for selecting different key for unicast and broadcast keys for the station (Managed) mode. with bcrx_sta_key 1 in the AP, one should be able to use this by changing the current transmit key to 1 even if the driver does not support key selection based on unicast/broadcast destination. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Fri Jan 10 00:42:49 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 21:42:49 -0800 Subject: 802.1X Unicast Keys? In-Reply-To: <20030109202607.ZLET5165.imta02a2.registeredsite.com@smtp.registeredsite.com> References: <20030109202607.ZLET5165.imta02a2.registeredsite.com@smtp.registeredsite.com> Message-ID: <20030110054249.GI2055@jm.kir.nu> On Thu, Jan 09, 2003 at 03:26:07PM -0500, Raghu wrote: > Jouni, > Since you made the unicast keys working, can I request > you to provide your setup so that I can replicate your > setup and make it work. > If possible, can you please provide a brief description > of how unicast keys work in hostAP? This will have to wait some time. My test setup at home is not currently exactly ready for easy 802.1x experiments.. Anyway, I try to remember to write step-by-step instructions both for Linux and WinXP clients when I get my 802.1x setup running next time. -- Jouni Malinen PGP id EFC895FA From venky at tataelxsi.co.in Fri Jan 10 01:39:21 2003 From: venky at tataelxsi.co.in (Venkatesh N) Date: Fri, 10 Jan 2003 12:09:21 +0530 Subject: Hostap on 802.11a Cards Message-ID: <3E1E6A99.50E1CAFE@tataelxsi.co.in> Hi All, I have a query. Does the Hostap-2002-10-12 works for PCI based 802.11a (prism 2/2.5/3) cards. Has anybody tried out this? If yes on what vendor cards does Hostap-2002-10-12 succesfully operate. If Hostap-2002-10-12 does not work with 802.11a cards, do we have any other similar software that could configure the card to behave as an access point. A help here is very much appreciated regards, Venkatesh N Tata Elxsi Ltd Bangalore -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030110/3878ee01/attachment.htm From jbreeden at plumhall.com Fri Jan 10 01:45:53 2003 From: jbreeden at plumhall.com (jbreeden at plumhall.com) Date: Thu, 9 Jan 2003 20:45:53 -1000 (HST) Subject: CVS -C switch Message-ID: <4024.192.168.1.12.1042181153.squirrel@webmail.plumhall.com> I'm trying to get the CVS "client login while ap" working. Everything goes great, fire off hostapd with -C set to the other ap's hdwr address and client logs in. shows up in other ap's /proc/net/hostap/wlan0. As soon as I send the first data frame (ping), the client deauthenticates with a reason_code=3 and sure 'nuff, it disappears out of /proc. I'm stumped. Tried sta 1.3.6, 1.4.9 and 1.5.6, same results. debug on ap side: Opening raw packet socket for ifindex 8 Using interface wlan0ap with hwaddr 00:90:d1:07:ed:02 and ssid 'fuzzy' Flushing old station entries Received 30 bytes management frame dump: b0 00 02 01 00 90 d1 07 ed 02 00 06 25 a8 46 01 00 90 d1 07 ed 02 80 44 0000 01 00 00 00 MGMT mgmt::auth authentication: STA=00:06:25:a8:46:01 auth_alg=0 auth_transaction=1 status_code=0 New STA Station 00:06:25:a8:46:01 authentication OK (open system) Received 30 bytes management frame dump: b2 00 02 01 00 06 25 a8 46 01 00 90 d1 07 ed 02 00 90 d1 07 ed 02 50 02 0000 02 00 00 00 MGMT (TX callback) ACK mgmt::auth cb Station 00:06:25:a8:46:01 authenticated Received 41 bytes management frame dump: 00 00 02 01 00 90 d1 07 ed 02 00 06 25 a8 46 01 00 90 d1 07 ed 02 90 44 0000 01 00 00 05 66 75 7a 7a 79 01 04 82 84 0b 16 MGMT mgmt::assoc_req association request: STA=00:06:25:a8:46:01 capab_info=0x00 listen_interval=1 new AID 1 Station 00:06:25:a8:46:01 association OK (aid 1) Received 36 bytes management frame dump: 12 00 02 01 00 06 25 a8 46 01 00 90 d1 07 ed 02 00 90 d1 07 ed 02 60 02 0100 00 00 01 c0 01 04 82 84 0b 16 MGMT (TX callback) ACK mgmt::assoc_resp cb Station 00:06:25:a8:46:01 associated (aid 1) *** FIRST data frame sent *** Received 26 bytes management frame dump: c0 00 02 01 00 90 d1 07 ed 02 00 06 25 a8 46 01 00 90 d1 07 ed 02 30 48 0300 MGMT mgmt::deauth deauthentication: STA=00:06:25:a8:46:01 reason_code=3 Station 00:06:25:a8:46:01 deauthenticated -- John R. Breeden jbreeden at plumhall.com Kailua-Kona, Hawaii --- This message (including any attachments) contains confidential information intended for a specific individual and purpose, and is protected by law. If you are not the intended recipient, you should delete this message and are hereby notified that any disclosure, copying, or distribution of this message, or the taking of any action based on it, is strictly prohibited. --- From jkmaline at cc.hut.fi Fri Jan 10 02:10:22 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 9 Jan 2003 23:10:22 -0800 Subject: CVS -C switch In-Reply-To: <4024.192.168.1.12.1042181153.squirrel@webmail.plumhall.com> References: <4024.192.168.1.12.1042181153.squirrel@webmail.plumhall.com> Message-ID: <20030110071022.GA3842@jm.kir.nu> On Thu, Jan 09, 2003 at 08:45:53PM -1000, jbreeden at plumhall.com wrote: > Everything goes great, fire off hostapd with -C set to the other > ap's hdwr address and client logs in. shows up in other ap's > /proc/net/hostap/wlan0. > > As soon as I send the first data frame (ping), the client deauthenticates with > a reason_code=3 and sure 'nuff, it disappears out of /proc. hostapd is not sending deauth frames with reason_code=3 (deauth, STA leaving), so this would need to come from firmware.. I did not actually test this with another Host AP device, but with an SMC AP. I'll test this with two hostapd's, but I would assume that this should not really change things at all (unless that SMC AP is ignoring deauth frame; I did not have sniffer running at the time, so I did not check whether there were any unexpected management frames being sent). Is the host that is running hostapd with -C option in Master mode? Does it show anything strange in the hostapd debug? -- Jouni Malinen PGP id EFC895FA From earl at stanfordalumni.org Fri Jan 10 04:48:07 2003 From: earl at stanfordalumni.org (earl) Date: Fri, 10 Jan 2003 01:48:07 -0800 Subject: Tx timed out! Resetting card (with OpenAP) References: <3E1BF9C4.4143A1F9@stanfordalumni.org> <007901c2b702$e52e9c90$53153b0a@nednet.co.za> <3E1C0812.36DF29D3@stanfordalumni.org> <008401c2b70d$61255690$53153b0a@nednet.co.za> Message-ID: <3E1E96D7.438445F5@stanfordalumni.org> Robert Botha wrote: > > > Have you tried using the external package seperately and disabling the > > > kernel drivers? > > > I've had endless hassles with the kernel pcmcia and most of them went away > > > > when I switched to the pcmcia-cs package. As you suggested (thanks!) i replaced kernel pcmcia with the pcmcia modules, and the behavior is different. I've only tried this so far in the case of wds_type=2 and *not* the case of doing the volatile firmware update with wds_type=4. (I don't understand whether a card-reset will cause the volatile firmware update to be lost, causing the card to revert to its permanent firmware --- anyone know what to expect?) While ping-flooding the "Tx timed out!" messages still happen, probably just as frequently as with kernel pcmcia, but the difference is that after most incidents the card is able to automatically reset successfully and the link comes back up again for a while, until the next timeout. Following the messages: NETDEV WATCHDOG: wlan0: transmit timed out wlan0 Tx timed out! Resetting card i'll see either this: hostap_cs: wlan0: resetting card or this: hostap_cs: wlan0: resetting card hostap_cs: card initialization timed out hostap_cs: Initialization failed hostap_cs: wlan0: resetting card or occasionally this: hostap_cs: wlan0: resetting card wlan0: Interrupt, but SWSUPPORT0 does not match: 0000 != 8A32 - card removed? hostap_cs: wlan0: resetting card hostap_cs: card initialization timed out hostap_cs: Initialization failed hostap_cs: wlan0: resetting card following which after a second or so the link comes back up. However, if i see this: hostap_cs: wlan0: resetting card hostap_cs: card initialization timed out hostap_cs: Initialization failed hostap_cs: wlan0: resetting card hostap_cs: card initialization timed out hostap_cs: Initialization failed or this: hostap_cs: wlan0: resetting card wlan0: Interrupt, but SWSUPPORT0 does not match: 0000 != 8A32 - card removed? hostap_cs: card initialization timed out hostap_cs: Initialization failed hostap_cs: wlan0: resetting card hostap_cs: card initialization timed out hostap_cs: Initialization failed then the link is permanently dead. After that point things can't even be revived with "iwpriv wlan0 reset 1" which merely responds with: hostap_cs: wlan0: resetting card hostap_cs: card initialization timed out hostap_cs: Initialization failed I guess this is slightly better than with kernel pcmcia but it's still bound to fail eventually. Any further suggestions? Some pcmcia parameters i should be setting to nondefaults? Oh yeah --- here are the pcmcia and hostap modules loading. Note the "ignore=1" argument is there because there's only one "slot" on this board, and without that argument the insmod causes a kernel oops crash. # insmod pcmcia_core insmod: /lib/modules/2.4.20: No such file or directory Using /lib/modules/pcmcia_core.o Linux PCMCIA Card Services 3.2.3 kernel build: 2.4.20 #1 Thu Jan 9 22:00:50 UTC 2003 options: none # insmod i82365 ignore=1 insmod: /lib/modules/2.4.20: No such file or directory Using /lib/modules/i82365.o Intel ISA/PCI/CardBus PCIC probe: Intel i82365sl A step ISA-to-PCMCIA at port 0x3e0 ofs 0x00 host opts [0]: none ISA irqs (default) = 3,5,7,9,10,11,12,15 polling interval = 1000 ms # insmod ds insmod: /lib/modules/2.<.20: No such file or directory Using /lib/modules/ds.o # /etc/init.d/wireless start Starting wireless: cardmgr[29]: watching 1 sockets Jan 1 00:12:11 cs: memory probe 0x0d0000-0x0dffff:cardmgr[30]: starting, version i clean. s 3.2.3 Jan 1 00:12:11 cardmgr[30]: socket 0: Longshine LCR-8531 11Mbps WLAN PCMCIA CARD Jan 1 00:12:11 cardmgr[30]: executing: 'insmod /lib/modules/preferred/pcmcia/hostap_crypt.o' Jan 1 00:12:12 cardmgr[30]: + Using /lib/modules/preferred/pcmcia/hostap_crypt.o Jan 1 00:12:12 cardmgr[30]: executing: 'insmod /lib/modules/preferred/pcmcia/hostap.o' Jan 1 00:12:13 cardmgr[30]: + Using /lib/modules/preferred/pcmcia/hostap.o Jan 1 00:12:13 cardmgr[30]: executing: 'insmod /lib/modules/preferred/pcmcia/hostap_cs.o ignore_cis_vcc=1' hostap_cs: 0.0.0 CVS (Jouni Malinen ) Jan 1 00:12:13 cs: IO port probe 0x0100-0x04ff:cardmgr[30]: + U clean. sing /lib/modulecs: IO port probe 0x0800-0x08ff:s/preferred/pcmc clean. cs: IO port probe 0x0c00-0x0cff: clean. hostap_cs: index 0x01: Vcc 5.0, irq 3, io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 wlan0: NIC: id=0x8002 v1.0.0 wlan0: PRI: id=0x15 v0.3.0 wlan0: STA: id=0x1f v0.7.5 Jan 1 00:12:14 cardmgr[30]: executing: './network start wlan0' Jan 1 00:12:15 cardmgr[30]: + wlan0 Link encap:Etherne? HWaddr 00:90:D1:00:C4:88 Jan 1 00:12:15 cardmgr[30]: + BROADCAST MULTICAST MTU:1500 Metric:1 Jan 1 00:12:15 cardmgr[30]: + RX packets:2 errors:0 dropped:0 overruns:0 frame:0 Jan 1 00:12:15 cardmgr[30]: + TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 Jan 1 00:12:15 cardmgr[30]: + collisions:0 txqueuelen:100 Jan 1 00:12:15 cardmgr[30]: + RX bytes:0 (0.0 iB) TX bytes:0 (0.0 iB) Jan 1 00:12:15 cardmgr[30]: + Interrupt:3 Base address:0x100 Jan 1 00:12:15 cardmgr[30]: + thanks earl From Peter.DeCleyn at ua.ac.be Fri Jan 10 05:03:56 2003 From: Peter.DeCleyn at ua.ac.be (Peter De Cleyn) Date: Fri, 10 Jan 2003 11:03:56 +0100 (CET) Subject: bitrate difference Message-ID: Hi, Can anyone tell me why my mobile station using an orinoco card (patched orinoco_cs driver from airsnort to enable monitoring, 8.10 firmware) can only connect to a hostap system using a WPC11 card at 2Mbit/s. The sta is in close range, the ap at full txpower. When I try the same with a WPC11 card in the mobile station, it indicates 11Mbit/s. Using iwconfig to force the card into 11Mbit/s doesn't work either. Comments on this are welkom! Peter De Cleyn _______________ / Peter De Cleyn\ ------------------------------------------------------- | Phone: +32 (0)3 820.24.18 | | Fax: +32 (0)3 820.24.21 | | email: peter.decleyn at ua.ac.be | | | | University of Antwerp (UIA) | | Dept. Mathematics and Computer Science | | PATS Research Group | | Universiteitsplein 1 | | B-2610 Wilrijk | | Belgium | ------------------------------------------------------- From linux at rocksolidnetworks.com Fri Jan 10 10:04:40 2003 From: linux at rocksolidnetworks.com (Linux Rocks!) Date: Fri, 10 Jan 2003 07:04:40 -0800 Subject: What to do now? Message-ID: <200301100704.40609.linux@rocksolidnetworks.com> My cards (Proxim RangeLan-DS) still dont work under linux :( They do respond under linux (I can change modes, and see the lights do different things...) but I dont have them doing anything usefull. They Do work under windows95/98. (but ... windows doesnt do what I need) They are 3.3v cards forced into 5v controllers. New cards arent an option New laptops arent an option hostap is really what I need to work (although just orinoco would be nice too) writing David Hinds (maintainer of pcmcia) bounces. and no-reply. All the linux software compiles ok, configs seem ok, dmesg looks good. Cards do not work under linux :( Jamie -- No microsoft products were used to produce this message. EUG-LUG Mailing List: http://mailman.efn.org/cgi-bin/listinfo/eug-lug From alessandro.bondoli at bondoli-campese.com Fri Jan 10 10:26:58 2003 From: alessandro.bondoli at bondoli-campese.com (Alessandro Bondoli) Date: Fri, 10 Jan 2003 16:26:58 +0100 Subject: wds encryption II Message-ID: <001401c2b8bc$b9761040$af9fa8c0@admi001> Yes Jouni, I realize that encryption tunneling can be done: it's only the way to do it that is in the darkness. If you should do tunneling for wlan0wds0 interface, what will you do to make it working ? regards, Alex From Jacques.Caron at IPsector.com Fri Jan 10 10:38:58 2003 From: Jacques.Caron at IPsector.com (Jacques Caron) Date: Fri, 10 Jan 2003 16:38:58 +0100 Subject: Hostap on 802.11a Cards In-Reply-To: <3E1E6A99.50E1CAFE@tataelxsi.co.in> Message-ID: <5.2.0.9.0.20030110163316.019a9b70@pop.mail.yahoo.com> Hi, The PRISM 2/2.5/3 chipsets only support 802.11b. 802.11a is supported by the PRISM Indigo and Duette chipsets only, but I don't think there are any cards with those chipsets yet. Let us know if you find any. Intersil said that hostap mode would continue to be supported in newer PRISM chipsets, so once the cards are out, chances are good they will be supported, but until then... And as far as I know, the only available 802.11a cards are based on the Broadcom chipset, the 802.11b version of which does not even have any drivers. And even if there were, there's no guarantee that the firmware of the card would allow anything like hostap to happen. Jacques. At 07:39 10/01/2003, Venkatesh N wrote: >Hi All, > >I have a query. >Does the Hostap-2002-10-12 works for PCI based 802.11a (prism 2/2.5/3) cards. >Has anybody tried out this? >If yes on what vendor cards does Hostap-2002-10-12 succesfully operate. > >If Hostap-2002-10-12 does not work with 802.11a cards, do we >have any other similar software that could configure the card to behave as an >access point. > >A help here is very much appreciated > >regards, >Venkatesh N >Tata Elxsi Ltd >Bangalore > > > -- Jacques Caron, IP Sector Technologies Join the discussion on public WLAN open global roaming: http://lists.ipsector.com/listinfo/openroaming From sam at errno.com Fri Jan 10 12:16:26 2003 From: sam at errno.com (Sam Leffler) Date: Fri, 10 Jan 2003 09:16:26 -0800 Subject: Hostap on 802.11a Cards References: <5.2.0.9.0.20030110163316.019a9b70@pop.mail.yahoo.com> Message-ID: <129e01c2b8cc$02f5af50$52557f42@errno.com> > And as far as I know, the only available 802.11a cards are based on the > Broadcom chipset, the 802.11b version of which does not even have any > drivers. And even if there were, there's no guarantee that the firmware of > the card would allow anything like hostap to happen. > I believe the majority of the 802.11a products currently on the market have Atheros chips in them. Open source driver support for Atheros h/w is going on but not yet available. Sam From Jacques.Caron at IPsector.com Fri Jan 10 12:19:51 2003 From: Jacques.Caron at IPsector.com (Jacques Caron) Date: Fri, 10 Jan 2003 18:19:51 +0100 Subject: Hostap on 802.11a Cards In-Reply-To: <129e01c2b8cc$02f5af50$52557f42@errno.com> References: <5.2.0.9.0.20030110163316.019a9b70@pop.mail.yahoo.com> Message-ID: <5.2.0.9.0.20030110181846.01c85a90@wheresmymailserver.com> Ooops indeed, my mistake, Broadcom is the only one with (draft) 802.11g, not a. Sorry about the confusion. Jacques. At 18:16 10/01/2003, Sam Leffler wrote: > > And as far as I know, the only available 802.11a cards are based on the > > Broadcom chipset, the 802.11b version of which does not even have any > > drivers. And even if there were, there's no guarantee that the firmware of > > the card would allow anything like hostap to happen. > > > >I believe the majority of the 802.11a products currently on the market have >Atheros chips in them. Open source driver support for Atheros h/w is going >on but not yet available. > > Sam -- Jacques Caron, IP Sector Technologies Join the discussion on public WLAN open global roaming: http://lists.ipsector.com/listinfo/openroaming From sam at errno.com Fri Jan 10 12:32:01 2003 From: sam at errno.com (Sam Leffler) Date: Fri, 10 Jan 2003 09:32:01 -0800 Subject: Hostap on 802.11a Cards References: <5.2.0.9.0.20030110163316.019a9b70@pop.mail.yahoo.com> <5.2.0.9.0.20030110181846.01c85a90@wheresmymailserver.com> Message-ID: <132001c2b8ce$2fa15070$52557f42@errno.com> > Ooops indeed, my mistake, Broadcom is the only one with (draft) 802.11g, > not a. Sorry about the confusion. The Atheros 5212 parts do a/b/g. > > At 18:16 10/01/2003, Sam Leffler wrote: > > > And as far as I know, the only available 802.11a cards are based on the > > > Broadcom chipset, the 802.11b version of which does not even have any > > > drivers. And even if there were, there's no guarantee that the firmware of > > > the card would allow anything like hostap to happen. > > > > > > >I believe the majority of the 802.11a products currently on the market have > >Atheros chips in them. Open source driver support for Atheros h/w is going > >on but not yet available. > > > > Sam > > > -- Jacques Caron, IP Sector Technologies > Join the discussion on public WLAN open global roaming: > http://lists.ipsector.com/listinfo/openroaming > > > > From hostapd at cybercity.sk Fri Jan 10 14:46:34 2003 From: hostapd at cybercity.sk (hostapd at cybercity.sk) Date: Fri, 10 Jan 2003 20:46:34 +0100 Subject: System freeze on higher troughtput than 64kbps Message-ID: <0865.030110@cybercity.sk> Hello hostap, My system that's act like AP sometimes freeze where i send more than 64kbps stream here is log Jan 10 19:51:17 palaska kernel: wlan0: prism2_tx - to BAP0 failed Jan 10 19:51:17 palaska kernel: hostap_pci: wlan0: resetting card Jan 10 19:51:17 palaska kernel: wlan0: prism2_tx - to BAP0 failed Jan 10 19:51:17 palaska kernel: hostap_pci: wlan0: resetting card any suggestion? Best regards, hostapd mailto:hostapd at cybercity.sk From jbreeden at plumhall.com Fri Jan 10 16:58:33 2003 From: jbreeden at plumhall.com (jbreeden at plumhall.com) Date: Fri, 10 Jan 2003 11:58:33 -1000 (HST) Subject: CVS -C switch In-Reply-To: <20030110071022.GA3842@jm.kir.nu> References: <4024.192.168.1.12.1042181153.squirrel@webmail.plumhall.com> <20030110071022.GA3842@jm.kir.nu> Message-ID: <1352.192.168.1.12.1042235913.squirrel@webmail.plumhall.com> \> Is the host that is running hostapd with -C option in Master mode? Does it > show anything strange in the hostapd debug? > Jouni, thanks for the quick response, and yes hostapd -C is in master mode. Tried a few more tings today, first updated my CVS to today and tried connecting hostapd -C to a Dlink 900AP with a number of sta versions, this is what I got: sta 0.7.6 - connected sta 1.3.6 - would not connect sta 1.4.9 - would not connect sta 1.5.6 - connected What I mean by connected is the log message "Associated (as station) with AP 00:05:5d:25:d5:db (aid=0)" after hostapd -C wlan0 is fired off. as soon as the first data pdu (ping) issent, I get a deauthenticate an all cases (0.7.6 and 1.5.6). This same behavior happens if hostapd -C is talking to the Dlink or another hostapd runnimg in Master. The logs on hostapd -C talking to the dlink shows a couple of things, one is: handle_auth_cb: STA 00:05:5d:25:d5:db not found and the other is: unknown mgmt cb frame subtype 0 Both of these messages show up connecting to both the dlink and hostap. Unique to the dlink is the following: Association (as station) with AP 00:05:5d:25:d5:db failed (status_code=10) followed by a successful association. The full log (hostapd client in Master connecting to dlink): Using interface wlan0ap with hwaddr 00:90:d1:07:27:1c and ssid 'test' Flushing old station entries Deauthenticate all stations Authenticate with AP 00:05:5d:25:d5:db SSID=fuzzy2 (as station) handle_auth_cb: STA 00:05:5d:25:d5:db not found Authenticated (as station) with AP 00:05:5d:25:d5:db Associate with AP 00:05:5d:25:d5:db SSID=fuzzy2 (as station) unknown mgmt cb frame subtype 0 Association (as station) with AP 00:05:5d:25:d5:db failed (status_code=10) Associated (as station) with AP 00:05:5d:25:d5:db (aid=0) ** first pdu (ping) sent from this client to dlink) ** Station 00:05:5d:25:d5:db trying to disassociate, but it is not associated. Station 00:05:5d:25:d5:db trying to disassociate, but it is not associated. ... The same with -d flag: Opening raw packet socket for ifindex 4 Using interface wlan0ap with hwaddr 00:90:d1:07:27:1c and ssid 'test' Flushing old station entries Deauthenticate all stations MGMT mgmt::beacon Authenticate with AP 00:05:5d:25:d5:db SSID=fuzzy2 (as station) Beacon from 00:05:5d:25:d5:db SSID='fuzzy2' CHAN=3 Received 30 bytes management frame dump: b2 00 02 01 00 05 5d 25 d5 db 00 90 d1 07 27 1c 00 05 5d 25 d5 db 10 89 00 00 01 00 00 00MGMT (TX callback) ACK mgmt::auth cb handle_auth_cb: STA 00:05:5d:25:d5:db not found Received 30 bytes management frame dump: b0 00 d5 00 00 90 d1 07 27 1c 00 05 5d 25 d5 db 00 05 5d 25 d5 db b0 c1 00 00 02 00 00 00MGMT mgmt::auth authentication: STA=00:05:5d:25:d5:db auth_alg=0 auth_transaction=2 status_code=0 Authenticated (as station) with AP 00:05:5d:25:d5:db Associate with AP 00:05:5d:25:d5:db SSID=fuzzy2 (as station) Received 42 bytes management frame dump: 02 00 02 01 00 05 5d 25 d5 db 00 90 d1 07 27 1c 00 05 5d 25 d5 db 20 89 00 00 01 00 00 06 66 75 7a 7a 79 32 01 04 82 84 0b 16MGMT (TX callback) ACK unknown mgmt cb frame subtype 0 Received 36 bytes management frame dump: 10 00 d5 00 00 90 d1 07 27 1c 00 05 5d 25 d5 db 00 05 5d 25 d5 db c0 c1 01 00 0a 00 00 00 01 04 02 04 0b 96MGMT mgmt::assoc_resp Association (as station) with AP 00:05:5d:25:d5:db failed (status_code=10) Associated (as station) with AP 00:05:5d:25:d5:db (aid=0) ... You are right, it's a fireware issue. the first time I set it up it worked. I was using cards with verious older sta versions, Some cards worked, others didn't. In a flash of brillance I re-flashed them all to 1.4.9. Yep, you're right - I don't know what those "older versions" where :-) duh... When it worked, it was great - far better thruput than I was getting with WDS. BTW, it's the same trick pre Cisco Aironet used on their old 350 APs. -- John R. Breeden jbreeden at plumhall.com Kailua-Kona, Hawaii --- This message (including any attachments) contains confidential information intended for a specific individual and purpose, and is protected by law. If you are not the intended recipient, you should delete this message and are hereby notified that any disclosure, copying, or distribution of this message, or the taking of any action based on it, is strictly prohibited. --- From cijoml at volny.cz Fri Jan 10 17:24:27 2003 From: cijoml at volny.cz (Michal Semler) Date: Fri, 10 Jan 2003 23:24:27 +0100 Subject: What to do now? In-Reply-To: <200301100704.40609.linux@rocksolidnetworks.com> References: <200301100704.40609.linux@rocksolidnetworks.com> Message-ID: Dne p? 10. leden 2003 16:04 jste napsal(a): > My cards (Proxim RangeLan-DS) still dont work under linux :( Proxim RangeLan drivers are here: http://www.komacke.com/distribution.html Michal > They do respond under linux (I can change modes, and see the lights do > different things...) but I dont have them doing anything usefull. > They Do work under windows95/98. (but ... windows doesnt do what I need) > They are 3.3v cards forced into 5v controllers. > New cards arent an option > New laptops arent an option > hostap is really what I need to work (although just orinoco would be nice > too) writing David Hinds (maintainer of pcmcia) bounces. and no-reply. > All the linux software compiles ok, configs seem ok, dmesg looks good. > Cards do not work under linux :( > > Jamie From cijoml at volny.cz Fri Jan 10 17:46:34 2003 From: cijoml at volny.cz (Michal Semler) Date: Fri, 10 Jan 2003 23:46:34 +0100 Subject: unresolved symbols Message-ID: hello, i downloaded actual CVS version of hostap and now I am getting this ;( notas:/usr/src/wireless/hostap# make install_pci Installing hostap_crypt*.o to /lib/modules/2.4.20/net mkdir -p /lib/modules/2.4.20/net cp -f driver/modules/hostap_crypt*.o /lib/modules/2.4.20/net Installing hostap.o to /lib/modules/2.4.20/net mkdir -p /lib/modules/2.4.20/net cp -f driver/modules/hostap.o /lib/modules/2.4.20/net Installing hostap_pci.o to /lib/modules/2.4.20/net mkdir -p /lib/modules/2.4.20/net cp -f driver/modules/hostap_pci.o /lib/modules/2.4.20/net /sbin/depmod -a depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o make: *** [install_pci] Error 1 Previous versions were OK Michal From linux at rocksolidnetworks.com Fri Jan 10 20:56:16 2003 From: linux at rocksolidnetworks.com (Linux Rocks!) Date: Fri, 10 Jan 2003 17:56:16 -0800 Subject: What to do now? In-Reply-To: References: <200301100704.40609.linux@rocksolidnetworks.com> Message-ID: <200301101756.16893.linux@rocksolidnetworks.com> Does that driver work with prism2 chipset? I thought it worked with prism2.5 chipset. I have a Proxim RangeLan-DS, Not a Proxim RangeLan2 card.. Mine is model number 8433-05 (internal antenna) Ive looked but havnt found what chipset the RangeLan2 cards have... I thought I read (was maybe 6 months ago...) that it was the Prism2.5 chip, and figured the rlmod driver wouldnt work with the prism2 chip. PS, I tried the linux-wlan driver,but never got it to compile right... Jamie On Friday 10 January 2003 02:24 pm, Michal Semler wrote: : Dne p? 10. leden 2003 16:04 jste napsal(a): : > My cards (Proxim RangeLan-DS) still dont work under linux :( : : Proxim RangeLan drivers are here: : : http://www.komacke.com/distribution.html : : Michal : : > They do respond under linux (I can change modes, and see the lights do : > different things...) but I dont have them doing anything usefull. : > They Do work under windows95/98. (but ... windows doesnt do what I need) : > They are 3.3v cards forced into 5v controllers. : > New cards arent an option : > New laptops arent an option : > hostap is really what I need to work (although just orinoco would be nice : > too) writing David Hinds (maintainer of pcmcia) bounces. and no-reply. : > All the linux software compiles ok, configs seem ok, dmesg looks good. : > Cards do not work under linux :( : > : > Jamie : : _______________________________________________ : HostAP mailing list : HostAP at shmoo.com : http://lists.shmoo.com/mailman/listinfo/hostap -- No microsoft products were used to produce this message. EUG-LUG Mailing List: http://mailman.efn.org/cgi-bin/listinfo/eug-lug From jkmaline at cc.hut.fi Fri Jan 10 22:34:12 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Fri, 10 Jan 2003 19:34:12 -0800 Subject: CVS -C switch In-Reply-To: <1352.192.168.1.12.1042235913.squirrel@webmail.plumhall.com> References: <4024.192.168.1.12.1042181153.squirrel@webmail.plumhall.com> <20030110071022.GA3842@jm.kir.nu> <1352.192.168.1.12.1042235913.squirrel@webmail.plumhall.com> Message-ID: <20030111033412.GA2224@jm.kir.nu> On Fri, Jan 10, 2003 at 11:58:33AM -1000, jbreeden at plumhall.com wrote: > Tried a few more tings today, first updated my CVS to today and tried > connecting hostapd -C to a Dlink 900AP with a number of sta versions, this is > what I got: > sta 0.7.6 - connected > sta 1.3.6 - would not connect > sta 1.4.9 - would not connect > sta 1.5.6 - connected That's unfortunate.. I was using version 0.8.0 when experimenting with the code and did not notice any problems. Indeed, there are multiple issues with different firmware versions.. 0.8.0 seems to be OK, 1.4.9 does not let beacon frames through so hostapd does not even try to authenticate (if I force this, 1.4.9 will associate, but it will deauth on any receiveded frame), 1.5.6 associates and works for TX, but deauthenticates for every received frame. Btw, it is working for me also with version 0.7.6.. I added client code handlers for deauth and disassoc frames, so at least hostapd now re-associates with the same AP after deauth. However, this does not help much, since all received frames are dropped. Sending frames works fine, though. > What I mean by connected is the log message "Associated (as station) with AP > 00:05:5d:25:d5:db (aid=0)" after hostapd -C wlan0 is fired off. > as soon as the first data pdu (ping) issent, I get a deauthenticate an all > cases (0.7.6 and 1.5.6). Sending seems to work, but first received frame causes the firmware to send deauth frame. > handle_auth_cb: STA 00:05:5d:25:d5:db not found > > and the other is: > > unknown mgmt cb frame subtype 0 Yes, these are OK. I did not yet implement TX callback handlers for client operations. > Both of these messages show up connecting to both the dlink and hostap. Unique > to the dlink is the following: > Association (as station) with AP 00:05:5d:25:d5:db failed (status_code=10) > > followed by a successful association. That's quite odd; 10 = unsupported capabilities and capabilities should not change between association requests.. > You are right, it's a fireware issue. the first time I set it up it worked. I > was using cards with verious older sta versions, Some cards worked, others > didn't. In a flash of brillance I re-flashed them all to 1.4.9. Yep, you're > right - I don't know what those "older versions" where :-) duh... At least 0.6.2, 0.7.6, and 0.8.0 seem to work fine; everything else I tested seems to deauth on received frames. I have to admit that this is not exactly the most normal operation in Host AP mode, but I would not have expected the firmware to care about this or at least not to send deauthentication frames. I have to research this a bit more, but it may well be a firmware issue for which there is no workaround (well, other than using old 0.6.2-0.8.0 firmware versions).. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Fri Jan 10 22:52:07 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Fri, 10 Jan 2003 19:52:07 -0800 Subject: CVS -C switch In-Reply-To: <20030111033412.GA2224@jm.kir.nu> References: <4024.192.168.1.12.1042181153.squirrel@webmail.plumhall.com> <20030110071022.GA3842@jm.kir.nu> <1352.192.168.1.12.1042235913.squirrel@webmail.plumhall.com> <20030111033412.GA2224@jm.kir.nu> Message-ID: <20030111035207.GA2913@jm.kir.nu> On Fri, Jan 10, 2003 at 07:34:12PM -0800, Jouni Malinen wrote: > deauthentication frames. I have to research this a bit more, but it may > well be a firmware issue for which there is no workaround (well, other > than using old 0.6.2-0.8.0 firmware versions).. Luckily, I was wrong.. There seems to be a workaround for this. Setting promiscuous mode on removes unwanted deauthentications on 1.4.9 and 1.5.6. This can be configured with 'iwpriv wlan0 set_rid_word 64645 1'. This is somewhat odd, since promiscuous mode should not really allow frames from other BSSIDs, but then again, this station operations in Host AP mode is certainly undocumented "feature" ;-). If only the firmware source code would be available so that one would not need to guess and experiment with multiple firmware versions to figure out what might happen.. I haven't yet resolved problems with 1.4.9 not passing beacon frames through. Anyway, these could be skipped by just passing the SSID of the other AP and starting authentication attempts without waiting for beacons. -- Jouni Malinen PGP id EFC895FA From james.harper at bigpond.com Fri Jan 10 23:37:09 2003 From: james.harper at bigpond.com (James Harper) Date: Sat, 11 Jan 2003 15:37:09 +1100 Subject: hostap startup bug Message-ID: <3E1F9F75.3000606@bigpond.com> i posted a few weeks ago about a problem i was having where my dlink520 card would fail to start about half the time. i'm running 2.5.55 and was maybe thinking it was something to do with that, or maybe because i'm using an smp system. i patched the modules so that you can unload them under the latest kernels (hostap.c uses a deprecated module call and can't be unloaded) and did some experimentation. if i load the modules then it gets stuck, pretty much 80% of the time and fails at various points, normally by the time it tries to enable the mac port. the funny thing i noticed though, was that when i did an rmmod hostap_pci, it would unload the module, then reload it all by itself, and then everything would work fine thereafter. if i type 'ifup wlan0' it seems to work, if i type 'ifconfig wlan0' it falls flat on it's head. i think the interface is being brought up before the ifupdown scripts would normally bring it up. i noticed that if i comment out the wlan0 section in /etc/network/interfaces everything worked fine. it just _never_ fails to startup! then i found that if I do a 'pre-up' script of 'sleep 1' it also appears to work. my theory therefore is that ifupdown is trying to do something with the driver before it is properly initialised, if it is initialised by other means (ifconfig/modules being loaded). i noticed that some of the init code doesn't use the normal serialising methods so the problem could lie in there... i'm quite happy to spend time on this, i've already done a bit of patching to make it work under 2.5.55, but can anyone give me some hints on where to start? thanks James From raghu at sraamerica.com Sat Jan 11 01:10:09 2003 From: raghu at sraamerica.com (Raghu) Date: Sat, 11 Jan 2003 1:10:09 -0500 Subject: 802.1X Unicast Keys? Message-ID: <20030111061009.KHWC2474.imta01a2.registeredsite.com@smtp.registeredsite.com> > > Current Transmit Key: [2] > > Hmm.. If I remember correctly, transmit key should be 1 so that the > individual key would be used. Otherwise this looks OK. You should see > matching key1 in the AP with hostap_crypt_conf. > Thankyou very much, Jouni, This was my mistake. After I set the station's default key using iwconfig wlan0 key [1], It started pinging AP. Though I still need to resolve some more minor problems, atleast now it is working great. Once again thankyou very much for all your help. -Raghu From cijoml at volny.cz Sat Jan 11 04:28:39 2003 From: cijoml at volny.cz (Michal Semler) Date: Sat, 11 Jan 2003 10:28:39 +0100 Subject: What to do now? In-Reply-To: <200301101756.16893.linux@rocksolidnetworks.com> References: <200301100704.40609.linux@rocksolidnetworks.com> <200301101756.16893.linux@rocksolidnetworks.com> Message-ID: Easyest way is to try compile drivers and then modprobe them :)))) Michal Dne so 11. leden 2003 02:56 jste napsal(a): > Does that driver work with prism2 chipset? I thought it worked with > prism2.5 chipset. > I have a Proxim RangeLan-DS, Not a Proxim RangeLan2 card.. > Mine is model number 8433-05 (internal antenna) > > Ive looked but havnt found what chipset the RangeLan2 cards have... I > thought I read (was maybe 6 months ago...) that it was the Prism2.5 chip, > and figured the rlmod driver wouldnt work with the prism2 chip. > > PS, I tried the linux-wlan driver,but never got it to compile right... > > Jamie > > On Friday 10 January 2003 02:24 pm, Michal Semler wrote: > : Dne p? 10. leden 2003 16:04 jste napsal(a): > : > My cards (Proxim RangeLan-DS) still dont work under linux :( > : > : Proxim RangeLan drivers are here: > : > : http://www.komacke.com/distribution.html > : > : Michal > : > : > They do respond under linux (I can change modes, and see the lights do > : > different things...) but I dont have them doing anything usefull. > : > They Do work under windows95/98. (but ... windows doesnt do what I > : > need) They are 3.3v cards forced into 5v controllers. > : > New cards arent an option > : > New laptops arent an option > : > hostap is really what I need to work (although just orinoco would be > : > nice too) writing David Hinds (maintainer of pcmcia) bounces. and > : > no-reply. All the linux software compiles ok, configs seem ok, dmesg > : > looks good. Cards do not work under linux :( > : > > : > Jamie > : > : _______________________________________________ > : HostAP mailing list > : HostAP at shmoo.com > : http://lists.shmoo.com/mailman/listinfo/hostap From cijoml at volny.cz Sat Jan 11 04:38:10 2003 From: cijoml at volny.cz (Michal Semler) Date: Sat, 11 Jan 2003 10:38:10 +0100 Subject: still compiling problems ;( Message-ID: Hello, I have still compiling problem with today's hostap: Michal notas:/usr/src/wireless# cvs -z3 -d:pserver:anonymous at hostap.epitest.fi:/cvs co hostap ? hostap/driver/modules/hostap.ver ? hostap/driver/modules/hostap_crypt.ver cvs server: Updating hostap U hostap/Makefile cvs server: Updating hostap/driver cvs server: Updating hostap/driver/etc cvs server: Updating hostap/driver/modules U hostap/driver/modules/hostap.c U hostap/driver/modules/hostap_ap.c U hostap/driver/modules/hostap_compat.h U hostap/driver/modules/hostap_crypt_wep.c U hostap/driver/modules/hostap_hw.c U hostap/driver/modules/hostap_info.c cvs server: Updating hostap/hostapd U hostap/hostapd/hostapd.c U hostap/hostapd/hostapd.h U hostap/hostapd/ieee802_11.c cvs server: Updating hostap/inject cvs server: Updating hostap/kernel-patches U hostap/kernel-patches/hostap-linux-2.5.56.patch cvs server: Updating hostap/pcmcia-cs-patches cvs server: Updating hostap/sniff cvs server: Updating hostap/testing cvs server: Updating hostap/testing/wireless cvs server: Updating hostap/utils cvs server: Updating hostap/www notas:/usr/src/wireless# cd hostap notas:/usr/src/wireless/hostap# make clean mrm -f driver/modules/{hostap_cs.o,hostap_plx.o,hostap_pci.o} rm -f driver/modules/{hostap.o,hostap_crypt.o,hostap_crypt_wep.o} rm -f driver/modules/*.ver notas:/usr/src/wireless/hostap# make pci gcc -I/usr/src/linux/include -O2 -D__KERNEL__ -DMODULE -Wall -g -c -I/usr/src/linux/arch/i386/mach-generic -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -E -D__GENKSYMS__ driver/modules/hostap.c | \ /sbin/genksyms -k 2.4.20 > driver/modules/hostap.ver In file included from driver/modules/hostap.c:21: /usr/src/linux/include/linux/module.h:15:1: warning: "_set_ver" redefined /usr/src/linux/include/linux/modsetver.h:9:1: warning: this is the location of the previous definition gcc -I/usr/src/linux/include -O2 -D__KERNEL__ -DMODULE -Wall -g -c -I/usr/src/linux/arch/i386/mach-generic -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -E -D__GENKSYMS__ driver/modules/hostap_crypt.c | \ /sbin/genksyms -k 2.4.20 > driver/modules/hostap_crypt.ver In file included from driver/modules/hostap_crypt.c:17: /usr/src/linux/include/linux/module.h:15:1: warning: "_set_ver" redefined /usr/src/linux/include/linux/modsetver.h:9:1: warning: this is the location of the previous definition gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c -I/usr/src/linux/arch/i386/mach-generic -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o driver/modules/hostap_pci.o driver/modules/hostap_pci.c gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c -I/usr/src/linux/arch/i386/mach-generic -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o driver/modules/hostap.o driver/modules/hostap.c gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c -I/usr/src/linux/arch/i386/mach-generic -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o driver/modules/hostap_crypt.o driver/modules/hostap_crypt.c gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c -I/usr/src/linux/arch/i386/mach-generic -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o driver/modules/hostap_crypt_wep.o driver/modules/hostap_crypt_wep.c Run 'make install_pccard' as a root to install hostap_cs.o Run 'make install_plx' as a root to install hostap_plx.o Run 'make install_pci' as a root to install hostap_pci.o notas:/usr/src/wireless/hostap# make install_pci Installing hostap_crypt*.o to /lib/modules/2.4.20/net mkdir -p /lib/modules/2.4.20/net cp -f driver/modules/hostap_crypt*.o /lib/modules/2.4.20/net Installing hostap.o to /lib/modules/2.4.20/net mkdir -p /lib/modules/2.4.20/net cp -f driver/modules/hostap.o /lib/modules/2.4.20/net Installing hostap_pci.o to /lib/modules/2.4.20/net mkdir -p /lib/modules/2.4.20/net cp -f driver/modules/hostap_pci.o /lib/modules/2.4.20/net /sbin/depmod -a depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o make: *** [install_pci] Error 1 From earl at stanfordalumni.org Sat Jan 11 06:46:10 2003 From: earl at stanfordalumni.org (earl) Date: Sat, 11 Jan 2003 03:46:10 -0800 Subject: Tx timed out! Resetting card (with OpenAP) References: <3E1BF9C4.4143A1F9@stanfordalumni.org> <007901c2b702$e52e9c90$53153b0a@nednet.co.za> <3E1C0812.36DF29D3@stanfordalumni.org> <008401c2b70d$61255690$53153b0a@nednet.co.za> <3E1E96D7.438445F5@stanfordalumni.org> Message-ID: <3E200401.D46D8863@stanfordalumni.org> earl wrote: > I've only tried this so far in the case of > wds_type=2 and *not* the case of doing the volatile firmware update with > wds_type=4. (I don't understand whether a card-reset will cause the volatile > firmware update to be lost, causing the card to revert to its permanent firmware > --- anyone know what to expect?) After experimenting with this, i've found that an automatic card reset which would have been tolerated in the first case (wds_type=2, without volatile firmware update) will kill the link in the second case (wds_type=4, with volatile firmware update). I'm not going to investigate further right now but it looks like the card reset causes a volatile firmware update to be lost. After the reset, if i set wds_type to 2, then the link will work again, so presumably it's gone back to the permanent firmware which can't handle wds_type=4. Regarding my previous message (about recoverable and not-recoverable card-reset events), i notice that when the link is truly dead the output of iwconfig for that interface is missing the two lines that look something like this: Bit Rate:1Mb/s Tx-Power:1 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Since a dead wlan0 is fatal for this particular system (its only purpose is to be an AP), i have a dumb background script that checks iwconfig's output for that condition (iwconfig wlan0 | grep Rate) every once in a while, and if wlan0 is dead will reboot the system as a last resort. Yes it's crude, but it's better than letting an AP languish in that useless state. I'd prefer to just unload-reload the modules but that doesn't seem to be working. earl From jwickman at tawi.fi Sat Jan 11 06:45:06 2003 From: jwickman at tawi.fi (Joachim Wickman) Date: Sat, 11 Jan 2003 13:45:06 +0200 Subject: still compiling problems ;( References: Message-ID: <001e01c2b968$e561e290$010aa8c0@vasa> Hello, I had the same problem too... I don't know why, but after compiling the whole kernel, pcmcia-cs and hostap packages 3-4 times with several reboots between every stage and tadaa.. It worked. Strange thing.. had same problem with kernel-2.4.19, but the others played at the first try. // Joachim ----- Original Message ----- From: "Michal Semler" To: Sent: Saturday, January 11, 2003 11:38 AM Subject: still compiling problems ;( > Hello, > I have still compiling problem with today's hostap: > > Michal > > notas:/usr/src/wireless# cvs -z3 -d:pserver:anonymous at hostap.epitest.fi:/cvs > co hostap > ? hostap/driver/modules/hostap.ver > ? hostap/driver/modules/hostap_crypt.ver > cvs server: Updating hostap > U hostap/Makefile > cvs server: Updating hostap/driver > cvs server: Updating hostap/driver/etc > cvs server: Updating hostap/driver/modules > U hostap/driver/modules/hostap.c > U hostap/driver/modules/hostap_ap.c > U hostap/driver/modules/hostap_compat.h > U hostap/driver/modules/hostap_crypt_wep.c > U hostap/driver/modules/hostap_hw.c > U hostap/driver/modules/hostap_info.c > cvs server: Updating hostap/hostapd > U hostap/hostapd/hostapd.c > U hostap/hostapd/hostapd.h > U hostap/hostapd/ieee802_11.c > cvs server: Updating hostap/inject > cvs server: Updating hostap/kernel-patches > U hostap/kernel-patches/hostap-linux-2.5.56.patch > cvs server: Updating hostap/pcmcia-cs-patches > cvs server: Updating hostap/sniff > cvs server: Updating hostap/testing > cvs server: Updating hostap/testing/wireless > cvs server: Updating hostap/utils > cvs server: Updating hostap/www > notas:/usr/src/wireless# cd hostap > notas:/usr/src/wireless/hostap# make clean > mrm -f driver/modules/{hostap_cs.o,hostap_plx.o,hostap_pci.o} > rm -f driver/modules/{hostap.o,hostap_crypt.o,hostap_crypt_wep.o} > rm -f driver/modules/*.ver > notas:/usr/src/wireless/hostap# make pci > gcc -I/usr/src/linux/include -O2 -D__KERNEL__ -DMODULE -Wall -g -c > -I/usr/src/linux/arch/i386/mach-generic > -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include > /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -E > -D__GENKSYMS__ driver/modules/hostap.c | \ > /sbin/genksyms -k 2.4.20 > driver/modules/hostap.ver > In file included from driver/modules/hostap.c:21: > /usr/src/linux/include/linux/module.h:15:1: warning: "_set_ver" redefined > /usr/src/linux/include/linux/modsetver.h:9:1: warning: this is the location > of the previous definition > gcc -I/usr/src/linux/include -O2 -D__KERNEL__ -DMODULE -Wall -g -c > -I/usr/src/linux/arch/i386/mach-generic > -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include > /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -E > -D__GENKSYMS__ driver/modules/hostap_crypt.c | \ > /sbin/genksyms -k 2.4.20 > driver/modules/hostap_crypt.ver > In file included from driver/modules/hostap_crypt.c:17: > /usr/src/linux/include/linux/module.h:15:1: warning: "_set_ver" redefined > /usr/src/linux/include/linux/modsetver.h:9:1: warning: this is the location > of the previous definition > gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver > -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c > -I/usr/src/linux/arch/i386/mach-generic > -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include > /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o > driver/modules/hostap_pci.o driver/modules/hostap_pci.c > gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver > -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c > -I/usr/src/linux/arch/i386/mach-generic > -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include > /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o > driver/modules/hostap.o driver/modules/hostap.c > gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver > -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c > -I/usr/src/linux/arch/i386/mach-generic > -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS -include > /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o > driver/modules/hostap_crypt.o driver/modules/hostap_crypt.c > gcc -I/usr/src/linux/include -include driver/modules/hostap_crypt.ver > -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -g -c > -I/usr/src/linux/arch/i386/mach-generic > -I/usr/src/linux/include/asm/mach-default -DMODVERSIONS > -include /usr/src/linux/include/linux/modversions.h -fomit-frame-pointer -o > driver/modules/hostap_crypt_wep.o driver/modules/hostap_crypt_wep.c > > Run 'make install_pccard' as a root to install hostap_cs.o > Run 'make install_plx' as a root to install hostap_plx.o > Run 'make install_pci' as a root to install hostap_pci.o > > notas:/usr/src/wireless/hostap# make install_pci > Installing hostap_crypt*.o to /lib/modules/2.4.20/net > mkdir -p /lib/modules/2.4.20/net > cp -f driver/modules/hostap_crypt*.o /lib/modules/2.4.20/net > Installing hostap.o to /lib/modules/2.4.20/net > mkdir -p /lib/modules/2.4.20/net > cp -f driver/modules/hostap.o /lib/modules/2.4.20/net > Installing hostap_pci.o to /lib/modules/2.4.20/net > mkdir -p /lib/modules/2.4.20/net > cp -f driver/modules/hostap_pci.o /lib/modules/2.4.20/net > /sbin/depmod -a > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o > make: *** [install_pci] Error 1 > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > > From blaze at sh.pl Sat Jan 11 08:37:45 2003 From: blaze at sh.pl (=?iso-8859-2?Q?Rados=B3aw_Micho=F1?=) Date: Sat, 11 Jan 2003 14:37:45 +0100 Subject: WDS Problem ;( Message-ID: <005501c2b976$a2b97da0$48aea8c0@platix> I'am using latest CVS hostap on my linux servers (both linux 2.4.20) with two d-link's 520 (f/w: 1.4.9). My network looks like that : network1-[eth0]-Server1-[wlan0]**air**[wlan0]-Server2-[eth0]-network2 When im using hostap in ad-hoc mode, without any bridge, the transfer is about 650 kb/s and it looks that everything is fine. But when hostap is being reconfigured to use Menaged mode with WDS system the transfer hooks on 2Mbps (180 kb/s). I to set the 'prism2_param wlan0 other_ap_policy 2' option on both sides, but nothing happend, transfer was the same. Is there any chance to edit the sources and force WDS to use 11MBPS ? Server configuration: root at netium:/proc/net/hostap/wlan0# cat ap_debug BridgedUnicastFrames=0 BridgedMulticastFrames=0 max_inactivity=300 bridge_packets=0 nullfunc_ack=0 auth_algs=3 ap_queued_skb_len=0 root at netium:/proc/net/hostap/wlan0# cat debug next_txfid=7 next_alloc=7 FID: tx=0127 intransmit=FFFF FID: tx=013C intransmit=FFFF FID: tx=0151 intransmit=FFFF FID: tx=0166 intransmit=FFFF FID: tx=017B intransmit=FFFF FID: tx=0190 intransmit=FFFF FID: tx=01A5 intransmit=FFFF FID: tx=01BA intransmit=FFFF FW TX rate control: 0 beacon_int=1000 dtim_period=1 wds_max_connections=16 dev_enabled=1 Cheers. --- Radoslaw Michon From gbritton at doomcom.org Sat Jan 11 11:56:21 2003 From: gbritton at doomcom.org (Gerald Britton) Date: Sat, 11 Jan 2003 11:56:21 -0500 Subject: still compiling problems ;( In-Reply-To: ; from cijoml@volny.cz on Sat, Jan 11, 2003 at 10:38:10AM +0100 References: Message-ID: <20030111115621.A22382@nevermore.toe.doomcom.org> On Sat, Jan 11, 2003 at 10:38:10AM +0100, Michal Semler wrote: > /sbin/depmod -a > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o > make: *** [install_pci] Error 1 if you could run "/sbin/depmod -ae" it would tell which symbols were undefined and would probably help others figure out where the actual problem is. -- Gerald From cijoml at volny.cz Sat Jan 11 12:27:34 2003 From: cijoml at volny.cz (Michal Semler) Date: Sat, 11 Jan 2003 18:27:34 +0100 Subject: still compiling problems ;( In-Reply-To: <20030111115621.A22382@nevermore.toe.doomcom.org> References: <20030111115621.A22382@nevermore.toe.doomcom.org> Message-ID: Liboc01:/usr/src/hostap# depmod -ae depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o depmod: wireless_send_event_R9e8b822b depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o depmod: wireless_send_event_R9e8b822b with 2.4.21-pre3 worked everything well...but on server I can't have pre kernel... Michal Dne so 11. leden 2003 17:56 jste napsal(a): > On Sat, Jan 11, 2003 at 10:38:10AM +0100, Michal Semler wrote: > > /sbin/depmod -a > > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o > > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o > > make: *** [install_pci] Error 1 > > if you could run "/sbin/depmod -ae" it would tell which symbols were > undefined and would probably help others figure out where the actual > problem is. > > -- Gerald > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap From jkmaline at cc.hut.fi Sat Jan 11 13:39:03 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sat, 11 Jan 2003 10:39:03 -0800 Subject: hostap startup bug In-Reply-To: <3E1F9F75.3000606@bigpond.com> References: <3E1F9F75.3000606@bigpond.com> Message-ID: <20030111183903.GA8935@jm.kir.nu> On Sat, Jan 11, 2003 at 03:37:09PM +1100, James Harper wrote: > i posted a few weeks ago about a problem i was having where my dlink520 > card would fail to start about half the time. > > i'm running 2.5.55 and was maybe thinking it was something to do with > that, or maybe because i'm using an smp system. > > i patched the modules so that you can unload them under the latest > kernels (hostap.c uses a deprecated module call and can't be unloaded) > and did some experimentation. Current CVS version should have this fixed. At least it loads and unloads fine with Linux 2.5.56. However, I haven't been able to test this completely since generic pcmcia code in 2.5.56 does not work with my laptop. In addition, I added a new patch that can be used with 2.5.56. I haven't yet changed module compilation, so latest 2.5.x kernel versions need to use in-kernel-tree compilation. > my theory therefore is that ifupdown is trying to do something with the > driver before it is properly initialised, if it is initialised by other > means (ifconfig/modules being loaded). i noticed that some of the init > code doesn't use the normal serialising methods so the problem could lie > in there... Yes, this seems to be problem. I'm usually just loading the module manually so I do not hit such cases at all. I tried to kill initialization by changing configuration as soon as possible, but that did not seem to break anything with PC Card version. I'll try similar things with hostap_pci.o later. I don't have SMP system easily available for testing now, so I won't test that. I'm using pre-emptive kernel, but yes, this should also be tested on SMP since it could allow races with the initialization code more easily. Host AP driver registers netdevice before it has finished initializing the card. This makes it possible for user space programs trying to touch the interface before everything is ready and might cause these somewhat random errors during initialization. I have to check whether register_netdev() could be moved so that it would be performed only after everything has been initialized. That should remove these kinds of races. > i'm quite happy to spend time on this, i've already done a bit of > patching to make it work under 2.5.55, but can anyone give me some hints > on where to start? Do you have something else patched than the changes I did for the current CVS version to load with 2.5.56? I think that at least one problem is in allowing user space to interract with the driver during initialization and fixing it cleanly might require reordering something in the initialization. -- Jouni Malinen PGP id EFC895FA From ammirata at econointl.com Sat Jan 11 14:45:17 2003 From: ammirata at econointl.com (Sergio M. Ammirata) Date: Sat, 11 Jan 2003 14:45:17 -0500 Subject: hostapd forking and enhancements In-Reply-To: <20030108030233.GA1134@jm.kir.nu> Message-ID: <000001c2b9a9$f8e4a1e0$017ba8c0@ammirata.net> Thanks for adding the forking option. I will try it out today. Here is another PAP radius capture using -s 512. I think this time it has all the data. As far as using a standard 802.11 encoding instead of PAP, I don't particularly have a preference as long as we can use it quickly :) As you can see on the captures it is extremely simple to implement PAP ... And on the server side you can leverage the user name lists already available in radius servers to hold the station MAC addresses. Thanks in advance, Sergio > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] On Behalf Of > Jouni Malinen > Sent: Tuesday, January 07, 2003 10:03 PM > To: hostap at shmoo.com > Subject: Re: hostapd forking and enhancements > > On Tue, Jan 07, 2003 at 02:03:09AM -0500, Sergio Ammirata wrote: > > > I think the forking is rather critical since it prevents people from > > deploying the software in a production environment. Specially if the are > > using it in a custom distro with limited tools. > > I added it yesterday (-B argument for hostapd). > > > I like the idea of RADIUS. Perhaps we can use PAP authentication and > > send the station MAC in both the username and password attributes. > > I would like to an encoding that is used by other 802.11 APs (I would > assume that there exists one supporting this), but yes, it could be > something like that. > > > I have attached a tcpdump capture of a PAP radius communication between > > a pppd daemon and a Microsoft IAS Radius Server. It includes connect and > > disconnect (use tcpdump -r filename to read). > > The snaplength (-s option for tcpdump) seemed to be too short and the > packets were truncated so that the interesting part of the attributes > was not included. > > -- > Jouni Malinen PGP id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap -------------- next part -------------- A non-text attachment was scrubbed... Name: success.log Type: application/octet-stream Size: 798 bytes Desc: not available Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030111/fdcc2de8/attachment.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: failure.log Type: application/octet-stream Size: 233 bytes Desc: not available Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030111/fdcc2de8/attachment-0001.obj From jkmaline at cc.hut.fi Sat Jan 11 15:17:56 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sat, 11 Jan 2003 12:17:56 -0800 Subject: hostap startup bug In-Reply-To: <20030111183903.GA8935@jm.kir.nu> References: <3E1F9F75.3000606@bigpond.com> <20030111183903.GA8935@jm.kir.nu> Message-ID: <20030111201756.GA1097@jm.kir.nu> On Sat, Jan 11, 2003 at 10:39:03AM -0800, Jouni Malinen wrote: > unloads fine with Linux 2.5.56. However, I haven't been able to test > this completely since generic pcmcia code in 2.5.56 does not work with > my laptop. Actually, it does work; one just needs to remember not to remove ISA support from the kernel while using 16-bit PC Cards.. ;-) So, it looks like current CVS version of Host AP driver works fine with Linux 2.5.56 (as a module compiled within kernel tree). -- Jouni Malinen PGP id EFC895FA From linux at rocksolidnetworks.com Sat Jan 11 15:39:07 2003 From: linux at rocksolidnetworks.com (Linux Rocks!) Date: Sat, 11 Jan 2003 12:39:07 -0800 Subject: What to do now? In-Reply-To: References: <200301100704.40609.linux@rocksolidnetworks.com> <200301101756.16893.linux@rocksolidnetworks.com> Message-ID: <200301111239.07711.linux@rocksolidnetworks.com> well... So far compiling rlmod has been nothing but frustration... It sure would be a lot nicer to use the hostap module (especially since It has compiled and loaded w/out errors, and is known to do what I want (I want to make an ap...) Jamie On Saturday 11 January 2003 01:28 am, Michal Semler wrote: : Easyest way is to try compile drivers and then modprobe them :)))) : : Michal : : Dne so 11. leden 2003 02:56 jste napsal(a): : > Does that driver work with prism2 chipset? I thought it worked with : > prism2.5 chipset. : > I have a Proxim RangeLan-DS, Not a Proxim RangeLan2 card.. : > Mine is model number 8433-05 (internal antenna) : > : > Ive looked but havnt found what chipset the RangeLan2 cards have... I : > thought I read (was maybe 6 months ago...) that it was the Prism2.5 chip, : > and figured the rlmod driver wouldnt work with the prism2 chip. : > : > PS, I tried the linux-wlan driver,but never got it to compile right... : > : > Jamie : > : > On Friday 10 January 2003 02:24 pm, Michal Semler wrote: : > : Dne p? 10. leden 2003 16:04 jste napsal(a): : > : > My cards (Proxim RangeLan-DS) still dont work under linux :( : > : : > : Proxim RangeLan drivers are here: : > : : > : http://www.komacke.com/distribution.html : > : : > : Michal : > : : > : > They do respond under linux (I can change modes, and see the lights : > : > do different things...) but I dont have them doing anything usefull. : > : > They Do work under windows95/98. (but ... windows doesnt do what I : > : > need) They are 3.3v cards forced into 5v controllers. : > : > New cards arent an option : > : > New laptops arent an option : > : > hostap is really what I need to work (although just orinoco would be : > : > nice too) writing David Hinds (maintainer of pcmcia) bounces. and : > : > no-reply. All the linux software compiles ok, configs seem ok, dmesg : > : > looks good. Cards do not work under linux :( : > : > : > : > Jamie : > : : > : _______________________________________________ : > : HostAP mailing list : > : HostAP at shmoo.com : > : http://lists.shmoo.com/mailman/listinfo/hostap : : _______________________________________________ : HostAP mailing list : HostAP at shmoo.com : http://lists.shmoo.com/mailman/listinfo/hostap -- No microsoft products were used to produce this message. EUG-LUG Mailing List: http://mailman.efn.org/cgi-bin/listinfo/eug-lug From jkmaline at cc.hut.fi Sat Jan 11 16:22:55 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sat, 11 Jan 2003 13:22:55 -0800 Subject: What to do now? In-Reply-To: <200301100704.40609.linux@rocksolidnetworks.com> References: <200301100704.40609.linux@rocksolidnetworks.com> Message-ID: <20030111212254.GA1404@jm.kir.nu> On Fri, Jan 10, 2003 at 07:04:40AM -0800, Linux Rocks! wrote: > My cards (Proxim RangeLan-DS) still dont work under linux :( > They do respond under linux (I can change modes, and see the lights do > different things...) but I dont have them doing anything usefull. > They Do work under windows95/98. (but ... windows doesnt do what I need) Have you tested running one card in Linux as an AP (i.e., using Host AP driver in Master mode) and another card in Windows as a client that tries to associate with the AP? I haven't noticed anything in your dmesg logs etc. that would indicate cause for problems, so this is a bit difficult to solve. I would suggest setting up one Host AP, make sure it is in Master mode and that the netdevice is UP. After this, try to associate with it using another laptop with Windows (that has been tested to work with another AP). If this does not work, please send me output of following commands from the laptop running Host AP driver (run these after the failed test, before having changed anything in the Host AP configuration): dmesg cat /proc/interrupts iwconfig wlan0 ifconfig -a hostap_diag wlan0 -r wlan0 (hostap_diag is in the utils directory of the Host AP driver package). I would also ask you to repeat the test with latest CVS version of the Host AP driver if you are still using the latest release (2002-10-12). -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Sat Jan 11 16:27:33 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sat, 11 Jan 2003 13:27:33 -0800 Subject: still compiling problems ;( In-Reply-To: References: <20030111115621.A22382@nevermore.toe.doomcom.org> Message-ID: <20030111212733.GB1404@jm.kir.nu> On Sat, Jan 11, 2003 at 06:27:34PM +0100, Michal Semler wrote: > Liboc01:/usr/src/hostap# depmod -ae > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o > depmod: wireless_send_event_R9e8b822b Linux 2.4.20 should have wireless_send_event().. Please check: grep wireless_send_event /proc/ksyms grep wireless_send_event /usr/src/linux/include/linux/modules/netsyms.ver (these should show the avbove mentioned symbol with matching 32-bit number if you have identical configuration for both Linux and Host AP driver compilations) -- Jouni Malinen PGP id EFC895FA From james.harper at bigpond.com Sat Jan 11 18:46:01 2003 From: james.harper at bigpond.com (James Harper) Date: Sun, 12 Jan 2003 10:46:01 +1100 Subject: hostap startup bug In-Reply-To: <20030111183903.GA8935@jm.kir.nu> Message-ID: <000001c2b9cb$98bf2ac0$c702a8c0@int.sbss.com.au> I've compiled it... But I'm using my laptop at the moment so I can't test it. I'm using smp and pre-emptive kernel. Do you use ifupdown? If you do, try bringing up the interface with ifconfig, on my system that invokes the ifupdown script as well before the card is really ready for it. My machine is an ABIT BP6 with 2x Celeron 400 (not overclocked). A faster or slower machine might behave differently wrt how the init timings work together. I seem to be having much success with the line pre-up sleep in my /etc/network/interfaces file. This makes it wait for a second before it does anything to the card. Obviously this isn't ideal but it works. I'll give it a good test tonite and let you know how it went. Thanks James > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] > On Behalf Of Jouni Malinen > Sent: Sunday, 12 January 2003 05:39 > To: hostap at shmoo.com > Subject: Re: hostap startup bug > > > On Sat, Jan 11, 2003 at 03:37:09PM +1100, James Harper wrote: > > > i posted a few weeks ago about a problem i was having where my > > dlink520 > > card would fail to start about half the time. > > > > i'm running 2.5.55 and was maybe thinking it was something > to do with > > that, or maybe because i'm using an smp system. > > > > i patched the modules so that you can unload them under the latest > > kernels (hostap.c uses a deprecated module call and can't > be unloaded) > > and did some experimentation. > > Current CVS version should have this fixed. At least it loads > and unloads fine with Linux 2.5.56. However, I haven't been > able to test this completely since generic pcmcia code in > 2.5.56 does not work with my laptop. In addition, I added a > new patch that can be used with 2.5.56. I haven't yet changed > module compilation, so latest 2.5.x kernel versions need to > use in-kernel-tree compilation. > > > my theory therefore is that ifupdown is trying to do something with > > the > > driver before it is properly initialised, if it is > initialised by other > > means (ifconfig/modules being loaded). i noticed that some > of the init > > code doesn't use the normal serialising methods so the > problem could lie > > in there... > > Yes, this seems to be problem. I'm usually just loading the > module manually so I do not hit such cases at all. I tried to > kill initialization by changing configuration as soon as > possible, but that did not seem to break anything with PC > Card version. I'll try similar things with hostap_pci.o > later. I don't have SMP system easily available for testing > now, so I won't test that. I'm using pre-emptive kernel, but > yes, this should also be tested on SMP since it could allow > races with the initialization code more easily. > > Host AP driver registers netdevice before it has finished > initializing the card. This makes it possible for user space > programs trying to touch the interface before everything is > ready and might cause these somewhat random errors during > initialization. I have to check whether > register_netdev() could be moved so that it would be > performed only after everything has been initialized. That > should remove these kinds of races. > > > i'm quite happy to spend time on this, i've already done a bit of > > patching to make it work under 2.5.55, but can anyone give > me some hints > > on where to start? > > Do you have something else patched than the changes I did for > the current CVS version to load with 2.5.56? I think that at > least one problem is in allowing user space to interract with > the driver during initialization and fixing it cleanly might > require reordering something in the initialization. > > -- > Jouni Malinen PGP > id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > > From linux at rocksolidnetworks.com Sat Jan 11 19:34:17 2003 From: linux at rocksolidnetworks.com (Linux Rocks!) Date: Sat, 11 Jan 2003 16:34:17 -0800 Subject: Fwd: Re: What to do now? Message-ID: <200301111634.17229.linux@rocksolidnetworks.com> Ooops... used the wrong email address! ---------- Forwarded Message ---------- Subject: Re: What to do now? Date: Sat, 11 Jan 2003 16:28:20 -0800 From: Jamie To: Jouni Malinen , hostap at shmoo.com Jouni, Thanks for asking, This message is REALLY long, with a lot of info, I tried to be as breif as possible, without omitting anything necessary. On Saturday 11 January 2003 01:22 pm, Jouni Malinen wrote: : On Fri, Jan 10, 2003 at 07:04:40AM -0800, Linux Rocks! wrote: : > My cards (Proxim RangeLan-DS) still dont work under linux :( : > They do respond under linux (I can change modes, and see the lights do : > different things...) but I dont have them doing anything usefull. : > They Do work under windows95/98. (but ... windows doesnt do what I need) Yes (assuming all the things I trierd were correct!) Ive tested the laptops in many situations. Ive set 2 laptop up in windows, had success. tested the laptop with a real ap under windows, had success. tested the hostap, and the orinoco modules with the ap w/no success. I think Ive tried 1 laptop in windows, and one in linux, using all the various possible combonations of settings, with no success. : Have you tested running one card in Linux as an AP (i.e., using Host AP : driver in Master mode) and another card in Windows as a client that : tries to associate with the AP? I haven't noticed anything in your dmesg : logs etc. that would indicate cause for problems, so this is a bit : difficult to solve. Yes! I agree... atleast when I had errors... I had something to work on... Ive just setup the cards with one windows box, and one linux box. windows98 has these settings: Mode: Infrastructure (this is also known as managed right?) SSID: ANY (This is supposed to scan...), Ive also tried jamienet EtherNEt Conversion: RCF 1042 TXRate: Fully Automatic WEP: Disabled PS Mode: Disabled (ive also tried enabled) (This I think is what worked with the AP at the LUG) Linux box: (running hostap_cs, using lastest 2002-10-12): (I have 3 different kernels/modules on the machine, but Im currently using 2.4.20-rc-3) Ive editied my /etc/pcmcia/config to add: source ./hostap_cs.conf and Ive added in my /etc/pcmcia/hostap_cs.conf card "JamieTest PRISM RangeLAN-DS/LAN PC CARD" manfid 0x0126, 0x8000 bind "hostap_cs" # Optional configuration parameters for hostap_cs.o # module "hostap_cs" opts "channel=6 essid=any ignore_cis_vcc=1" module "hostap_cs" opts "channel=3 iw_mode=3 essid=jamienet ignore_cis_vcc=1" (this sets the card to Master Mode, use essid jamietest, and ignore the voltage issue (I do get errors with it set to 0) Heres my DMESG: cs: IO port probe 0x0a00-0x0aff: excluding 0xa68-0xa6f 0xaa0-0xaa7 0xaf8-0xaff cs: memory probe 0x0d0000-0x0dffff: clean. hostap_crypt: registered algorithm 'NULL' hostap_cs: hostap_cs.c 0.0.0 2002-10-12 (SSH Communications Security Corp, Jouni Malinen) hostap_cs: (c) Jouni Malinen hostap_cs: setting Vcc=33 (constant) hostap_cs: CS_EVENT_CARD_INSERTION prism2_config() hostap_cs: ignoring Vcc=50 (from config) Checking CFTABLE_ENTRY 0x01 (default 0x01) IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 io->flags = 0x0046, io.base=0x0000, len=64 hostap_cs: index 0x01: Vcc 5.0, irq 3, io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 prism2_hw_init() prism2_hw_config: initialized in 9094 iterations wlan0: NIC: id=0x8002 v1.0.0 wlan0: PRI: id=0x15 v0.3.0 wlan0: STA: id=0x1f v0.8.3 wlan0: defaulting to host-based encryption as a workaround for firmware bug in Host AP mode WEP wlan0: LinkStatus=2 (Disconnected) cat /proc/Interrupts bubba at lapdance:~$ cat /proc/interrupts CPU0 0: 733578 XT-PIC timer 1: 302 XT-PIC keyboard 2: 0 XT-PIC cascade 3: 135 XT-PIC hostap_cs 10: 563 XT-PIC NE2000 12: 1 XT-PIC PS/2 Mouse 14: 1773 XT-PIC ide0 NMI: 0 ERR: 0 root at lapdance:~# iwconfig wlan0 wlan0 IEEE 802.11-b ESSID:"jamienet" Mode:Master Frequency:2.422GHz Access Point: 00:60:B3:6F:63:CC Bit Rate:11Mb/s Tx-Power:-12 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:0 Signal level:0 Noise level:0 Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 ifconfig -a: root at lapdance:~# ifconfig -a eth0 Link encap:Ethernet HWaddr 00:A0:D1:01:0D:86 inet addr:192.168.14.20 Bcast:192.168.14.255 Mask:255.255.255.0 UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1 RX packets:394 errors:0 dropped:0 overruns:0 frame:0 TX packets:285 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:35154 (34.3 Kb) TX bytes:144579 (141.1 Kb) Interrupt:10 Base address:0x300 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) wlan0 Link encap:Ethernet HWaddr 00:60:B3:6F:63:CC BROADCAST MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) Interrupt:3 Base address:0x100 I hadnt built the hostap utils... so I cd'd to the directory, and did make. it made the utils! I tried : hostap_diag wlan0 -r wlan0, but that failed. so I tried: ./hostap_diag -r wlan0 and got the following(Very long!) root at lapdance:~/wirless/hostap-2002-10-12/utils# ./hostap_diag -r wlan0 Host AP driver diagnostics information for 'wlan0' NICID: id=0x8002 v1.0.0 (HWB3163-01,02,03,04 Rev A) PRIID: id=0x0015 v0.3.0 STAID: id=0x001f v0.8.3 (station firmware) Known RIDs (Resource IDentifiers) FC00=CNFPORTTYPE=6 FC01=CNFOWNMACADDR=00:60:b3:6f:63:cc FC02=CNFDESIREDSSID=jamienet FC03=CNFOWNCHANNEL=3 FC04=CNFOWNSSID=jamienet FC05=CNFOWNATIMWINDOW=0 FC06=CNFSYSTEMSCALE=1 FC07=CNFMAXDATALEN=2304 FC08=CNFWDSADDRESS=00:00:00:00:00:00 FC09=CNFPMENABLED=0 FC0A=CNFPMEPS=0 FC0B=CNFMULTICASTRECEIVE=1 FC0C=CNFMAXSLEEPDURATION=100 FC0D=CNFPMHOLDOVERDURATION=100 FC0E=CNFOWNNAME= FC10=CNFOWNDTIMPERIOD=1 FC20=UNKNOWN1=33 FC21=UNKNOWN2=33 FC23=CNFWEPDEFAULTKEYID=0 FC24=CNFDEFAULTKEY0=<21><00><24><00><00> FC25=CNFDEFAULTKEY1=<21><00><25><00><00> FC26=CNFDEFAULTKEY2=<21><00><26><00><00> FC27=CNFDEFAULTKEY3=<21><00><27><00><00> FC28=CNFWEPFLAGS=<00><00> FC2A=CNFAUTHENTICATION=1 FC2C=CNFTXCONTROL=512 FC2D=CNFROAMINGMODE=1 FC30=CNFRCVCRCERROR=0 FC31=CNFMMLIFE=65000 FC32=CNFALTRETRYCOUNT=7 FC33=CNFBEACONINT=100 FC35=CNFSTAPCFINFO=<01><00> FC40=CNFTIMCTRL=1 FC41=UNKNOWN3=<00><00><00><00><00><00> FC42=CNFTHIRTY2TALLY=0 FC80=GROUPADDRESSES= FC81=CREATEIBSS=1 FC82=FRAGMENTATIONTHRESHOLD=2346 FC83=RTSTHRESHOLD=2347 FC84=TXRATECONTROL=15 FC85=PROMISCUOUSMODE=0 FCB0=CNFSHORTPREAMBLE=0 FCB3=CNFBASICRATES=<03><00> FCB4=CNFSUPPORTEDRATES=<0f><00> FCE0=TICKTIME=0 FCE1=SCANREQUEST=<3f><20><00> FCE2=JOINREQUEST=<21><00><00><00><00><00> FD00=MAXLOADTIME=60000 FD01=DOWNLOADBUFFER=<60><00><00><00><00><10> FD02=PRIID=0x15 v0.3.0 FD03=PRISUPRANGE=0 0x03 1 1-2 FD04=CFIACTRANGES=1 0x02 1 1-1 FD0A=NICSERNUM=SA01000000 FD0B=NICID=0x8002 v1.0.0 FD0C=MFISUPRANGE=0 0x01 1 1-1 FD0D=CFISUPRANGE=0 0x02 1 1-1 FD10=CHANNELLIST=<07> FD11=REGULATORYDOMAINS=<10> FD12=TEMPTYPE=1 FD13=CIS=<01><03><00><00><17><04><67><5a><08><1d><05><03><67><5a><08> <1c><04><03><00><00><15><23><05><00><50><52><4f><58><49><4d><00><52>< 61><6e><67><65><4c><41><4e><2d><44><53><2f><4c><41><4e><20><50><43><20><43><4 1><52><44><00><00><20><04><26><01><00><80><21><02><06><00><22><02><01><07 ><22><05><02><40><42><0f><00><22><05><02><80><84><1e><00><22><02><03><07><22> <02><05><01><1a><07><03><02><03><00><00><01><1b><10><01><19><77>< 1e><35><3c><36><36><05><46><7f><00><00><00> + 352* FD20=STAID=0x1f v0.8.3 FD21=STASUPRANGE=0 0x04 1 1-6 FD22=MFIACTRANGES=1 0x01 1 1-1 FD23=CFIACTRANGES2=1 0x02 1 1-1 FD40=PORTSTATUS=8 FD41=CURRENTSSID=jamienet FD42=CURRENTBSSID=00:60:b3:6f:63:cc FD43=COMMSQUALITY=<00><00><1b><00><1b><00> FD44=CURRENTTXRATE=2 FD45=CURRENTBEACONINTERVAL=100 FD46=CURRENTSCALETHRESHOLDS=<36><00><36><00><36><00><0a><00><02><00><06><00> FD47=PROTOCOLRSPTIME=100 FD48=SHORTRETRYLIMIT=8 FD49=LONGRETRYLIMIT=5 FD4A=MAXTRANSMITLIFETIME=512 FD4B=MAXRECEIVELIFETIME=512 FD4C=CFPOLLABLE=0 FD4D=AUTHENTICATIONALGORITHMS=<01><00><01><00> FD4F=PRIVACYOPTIONIMPLEMENTED=1 FD88=SCANRESULTSTABLE=<01><00><03><00> FDC0=PHYTYPE=2 FDC1=CURRENTCHANNEL=3 FDC2=CURRENTPOWERSTATE=1 FDC3=CCAMODE=2 FDC6=SUPPORTEDDATARATES=<02><04><0b><16> FFFE=BUILDSEQ=<00><00><00><00> FFFF=FWID=0003C0.HEX<00><00><00> WOW Thats a lot of info! I have no idea what all of that means, but Id say that it suggests somethings working right! (hopefully you will understand what isnt working right! Ill try to get the current cvs, could you remind me how/where? Thanks in advance for help. Jamie : I would suggest setting up one Host AP, make sure it is in Master mode : and that the netdevice is UP. After this, try to associate with it using : another laptop with Windows (that has been tested to work with another : AP). If this does not work, please send me output of following commands : from the laptop running Host AP driver (run these after the failed test, : before having changed anything in the Host AP configuration): : : dmesg : cat /proc/interrupts : iwconfig wlan0 : ifconfig -a : hostap_diag wlan0 -r wlan0 : : (hostap_diag is in the utils directory of the Host AP driver package). : : I would also ask you to repeat the test with latest CVS version of the : Host AP driver if you are still using the latest release (2002-10-12). ------------------------------------------------------- -- No microsoft products were used to produce this message. EUG-LUG Mailing List: http://mailman.efn.org/cgi-bin/listinfo/eug-lug From jkmaline at cc.hut.fi Sat Jan 11 21:10:17 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sat, 11 Jan 2003 18:10:17 -0800 Subject: What to do now? In-Reply-To: <200301111628.20612.jamie@rocksolidnetworks.com> References: <200301100704.40609.linux@rocksolidnetworks.com> <20030111212254.GA1404@jm.kir.nu> <200301111628.20612.jamie@rocksolidnetworks.com> Message-ID: <20030112021017.GA2720@jm.kir.nu> On Sat, Jan 11, 2003 at 04:28:20PM -0800, Jamie wrote: > Mode: Infrastructure (this is also known as managed right?) Yes. > Heres my DMESG: > prism2_hw_config: initialized in 9094 iterations > wlan0: NIC: id=0x8002 v1.0.0 > wlan0: PRI: id=0x15 v0.3.0 > wlan0: STA: id=0x1f v0.8.3 > wlan0: defaulting to host-based encryption as a workaround for firmware bug in > Host AP mode WEP > wlan0: LinkStatus=2 (Disconnected) That looks fine as far as initialization is concerned, but it does not show any authentication/association attempts from a station. > bubba at lapdance:~$ cat /proc/interrupts > CPU0 > 3: 135 XT-PIC hostap_cs And interrupts are being delivered. Good. > root at lapdance:~# iwconfig wlan0 > wlan0 IEEE 802.11-b ESSID:"jamienet" > Mode:Master Frequency:2.422GHz Access Point: 00:60:B3:6F:63:CC And this is OK, too. > root at lapdance:~/wirless/hostap-2002-10-12/utils# ./hostap_diag -r wlan0 > FFFF=FWID=0003C0.HEX<00><00><00> Hmm.. That " WOW Thats a lot of info! I have no idea what all of that means, but Id say > that it suggests somethings working right! (hopefully you will understand > what isnt working right! Yes, everything seems to be OK as far as card initialization and configuration is concerned. Unfortunately, I cannot see anything that would indicate what is causing the problems. You could also enable all RX/TX packets dumps in the driver with 'iwpriv wlan0 dump 7' (or 'utils/prism2_param wlan0 dump 7' if you are not using wireless tools version 25 or newer). Next step would be to setup a third host as a wireless sniffer and make a dump of all frames sends by the station and the AP.. This should be able to confirm whether the AP is sending beacons and is responding to station's probe request. > Ill try to get the current cvs, could you remind me how/where? Easiest way is to download a snapshot of the CVS tree from http://hostap.epitest.fi/. Direct link for this is http://hostap.epitest.fi/cgi-bin/viewcvs.cgi/hostap/hostap.tar.gz?tarball=1 -- Jouni Malinen PGP id EFC895FA From keith at ksmith.com Sat Jan 11 23:36:46 2003 From: keith at ksmith.com (Keith E Smith) Date: Sat, 11 Jan 2003 21:36:46 -0700 Subject: Tx timed out! Resetting card (with OpenAP) References: <3E1BF9C4.4143A1F9@stanfordalumni.org> <007901c2b702$e52e9c90$53153b0a@nednet.co.za> <3E1C0812.36DF29D3@stanfordalumni.org> Message-ID: <3E20F0DE.1000101@ksmith.com> Well, you are right and wrong :) :) . The *original* openap code from instant 802 uses pcmcia-cs. The -ng code must have gone to the kernel drivers. I might mention the -ng code also has some broken patches for the MTD stuff. I tried a straight compile in of the kernel PCMCIA drivers to eliminate loading modules for linuxAP. This actually made the compressed kernel around 60K larger so I'm not sure it was worth it. Since I also couldn't then get cardmgr to initialize the PCMCIA slot correctly to load the hostap mods, . . . I cut linuxAP back over to pcmcia-cs. I'll just pay the 10K or so of extra space. pcmcia-cs compiled in flawlessly, and worked flawlessly under soekris. The new hostap code lets you compile straight out of the hostap directory, though the 'install' had a couple of gotcha's. If memory serves I had to make some small adjustments to get it to *install* the *additional* modules into my image. You now have 4 modules instead of 1. First go around I ended up with just hostap.o :(! earl wrote: > Robert Botha wrote: > > >>It seems from the Messages you pasted that you're using the builtin kernel >>pcmcia, and not the external package. >>Have you tried using the external package seperately and disabling the >>kernel drivers? >>I've had endless hassles with the kernel pcmcia and most of them went away >>when I switched to the pcmcia-cs package. > > > Thanks for the suggestion, i can give that a shot. It's possible that the > reason this was never done by the original OpenAP developers, with whose work > i'm starting, is that the pcmcia driver modules are too large (the boot media > for this platform is a mere 1MB of flash, and there's not much free space left > on it) in which case i'd be stuck with the kernel pcmcia. > > What kind of problems did you have with kernel pcmcia and hostap? Any way to > get a better idea whether this is a pcmcia issue or a hostap issue? > > thanks > earl > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap -- Keith Smith keith at ksmith.com 655 W Fremont Dr Tempe AZ 85282 it's hot From linux at rocksolidnetworks.com Sun Jan 12 00:06:56 2003 From: linux at rocksolidnetworks.com (Linux Rocks!) Date: Sat, 11 Jan 2003 21:06:56 -0800 Subject: What to do now? In-Reply-To: <20030112021017.GA2720@jm.kir.nu> References: <200301100704.40609.linux@rocksolidnetworks.com> <200301111628.20612.jamie@rocksolidnetworks.com> <20030112021017.GA2720@jm.kir.nu> Message-ID: <200301112106.56855.linux@rocksolidnetworks.com> On Saturday 11 January 2003 06:10 pm, Jouni Malinen wrote: : On Sat, Jan 11, 2003 at 04:28:20PM -0800, Jamie wrote: : > FFFF=FWID=0003C0.HEX<00><00><00> : : Hmm.. That " : : Next step would be to setup a third host as a wireless sniffer and make : a dump of all frames sends by the station and the AP.. This should be : able to confirm whether the AP is sending beacons and is responding to : station's probe request. Hrmmm... problems! I only have 2 cards. I have tried Network stumbler, and aerosol on the windows laptop, but they dont seem to work. I have another linux hard drive (its pretty small... ) with hostap, and orinoco modules on it, but since I dont have it working in linux, how could I use a sniffer with it? This Is what I have: 3 Laptops 2 PCMCIA wireless NIC's (Proxim RangeLan DS) Hard Drives (hard drives are interchangeable) 1 340 meg Windows98 1 720 meg (Slackware Linux) 1 340 Meg (OpenBSD 3.0) (does detect the card, but I dont know how to configure it..., or if it will work if I did...) 1 340 Meg Slackware Linux 1 workstation running low on diskspace (with no wireless equipment, but connected via ethernet..) I just downloaded the current cvs. maybe I can try it later... any ideas what I can do with what I have? At the LUG meetings, there is an AP, Ive connected to it via windows, it works (but I cant mess with thier AP...) Jamie : : > Ill try to get the current cvs, could you remind me how/where? : : Easiest way is to download a snapshot of the CVS tree from : http://hostap.epitest.fi/. Direct link for this is : http://hostap.epitest.fi/cgi-bin/viewcvs.cgi/hostap/hostap.tar.gz?tarball=1 -- No microsoft products were used to produce this message. EUG-LUG Mailing List: http://mailman.efn.org/cgi-bin/listinfo/eug-lug From beattie at beattie-home.net Sun Jan 12 14:07:19 2003 From: beattie at beattie-home.net (Brian Beattie) Date: 12 Jan 2003 11:07:19 -0800 Subject: wierd hardware Message-ID: <1042398439.30569.696.camel@coyote.pdx.beattie-home.net> I;ve got a couple of mini-PCI boards that are supposed to be 802.11b cards. When I put them into my soekris 4521 running Debian Linux the results I get are puzzleing. First it does not show up as a pci card, lspci does not show it, hostap_pci doe not find it. hostap_cs however does find it, sortof. Console log below. First question, why is this card showing up as a pccard and not a pci card. Second, coudl this juts be a firmware problem and how might I go about upgarding. Starting PCMCIA services: modulesLinux Kernel Card Services 3.1.22 options: [pci] [cardbus] [pm] Yenta IRQ list 0000, PCI irq10 Socket status: 30000011 Yenta IRQ list 0000, PCI irq11 Socket status: 30000006 Yenta IRQ list 0000, PCI irq11 Socket status: 30000006 cardmgr. Starting network management services:cs: IO port probe 0x0c00-0x0cff: clean. cs: IO port probe 0x0800-0x08ff: clean. cs: IO port probe 0x0100-0x04ff: clean. cs: IO port probe 0x0a00-0x0aff: clean. cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa00fffff snmpdhostap_cs: hostap_cs.c 0.0.0 2002-10-12 (SSH Communications Security Corp, Jouni Malinen) hostap_cs: (c) Jouni Malinen hostap_cs: index 0x01: Vcc 3.3, irq 10, io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 snmptrapd. Starting OpenBSD Secure Shell server: sshdwlan0: could not get own MAC address wlan0: Port type setting to 6 failed wlan0: MAC port 0 enabling failed -- Brian Beattie | Having had the honor of being selected beattie at beattie-home.net | for a Resource Action by my former employer, | it is my pleasure to announce my immediate www.beattie-home.net | availability, contract or permanent. Embedded Systems, Linux/Unix internals Software Engineer From jkmaline at cc.hut.fi Sun Jan 12 14:08:48 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 12 Jan 2003 11:08:48 -0800 Subject: hostapd forking and enhancements In-Reply-To: <000001c2b9a9$f8e4a1e0$017ba8c0@ammirata.net> References: <20030108030233.GA1134@jm.kir.nu> <000001c2b9a9$f8e4a1e0$017ba8c0@ammirata.net> Message-ID: <20030112190848.GA3160@jm.kir.nu> On Sat, Jan 11, 2003 at 02:45:17PM -0500, Sergio M. Ammirata wrote: > Here is another PAP radius capture using -s 512. I think this time it > has all the data. Thanks (and also thanks to Brian for Cisco 350 packet dumps). > As far as using a standard 802.11 encoding instead of PAP, I don't > particularly have a preference as long as we can use it quickly :) > As you can see on the captures it is extremely simple to implement PAP > ... And on the server side you can leverage the user name lists already > available in radius servers to hold the station MAC addresses. I added support for this to hostapd. Station MAC address is used both as User-Name and User-Password. Default ASCII format for the address matches with the one used by Cisco 350 (no delimiters between address octets). If needed, this can be changed in hostapd/radius.h (RADIUS_ADDR_FORMAT). I have tested RADIUS-based 802.11 authentication successfully with FreeRADIUS. Please let me know, if there are any problems with other RADIUS servers. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Sun Jan 12 14:28:04 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 12 Jan 2003 11:28:04 -0800 Subject: hostapd changes (configuration file and RADIUS without 802.1X) Message-ID: <20030112192804.GA3362@jm.kir.nu> I changed hostapd to use a configuration file instead of command line arguments. Debug and daemonize options can still be modified from command line, but everything else is read from a file. hostapd/hostapd.conf is an example configuration file with all available configuration items. Current code supports configuration reloading partially. If the process is signaled with HUP signal, hostapd will reread the configuration file and replaces the old configuration with new one. All dynamic data is not yet updated, so some changes are not taken into use without restarting the program. However, at least accept/deny lists for 802.11 authentication and RADIUS passwords can be changed without stopping the process (i.e., without losing data about associated stations). Please use full path to the configuration file when starting hostapd since working directory will be changed if the process is daemonized. I added support for using MAC address -based lists to accept/deny 802.11 authentication. In addition, external RADIUS server can be used to contain this information in a centralized (shared by multiple APs) location. With macaddr_acl=2, hostapd will query RADIUS server with Access-Request (User-Name and User-Password attributes set to station MAC address) if the MAC address is not found from accept/deny lists. RADIUS authentication for 802.11 can be used both with and without 802.1X. Likewise, RADIUS accounting does not require 802.1X. I'm planning to release next version of the Host AP driver, so I try to avoid adding new features and concentrate more on testing now. I have multiple contributed patches in my queue and I'll try to get these processed now, but other new features can wait for some time. -- Jouni Malinen PGP id EFC895FA From jamie at rocksolidnetworks.com Sat Jan 11 19:28:20 2003 From: jamie at rocksolidnetworks.com (Jamie) Date: Sat, 11 Jan 2003 16:28:20 -0800 Subject: What to do now? In-Reply-To: <20030111212254.GA1404@jm.kir.nu> References: <200301100704.40609.linux@rocksolidnetworks.com> <20030111212254.GA1404@jm.kir.nu> Message-ID: <200301111628.20612.jamie@rocksolidnetworks.com> Jouni, Thanks for asking, This message is REALLY long, with a lot of info, I tried to be as breif as possible, without omitting anything necessary. On Saturday 11 January 2003 01:22 pm, Jouni Malinen wrote: : On Fri, Jan 10, 2003 at 07:04:40AM -0800, Linux Rocks! wrote: : > My cards (Proxim RangeLan-DS) still dont work under linux :( : > They do respond under linux (I can change modes, and see the lights do : > different things...) but I dont have them doing anything usefull. : > They Do work under windows95/98. (but ... windows doesnt do what I need) Yes (assuming all the things I trierd were correct!) Ive tested the laptops in many situations. Ive set 2 laptop up in windows, had success. tested the laptop with a real ap under windows, had success. tested the hostap, and the orinoco modules with the ap w/no success. I think Ive tried 1 laptop in windows, and one in linux, using all the various possible combonations of settings, with no success. : : Have you tested running one card in Linux as an AP (i.e., using Host AP : driver in Master mode) and another card in Windows as a client that : tries to associate with the AP? I haven't noticed anything in your dmesg : logs etc. that would indicate cause for problems, so this is a bit : difficult to solve. Yes! I agree... atleast when I had errors... I had something to work on... Ive just setup the cards with one windows box, and one linux box. windows98 has these settings: Mode: Infrastructure (this is also known as managed right?) SSID: ANY (This is supposed to scan...), Ive also tried jamienet EtherNEt Conversion: RCF 1042 TXRate: Fully Automatic WEP: Disabled PS Mode: Disabled (ive also tried enabled) (This I think is what worked with the AP at the LUG) Linux box: (running hostap_cs, using lastest 2002-10-12): (I have 3 different kernels/modules on the machine, but Im currently using 2.4.20-rc-3) Ive editied my /etc/pcmcia/config to add: source ./hostap_cs.conf and Ive added in my /etc/pcmcia/hostap_cs.conf card "JamieTest PRISM RangeLAN-DS/LAN PC CARD" manfid 0x0126, 0x8000 bind "hostap_cs" # Optional configuration parameters for hostap_cs.o # module "hostap_cs" opts "channel=6 essid=any ignore_cis_vcc=1" module "hostap_cs" opts "channel=3 iw_mode=3 essid=jamienet ignore_cis_vcc=1" (this sets the card to Master Mode, use essid jamietest, and ignore the voltage issue (I do get errors with it set to 0) Heres my DMESG: cs: IO port probe 0x0a00-0x0aff: excluding 0xa68-0xa6f 0xaa0-0xaa7 0xaf8-0xaff cs: memory probe 0x0d0000-0x0dffff: clean. hostap_crypt: registered algorithm 'NULL' hostap_cs: hostap_cs.c 0.0.0 2002-10-12 (SSH Communications Security Corp, Jouni Malinen) hostap_cs: (c) Jouni Malinen hostap_cs: setting Vcc=33 (constant) hostap_cs: CS_EVENT_CARD_INSERTION prism2_config() hostap_cs: ignoring Vcc=50 (from config) Checking CFTABLE_ENTRY 0x01 (default 0x01) IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 io->flags = 0x0046, io.base=0x0000, len=64 hostap_cs: index 0x01: Vcc 5.0, irq 3, io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 prism2_hw_init() prism2_hw_config: initialized in 9094 iterations wlan0: NIC: id=0x8002 v1.0.0 wlan0: PRI: id=0x15 v0.3.0 wlan0: STA: id=0x1f v0.8.3 wlan0: defaulting to host-based encryption as a workaround for firmware bug in Host AP mode WEP wlan0: LinkStatus=2 (Disconnected) cat /proc/Interrupts bubba at lapdance:~$ cat /proc/interrupts CPU0 0: 733578 XT-PIC timer 1: 302 XT-PIC keyboard 2: 0 XT-PIC cascade 3: 135 XT-PIC hostap_cs 10: 563 XT-PIC NE2000 12: 1 XT-PIC PS/2 Mouse 14: 1773 XT-PIC ide0 NMI: 0 ERR: 0 root at lapdance:~# iwconfig wlan0 wlan0 IEEE 802.11-b ESSID:"jamienet" Mode:Master Frequency:2.422GHz Access Point: 00:60:B3:6F:63:CC Bit Rate:11Mb/s Tx-Power:-12 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:0 Signal level:0 Noise level:0 Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 ifconfig -a: root at lapdance:~# ifconfig -a eth0 Link encap:Ethernet HWaddr 00:A0:D1:01:0D:86 inet addr:192.168.14.20 Bcast:192.168.14.255 Mask:255.255.255.0 UP BROADCAST NOTRAILERS RUNNING MULTICAST MTU:1500 Metric:1 RX packets:394 errors:0 dropped:0 overruns:0 frame:0 TX packets:285 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:35154 (34.3 Kb) TX bytes:144579 (141.1 Kb) Interrupt:10 Base address:0x300 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) wlan0 Link encap:Ethernet HWaddr 00:60:B3:6F:63:CC BROADCAST MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b) Interrupt:3 Base address:0x100 I hadnt built the hostap utils... so I cd'd to the directory, and did make. it made the utils! I tried : hostap_diag wlan0 -r wlan0, but that failed. so I tried: ./hostap_diag -r wlan0 and got the following(Very long!) root at lapdance:~/wirless/hostap-2002-10-12/utils# ./hostap_diag -r wlan0 Host AP driver diagnostics information for 'wlan0' NICID: id=0x8002 v1.0.0 (HWB3163-01,02,03,04 Rev A) PRIID: id=0x0015 v0.3.0 STAID: id=0x001f v0.8.3 (station firmware) Known RIDs (Resource IDentifiers) FC00=CNFPORTTYPE=6 FC01=CNFOWNMACADDR=00:60:b3:6f:63:cc FC02=CNFDESIREDSSID=jamienet FC03=CNFOWNCHANNEL=3 FC04=CNFOWNSSID=jamienet FC05=CNFOWNATIMWINDOW=0 FC06=CNFSYSTEMSCALE=1 FC07=CNFMAXDATALEN=2304 FC08=CNFWDSADDRESS=00:00:00:00:00:00 FC09=CNFPMENABLED=0 FC0A=CNFPMEPS=0 FC0B=CNFMULTICASTRECEIVE=1 FC0C=CNFMAXSLEEPDURATION=100 FC0D=CNFPMHOLDOVERDURATION=100 FC0E=CNFOWNNAME= FC10=CNFOWNDTIMPERIOD=1 FC20=UNKNOWN1=33 FC21=UNKNOWN2=33 FC23=CNFWEPDEFAULTKEYID=0 FC24=CNFDEFAULTKEY0=<21><00><24><00><00> FC25=CNFDEFAULTKEY1=<21><00><25><00><00> FC26=CNFDEFAULTKEY2=<21><00><26><00><00> FC27=CNFDEFAULTKEY3=<21><00><27><00><00> FC28=CNFWEPFLAGS=<00><00> FC2A=CNFAUTHENTICATION=1 FC2C=CNFTXCONTROL=512 FC2D=CNFROAMINGMODE=1 FC30=CNFRCVCRCERROR=0 FC31=CNFMMLIFE=65000 FC32=CNFALTRETRYCOUNT=7 FC33=CNFBEACONINT=100 FC35=CNFSTAPCFINFO=<01><00> FC40=CNFTIMCTRL=1 FC41=UNKNOWN3=<00><00><00><00><00><00> FC42=CNFTHIRTY2TALLY=0 FC80=GROUPADDRESSES= FC81=CREATEIBSS=1 FC82=FRAGMENTATIONTHRESHOLD=2346 FC83=RTSTHRESHOLD=2347 FC84=TXRATECONTROL=15 FC85=PROMISCUOUSMODE=0 FCB0=CNFSHORTPREAMBLE=0 FCB3=CNFBASICRATES=<03><00> FCB4=CNFSUPPORTEDRATES=<0f><00> FCE0=TICKTIME=0 FCE1=SCANREQUEST=<3f><20><00> FCE2=JOINREQUEST=<21><00><00><00><00><00> FD00=MAXLOADTIME=60000 FD01=DOWNLOADBUFFER=<60><00><00><00><00><10> FD02=PRIID=0x15 v0.3.0 FD03=PRISUPRANGE=0 0x03 1 1-2 FD04=CFIACTRANGES=1 0x02 1 1-1 FD0A=NICSERNUM=SA01000000 FD0B=NICID=0x8002 v1.0.0 FD0C=MFISUPRANGE=0 0x01 1 1-1 FD0D=CFISUPRANGE=0 0x02 1 1-1 FD10=CHANNELLIST=<07> FD11=REGULATORYDOMAINS=<10> FD12=TEMPTYPE=1 FD13=CIS=<01><03><00><00><17><04><67><5a><08><1d><05><03><67><5a><08><1c><04><03><00><00><15><23><05><00><50><52><4f><58><49><4d><00><52><61><6e><67><65><4c><41><4e><2d><44><53><2f><4c><41><4e><20><50><43><20><43><41><52><44><00><00><20><04><26><01><00><80><21><02><06><00><22><02><01><07><22><05><02><40><42><0f><00><22><05><02><80><84><1e><00><22><02><03><07><22><02><05><01><1a><07><03><02><03><00><00><01><1b><10><01><19><77><1e><35><3c><36><36><05><46><7f><00><00><00> + 352* FD20=STAID=0x1f v0.8.3 FD21=STASUPRANGE=0 0x04 1 1-6 FD22=MFIACTRANGES=1 0x01 1 1-1 FD23=CFIACTRANGES2=1 0x02 1 1-1 FD40=PORTSTATUS=8 FD41=CURRENTSSID=jamienet FD42=CURRENTBSSID=00:60:b3:6f:63:cc FD43=COMMSQUALITY=<00><00><1b><00><1b><00> FD44=CURRENTTXRATE=2 FD45=CURRENTBEACONINTERVAL=100 FD46=CURRENTSCALETHRESHOLDS=<36><00><36><00><36><00><0a><00><02><00><06><00> FD47=PROTOCOLRSPTIME=100 FD48=SHORTRETRYLIMIT=8 FD49=LONGRETRYLIMIT=5 FD4A=MAXTRANSMITLIFETIME=512 FD4B=MAXRECEIVELIFETIME=512 FD4C=CFPOLLABLE=0 FD4D=AUTHENTICATIONALGORITHMS=<01><00><01><00> FD4F=PRIVACYOPTIONIMPLEMENTED=1 FD88=SCANRESULTSTABLE=<01><00><03><00> FDC0=PHYTYPE=2 FDC1=CURRENTCHANNEL=3 FDC2=CURRENTPOWERSTATE=1 FDC3=CCAMODE=2 FDC6=SUPPORTEDDATARATES=<02><04><0b><16> FFFE=BUILDSEQ=<00><00><00><00> FFFF=FWID=0003C0.HEX<00><00><00> WOW Thats a lot of info! I have no idea what all of that means, but Id say that it suggests somethings working right! (hopefully you will understand what isnt working right! Ill try to get the current cvs, could you remind me how/where? Thanks in advance for help. Jamie : : I would suggest setting up one Host AP, make sure it is in Master mode : and that the netdevice is UP. After this, try to associate with it using : another laptop with Windows (that has been tested to work with another : AP). If this does not work, please send me output of following commands : from the laptop running Host AP driver (run these after the failed test, : before having changed anything in the Host AP configuration): : : dmesg : cat /proc/interrupts : iwconfig wlan0 : ifconfig -a : hostap_diag wlan0 -r wlan0 : : (hostap_diag is in the utils directory of the Host AP driver package). : : I would also ask you to repeat the test with latest CVS version of the : Host AP driver if you are still using the latest release (2002-10-12). From cupid90 at samsung.co.kr Sun Jan 12 21:02:33 2003 From: cupid90 at samsung.co.kr (cupid90 at samsung.co.kr) Date: Mon, 13 Jan 2003 11:02:33 +0900 Subject: Transmit timed out occurred! Message-ID: Hi, This is Mr.Roh. I am porting to embedded wireless AP router by using HostAP driver. ping is good. but when I have performance test by using FTP program, the following message was displayed. the messages are like this: NETDEV WATCHDOG: wlan0ap: transmit timed out wlan0ap Tx timed out! Resetting card hostap_cs: wlan0: resetting card prism2_hw_init() I looked for the old mailing list, and read Jouni Malinen's reply. but I did not understand irq_mode and PCIC_OPTS. Our pcmicia does not use PCI bus. our chipset is ARM-9 family chipset. and my kernel version is 2.4.17 and prism2_param command also does not working. If I type the command, "syntax error" messages was displayed. I hope somebody who have been experienced the above errors could help me. Thanks Roh you-chang Application S/W Engineer System Solution Group Access Network SOC Team System LSI Business Device Solution Network SAMSUNG ELECTRONICS CO., LTD. phone:82-31-209-8314, fax:82-31-209-8309 e-mail:cupid90 at samsung.co.kr From jkmaline at cc.hut.fi Mon Jan 13 00:06:25 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 12 Jan 2003 21:06:25 -0800 Subject: problems with EN2511CD / NL2511CD In-Reply-To: References: <20021203062656.GA12480@jm.kir.nu> <20021204032243.GA1232@jm.kir.nu> Message-ID: <20030113050625.GD28005@jm.kir.nu> On Sat, Dec 21, 2002 at 07:28:02PM -0800, dean gaudet wrote: > the attached file was collected by "tethereal -s 1600"... i hope it helps! > > > On Mon, Dec 02, 2002 at 10:59:24PM -0800, dean gaudet wrote: > > > > Dec 2 22:46:51 mewlark kernel: wlan0: 00:40:96:5a:5d:ce auth (alg=0 trans#=1 stat=0 len=6 fc=00b0) ==> 0 (authOK) > > > > Dec 2 22:46:51 mewlark kernel: wlan0: 00:40:96:5a:5d:ce auth_cb - frame was not ACKed The capture log seemed to miss lot of messages. There was only one beacon frame although there should have been about 10 beacons a second. In addition, more or less all ACKs were missing. If these frames were really missing, it would explain problems. However, I do not know why the station even tried to authenticate with the AP. It was sending Probe Requests for SSID 'groove kitchen' (i.e., not wildcard accepting any SSID) and the AP was (based on that one beacon frame) configured to use SSID 'test'. AP did not reply to probe requests because of this. Based on the numerous retries of authentication messages it looks like there are some problems in packet acknowledging. The packet dump did not show ACKs, so this might have been due to some problems in capturing the frames or AP (and actually even station) not ACKing them. I cannot determine cause for the main problem based on that packet dump. I would try making sure that the client station is configured to use the same SSID as the AP. It might be possible that the wlan card in AP does not ACK all frames for unknown reason. I have seen this happening with Prism2 cards, but usually in station mode. I do not know what is the real reason for this. Sometimes it is enough to reset the card ('iwpriv wlan0 reset 0' or 'iwpriv wlan0 reset 1') to get ACKing back; sometimes this is not enough.. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Mon Jan 13 00:37:02 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 12 Jan 2003 21:37:02 -0800 Subject: 2002-10-12 monitor mode In-Reply-To: References: Message-ID: <20030113053702.GE28005@jm.kir.nu> On Tue, Dec 10, 2002 at 11:57:48AM -0500, Bichot Guillaume (Princeton) wrote: > I found that some time, switching to the monitor mode, the incoming frames > are wrong. I mean running wlansniff I found that there is a shift in the > data packet delivered to the user space in such a way the wlansniff > misinterprets the header. Ethereal is also unable to interprets correctly > the frames as well. To solve the problem I have to switch back to the > managed mode and switch back to the monitor mode (need sometime several back > and forth until the things go right). I have to add that I use the monitor > mode with the frequency I want to listen (i.e. iwconfig wlan0 freq > "2.437ghz" mode "monitor"). Can you repeat this easily? With current CVS version of Host AP driver? If yes, could you please describe all commands you used to change the card into monitor mode up to the point when you received invalid frames? I can think of only one way to get invalid frame for wlansniff (or Ethereal) and that is to change the netdev type during sniffing session (i.e., without restarting wlansniff). wlansniff does not update interface type per packet, so it requires restarting. Other than this, I haven't seen such errors with the latest driver version. > I'm using the Linksys WPC 11 PCMCIA card (manfid 0x274, 0x1613). Today I did some testing with the same card model (and STA f/w 1.4.9 and latest CVS version of Host AP driver) and was unable to duplicate this. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Mon Jan 13 00:44:25 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 12 Jan 2003 21:44:25 -0800 Subject: DHCP Problems In-Reply-To: <00a301c2a09c$f87eff50$6df058db@PC2> References: <00a301c2a09c$f87eff50$6df058db@PC2> Message-ID: <20030113054425.GF28005@jm.kir.nu> On Wed, Dec 11, 2002 at 11:39:16AM +1300, Craig Whitmore wrote: > We have hostap up and running and if we set an ip address on the remote card > (winXP machine) we can ping the linux box and internet access works 100%, > But if we run DHCPD on the linux box to give the ip address to the WinXP > machine we get lots of errors as below. Can anyone know what is causing this > and can help? Does the network access work even with dhcpd (i.e., are the error messages only thing that tells there is something odd going on) or are there noticeable problems in connection? Do you get error messages even without the station (i.e., just running dhcpd)? > I've tried the latest stable version and the latest CVS version (upto about > 5 mins ago) with the same problem. I have seen similar errors when promiscuous mode was used, but I commented it out Dec 9, 2002 so it should have been of in that latest CVS version test you did. Anyway, DHCP server should not really need promiscuous mode. If you can still repeat the problem with the current CVS version, please tell me which DHCP server you are using and I'll try to duplicate this. You could also send dmesg output from the first error messages (i.e., including the card initialization and the first few error messages). -- Jouni Malinen PGP id EFC895FA From dave at cc0.net Mon Jan 13 04:19:28 2003 From: dave at cc0.net (Dave) Date: Mon, 13 Jan 2003 01:19:28 -0800 Subject: still compiling problems ;( References: <20030111115621.A22382@nevermore.toe.doomcom.org> Message-ID: <000401c2bae4$df87e450$0200a8c0@ws7> Hello, I had same problem however my setup might be a little different. The way I solved it was to run the full depmod command for the kernel I was planning on using hostap with. Typically the kernel I was to be running hostap was not the current running kernel...so to be sure I would point it to the correct system map file via: depmod -ae -F System.map-2.4.x 2.4.x Where System.map-2.4.x is the System.map generated when your kernel was built and 2.4.x is the actual kernel version matching that System.map When doing that I never had a problem with unresolved symbols... I don't know for sure if that is the correct way of resolving this problem...but it has worked for me at least 10 times...;) -Dave > Liboc01:/usr/src/hostap# depmod -ae > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o > depmod: wireless_send_event_R9e8b822b > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o > depmod: wireless_send_event_R9e8b822b > > with 2.4.21-pre3 worked everything well...but on server I can't have pre > kernel... > > Michal > > Dne so 11. leden 2003 17:56 jste napsal(a): > > On Sat, Jan 11, 2003 at 10:38:10AM +0100, Michal Semler wrote: > > > /sbin/depmod -a > > > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap.o > > > depmod: *** Unresolved symbols in /lib/modules/2.4.20/net/hostap_pci.o > > > make: *** [install_pci] Error 1 > > > > if you could run "/sbin/depmod -ae" it would tell which symbols were > > undefined and would probably help others figure out where the actual > > problem is. > > > > -- Gerald > > _______________________________________________ > > HostAP mailing list > > HostAP at shmoo.com > > http://lists.shmoo.com/mailman/listinfo/hostap > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From thed at intracom.gr Mon Jan 13 10:13:25 2003 From: thed at intracom.gr (Dimitris Theofilatos) Date: Mon, 13 Jan 2003 15:13:25 +0000 Subject: HostAP and xsupplicant Message-ID: <200301131513.25081.thed@intracom.gr> Hello, Does anyone know if the HostAP code interworks with the xsupplicant code from open1x? I followed the instructions from the site www.missl.cs.umd.edu/wireless/eaptls/. I used 3 linux machines with RedHat 7.3 and kernel 2.4.18. I used hostap-2002-10-12 driver and the latest version of xsupplicant (xsupplicant-0.5) but when the xsupplicant sends the first EAPOL Start message the hostapd daemon unauthorize the station. Thanks in advance, Dimitris From jirasek at egerius.cz Mon Jan 13 06:54:37 2003 From: jirasek at egerius.cz (=?iso-8859-2?Q?Jaroslav_Jir=E1sek?=) Date: Mon, 13 Jan 2003 12:54:37 +0100 Subject: slow data transfer over hostap Message-ID: <008101c2bafa$903cf6e0$0201a8c0@cheb.egerius.cz> Hi, i have problem with data transfer between clients over hostap. Data transfer from hostap to client is OK, i reach about 500kB/s (from both), but between clients is speed about 80kB/s (tested with scp). I also tried this tests with ping: -from both clients to hostap, no lost packet ping -s 1400 -f 192.168.1.1 -from client to client over hostap: ping -s 1400 -i 0.02 192.168.1.4 results: 1408 bytes from 192.168.1.4: icmp_seq=0 ttl=62 time=21.280 msec 1408 bytes from 192.168.1.4: icmp_seq=1 ttl=62 time=29.031 msec 1408 bytes from 192.168.1.4: icmp_seq=2 ttl=62 time=25.872 msec 1408 bytes from 192.168.1.4: icmp_seq=3 ttl=62 time=26.896 msec 1408 bytes from 192.168.1.4: icmp_seq=4 ttl=62 time=17.874 msec 1408 bytes from 192.168.1.4: icmp_seq=5 ttl=62 time=19.085 msec ... 1408 bytes from 192.168.1.4: icmp_seq=72 ttl=62 time=97.820 msec 1408 bytes from 192.168.1.4: icmp_seq=73 ttl=62 time=87.663 msec 1408 bytes from 192.168.1.4: icmp_seq=74 ttl=62 time=70.793 msec 1408 bytes from 192.168.1.4: icmp_seq=76 ttl=62 time=125.768 msec 1408 bytes from 192.168.1.4: icmp_seq=77 ttl=62 time=148.545 msec 1408 bytes from 192.168.1.4: icmp_seq=78 ttl=62 time=275.824 msec ... 1408 bytes from 192.168.1.4: icmp_seq=315 ttl=62 time=7.125 sec 1408 bytes from 192.168.1.4: icmp_seq=319 ttl=62 time=7.597 sec 1408 bytes from 192.168.1.4: icmp_seq=322 ttl=62 time=7.699 sec 1408 bytes from 192.168.1.4: icmp_seq=323 ttl=62 time=7.731 sec with 66% packet loss. I can see in /proc/net/hostap, that rx packets goes from both clients only on 1Mbit. Tx to both clients keeps at 11Mbit. Few seconds after stoping test rx packets to hostap goes again in 11Mbit. Hostap is running on C366, 64MB, intel lx chipset and ZI-626, clients is 486/80mhz, 16Mb ram with orinoco. All computers have redhat linux 7.3 with latest redhat kernel (2.4.18-18.7.x), lastest wireless tools (wireless-tools-25-1). Hostap is CVS version from 7.1.2003. (I also tried version from 12.10.2002 with identical results) Clients are routers for 2 local networks, where are computes with Windows. They cannot copy data over hostap, because data transfer causes overload of wireles network, even often file data transfer failed. Slow data transfers (browsing internet) is without problem. Signal quality on first klient is 18, on second is 15. Hostap have signal quality for first client 17, for second 13. Thank you for any suggestion. Jarda From hairong at ee.duke.edu Mon Jan 13 18:30:07 2003 From: hairong at ee.duke.edu (Hairong Sun) Date: Mon, 13 Jan 2003 18:30:07 -0500 (EST) Subject: about noise level, signal level and quality Message-ID: Hello, I set up a testing environment and tried to measure the noise,signal and quality of each channel. I have two WLANs in lab (same room, close to each other): WLAN 1: one linksys AP and one STA associated with the linksys AP, set channel # =8, and download a long file. WLAN 2: one hostap and one STA associated with the hostap I ran the site survey and confirmed there is no other WLAN nearby. I printed out the noise, signal and quality of the STA associated with the hostap (by inserting printk somewhere in source code, and using dmesg to read). What I got is kind of confusing: 1. When I set the channel # of WLAN 2 the same as WLAN 1 (i.,e, channel 8 ), I printed out the noise level as around 157~159. When I set the channel # of WLAN 2 to be 1, the noise level is surprisingly still 157~159. I shut down WLAN 1, and the noise level is still ~158. NOTE: I restarted hostap and reassociated STA each time I measured the noise level in order to ensure the noise level reading are updated. I repeated the testing a couple of times, and still got the same results (noise level around 157~159). I am puzzled by the results. WLAN1 should have interference to WLAN 2 if the two channels are close, esp. choosing the same channel. The noise level should decrease if WLAN2 is shut down or chooses channel 1 instead of channel 8. 2. When "signal" - "noise"<35, the quality would be larger than 200 (say 240). It is explanable according to the definition of "quality" ("signal"-"noise"-35, being u8), though "quality" is defined between 0~92. Is it a bug need to be fixed? Since "signal" and "noise" are converted by (1x100+v*100/255-100) and "quality" is converted by (signal-noise-35) in "prism2_ap_get_sta_qual()" from their original values (last_rx_signal and last_rx_silence) in the firmware, my questions are 1. what do "signal","noise", last_rx_signal, and last_rx_silence really mean in the driver AND firmware? 2. how to explain the noise level reading in my experiments (i.e.,noise level changes little with the interference from neighboring AP)? -- Regards Hairong From jkmaline at cc.hut.fi Mon Jan 13 22:00:06 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 13 Jan 2003 19:00:06 -0800 Subject: HostAP and xsupplicant In-Reply-To: <200301131513.25081.thed@intracom.gr> References: <200301131513.25081.thed@intracom.gr> Message-ID: <20030114030006.GA1456@jm.kir.nu> On Mon, Jan 13, 2003 at 03:13:25PM +0000, Dimitris Theofilatos wrote: > Does anyone know if the HostAP code interworks with the xsupplicant code from > open1x? I followed the instructions from the site > www.missl.cs.umd.edu/wireless/eaptls/. I used 3 linux machines with RedHat > 7.3 and kernel 2.4.18. I used hostap-2002-10-12 driver and the latest version > of xsupplicant (xsupplicant-0.5) but when the xsupplicant sends the first > EAPOL Start message the hostapd daemon unauthorize the station. I tested xsupplicant v0.5 today, but it segfaulted when I tried to authenticate.. I didn't yet have time to check why. Anyway, older version of xsupplicant work fine with the same certificates and configuration using hostapd as Authenticator and FreeRADIUS as Authentication Server. Authenticator state machines set the port into unauthorized state when a new station is added so it is OK to get that messages after EAPOL-Start. The port is authorized for the first time only after successful authentication has been completed. -- Jouni Malinen PGP id EFC895FA From imp at airlinktek.com Tue Jan 14 02:14:31 2003 From: imp at airlinktek.com (=?utf-7?B?K3ZCWFhaY1gwLQ==?=) Date: Tue, 14 Jan 2003 16:14:31 +0900 Subject: snmp Message-ID: Dear all.. These day, I'm try to porting ucd-snmp code for the 802.11 MIB for snmp management. Of course, I use the hostap and hostapd for the wireless communication.. When I finished it for 802.11 MIB for hostap I will send it to Jouni Malinen for the future usage.. by the way, the 802.11 mib have defined the dot11AuthenticationResponseTimeOut as a read-write I think It match for the REAUTH_TIMER_DEFAULT_reAuthPeriod in hostapd at eapol_sm.h file the problem is it used as definition not variable... and I will find more definition for this kinds of variables in 802.11 MIB.. In this case, what is the best way to avoid future conflict in source code level ? Thanks for your attention.. From thed at intracom.gr Tue Jan 14 06:18:42 2003 From: thed at intracom.gr (Dimitris Theofilatos) Date: Tue, 14 Jan 2003 11:18:42 +0000 Subject: HostAP and xsupplicant In-Reply-To: <20030114030006.GA1456@jm.kir.nu> References: <200301131513.25081.thed@intracom.gr> <20030114030006.GA1456@jm.kir.nu> Message-ID: <200301141118.42960.thed@intracom.gr> Hi, The problem still exists with older version of xsupplicant. I think that it is not a problem of xsupplicant but a problem of Host AP (or i am doing something wrong). I believe so, because when i follow the next steps: make plx EXTRA_CFLAGS="-DPRISM2_HOSTAPD" make install_plx modprobe hostap_plx cd hostapd make ./hostapd wlan0 which means that i dont use 802.1x (am i correct?) the mobile station becomes authenticated and associated but i can't make a ping from the mobile station to the access point. When i follow the next steps (not 802.1x functionality): make plx make install_plx modprobe hostap_plx everything works fine. I use a 3COM 3CRWE777A-E1 card at the access point and a Orinoco silver card at the mobile station. Thanks Dimitris > On Mon, Jan 13, 2003 at 03:13:25PM +0000, Dimitris Theofilatos wrote: > > Does anyone know if the HostAP code interworks with the xsupplicant code > > from open1x? I followed the instructions from the site > > www.missl.cs.umd.edu/wireless/eaptls/. I used 3 linux machines with > > RedHat 7.3 and kernel 2.4.18. I used hostap-2002-10-12 driver and the > > latest version of xsupplicant (xsupplicant-0.5) but when the xsupplicant > > sends the first EAPOL Start message the hostapd daemon unauthorize the > > station. > > I tested xsupplicant v0.5 today, but it segfaulted when I tried to > authenticate.. I didn't yet have time to check why. Anyway, older > version of xsupplicant work fine with the same certificates and > configuration using hostapd as Authenticator and FreeRADIUS as > Authentication Server. > > Authenticator state machines set the port into unauthorized state when a > new station is added so it is OK to get that messages after EAPOL-Start. > The port is authorized for the first time only after successful > authentication has been completed. From ddttpp at hotmail.com Tue Jan 14 06:06:31 2003 From: ddttpp at hotmail.com (ddttpp ddttpp) Date: Tue, 14 Jan 2003 11:06:31 +0000 Subject: Compile error Message-ID: Hi everyone, I can't compile the newest hotsap release (2002-10-12) using "make pci". The error message is as follows. I'm using RedHat 8.0 (kernel 2.4.18-14) and have compiled the kernel before this. I simply don't know what I should do with this. Thank you very much for help in advance. Best regards! David T Poll make pci gcc -I/usr/src/linux-2.4/include -include driver/modules/hostap_crypt.ver -include driver/modules/hostap.ver -O2 -D__KERNEL__ -DMODULE -Wall -c -DMODVERSIONS -include /usr/src/linux-2.4/include/linux/modversions.h -fomit-frame-pointer -o driver/modules/hostap_crypt.o driver/modules/hostap_crypt.c In file included from /usr/src/linux-2.4/include/linux/prefetch.h:13, from /usr/src/linux-2.4/include/linux/list.h:6, from /usr/src/linux-2.4/include/linux/module.h:12, from driver/modules/hostap_crypt.c:15: /usr/src/linux-2.4/include/asm/processor.h:51: warning: parameter names (without types) in function declaration /usr/src/linux-2.4/include/asm/processor.h:51: field `loops_per_jiffy_R_ver_str' declared as a function /usr/src/linux-2.4/include/asm/processor.h:73: invalid suffix on integer constant /usr/src/linux-2.4/include/asm/processor.h:73: parse error before numeric constant In file included from /usr/src/linux-2.4/include/linux/prefetch.h:13, from /usr/src/linux-2.4/include/linux/list.h:6, from /usr/src/linux-2.4/include/linux/module.h:12, from driver/modules/hostap_crypt.c:15: /usr/src/linux-2.4/include/asm/processor.h:258:1: warning: "EISA_bus" redefined In file included from /usr/src/linux-2.4/include/linux/modversions.h:128, from :3: /usr/src/linux-2.4/include/linux/modules/i386_ksyms.ver:88:1: warning: this is the location of the previous definition /usr/src/linux-2.4/include/asm/processor.h:260: warning: parameter names (without types) in function declaration /usr/src/linux-2.4/include/asm/processor.h:436: invalid suffix on floating constant /usr/src/linux-2.4/include/asm/processor.h:436: parse error before numeric constant /usr/src/linux-2.4/include/asm/processor.h:436: `kernel_thread_R_ver_str' declared as function returning a function In file included from driver/modules/hostap_crypt.c:15: /usr/src/linux-2.4/include/linux/module.h:183: invalid suffix on integer constant /usr/src/linux-2.4/include/linux/module.h:183: parse error before numeric constant /usr/src/linux-2.4/include/linux/module.h:183: `inter_module_register_R_ver_str' declared as function returning a function /usr/src/linux-2.4/include/linux/module.h:184: invalid suffix on integer constant /usr/src/linux-2.4/include/linux/module.h:184: parse error before numeric constant /usr/src/linux-2.4/include/linux/module.h:184: `inter_module_unregister_R_ver_str' declared as function returning a function /usr/src/linux-2.4/include/linux/module.h:185: `inter_module_get_R_ver_str' declared as function returning a function /usr/src/linux-2.4/include/linux/module.h:185: warning: parameter names (without types) in function declaration /usr/src/linux-2.4/include/linux/module.h:186: `inter_module_get_request_R_ver_str' declared as function returning a function /usr/src/linux-2.4/include/linux/module.h:186: warning: parameter names (without types) in function declaration /usr/src/linux-2.4/include/linux/module.h:187: invalid suffix on integer constant /usr/src/linux-2.4/include/linux/module.h:187: parse error before numeric constant /usr/src/linux-2.4/include/linux/module.h:187: `inter_module_put_R_ver_str' declared as function returning a function /usr/src/linux-2.4/include/linux/module.h:196: `try_inc_mod_count_R_ver_str' declared as function returning a function /usr/src/linux-2.4/include/linux/module.h:196: warning: parameter names (without types) in function declaration make: *** [driver/modules/hostap_crypt.o] Error 1 _________________________________________________________________ ???? MSN Explorer: http://explorer.msn.com/lccn From david.goodenough at btconnect.com Tue Jan 14 14:43:57 2003 From: david.goodenough at btconnect.com (David Goodenough) Date: Tue, 14 Jan 2003 19:43:57 +0000 Subject: Prefered wireless hardware Message-ID: <200301141943.KAA79014@archimedes.shmoo.com> I am embarking on a project which will either use PCMCIA or mini-PCI Wifi cards. I am obviously going to use Prism 2/2.5/3 cards as those are the only ones that seem to have Open Source access to the AP functions. I am however using external aerials, over as great a distance as possible, so I need cards that give as good power output and receive sensitivity, and do not rely on internal aerials. In particular for the PCMCIA cards I need to have more than one card in each node, and I need to be able to put two cards into adjacent sockets (assume a physical configuration like a laptop with two PCMCIA slots), without the pigtails interfering with each other. I am doing this in the UK, so I will need cards that are available and legal in this country (I understand for instance that 200mW cards are available in the US but are not legal in the UK). Any suggestions. Thanks in advance David From geeojr at krombie.net Tue Jan 14 14:58:47 2003 From: geeojr at krombie.net (geeojr at krombie.net) Date: 14 Jan 2003 19:58:47 -0000 Subject: Prefered wireless hardware Message-ID: <20030114195847.16357.qmail@aphrodite.empoweringmedia.net> I've not tried one as of yet; but this card (100mW): http://www.demarctech.com/products/reliawave-rwz/reliawave-rwz-100mw-prism2-5-pcmcia-card.html shows some promising numbers on both ends... I've been eyeing the 200mW version for myself... Jeremy ---- Original Message ---- from: "David Goodenough" to: sent: Jan 14 2003 02:43:57 PM subject: Prefered wireless hardware I am embarking on a project which will either use PCMCIA or mini-PCI Wifi cards. I am obviously going to use Prism 2/2.5/3 cards as those are the only ones that seem to have Open Source access to the AP functions. I am however using external aerials, over as great a distance as possible, so I need cards that give as good power output and receive sensitivity, and do not rely on internal aerials. In particular for the PCMCIA cards I need to have more than one card in each node, and I need to be able to put two cards into adjacent sockets (assume a physical configuration like a laptop with two PCMCIA slots), without the pigtails interfering with each other. I am doing this in the UK, so I will need cards that are available and legal in this country (I understand for instance that 200mW cards are available in the US but are not legal in the UK). Any suggestions. Thanks in advance David _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap ____________________________________________________________ Krombie Networks -- http://www.krombie.net From dave at interepoch.com.tw Tue Jan 14 21:43:51 2003 From: dave at interepoch.com.tw (David Chen) Date: Wed, 15 Jan 2003 10:43:51 +0800 Subject: EAP-TLS and WEP together with hostapd! Message-ID: Hi all I am now trying EAP-TLS with hostap driver and hostapd. EAP-TLS works fine with windows 2000 IAS server if the WEP is not turned on as followed: hostapd -x -o192.168.168.200 -a192.168.168.220 -stest wlan0 but the hostapd doesn't receive even authentication request if the individual key mapping is turned on as followed: hostapd -x -b5 -i5 -o192.168.168.200 -a192.168.168.220 -stest wlan0 does anyone has similar experience and is able to give me a few pointers? thanks dave From M.F.A.Coenen at student.tue.nl Wed Jan 15 10:39:24 2003 From: M.F.A.Coenen at student.tue.nl (M.F.A.Coenen at student.tue.nl) Date: Wed, 15 Jan 2003 16:39:24 +0100 Subject: 3Com PCI and PCMCIA cards support Message-ID: <20030115153924.GA811@finwe.zillian.org> Hello all, Has anybody ever tried to get the 3com 3CRSHPW196 wireless PCMCIA network-card or the 3CRDW696 PCI card working with host ap? We are considering to get these cards, but host ap support is a pre. Any anwer is highly appreciated. Best Regards, Martijn Coenen From msmith at cbnco.com Wed Jan 15 10:46:23 2003 From: msmith at cbnco.com (Michael Smith) Date: Wed, 15 Jan 2003 10:46:23 -0500 (EST) Subject: about noise level, signal level and quality In-Reply-To: Message-ID: On Mon, 13 Jan 2003, Hairong Sun wrote: > I have two WLANs in lab (same room, close to each other): > WLAN 1: one linksys AP and one STA associated with the linksys AP, > set channel # =8, and download a long file. > WLAN 2: one hostap and one STA associated with the hostap > I ran the site survey and confirmed there is no other WLAN nearby. I don't think the noise stats that the AP on WLAN 2 sees are meaningful unless there's a lot of traffic going over both networks. Signal and noise levels in STA mode are given in dBm with this driver I think. Unfortunately the levels in AP mode are in some unknown (to me) unit from the firmware. AP mode also doesn't seem to generate a link quality for its STAs, but you can simulate it pretty easily with a signal:noise ratio on a logarithmic scale (or so I've found). Mike From jar at pcuf.fi Wed Jan 15 15:07:21 2003 From: jar at pcuf.fi (jar) Date: Wed, 15 Jan 2003 22:07:21 +0200 Subject: FW Flashing utilities Message-ID: <3E25BF79.3BB285B3@pcuf.fi> Is it safe to use the prism2_srec to flash the firmware or is it safer to use some Windows utility e.g. the Intersil one ? Regards, Jar From BichotG at tce.com Wed Jan 15 15:43:27 2003 From: BichotG at tce.com (Bichot Guillaume (Princeton)) Date: Wed, 15 Jan 2003 15:43:27 -0500 Subject: Monitor mode and FCS Message-ID: <0BCAE6FBFB16D511922A0002A507DACE030C8E27@INDYEXCH24> Hi all, We are working in monitor mode with Hostap-2002-10-12 and a linksys WPC11 pccard v3. We tried the set-up the card in order to receive the frame with a wrong FCS (prism2_param wlan0 monitor_allow_fcserr 1). We then use wlansniff. The packets belong to a downlink multicast RTP stream forwarded by an access point (PC + hostap). We added in sniff the display of the RTP sequence number.Being closed to the AP, all received frames are good (status : 0000) and the RTP sequence number is sequential. In order to create some wrong frames, we move away from the access point. What we observed is that wlansniff display only good (status : 0000) frames although our RTP sequence number was not anymore continuous. We got the same result with different prism2x cards. QUESTION : Does prism2_param wlan0 monitor_allow_fcserr works with any card ? Did we missed something in our configuration ? Thanks Guillaume Bichot THOMSON -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030115/4c0d6970/attachment.htm From jt at bougret.hpl.hp.com Wed Jan 15 16:13:09 2003 From: jt at bougret.hpl.hp.com (Jean Tourrilhes) Date: Wed, 15 Jan 2003 13:13:09 -0800 Subject: Monitor mode and FCS Message-ID: <20030115211308.GA4518@bougret.hpl.hp.com> Bichot Guillaume (Princeton) wrote : > In order to create some wrong frames, we move away from the access point. > What we observed is that wlansniff display only good (status : 0000) frames > although our RTP sequence number was not anymore continuous. Slightly off topic, but this is my take on why trying to capture packet with incorrect FCS is mostly a waste of time : http://lists.shmoo.com/pipermail/hostap/2002-December/000596.html Of course, I would *love* to be proven wrong ;-) Jean From bogin at shortcircut.org Wed Jan 15 17:19:01 2003 From: bogin at shortcircut.org (Mike Bohan) Date: 15 Jan 2003 17:19:01 -0500 Subject: constant disconnection/reconnection between Netgear MA401 AP and SMC2632W V.2 client Message-ID: <1042669141.10883.39.camel@diesel> Hello there, I recently setup a Toshiba Satellite as a WAP/router for my home network. The setup uses a Netgear MA401 PrismII based card in the laptop, which is binded to the HostAP driver. The configuration was very strait forward (no special firmware needed), and it works well for the most part. A problem did emerge when I tried using a SMC2632W V.2 (Amtel chipset) based client on the network. The SMC client becomes associated on the network without a problem, and even had a very strong signal. However, every minute or so the SMC client laptop (running Windows XP) displays a network disconnection of the wireless NIC, and then reconnects. These disconnect/reconnect intervals seem to be equal, though I have not proved this assumption. The simple and most obvious answer is that the client is too far away from the WAP. However, the signal on the SMC xp client is categorized as very "very good". In fact, it's much better than any of the Prism II clients have ever been (I have several machines that access this WAP. All the rest are PrismII based). The transfer rates (when actually connected) are indicative of this too, as they are excellent. Most importantly, the other PrismII clients perform without disconnection at equal distance from the WAP with a lower signal. I have also tried using both SMC's windows utility and Windows XP's integrated "zero" configuration system. Both yield equal results. I should also mention two factors.. 1. This is the first non-prismII client I've used with this WAP system, and it's the first with problems. 2. I upgraded the firmware on the MA401 to the most recent version (standard firmware, not AP firmware) in hopes this would solve the problem. It did not. Are there known incompatabilities with non-PrismII chipsets acting as clients to the hostap driver? I'm wondering if anyone has any insights to the cause of this troublesum problem. Perhaps there is a way to tweak either the hostap driver or the smc client settings? -- Mike Bohan From bogin at shortcircut.org Wed Jan 15 17:24:33 2003 From: bogin at shortcircut.org (Mike Bohan) Date: 15 Jan 2003 17:24:33 -0500 Subject: constant disconnection/reconnection between Netgear MA401 AP and SMC2632W V.2 client In-Reply-To: <1042669141.10883.39.camel@diesel> References: <1042669141.10883.39.camel@diesel> Message-ID: <1042669473.10884.44.camel@diesel> I'm deeply sorry for the second message, but I felt I left out some important information. When the SMC client disconnects from the Linux WAP (using the netgear MA401 with hostap driver), the driver displays the following data... wlan0: 00:04:e2:46:c0:07 auth (alg=0 trans#=1 stat=0 len=6fc=00b0) ==> 0 (authenticated) wlan0: 00:04:e2:46:c0:07 reassoc (len=23 prev_ap=00:30:ab:09:4f:9e) => 0(0) (OK, old AID) wlan0: 00:04:e2:46:c0:07 auth (alg=0 trans#=1 stat=0 len=6fc=00b0) ==> 0 (authenticated) Apparently, every time the client disconnects/reconnects, a coresponding "reassoc" is issued. I thought perhaps this debug info would be useful. Thanks! On Wed, 2003-01-15 at 17:19, Mike Bohan wrote: > Hello there, > > I recently setup a Toshiba Satellite as a WAP/router for my home > network. The setup uses a Netgear MA401 PrismII based card in the > laptop, which is binded to the HostAP driver. The configuration was very > strait forward (no special firmware needed), and it works well for the > most part. > A problem did emerge when I tried using a SMC2632W V.2 (Amtel chipset) > based client on the network. The SMC client becomes associated on the > network without a problem, and even had a very strong signal. However, > every minute or so the SMC client laptop (running Windows XP) displays a > network disconnection of the wireless NIC, and then reconnects. These > disconnect/reconnect intervals seem to be equal, though I have not > proved this assumption. The simple and most obvious answer is that the > client is too far away from the WAP. However, the signal on the SMC xp > client is categorized as very "very good". In fact, it's much better > than any of the Prism II clients have ever been (I have several machines > that access this WAP. All the rest are PrismII based). The transfer > rates (when actually connected) are indicative of this too, as they are > excellent. Most importantly, the other PrismII clients perform without > disconnection at equal distance from the WAP with a lower signal. I have > also tried using both SMC's windows utility and Windows XP's integrated > "zero" configuration system. Both yield equal results. > I should also mention two factors.. > > 1. This is the first non-prismII client I've used with this WAP system, > and it's the first with problems. > 2. I upgraded the firmware on the MA401 to the most recent version > (standard firmware, not AP firmware) in hopes this would solve the > problem. It did not. > > Are there known incompatabilities with non-PrismII chipsets acting as > clients to the hostap driver? > > I'm wondering if anyone has any insights to the cause of this > troublesum problem. Perhaps there is a way to tweak either the hostap > driver or the smc client settings? -- Mike Bohan From sneeze at alphalink.com.au Wed Jan 15 19:51:18 2003 From: sneeze at alphalink.com.au (Ryan Abbenhuys) Date: Thu, 16 Jan 2003 11:51:18 +1100 Subject: constant disconnection/reconnection between Netgear MA401 AP and SMC2632W V.2 client Message-ID: <3e260206.b60.0@alphalink.com.au> This is a problem with WinXP. Apparently this is a WinXP update or service pack that fixes it. >Hello there, > > I recently setup a Toshiba Satellite as a WAP/router for my home >network. The setup uses a Netgear MA401 PrismII based card in the >laptop, which is binded to the HostAP driver. The configuration was very >strait forward (no special firmware needed), and it works well for the >most part. > A problem did emerge when I tried using a SMC2632W V.2 (Amtel chipset) >based client on the network. The SMC client becomes associated on the >network without a problem, and even had a very strong signal. However, >every minute or so the SMC client laptop (running Windows XP) displays a >network disconnection of the wireless NIC, and then reconnects. These >disconnect/reconnect intervals seem to be equal, though I have not >proved this assumption. The simple and most obvious answer is that the >client is too far away from the WAP. However, the signal on the SMC xp >client is categorized as very "very good". In fact, it's much better >than any of the Prism II clients have ever been (I have several machines >that access this WAP. All the rest are PrismII based). The transfer >rates (when actually connected) are indicative of this too, as they are >excellent. Most importantly, the other PrismII clients perform without >disconnection at equal distance from the WAP with a lower signal. I have >also tried using both SMC's windows utility and Windows XP's integrated >"zero" configuration system. Both yield equal results. > I should also mention two factors.. > >1. This is the first non-prismII client I've used with this WAP system, >and it's the first with problems. >2. I upgraded the firmware on the MA401 to the most recent version >(standard firmware, not AP firmware) in hopes this would solve the >problem. It did not. > > Are there known incompatabilities with non-PrismII chipsets acting as >clients to the hostap driver? > > I'm wondering if anyone has any insights to the cause of this >troublesum problem. Perhaps there is a way to tweak either the hostap >driver or the smc client settings? > >-- >Mike Bohan > >_______________________________________________ >HostAP mailing list >HostAP at shmoo.com >http://lists.shmoo.com/mailman/listinfo/hostap > > From bogin at shortcircut.org Wed Jan 15 20:20:12 2003 From: bogin at shortcircut.org (Mike Bohan) Date: 15 Jan 2003 20:20:12 -0500 Subject: constant disconnection/reconnection between Netgear MA401 AP and SMC2632W V.2 client In-Reply-To: <3e260206.b60.0@alphalink.com.au> References: <3e260206.b60.0@alphalink.com.au> Message-ID: <1042680011.10884.55.camel@diesel> I actually have the machine up to date with "Windows Update". This includes sp1 and all patches after that. Also, I borrowed a friend's prismII pcmcia card and used it in the same machine. This yielded a flawless connection. The one conflict I read about with win XP post sp1 is that you must check the box "Allow Connection without WEP Encryption", or something to the same effect. So unless there's a patch that is not part of Windows Update (or aware of), I don't think that's the issue unfortunately. On Wed, 2003-01-15 at 19:51, Ryan Abbenhuys wrote: > This is a problem with WinXP. > Apparently this is a WinXP update or service pack that fixes it. > > > >Hello there, > > > > I recently setup a Toshiba Satellite as a WAP/router for my home > >network. The setup uses a Netgear MA401 PrismII based card in the > >laptop, which is binded to the HostAP driver. The configuration was very > >strait forward (no special firmware needed), and it works well for the > >most part. > > A problem did emerge when I tried using a SMC2632W V.2 (Amtel chipset) > >based client on the network. The SMC client becomes associated on the > >network without a problem, and even had a very strong signal. However, > >every minute or so the SMC client laptop (running Windows XP) displays a > >network disconnection of the wireless NIC, and then reconnects. These > >disconnect/reconnect intervals seem to be equal, though I have not > >proved this assumption. The simple and most obvious answer is that the > >client is too far away from the WAP. However, the signal on the SMC xp > >client is categorized as very "very good". In fact, it's much better > >than any of the Prism II clients have ever been (I have several machines > >that access this WAP. All the rest are PrismII based). The transfer > >rates (when actually connected) are indicative of this too, as they are > >excellent. Most importantly, the other PrismII clients perform without > >disconnection at equal distance from the WAP with a lower signal. I have > >also tried using both SMC's windows utility and Windows XP's integrated > >"zero" configuration system. Both yield equal results. > > I should also mention two factors.. > > > >1. This is the first non-prismII client I've used with this WAP system, > >and it's the first with problems. > >2. I upgraded the firmware on the MA401 to the most recent version > >(standard firmware, not AP firmware) in hopes this would solve the > >problem. It did not. > > > > Are there known incompatabilities with non-PrismII chipsets acting as > >clients to the hostap driver? > > > > I'm wondering if anyone has any insights to the cause of this > >troublesum problem. Perhaps there is a way to tweak either the hostap > >driver or the smc client settings? > > > >-- > >Mike Bohan > > > >_______________________________________________ > >HostAP mailing list > >HostAP at shmoo.com > >http://lists.shmoo.com/mailman/listinfo/hostap > > > > -- Mike Bohan From jkmaline at cc.hut.fi Wed Jan 15 21:02:55 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 18:02:55 -0800 Subject: constant disconnection/reconnection between Netgear MA401 AP and SMC2632W V.2 client In-Reply-To: <1042669141.10883.39.camel@diesel> References: <1042669141.10883.39.camel@diesel> Message-ID: <20030116020255.GA1200@jm.kir.nu> On Wed, Jan 15, 2003 at 05:19:01PM -0500, Mike Bohan wrote: > every minute or so the SMC client laptop (running Windows XP) displays a > network disconnection of the wireless NIC, and then reconnects. These > disconnect/reconnect intervals seem to be equal, though I have not > proved this assumption. The simple and most obvious answer is that the Can that "every minute or so" be exactly 5 minutes? If yes, this could be related to inactivity timeout in the Host AP driver. Is there any additional log information about this event (apart from the new auth & reassoc sent by the STA), like "sending deauth/disassoc info to STA"? I'm assuming the prev_ap value in reassoc messages is the same as the AP's own MAC address(?) Other option would be some timer in WinXP expiring and causing link down event and eventual new auth/assoc. I have seen this mainly with IEEE 802.1X enabled, but not without it. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Wed Jan 15 21:06:02 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 18:06:02 -0800 Subject: FW Flashing utilities In-Reply-To: <3E25BF79.3BB285B3@pcuf.fi> References: <3E25BF79.3BB285B3@pcuf.fi> Message-ID: <20030116020602.GB1200@jm.kir.nu> On Wed, Jan 15, 2003 at 10:07:21PM +0200, jar wrote: > Is it safe to use the prism2_srec to flash the firmware or > is it safer to use some Windows utility e.g. the > Intersil one ? Well, it is quite safe to _try_ to use prism2_srec to flash the firmware since this has not been implemented in the driver and nothing really happens.. ;-). Anyway, I would recommend using tools supported by the card manufacturer. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Wed Jan 15 21:09:51 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 18:09:51 -0800 Subject: EAP-TLS and WEP together with hostapd! In-Reply-To: References: Message-ID: <20030116020951.GC1200@jm.kir.nu> On Wed, Jan 15, 2003 at 10:43:51AM +0800, David Chen wrote: > but the hostapd doesn't receive even authentication request if the individual key mapping is turned on as followed: > > hostapd -x -b5 -i5 -o192.168.168.200 -a192.168.168.220 -stest wlan0 Could you please be more specific on what packet you are refering to with "authentication request"? IEEE 802.11 authentication frame? Some IEEE 802.1X EAPOL/EAP packet? Which version of the Host AP driver are you using? What driver and supplicant are you using in the station? Have you verified whether station is really sending the frame that hostapd does not receive? -- Jouni Malinen PGP id EFC895FA From dsen at homemail.com Wed Jan 15 21:54:16 2003 From: dsen at homemail.com (D. Sen) Date: Wed, 15 Jan 2003 21:54:16 -0500 Subject: constant disconnection/reconnection between Netgear MA401 AP and SMC2632W V.2 client References: <3e260206.b60.0@alphalink.com.au> Message-ID: <3E261ED8.1010007@homemail.com> Where can I find out more about this WinXP problem? An URL? Ryan Abbenhuys wrote: > This is a problem with WinXP. > Apparently this is a WinXP update or service pack that fixes it. > > > >>Hello there, >> >> I recently setup a Toshiba Satellite as a WAP/router for my home >>network. The setup uses a Netgear MA401 PrismII based card in the >>laptop, which is binded to the HostAP driver. The configuration was very >>strait forward (no special firmware needed), and it works well for the >>most part. >> A problem did emerge when I tried using a SMC2632W V.2 (Amtel chipset) >>based client on the network. The SMC client becomes associated on the >>network without a problem, and even had a very strong signal. However, >>every minute or so the SMC client laptop (running Windows XP) displays a >>network disconnection of the wireless NIC, and then reconnects. These >>disconnect/reconnect intervals seem to be equal, though I have not >>proved this assumption. The simple and most obvious answer is that the >>client is too far away from the WAP. However, the signal on the SMC xp >>client is categorized as very "very good". In fact, it's much better >>than any of the Prism II clients have ever been (I have several machines >>that access this WAP. All the rest are PrismII based). The transfer >>rates (when actually connected) are indicative of this too, as they are >>excellent. Most importantly, the other PrismII clients perform without >>disconnection at equal distance from the WAP with a lower signal. I have >>also tried using both SMC's windows utility and Windows XP's integrated >>"zero" configuration system. Both yield equal results. >> I should also mention two factors.. >> >>1. This is the first non-prismII client I've used with this WAP system, >>and it's the first with problems. >>2. I upgraded the firmware on the MA401 to the most recent version >>(standard firmware, not AP firmware) in hopes this would solve the >>problem. It did not. >> >>Are there known incompatabilities with non-PrismII chipsets acting as >>clients to the hostap driver? >> >> I'm wondering if anyone has any insights to the cause of this >>troublesum problem. Perhaps there is a way to tweak either the hostap >>driver or the smc client settings? >> >>-- >>Mike Bohan >> >>_______________________________________________ >>HostAP mailing list >>HostAP at shmoo.com >>http://lists.shmoo.com/mailman/listinfo/hostap >> >> > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > > -- --------------------------------------- D. Sen, PhD 21 Woodmont Drive Randolph NJ 07869 Home Email: dsen at homemail.com Tel: 973 216 2326 Work Email: dsen at ieee.org Web: http://www.auditorymodels.org/~dsen From jkmaline at cc.hut.fi Wed Jan 15 22:52:47 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 19:52:47 -0800 Subject: Too slow then previous In-Reply-To: References: Message-ID: <20030116035247.GA3847@jm.kir.nu> On Tue, Dec 17, 2002 at 04:36:20PM +0900, ?vBXXZcX0-? wrote: > I thinks the software interrupt handler work too slow then previous host AP. > > the previous one is before 2002-10-12. I measure the througput test from client PC with pcmcia card > to hostap and internet speed test site... > > and the result is the previous one is about 4.5Mbps for download through ADSL line > but the current one is only 3.0 or 3.1 Mbps ... In what kind of host platform are you using Host AP driver? I did some throughput testing (mostly on Intel P4 laptop and minimal testing on embedded PowerPC) before changing the code and I did not notice any performance drop. I was still able to achieve about 6 Mbps with the new code. I'll try to confirm this with the latest version and bit more test systems at some point. PS. SpamAssassin does not seem to like one of your mail servers (RBL reports it as confirmed spam source). Therefore, don't be surprised if there is quite a delay in me reading these messages since I do not go through my spam folder that often.. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Wed Jan 15 22:57:12 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 19:57:12 -0800 Subject: hostapd help.. In-Reply-To: References: Message-ID: <20030116035712.GB3847@jm.kir.nu> On Mon, Jan 06, 2003 at 06:23:14PM +0900, ?vBXXZcX0-? wrote: > When I run hostapd as #hostapd wlan0 or #hostapd -xm wlan0 > the hostapd message always show bellow message.. and the client wireless PC > connect to AP then disconnect again and again.. Which version of the Host AP driver are you using? > handle_assoc_cb(reassoc=1) - too short payload (len=30) Could you please re-test with the latest CVS version of hostapd? I remember fixing a bug in something that might have been related to this. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Wed Jan 15 23:02:00 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 20:02:00 -0800 Subject: The IAPP In-Reply-To: References: Message-ID: <20030116040200.GC3847@jm.kir.nu> On Thu, Jan 09, 2003 at 04:42:30PM +0900, ?vBXXZcX0-? wrote: > Could I ask when the IAPP protocol fully implemented ? You can always ask, but you won't get an answer from me ;-). I try not to give any estimates on when something will be added to free software projects. > As I know iapp.c only implement about the associate request action.. isn't it ? Yes. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Wed Jan 15 23:14:46 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 20:14:46 -0800 Subject: snmp In-Reply-To: References: Message-ID: <20030116041446.GD3847@jm.kir.nu> On Tue, Jan 14, 2003 at 04:14:31PM +0900, ?vBXXZcX0-? wrote: > These day, I'm try to porting ucd-snmp code for the 802.11 MIB for snmp management. That could be useful addition and any contribution on it will be appreciated. > by the way, the 802.11 mib have defined the dot11AuthenticationResponseTimeOut as a read-write > I think It match for the REAUTH_TIMER_DEFAULT_reAuthPeriod in hostapd at eapol_sm.h file Are you sure that this 802.11 mib item would match with 802.1x value? reAuthPeriod is dot1xAuthReAuthPeriod in 802.1x MIB.. (but yes, it is read-write). dot11AuthenticationResponseTimeOut is the maximum time that STA should wait for next authentication frame. I think it is mostly for station functionality, but it could actually affect APs when shared key authentication is used (i.e., delay between frames 2 and 3). However, current Host AP driver does not implement such limit. Station data will eventually expire, but it is not exactly the same thing as this timeout. > the problem is it used as definition not variable... and I will find more definition for this kinds of > variables in 802.11 MIB.. REAUTH_TIMER_DEFAULT_reAuthPeriod is the default value assigned for sm->reauth_timer.reAuthPeriod that is a variable and can be modified.. > In this case, what is the best way to avoid future conflict in source code level ? For this particular variable, using correct mib would help in avoidinf conflicts ;-). For other items, reporting possibly missing values or conflicts early would probably be the best way to avoid wasting more time on conflicts later. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Wed Jan 15 23:38:20 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 15 Jan 2003 20:38:20 -0800 Subject: Monitor mode and FCS In-Reply-To: <0BCAE6FBFB16D511922A0002A507DACE030C8E27@INDYEXCH24> References: <0BCAE6FBFB16D511922A0002A507DACE030C8E27@INDYEXCH24> Message-ID: <20030116043820.GA4227@jm.kir.nu> On Wed, Jan 15, 2003 at 03:43:27PM -0500, Bichot Guillaume (Princeton) wrote: > We are working in monitor mode with Hostap-2002-10-12 and a linksys WPC11 > pccard v3. We tried the set-up the card in order to receive the frame with a > wrong FCS (prism2_param wlan0 monitor_allow_fcserr 1). > In order to create some wrong frames, we move away from the access point. > What we observed is that wlansniff display only good (status : 0000) frames > although our RTP sequence number was not anymore continuous. Like Jean wrote earlier, errors might occur during the frame preamble and there is not much that can be done in such cases since the receiver does not really know what kind of encoding it should expect to receive. Moving the sniffer away from the AP could produce errors during the preamble and the frame is dropped completely. > We got the same result with different prism2x cards. > > QUESTION : Does prism2_param wlan0 monitor_allow_fcserr works with any card > ? Did we missed something in our configuration ? Yes, it does work at least with the cards I use in testing. One of the easiest tests for this is to change the sniffer to incorrect channel (e.g., 5 when AP is using channel 6) and compare the number of received frames when allow_fcserr is 0 and 1. However, even the frames with FCS errors seem to come with "status 0000" in wlansniff.. This could be a bug in the driver or wlansniff, but I did not notice anything wrong in this code path during a quick look.. -- Jouni Malinen PGP id EFC895FA From B.Zierath at ebv.com Thu Jan 16 01:42:16 2003 From: B.Zierath at ebv.com (Zierath Bernd) Date: Thu, 16 Jan 2003 07:42:16 +0100 Subject: Dead card Message-ID: <51420C40A341F14499B8C5A488BEDA580440F9@EXSRV02.int.ebv.com> Hi, I updated a Prism2 (0x8002) PCMCIA 0.3.0/0.8.0 und run in the same problem. Now I'm looking for a initial firmware to get this card alive, can anyone help? CU Bernd Zierath > -----Original Message----- > From: Yon Uriarte [mailto:havanna_moon at gmx.net] > Sent: Monday, January 06, 2003 8:34 PM > To: hostap at shmoo.com > Subject: Dead card > > > Hi, > > I'm having problems with an SMC2632W (prism2) card. After > forgetting to > plug the AC-cord on the laptop, it run until its battery > drained and it > automatically shut down. Afterward, the card wont associate > with the AP > anymore. Monitor mode works (!). This was under linux. I've tested it > under windows, and it doesnt work either. Updating to sta > 1.4.9 didnt help > either (linux or windows). > > I'm not susbscribed, so please, CC me.. I'll happily provide any > information required. > > BTW, I cant get the last version of hostap to work on my kernel (fails > when assigning an IRQ), but I suspect a pcmcia-cs/kernel > version mismatch. > > TIA, regards, > yon > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From landreani at seltatel.it Thu Jan 16 09:16:44 2003 From: landreani at seltatel.it (Andreani Luca) Date: Thu, 16 Jan 2003 15:16:44 +0100 Subject: which card (in the world) ? Message-ID: <79A8E184F948A54C9BA8A3DDB7DE6E2813C2EC@SELTATOR> Hi all, I want to try the possibility of built a cheap Acess point using only a terminal card. I know it is possible to do this with the hostap driver. I spent the last week searching the right cards to buy, but it seems that all vendors are migrating from prism2 to different chips. It is impossible to find d-link 520 or 640 cards. You can find only 520+ and 640+ (non prism based) Linksys seems to use a broadcom chip now in its WMP11 card. Compaq/HP migrated to new 22Mbps chip based line of product. Is there in the world a card that is still based in prism2 chip and works as AP with the driver? Please give me an answer, Luca Andreani From BichotG at tce.com Thu Jan 16 09:24:20 2003 From: BichotG at tce.com (Bichot Guillaume (Princeton)) Date: Thu, 16 Jan 2003 09:24:20 -0500 Subject: which card (in the world) ? Message-ID: <0BCAE6FBFB16D511922A0002A507DACE030C8E2C@INDYEXCH24> The following links can help you: http://www.seattlewireless.net/index.cgi/HardwareComparison?action=SpellChec k http://www.personaltelco.net/index.cgi/Prism2Card http://www.linux-wlan.org/index.html Guillaume Bichot THOMSON > -----Original Message----- > From: Andreani Luca [mailto:landreani at seltatel.it] > Sent: Thursday, January 16, 2003 9:17 AM > To: 'hostap at shmoo.com' > Subject: which card (in the world) ? > > > Hi all, > > I want to try the possibility of built a cheap Acess point > using only a terminal card. I know it is possible to do this > with the hostap driver. I spent the last week searching the > right cards to buy, but it seems that all vendors are > migrating from prism2 to different chips. It is impossible to > find d-link 520 or 640 cards. You can find only 520+ and > 640+ (non prism based) > Linksys seems to use a broadcom chip now in its WMP11 card. > Compaq/HP migrated to new 22Mbps chip based line of product. > > Is there in the world a card that is still based in prism2 > chip and works as AP with the driver? Please give me an answer, > > Luca Andreani > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030116/c6a3d9d2/attachment.htm From kenw at keeper.org.uk Thu Jan 16 12:09:16 2003 From: kenw at keeper.org.uk (kenw) Date: Thu, 16 Jan 2003 17:09:16 +0000 Subject: hostapd does not detach from the terminal Message-ID: <3E26E73C.7080109@keeper.org.uk> Hi, When I start hostapd it does not detach from the terminal. Is this normal behaviour, or should it fork giving me the terminal back. A couple of example command line I am using are below. Simple command line: [root at linux01 hostapd]# ./hostapd wlan0 Using interface wlan0ap with hwaddr 00:80:c6:e3:5c:94 and ssid 'test' Flushing old station entries More complex command line: [root at linux01 hostapd]# ./hostapd -x -S KEBAB -o 192.168.0.254 -a 192.168.0.254 -p1645 -s secret wlan0 Using interface wlan0ap with hwaddr 00:80:c6:e3:5c:94 and ssid 'KEBAB' Flushing old station entries Thanks and best regards, Ken www.wnao.org.uk From anthony at magix.com.sg Thu Jan 16 10:10:48 2003 From: anthony at magix.com.sg (Anthony Heading) Date: Fri, 17 Jan 2003 00:10:48 +0900 Subject: Prism wireless lan card is causing time-travel Message-ID: <20030116151048.GA978@begemot.ajrh.net> Hello, I have a wireless lan card based on the Prism 2 chipset, branded by Corega, who are quite a big brand here in Japan. (details below) Whenever I install the hostap driver (http://hostap.epitest.fi/), using "modprobe hostap_plx", the module loads fine, but the whole machine's sense of time gets very confused. The problem continues even when I remove the module. Asking on the hostap list a month or two ago brought no joy, and I can't find anyone else reporting a similar problem, so I'm wondering if it's just broken hardware. The effect can be seen with a shell loop to call /bin/date. Jan-16 23:07:00.947547000 Jan-16 23:07:00.952915000 Jan-16 23:07:00.958982000 Jan-16 23:07:00.964766000 Jan-16 23:07:00.970666000 Jan-16 23:07:00.976285000 Jan-17 00:18:35.949607000 Jan-17 00:18:35.955582000 Jan-17 00:18:35.961650000 Jan-16 23:07:01.000077000 Jan-16 23:07:01.005745000 Jan-16 23:07:01.011856000 Jan-16 23:07:01.017729000 Jan-16 23:07:01.023490000 Jan-16 23:07:01.029346000 Jan-16 23:07:01.035519000 Jan-16 23:07:01.041152000 For a few milliseconds around each second, the time jumps forward more than an hour, and then resets itself. Updating the firmware on the card doesn't improve things, nor does installing the card in a different PC. Can anyone suggest what the problem might be, or how I could debug it further? Thanks! Anthony Linux begemot.ajrh.net 2.4.20 #20 SMP Thu Jan 16 23:04:42 JST 2003 i686 Intel(R) Xeon(TM) CPU 2.66GHz GenuineIntel GNU/Linux 05:0d.0 Network controller: National Datacomm Corp Wireless PCI Card (rev 02) Subsystem: Allied Telesyn International: Unknown device a113 Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- I noticed there is a hostap/inject directory in the CVS archives that contains no files. Any plans on making some example injection code available, similar to the sniffer code? Thanks. -Joshua Wright Team Leader, Networks and Systems Johnson & Wales University Joshua.Wright at jwu.edu http://home.jwu.edu/jwright/ pgpkey: http://home.jwu.edu/jwright/pgpkey.htm fingerprint: FDA5 12FC F391 3740 E0AE BDB6 8FE2 FC0A D44B 4A73 From jdl at jdl.com Thu Jan 16 13:13:28 2003 From: jdl at jdl.com (Jon Loeliger) Date: Thu, 16 Jan 2003 10:13:28 -0800 Subject: State of the sources? Message-ID: Folks, I only semi-recently started looking at the HostAP source base. I am seeing a few difficulties, and I want to eliminate the traditional stupid problems first if possible. I currently have the hostap-2002-10-12 code base being used on a Debian 2.4.20 box with some ZoomAir card of dubious heritage. I'm using pcmcia-cs and Yenta built into he kernel, with parts like the orinoco_cs and hostap driver built as separate modules. Can someone give me a feel for how much difference there is between the 2002-10-12 release and the current top-of-CVS-tree? Are there enough significant changes or bug fixes that I should grab from the CVS tree rather than re-stumble across each known 2002-10-12 bug/issue? How stable is the CVS development tree? Does it work on a daily basis, or is it occasionally broken? What is the $CVSROOT I would need to access (grab) the tree? (That's a FAQ, somwhere too, right? :-) I am reluctant to abuse the list with a huge trace file of the problems I'm seeing until I've done enough homework here to eliminate the stupid problems first... :-) Thanks, jdl From hozturk at nps.navy.mil Thu Jan 16 14:32:23 2003 From: hozturk at nps.navy.mil (Selcuk OZTURK) Date: Thu, 16 Jan 2003 11:32:23 -0800 Subject: Help 802.1X support with Freeradius Message-ID: Hi all, When I tried to authenticate a WinXP client with Freeradius , Freeradius threw a segmnentation fault and terminate. I thought framed-MTU size caused this problem , so I tried to changed it with ' iwconfig eth0 frag 1400" , nothing changed at the radius side, framed MTU was still 2304 . Could someone help me? Thanks... SECO + LD_LIBRARY_PATH=/usr/local/openssl/lib + LD_PRELOAD=/usr/local/openssl/lib/libcrypto.so + export LD_LIBRARY_PATH LD_PRELOAD + /usr/local/sbin/radiusd -X -A Starting - reading configuration files ... reread_config: reading radiusd.conf Config: including file: /etc/raddb/proxy.conf Config: including file: /etc/raddb/clients.conf Config: including file: /etc/raddb/snmp.conf Config: including file: /etc/raddb/sql.conf main: prefix = "/usr/local" main: localstatedir = "/usr/local/var" main: logdir = "/usr/local/var/log/radius" main: libdir = "/usr/local/lib" main: radacctdir = "/usr/local/var/log/radius/radacct" main: hostname_lookups = no read_config_files: reading dictionary read_config_files: reading naslist read_config_files: reading clients read_config_files: reading realms main: max_request_time = 30 main: cleanup_delay = 5 main: max_requests = 1024 main: delete_blocked_requests = 0 main: port = 0 main: allow_core_dumps = no main: log_stripped_names = no main: log_file = "/usr/local/var/log/radius/radius.log" main: log_auth = no main: log_auth_badpass = no main: log_auth_goodpass = no main: pidfile = "/usr/local/var/run/radiusd/radiusd.pid" main: user = "(null)" main: group = "(null)" main: usercollide = no main: lower_user = "no" main: lower_pass = "no" main: nospace_user = "no" main: nospace_pass = "no" main: checkrad = "/usr/local/sbin/checkrad" main: proxy_requests = yes proxy: retry_delay = 5 proxy: retry_count = 3 proxy: synchronous = no proxy: default_fallback = yes proxy: dead_time = 120 proxy: servers_per_realm = 15 security: max_attributes = 200 security: reject_delay = 1 main: debug_level = 0 read_config_files: entering modules setup Module: Library search path is /usr/local/lib Module: Loaded expr Module: Instantiated expr (expr) Module: Loaded System unix: cache = yes unix: passwd = "/etc/passwd" unix: shadow = "/etc/shadow" unix: group = "/etc/group" unix: radwtmp = "/usr/local/var/log/radius/radwtmp" unix: usegroup = no unix: cache_reload = 600 HASH: Reinitializing hash structures and lists for caching... HASH: user root found in hashtable bucket 11726 HASH: user bin found in hashtable bucket 86651 HASH: user daemon found in hashtable bucket 11668 HASH: user adm found in hashtable bucket 26466 HASH: user lp found in hashtable bucket 54068 HASH: user sync found in hashtable bucket 42895 HASH: user shutdown found in hashtable bucket 71746 HASH: user halt found in hashtable bucket 7481 HASH: user mail found in hashtable bucket 79471 HASH: user news found in hashtable bucket 5375 HASH: user uucp found in hashtable bucket 38541 HASH: user operator found in hashtable bucket 21748 HASH: user games found in hashtable bucket 47657 HASH: user gopher found in hashtable bucket 47357 HASH: user ftp found in hashtable bucket 56226 HASH: user nobody found in hashtable bucket 99723 HASH: user mailnull found in hashtable bucket 78086 HASH: user rpm found in hashtable bucket 72383 HASH: user xfs found in hashtable bucket 17213 HASH: user ntp found in hashtable bucket 21418 HASH: user rpc found in hashtable bucket 72373 HASH: user rpcuser found in hashtable bucket 552 HASH: user nfsnobody found in hashtable bucket 51830 HASH: user nscd found in hashtable bucket 36306 HASH: user ident found in hashtable bucket 40304 HASH: user radvd found in hashtable bucket 66743 HASH: user pcap found in hashtable bucket 55326 HASH: user hozturk found in hashtable bucket 95961 HASH: Stored 28 entries from /etc/passwd HASH: Stored 38 entries from /etc/group Module: Instantiated unix (unix) Module: Loaded eap eap: default_eap_type = "tls" eap: timer_expire = 60 tls: rsa_key_exchange = no tls: dh_key_exchange = yes tls: rsa_key_length = 512 tls: dh_key_length = 512 tls: verify_depth = 0 tls: CA_path = "(null)" tls: pem_file_type = yes tls: private_key_file = "/etc/1x/srv.pem" tls: certificate_file = "/etc/1x/srv.pem" tls: CA_file = "/etc/1x/root.pem" tls: private_key_password = "whatever" tls: dh_file = "/etc/1x/dh" tls: random_file = "/etc/1x/random" tls: fragment_size = 1024 tls: include_length = yes rlm_eap_tls: conf N ctx stored rlm_eap: Loaded and initialized the type tls Module: Instantiated eap (eap) Module: Loaded preprocess preprocess: huntgroups = "/etc/raddb/huntgroups" preprocess: hints = "/etc/raddb/hints" preprocess: with_ascend_hack = no preprocess: ascend_channels_per_line = 23 preprocess: with_ntdomain_hack = no preprocess: with_specialix_jetstream_hack = no preprocess: with_cisco_vsa_hack = no Module: Instantiated preprocess (preprocess) Module: Loaded realm realm: format = "suffix" realm: delimiter = "@" Module: Instantiated realm (suffix) Module: Loaded files files: usersfile = "/etc/raddb/users" files: acctusersfile = "/etc/raddb/acct_users" files: preproxy_usersfile = "/etc/raddb/preproxy_users" files: compat = "no" Module: Instantiated files (files) Module: Loaded detail detail: detailfile = "/usr/local/var/log/radius/radacct/%{Client-IP-Address}/detail-%Y%m%d" detail: detailperm = 384 detail: dirperm = 493 detail: locking = no Module: Instantiated detail (detail) Module: Loaded radutmp radutmp: filename = "/usr/local/var/log/radius/radutmp" radutmp: username = "%{User-Name}" radutmp: perm = 384 radutmp: callerid = yes Module: Instantiated radutmp (radutmp) Listening on IP address *, ports 1812/udp and 1813/udp, with proxy on 1814/udp. Ready to process requests. rad_recv: Access-Request packet from host 131.XXX.XXX.XXX:32772, id=0, length=152 User-Name = "hozturk" NAS-IP-Address = 131.xxx.xxx.xxx NAS-Port = 1 Called-Station-Id = "00-05-5D-D9-55-A5:test" Calling-Station-Id = "00-05-5D-D9-57-59" Framed-MTU = 2304 NAS-Port-Type = Wireless-802.11 Connect-Info = "CONNECT 11Mbps 802.11b" EAP-Message = "\002\001\000\014\001hozturk" Message-Authenticator = 0x2c95623afbed26eaa60af0612611fb50 modcall: entering group authorize modcall[authorize]: module "preprocess" returns ok modcall[authorize]: module "eap" returns updated rlm_realm: No '@' in User-Name = "hozturk", looking up realm NULL rlm_realm: No such realm NULL modcall[authorize]: module "suffix" returns noop users: Matched hozturk at 99 modcall[authorize]: module "files" returns ok modcall: group authorize returns updated rad_check_password: Found Auth-Type EAP auth: type "EAP" modcall: entering group authenticate rlm_eap: processing type tls modcall[authenticate]: module "eap" returns ok modcall: group authenticate returns ok Sending Access-Challenge of id 0 to 131.xxx.xxx.xxx:32772 EAP-Message = "\001\002\000\006\r " Message-Authenticator = 0x00000000000000000000000000000000 State = 0xaf3410d4e62fbb79a1568e293bc5af4ae03e263efc1e539a82a8ea158648e65ef58cbfd3 Finished request 0 Going to the next request --- Walking the entire request list --- Waking up in 6 seconds... rad_recv: Access-Request packet from host 131.xxx.xxx.xxx:32772, id=1, length=258 User-Name = "hozturk" NAS-IP-Address = 131.xxx.xxx.xxx NAS-Port = 1 Called-Station-Id = "00-05-5D-D9-55-A5:test" Calling-Station-Id = "00-05-5D-D9-57-59" Framed-MTU = 2304 NAS-Port-Type = Wireless-802.11 Connect-Info = "CONNECT 11Mbps 802.11b" EAP-Message = "\002\002\000P\r\200\000\000\000F\026\003\001\000A\001\000\000=\003\001>&>_\263\305\211#\377\030#\335]\263\025\014\221\247 \332]\211|\264\333\361XH\364S:\301\000\000\026\000\004\000\005\000\n\000\t\000d\000b\000\003\00 State = 0xaf3410d4e62fbb79a1568e293bc5af4ae03e263efc1e539a82a8ea158648e6 Message-Authenticator = 0xf016778b1cf9026ae4478d25dab79c87 modcall: entering group authorize modcall[authorize]: module "preprocess" returns ok modcall[authorize]: module "eap" returns updated rlm_realm: No '@' in User-Name = "hozturk", looking up realm NULL rlm_realm: No such realm NULL modcall[authorize]: module "suffix" returns noop users: Matched hozturk at 99 modcall[authorize]: module "files" returns ok modcall: group authorize returns updated rad_check_password: Found Auth-Type EAP auth: type "EAP" modcall: entering group authenticate rlm_eap: Request found, released from the list rlm_eap: EAP_TYPE - tls rlm_eap: processing type tls rlm_eap_tls: Length Included undefined: before/accept initialization TLS_accept: before/accept initialization ./run-radiusd: line 8: 2388 Segmentation fault /usr/local/sbin/radiusd -X From jkmaline at cc.hut.fi Thu Jan 16 22:28:23 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 16 Jan 2003 19:28:23 -0800 Subject: hostapd does not detach from the terminal In-Reply-To: <3E26E73C.7080109@keeper.org.uk> References: <3E26E73C.7080109@keeper.org.uk> Message-ID: <20030117032823.GA1096@jm.kir.nu> On Thu, Jan 16, 2003 at 05:09:16PM +0000, kenw wrote: > When I start hostapd it does not detach from the terminal. Is this > normal behaviour, or should it fork giving me the terminal back. A > couple of example command line I am using are below. It is normal behavior for the latest release, but CVS version has support for forking into background. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Thu Jan 16 22:29:54 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 16 Jan 2003 19:29:54 -0800 Subject: injection question In-Reply-To: <156AA90517398D479A3473C35CE99EDD0157A6C1@pvdexc99.jwu.edu> References: <156AA90517398D479A3473C35CE99EDD0157A6C1@pvdexc99.jwu.edu> Message-ID: <20030117032954.GB1096@jm.kir.nu> On Thu, Jan 16, 2003 at 12:40:40PM -0500, Joshua Wright wrote: > I noticed there is a hostap/inject directory in the CVS archives that contains no files. Any plans on making some example injection code available, similar to the sniffer code? There was (and is in attic, but it won't work with the current driver) an example program in that directory. The interface has since changed and hostapd has examples on how to send frames with 802.11 headers using wlan0ap interface (kernel driver compiled with PRISM2_HOSTAPD defined). -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Thu Jan 16 22:34:38 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 16 Jan 2003 19:34:38 -0800 Subject: State of the sources? In-Reply-To: References: Message-ID: <20030117033438.GC1096@jm.kir.nu> On Thu, Jan 16, 2003 at 10:13:28AM -0800, Jon Loeliger wrote: > Can someone give me a feel for how much difference there is > between the 2002-10-12 release and the current top-of-CVS-tree? A lot.. > Are there enough significant changes or bug fixes that I should > grab from the CVS tree rather than re-stumble across each > known 2002-10-12 bug/issue? Yes. > How stable is the CVS development tree? Does it work on a > daily basis, or is it occasionally broken? I try to keep it always in working condition, i.e., I do test my changes before committing; well. However, I'm usually testing only one configuration (at the moment Linux 2.5.x and hostapd) for CVS commits and I may miss some changes that break, e.g., Linux 2.4 compilation. These are normally fixed as soon as someone reports them. > What is the $CVSROOT I would need to access (grab) the tree? > (That's a FAQ, somwhere too, right? :-) Yes, see http://hostap.epitest.fi/ for alternative methods of accessing CVS tree. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Thu Jan 16 23:19:15 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Thu, 16 Jan 2003 20:19:15 -0800 Subject: Help 802.1X support with Freeradius In-Reply-To: References: Message-ID: <20030117041915.GA2555@jm.kir.nu> On Thu, Jan 16, 2003 at 11:32:23AM -0800, Selcuk OZTURK wrote: > When I tried to authenticate a WinXP client with Freeradius , Freeradius > threw a segmnentation fault and terminate. I thought framed-MTU size caused > this problem , so I tried to changed it with ' iwconfig eth0 frag 1400" , > nothing changed at the radius side, framed MTU was still 2304 . Which version of FreeRADIUS and OpenSSL are you using? Why do you think Framed-MTU would cause this problem? I do not understand how that iwconfig command is related to this. Changing 802.11 fragmentation limit does not have any effect on MTU.. Do you mean 'ifconfig wlan0 mtu 1400'? hostapd sends EAPOL frames using wlan0ap interface. You should also note that 2304 is hardcoded value for Framed-MTU in the current hostapd implementation, so it cannot really be changed without editing source code (ieee802_1x.c).. > rad_recv: Access-Request packet from host 131.xxx.xxx.xxx:32772, id=1, > length=258 User-Name = "hozturk" > NAS-IP-Address = 131.xxx.xxx.xxx > NAS-Port = 1 > Called-Station-Id = "00-05-5D-D9-55-A5:test" > Calling-Station-Id = "00-05-5D-D9-57-59" > Framed-MTU = 2304 ... so radiusd successfully processed some of the packets. > rlm_eap: Request found, released from the list > rlm_eap: EAP_TYPE - tls > rlm_eap: processing type tls > rlm_eap_tls: Length Included > undefined: before/accept initialization > TLS_accept: before/accept initialization > ./run-radiusd: line 8: 2388 Segmentation fault /usr/local/sbin/radiusd > -X Sounds like a bug in FreeRADIUS TLS implementation (it is marked as experimental) or some sort of mismatch with openssl version etc. -- Jouni Malinen PGP id EFC895FA From vda at port.imtp.ilyichevsk.odessa.ua Fri Jan 17 03:08:09 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Fri, 17 Jan 2003 10:08:09 +0200 Subject: DLink DWL-520 + hostap in STA mode Message-ID: <200301170808.h0H88gs20786@Port.imtp.ilyichevsk.odessa.ua> DLink DWL-520 2.4.20-pre11, wireless extensions 14, hostap-2002-10-12 We (my father and me) made a fairly long helical aerial. We are trying to communicate over ~15 km with a small wireless cell. (~10 hosts, one AP). Mainline kernel's orinoco_pci module works but seems to lose an association within ~30 secs and is not able to restore it without ifconfig down/up cycle. Win98 driver fares better but dies too after ~3 mins. I am trying hostap now. It loads without problems but does not work, I am unable to switch the node into STA mode. Looks like card's firmware is old. Where can I find newer one? (I'll go googling to try and find it myself...) -- vda modprobe hostap_pci =================== kernel: hostap_crypt: registered algorithm 'NULL' kernel: hostap_pci: hostap_pci.c 0.0.0 2002-10-12 (SSH Communications Security Corp, Jouni Malinen) kernel: hostap_pci: (c) Jouni Malinen kernel: PCI: Found IRQ 11 for device 00:0f.0 kernel: PCI: Sharing IRQ 11 with 00:07.5 *** 00:07.5 is an onboard sound. Not used (sound modules aren't loaded) kernel: hostap_pci: Registered netdevice wlan0 kernel: prism2_hw_init() kernel: prism2_hw_config: initialized in 19879 iterations kernel: wlan0: NIC: id=0x8013 v1.0.0 kernel: wlan0: PRI: id=0x15 v1.0.7 kernel: wlan0: STA: id=0x1f v1.3.6 *** This looks like firmware revisions. Firmware is old. Where can I find newer ones? kernel: wlan0: defaulting to host-based encryption as a workaround for firmware bug in Host AP mode WEP kernel: wlan0: LinkStatus=2 (Disconnected) kernel: wlan0: Intersil Prism2.5 PCI: mem=0xd8000000, irq=11 iwconfig wlan0 mode monitor 3 (forgot to capture exact error message, syslog only :( ) ======================================================== kernel: prism2: wlan0: operating mode changed 3 -> 6 kernel: Enabling monitor mode kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf6d4854, type=0, res=0) kernel: wlan0: hfa384x_cmd: command was not completed (res=0, entry=cf6d4854, type=0, cmd=0x0002, param0=0x0000) kernel: Setting monitor mode failed kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf6d4854, type=0, res=0) kernel: wlan0: hfa384x_cmd: command was not completed (res=0, entry=cf6d4854, type=0, cmd=0x0121, param0=0xfc00) kernel: wlan0: hfa384x_set_rid: CMDCODE_ACCESS_WRITE failed (res=-110, rid=fc00, len=2) I rebooted and decided to go straight to switching the card in Managed mode. No luck: iwconfig wlan0 mode managed =========================== Error for wireless request "Set Mode" (8B06) : SET failed on device wlan0 ; Invalid argument. in syslog: kernel: prism2: wlan0: operating mode changed 3 -> 2 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61089c, type=0, res=0) kernel: wlan0: hfa384x_cmd: command was not completed (res=0, entry=cf61089c, type=0, cmd=0x0002, param0=0x0000) kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61089c, type=0, res=0) kernel: wlan0: hfa384x_cmd: command was not completed (res=0, entry=cf61089c, type=0, cmd=0x0021, param0=0xfdc6) kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fdc6, len=12) kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61089c, type=0, res=-1) kernel: wlan0: hfa384x_cmd: interrupted; err=-110 kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fdc6, len=12) kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61089c, type=0, res=-1) kernel: wlan0: hfa384x_cmd: interrupted; err=-110 kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fdc1, len=2) kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61089c, type=0, res=-1) kernel: wlan0: hfa384x_cmd: interrupted; err=-110 kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc06, len=2) kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61080c, type=0, res=-1) kernel: wlan0: hfa384x_cmd: interrupted; err=-110 kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd41, len=34) kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61089c, type=0, res=-1) kernel: wlan0: hfa384x_cmd: interrupted; err=-110 kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd42, len=6) kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61080c, type=0, res=-1) kernel: wlan0: hfa384x_cmd: interrupted; err=-110 kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc0e, len=34) kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 kernel: wlan0: hfa384x_cmd: entry still in list? (entry=cf61089c, type=0, res=-1) kernel: wlan0: hfa384x_cmd: interrupted; err=-110 kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc84, len=2) *** this continues until I rmmod From Joshua.Wright at jwu.edu Fri Jan 17 09:20:05 2003 From: Joshua.Wright at jwu.edu (Joshua Wright) Date: Fri, 17 Jan 2003 09:20:05 -0500 Subject: Trouble with enGenius card Message-ID: <156AA90517398D479A3473C35CE99EDD0195AB9A@pvdexc99.jwu.edu> I recently ordered a 100mW enGenius PrismII card and am having trouble using it with the HostAP drivers. The drivers modprobe without a problem and I can configure the card with the standard iwconfig and prism2_param tools, but dmesg keeps reporting: wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa287 ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa32b ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa7f1 ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa9e8 ScanReason=2 wlan0: LinkStatus=5 (Access point in range) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: JoinRequest: BSSID=00:e0:63:82:19:c6 channel=11 wlan0: LinkStatus=6 (Association failed) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa88f ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa89d ScanReason=2 wlan0: LinkStatus=4 (Access point out of range) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=6 (Association failed) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0x0000 ScanReason=2 wlan0: LinkStatus=5 (Access point in range) wlan0: LinkStatus: BSSID=44:44:44:44:44:44 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0x5c47 ScanReason=2 wlan0: LinkStatus=4 (Access point out of range) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) "iwconfig wlan0" seems to indicate I am associated with my AP (which is very close to my desk) but I keep seeing the LED on the enGenius card flashing which I presume means it is diassociating and scanning for a new connection. Any thoughts on making this card work? I am using the current CVS HostAP drivers with a 2.4.19 kernel, wireless tools 25 and wireless extensions 14. Thanks. -Joshua Wright Team Leader, Networks and Systems Johnson & Wales University Joshua.Wright at jwu.edu http://home.jwu.edu/jwright/ pgpkey: http://home.jwu.edu/jwright/pgpkey.htm fingerprint: FDA5 12FC F391 3740 E0AE BDB6 8FE2 FC0A D44B 4A73 iwconfig: wlan0 IEEE 802.11b ESSID:"JWU" Mode:Managed Frequency:2.422GHz Access Point: 00:E0:63:82:0D:27 Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Power Management:off Link Quality:0/92 Signal level:-90 dBm Noise level:-100 dBm Rx invalid nwid:0 Rx invalid crypt:39 Rx invalid frag:0 Tx excessive retries:23 Invalid misc:3912 Missed beacon:0 wlan0ap IEEE 802.11b ESSID:"JWU" Mode:Managed Frequency:2.422GHz Access Point: 00:E0:63:82:0D:27 Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Power Management:off wlan0sta IEEE 802.11b ESSID:"JWU" Mode:Managed Frequency:2.422GHz Access Point: 00:00:00:00:00:00 Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Power Management:off From wireless at douglasheld.net Fri Jan 17 13:23:05 2003 From: wireless at douglasheld.net (wireless at douglasheld.net) Date: Fri, 17 Jan 2003 10:23:05 -0800 Subject: Trouble with enGenius card References: <156AA90517398D479A3473C35CE99EDD0195AB9A@pvdexc99.jwu.edu> Message-ID: <000401c2be55$7e3eaed0$0200a8c0@buttrump> Darn, I just bought one of those... Joshua, can you clarify: Is your enGenius functioning as as station or as the AP? --Doug ----- Original Message ----- From: "Joshua Wright" To: Sent: Friday, January 17, 2003 6:20 AM Subject: Trouble with enGenius card I recently ordered a 100mW enGenius PrismII card and am having trouble using it with the HostAP drivers. The drivers modprobe without a problem and I can configure the card with the standard iwconfig and prism2_param tools, but dmesg keeps reporting: wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa287 ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa32b ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa7f1 ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa9e8 ScanReason=2 wlan0: LinkStatus=5 (Access point in range) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: JoinRequest: BSSID=00:e0:63:82:19:c6 channel=11 wlan0: LinkStatus=6 (Association failed) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa88f ScanReason=2 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0xa89d ScanReason=2 wlan0: LinkStatus=4 (Access point out of range) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=6 (Association failed) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0x0000 ScanReason=2 wlan0: LinkStatus=5 (Access point in range) wlan0: LinkStatus: BSSID=44:44:44:44:44:44 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: ScanResults: Reserved=0x5c47 ScanReason=2 wlan0: LinkStatus=4 (Access point out of range) wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 wlan0: LinkStatus=1 (Connected) "iwconfig wlan0" seems to indicate I am associated with my AP (which is very close to my desk) but I keep seeing the LED on the enGenius card flashing which I presume means it is diassociating and scanning for a new connection. Any thoughts on making this card work? I am using the current CVS HostAP drivers with a 2.4.19 kernel, wireless tools 25 and wireless extensions 14. Thanks. -Joshua Wright Team Leader, Networks and Systems Johnson & Wales University Joshua.Wright at jwu.edu http://home.jwu.edu/jwright/ pgpkey: http://home.jwu.edu/jwright/pgpkey.htm fingerprint: FDA5 12FC F391 3740 E0AE BDB6 8FE2 FC0A D44B 4A73 iwconfig: wlan0 IEEE 802.11b ESSID:"JWU" Mode:Managed Frequency:2.422GHz Access Point: 00:E0:63:82:0D:27 Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Power Management:off Link Quality:0/92 Signal level:-90 dBm Noise level:-100 dBm Rx invalid nwid:0 Rx invalid crypt:39 Rx invalid frag:0 Tx excessive retries:23 Invalid misc:3912 Missed beacon:0 wlan0ap IEEE 802.11b ESSID:"JWU" Mode:Managed Frequency:2.422GHz Access Point: 00:E0:63:82:0D:27 Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Power Management:off wlan0sta IEEE 802.11b ESSID:"JWU" Mode:Managed Frequency:2.422GHz Access Point: 00:00:00:00:00:00 Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Power Management:off _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap From Joshua.Wright at jwu.edu Fri Jan 17 11:21:09 2003 From: Joshua.Wright at jwu.edu (Joshua Wright) Date: Fri, 17 Jan 2003 11:21:09 -0500 Subject: Trouble with enGenius card Message-ID: <156AA90517398D479A3473C35CE99EDD0157A6CE@pvdexc99.jwu.edu> I was using it in managed mode (STA). After a little more analysis, I'm thinking that the enGenius card doesn't actually have a built-in antenna at all - just the MMCX connectors. That may be my problem. :) -Joshua Wright Team Leader, Networks and Systems Johnson & Wales University Joshua.Wright at jwu.edu http://home.jwu.edu/jwright/ pgpkey: http://home.jwu.edu/jwright/pgpkey.htm fingerprint: FDA5 12FC F391 3740 E0AE BDB6 8FE2 FC0A D44B 4A73 > -----Original Message----- > From: wireless at douglasheld.net [mailto:wireless at douglasheld.net] > Sent: Friday, January 17, 2003 1:23 PM > To: Joshua Wright; hostap at shmoo.com > Subject: Re: Trouble with enGenius card > > > Darn, I just bought one of those... > > Joshua, can you clarify: Is your enGenius functioning as as > station or as > the AP? > > --Doug > > ----- Original Message ----- > From: "Joshua Wright" > To: > Sent: Friday, January 17, 2003 6:20 AM > Subject: Trouble with enGenius card > > > I recently ordered a 100mW enGenius PrismII card and am > having trouble using > it with the HostAP drivers. > > The drivers modprobe without a problem and I can configure > the card with the > standard iwconfig and prism2_param tools, but dmesg keeps reporting: > > wlan0: LinkStatus=1 (Connected) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0xa287 ScanReason=2 > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > wlan0: LinkStatus=1 (Connected) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0xa32b ScanReason=2 > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > wlan0: LinkStatus=1 (Connected) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0xa7f1 ScanReason=2 > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > wlan0: LinkStatus=1 (Connected) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0xa9e8 ScanReason=2 > wlan0: LinkStatus=5 (Access point in range) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: JoinRequest: BSSID=00:e0:63:82:19:c6 channel=11 > wlan0: LinkStatus=6 (Association failed) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0xa88f ScanReason=2 > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > wlan0: LinkStatus=1 (Connected) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0xa89d ScanReason=2 > wlan0: LinkStatus=4 (Access point out of range) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > wlan0: LinkStatus=6 (Association failed) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0x0000 ScanReason=2 > wlan0: LinkStatus=5 (Access point in range) > wlan0: LinkStatus: BSSID=44:44:44:44:44:44 > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > wlan0: LinkStatus=1 (Connected) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: ScanResults: Reserved=0x5c47 ScanReason=2 > wlan0: LinkStatus=4 (Access point out of range) > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > wlan0: LinkStatus=1 (Connected) > > > "iwconfig wlan0" seems to indicate I am associated with my AP > (which is very > close to my desk) but I keep seeing the LED on the enGenius > card flashing > which I presume means it is diassociating and scanning for a > new connection. > > Any thoughts on making this card work? I am using the > current CVS HostAP > drivers with a 2.4.19 kernel, wireless tools 25 and wireless > extensions 14. > > Thanks. > > -Joshua Wright > Team Leader, Networks and Systems > Johnson & Wales University > Joshua.Wright at jwu.edu > http://home.jwu.edu/jwright/ > > pgpkey: http://home.jwu.edu/jwright/pgpkey.htm > fingerprint: FDA5 12FC F391 3740 E0AE BDB6 8FE2 FC0A D44B 4A73 > > > > > iwconfig: > > wlan0 IEEE 802.11b ESSID:"JWU" > Mode:Managed Frequency:2.422GHz Access Point: > 00:E0:63:82:0D:27 > Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 > Retry min limit:8 RTS thr:off Fragment thr:off > Power Management:off > Link Quality:0/92 Signal level:-90 dBm Noise > level:-100 dBm > Rx invalid nwid:0 Rx invalid crypt:39 Rx invalid frag:0 > Tx excessive retries:23 Invalid misc:3912 Missed beacon:0 > > wlan0ap IEEE 802.11b ESSID:"JWU" > Mode:Managed Frequency:2.422GHz Access Point: > 00:E0:63:82:0D:27 > Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 > Retry min limit:8 RTS thr:off Fragment thr:off > Power Management:off > > wlan0sta IEEE 802.11b ESSID:"JWU" > Mode:Managed Frequency:2.422GHz Access Point: > 00:00:00:00:00:00 > Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 > Retry min limit:8 RTS thr:off Fragment thr:off > Power Management:off > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > > From cas at taz.net.au Fri Jan 17 11:38:00 2003 From: cas at taz.net.au (Craig Sanders) Date: Sat, 18 Jan 2003 03:38:00 +1100 Subject: hostap and compaq wl-200 Message-ID: <20030117163800.GP27132@taz.net.au> ok, i'm stumped. i've tried three different versions of the hostap drivers (2002.09.12, 2002.10.12, and the CVS tarball fetched about an hour ago). i get the same problem with them all. kernel version is 2.4.20 wireless extensions patches v15 and v16 applied. (same problem without the patches) pcmcia-cs is 3.2.2 (i've also tried 3.1.33) wireless tools v25 dist is debian (unstable, from my local mirror last updated about a week ago) the hardware is an Intel BX motherboard with a P2-233 and 128MB. 1 ethernet card, and 2 compaq WL200 wireless cards. (i get the same problem with only one card installed). any clues or pointers in the right direction would be appreciated. dmesg shows pcmcia-cs starts up OK: Linux PCMCIA Card Services 3.2.2 kernel build: 2.4.20 unknown options: [pci] [cardbus] [apm] Intel ISA/PCI/CardBus PCIC probe: PCI: Found IRQ 11 for device 00:0d.0 PCI: Sharing IRQ 11 with 01:00.0 Cirrus PD6729 rev 00 PCI-to-PCMCIA at slot 00:0d, port 0x1430 host opts [0]: [ring] [1/5/0] [1/20/0] host opts [1]: [ring] [1/5/0] [1/20/0] ISA irqs (default) = 3,4,5,7,12 polling interval = 1000 ms PCI: Found IRQ 10 for device 00:0e.0 Cirrus PD6729 rev 00 PCI-to-PCMCIA at slot 00:0e, port 0x1434 host opts [0]: [ring] [1/5/0] [1/20/0] host opts [1]: [ring] [1/5/0] [1/20/0] ISA irqs (default) = 3,4,5,7,12 polling interval = 1000 ms cs: memory probe 0x0d0000-0x0dffff: clean. the hostap driver loads ok. it recognises but fails to initialise the cards. looks like it can't get or set any config data from them. i suspect (but am far from sure) that it may be some IRQ weirdness. note that pcmcia-cs above detects IRQs 11 and 10, while hostap_cs below ends up using IRQs 3 and 5. or this may be perfectly normal for these pci-to-pcmcia adaptors. dunno. hostap_crypt: registered algorithm 'NULL' hostap_cs: 0.0.0 CVS (Jouni Malinen ) hostap_cs: setting Vcc=33 (constant) hostap_cs: CS_EVENT_CARD_INSERTION prism2_config() hostap_cs: setting Vcc=33 (from config) Checking CFTABLE_ENTRY 0x01 (default 0x01) Config has no IRQ info, but trying to enable IRQ anyway.. IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 io->flags = 0x0046, io.base=0x0000, len=64 cs: IO port probe 0x0100-0x04ff: excluding 0x378-0x37f 0x3c0-0x3df 0x4d0-0x4d7 cs: IO port probe 0x0380-0x03bf: clean. cs: IO port probe 0x03e0-0x04cf: clean. cs: IO port probe 0x04d8-0x04ff: clean. cs: IO port probe 0x0800-0x08ff: clean. cs: IO port probe 0x0a00-0x0aff: clean. cs: IO port probe 0x0c00-0x0cff: clean. hostap_cs: index 0x01: Vcc 3.3, irq 3, io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 prism2_hw_init() prism2_hw_config: initialized in 9168 iterations wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) wlan0: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd0b) wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd0b, len=8) Could not get RID for component NIC wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) wlan0: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd02) wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd02, len=8) Could not get RID for component PRI wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) Could not get RID for component STA wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc01, len=6) wlan0: could not get own MAC address wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_set_rid: CMDCODE_ACCESS_WRITE failed (res=-110, rid=fc00, len=2) wlan0: Port type setting to 6 failed wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: MAC port 0 enabling failed same thing for the wlan1 card. hostap_cs: setting Vcc=33 (constant) hostap_cs: CS_EVENT_CARD_INSERTION prism2_config() hostap_cs: setting Vcc=33 (from config) Checking CFTABLE_ENTRY 0x01 (default 0x01) Config has no IRQ info, but trying to enable IRQ anyway.. IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 io->flags = 0x0046, io.base=0x0000, len=64 hostap_cs: index 0x01: Vcc 3.3, irq 5, io 0x0140-0x017f hostap_cs: Registered netdevice wlan1 prism2_hw_init() prism2_hw_config: initialized in 9171 iterations wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) wlan1: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd0b) wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd0b, len=8) Could not get RID for component NIC wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) wlan1: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd02) wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd02, len=8) Could not get RID for component PRI wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan1: hfa384x_cmd_issue - timeout - reg=0x8021 wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan1: hfa384x_cmd: interrupted; err=-110 wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) Could not get RID for component STA wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan1: hfa384x_cmd_issue - timeout - reg=0x8021 wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan1: hfa384x_cmd: interrupted; err=-110 wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan1: hfa384x_cmd_issue - timeout - reg=0x8021 wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan1: hfa384x_cmd: interrupted; err=-110 wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc01, len=6) wlan1: could not get own MAC address wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan1: hfa384x_cmd_issue - timeout - reg=0x8000 wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan1: hfa384x_cmd: interrupted; err=-110 wlan1: hfa384x_set_rid: CMDCODE_ACCESS_WRITE failed (res=-110, rid=fc00, len=2) wlan1: Port type setting to 6 failed wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan1: hfa384x_cmd_issue - timeout - reg=0x8000 wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) wlan1: hfa384x_cmd: interrupted; err=-110 wlan1: MAC port 0 enabling failed not surprisingly, any attempt to use iwconfig results in the same errors: hanuman:~# iwconfig wlan0 essid foo Error for wireless request "Set ESSID" (8B1A) : SET failed on device wlan0 ; Invalid argument. ... wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc09, len=2) wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 wlan0: hfa384x_cmd: entry still in list? (entry=c74bb7a0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 wlan0: hfa384x_cmd: entry still in list? (entry=c74bb7a0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd48, len=2) wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 wlan0: hfa384x_cmd: entry still in list? (entry=c74bb7a0, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd43, len=6) wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 wlan0: hfa384x_cmd: entry still in list? (entry=c74bb760, type=0, res=-1) wlan0: hfa384x_cmd: interrupted; err=-110 wlan0: hfa384x_set_rid: CMDCODE_ACCESS_WRITE failed (res=-110, rid=fc02, len=34) thanks for any help, suggestions, or ideas. craig -- craig sanders Fabricati Diem, PVNC. -- motto of the Ankh-Morpork City Watch From jwickman at tawi.fi Fri Jan 17 11:54:35 2003 From: jwickman at tawi.fi (Joachim Wickman) Date: Fri, 17 Jan 2003 18:54:35 +0200 Subject: hostap and compaq wl-200 References: <20030117163800.GP27132@taz.net.au> Message-ID: <002501c2be49$1f03d4b0$010aa8c0@vasa> I had some familiar problems, don't remember exactly how it did behave. But.. putting PCIC_OPTS="irq_mode=1" in rc.pcmcia did the trick. Maybe samething with yours. // Joachim ----- Original Message ----- From: "Craig Sanders" To: Sent: Friday, January 17, 2003 6:38 PM Subject: hostap and compaq wl-200 > ok, i'm stumped. i've tried three different versions of the hostap > drivers (2002.09.12, 2002.10.12, and the CVS tarball fetched about an > hour ago). i get the same problem with them all. > > kernel version is 2.4.20 > wireless extensions patches v15 and v16 applied. (same problem without the patches) > pcmcia-cs is 3.2.2 (i've also tried 3.1.33) > wireless tools v25 > dist is debian (unstable, from my local mirror last updated about a week ago) > > > the hardware is an Intel BX motherboard with a P2-233 and 128MB. 1 > ethernet card, and 2 compaq WL200 wireless cards. (i get the same > problem with only one card installed). > > > any clues or pointers in the right direction would be appreciated. > > > dmesg shows pcmcia-cs starts up OK: > > Linux PCMCIA Card Services 3.2.2 > kernel build: 2.4.20 unknown > options: [pci] [cardbus] [apm] > Intel ISA/PCI/CardBus PCIC probe: > PCI: Found IRQ 11 for device 00:0d.0 > PCI: Sharing IRQ 11 with 01:00.0 > Cirrus PD6729 rev 00 PCI-to-PCMCIA at slot 00:0d, port 0x1430 > host opts [0]: [ring] [1/5/0] [1/20/0] > host opts [1]: [ring] [1/5/0] [1/20/0] > ISA irqs (default) = 3,4,5,7,12 polling interval = 1000 ms > PCI: Found IRQ 10 for device 00:0e.0 > Cirrus PD6729 rev 00 PCI-to-PCMCIA at slot 00:0e, port 0x1434 > host opts [0]: [ring] [1/5/0] [1/20/0] > host opts [1]: [ring] [1/5/0] [1/20/0] > ISA irqs (default) = 3,4,5,7,12 polling interval = 1000 ms > cs: memory probe 0x0d0000-0x0dffff: clean. > > > the hostap driver loads ok. it recognises but fails to initialise the > cards. looks like it can't get or set any config data from them. > > i suspect (but am far from sure) that it may be some IRQ weirdness. > note that pcmcia-cs above detects IRQs 11 and 10, while hostap_cs below > ends up using IRQs 3 and 5. or this may be perfectly normal for these > pci-to-pcmcia adaptors. dunno. > > > hostap_crypt: registered algorithm 'NULL' > hostap_cs: 0.0.0 CVS (Jouni Malinen ) > hostap_cs: setting Vcc=33 (constant) > hostap_cs: CS_EVENT_CARD_INSERTION > prism2_config() > hostap_cs: setting Vcc=33 (from config) > Checking CFTABLE_ENTRY 0x01 (default 0x01) > Config has no IRQ info, but trying to enable IRQ anyway.. > IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 > io->flags = 0x0046, io.base=0x0000, len=64 > cs: IO port probe 0x0100-0x04ff: excluding 0x378-0x37f 0x3c0-0x3df 0x4d0-0x4d7 > cs: IO port probe 0x0380-0x03bf: clean. > cs: IO port probe 0x03e0-0x04cf: clean. > cs: IO port probe 0x04d8-0x04ff: clean. > cs: IO port probe 0x0800-0x08ff: clean. > cs: IO port probe 0x0a00-0x0aff: clean. > cs: IO port probe 0x0c00-0x0cff: clean. > hostap_cs: index 0x01: Vcc 3.3, irq 3, io 0x0100-0x013f > hostap_cs: Registered netdevice wlan0 > prism2_hw_init() > prism2_hw_config: initialized in 9168 iterations > wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) > wlan0: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd0b) > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd0b, len=8) > Could not get RID for component NIC > wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) > wlan0: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd02) > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd02, len=8) > Could not get RID for component PRI > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 > wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) > Could not get RID for component STA > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 > wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8021 > wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc01, len=6) > wlan0: could not get own MAC address > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_set_rid: CMDCODE_ACCESS_WRITE failed (res=-110, rid=fc00, len=2) > wlan0: Port type setting to 6 failed > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan0: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: MAC port 0 enabling failed > > > same thing for the wlan1 card. > > > hostap_cs: setting Vcc=33 (constant) > hostap_cs: CS_EVENT_CARD_INSERTION > prism2_config() > hostap_cs: setting Vcc=33 (from config) > Checking CFTABLE_ENTRY 0x01 (default 0x01) > Config has no IRQ info, but trying to enable IRQ anyway.. > IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 > io->flags = 0x0046, io.base=0x0000, len=64 > hostap_cs: index 0x01: Vcc 3.3, irq 5, io 0x0140-0x017f > hostap_cs: Registered netdevice wlan1 > prism2_hw_init() > prism2_hw_config: initialized in 9171 iterations > wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) > wlan1: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd0b) > wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd0b, len=8) > Could not get RID for component NIC > wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=0) > wlan1: hfa384x_cmd: command was not completed (res=0, entry=c4074da0, type=0, cmd=0x0021, param0=0xfd02) > wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd02, len=8) > Could not get RID for component PRI > wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan1: hfa384x_cmd_issue - timeout - reg=0x8021 > wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan1: hfa384x_cmd: interrupted; err=-110 > wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) > Could not get RID for component STA > wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan1: hfa384x_cmd_issue - timeout - reg=0x8021 > wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan1: hfa384x_cmd: interrupted; err=-110 > wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd20, len=8) > wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan1: hfa384x_cmd_issue - timeout - reg=0x8021 > wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan1: hfa384x_cmd: interrupted; err=-110 > wlan1: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc01, len=6) > wlan1: could not get own MAC address > wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan1: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan1: hfa384x_cmd: interrupted; err=-110 > wlan1: hfa384x_set_rid: CMDCODE_ACCESS_WRITE failed (res=-110, rid=fc00, len=2) > wlan1: Port type setting to 6 failed > wlan1: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan1: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan1: hfa384x_cmd: entry still in list? (entry=c4074da0, type=0, res=-1) > wlan1: hfa384x_cmd: interrupted; err=-110 > wlan1: MAC port 0 enabling failed > > > > > not surprisingly, any attempt to use iwconfig results in the same > errors: > > hanuman:~# iwconfig wlan0 essid foo > Error for wireless request "Set ESSID" (8B1A) : > SET failed on device wlan0 ; Invalid argument. > > > ... > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fc09, len=2) > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan0: hfa384x_cmd: entry still in list? (entry=c74bb7a0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan0: hfa384x_cmd: entry still in list? (entry=c74bb7a0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd48, len=2) > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan0: hfa384x_cmd: entry still in list? (entry=c74bb7a0, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd43, len=6) > wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec > wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 > wlan0: hfa384x_cmd: entry still in list? (entry=c74bb760, type=0, res=-1) > wlan0: hfa384x_cmd: interrupted; err=-110 > wlan0: hfa384x_set_rid: CMDCODE_ACCESS_WRITE failed (res=-110, rid=fc02, len=34) > > > > > thanks for any help, suggestions, or ideas. > > > craig > > > -- > craig sanders > > Fabricati Diem, PVNC. > -- motto of the Ankh-Morpork City Watch > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From cas at taz.net.au Fri Jan 17 12:18:08 2003 From: cas at taz.net.au (Craig Sanders) Date: Sat, 18 Jan 2003 04:18:08 +1100 Subject: hostap and compaq wl-200 In-Reply-To: <002501c2be49$1f03d4b0$010aa8c0@vasa> References: <20030117163800.GP27132@taz.net.au> <002501c2be49$1f03d4b0$010aa8c0@vasa> Message-ID: <20030117171808.GQ27132@taz.net.au> On Fri, Jan 17, 2003 at 06:54:35PM +0200, Joachim Wickman wrote: > I had some familiar problems, don't remember exactly how it did > behave. But.. putting PCIC_OPTS="irq_mode=1" in rc.pcmcia did the > trick. Maybe samething with yours. thanks, looks like that did the trick. although in debian, the file to edit is /etc/default/pcmcia rather than rc.pcmcia craig -- craig sanders Fabricati Diem, PVNC. -- motto of the Ankh-Morpork City Watch From raghu at sraamerica.com Fri Jan 17 14:19:35 2003 From: raghu at sraamerica.com (Raghu) Date: Fri, 17 Jan 2003 14:19:35 -0500 Subject: wlansniff setup Message-ID: <20030117191935.RNNA2474.imta01a2.registeredsite.com@smtp.registeredsite.com> Hi, I am using HostAP(STA) & HostAP(AP). Can someone please let me know how to setup in monitoring mode. Basically I should be able to parse all the 802.11 packets that are sent over the air between STA & AP. I tried the following (at both AP & STA) cd sniff make ./wlansniff -b0 wlan0 Unsupported arptype 0x0001 I searched the mailing list and didnot find any matches. I think I am missing some basic setup. Can someone please point me in the right direction. -Raghu From BichotG at tce.com Fri Jan 17 14:40:12 2003 From: BichotG at tce.com (Bichot Guillaume (Princeton)) Date: Fri, 17 Jan 2003 14:40:12 -0500 Subject: wlansniff setup Message-ID: <0BCAE6FBFB16D511922A0002A507DACE030C8E31@INDYEXCH24> Raghu, Basically : iwconfig wlan0 mode "Monitor" And to select the type of header included in the monitored frames: Prism2_param wlan0 monitor_type (0: 802.11 / 1: prism2 + 802.11) If it does not work (wireless tools < v25) then Iwpriv wlan0 monitor ((0: off /2: 802.11 / 3: prism2 + 802.11) Guillaume > -----Original Message----- > From: Raghu [mailto:raghu at sraamerica.com] > Sent: Friday, January 17, 2003 2:20 PM > To: hostap at shmoo.com > Subject: wlansniff setup > > > Hi, > > I am using HostAP(STA) & HostAP(AP). > Can someone please let me know how to setup in monitoring > mode. Basically I should be able to parse all the 802.11 > packets that are sent over the air between STA & AP. > > I tried the following (at both AP & STA) > cd sniff > make > ./wlansniff -b0 wlan0 > Unsupported arptype 0x0001 > > I searched the mailing list and didnot find any matches. > I think I am missing some basic setup. > Can someone please point me in the right direction. > > -Raghu > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030117/627ae524/attachment.htm From pzz at pzz.msk.ru Fri Jan 17 14:56:43 2003 From: pzz at pzz.msk.ru (Alexander Pevzner) Date: 17 Jan 2003 22:56:43 +0300 Subject: State of the sources? References: <20030117033438.GC1096@jm.kir.nu> Message-ID: <87wul3zh1g.fsf@toy.pzz.msk.ru> Hello, Jouni Malinen! Thu, 16 Jan 2003 19:34:38 -0800 you wrote: JM> I try to keep it always in working condition, i.e., I do test my JM> changes before committing; well. However, I'm usually testing only one JM> configuration (at the moment Linux 2.5.x and hostapd) for CVS commits JM> and I may miss some changes that break, e.g., Linux 2.4 compilation. JM> These are normally fixed as soon as someone reports them. BTW, could you please incorporate a minor fix into a top-level Makefile? My Linux distribution, which is more or less standard RedHat 6.2, doesn't have cpp command. Please, replace cpp with gcc -E -. It will not affect the users of newer distributions. -- Wishes, Alexander Pevzner (pzz at pzz.msk.ru) From lsantos at pd3.com.br Fri Jan 17 15:07:02 2003 From: lsantos at pd3.com.br (Leonardo Pereira Santos) Date: Fri, 17 Jan 2003 18:07:02 -0200 Subject: interrupt problem? Message-ID: <200301171807.02721.lsantos@pd3.com.br> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello There! I'm using hostap_cs in a embedded linux board (Motorola's FADS) with kernel 2.4.20-pre11 and pcmcia-cs 3.1.22 or pcmcia-cs 3.2.3. After compiling everything without a glitch and loading up all modules, a win2k laptop is able to connecto to my acess point. After some ramdom time, I get the following message: # NETDEV WATCHDOG: wlan0: transmit timed out wlan0 Tx timed out! Resetting card hostap_cs: wlan0: resetting card wlan0: Interrupt, but SWSUPPORT0 does not match: 0000 != 8A32 - card removed? wlan0: Interrupt, but SWSUPPORT0 does not match: 0000 != 8A32 - card removed? wlan0: Interrupt, but SWSUPPORT0 does not match: 0008 != 8A32 - card removed? wlan0: Interrupt, but SWSUPPORT0 does not match: 0000 != 8A32 - card removed? While loading, the hostap shows these messages: hostap_cs: index 0x01: Vcc 5.0, irq 9, io 0x0100-0x013f Cat on /proc/interrupts: # cat /proc/interrupts CPU0 5: 630 8xx SIU Edge cpm 9: 24 8xx SIU Edge hostap_cs 11: 0 8xx SIU Edge m8xx_pcmcia 15: 0 8xx SIU Edge tbint The time to the error is random, sometimes I'm able to download some data (1 meg), sometimes it takes just a sec to give the error. I've already read some messages on the list about similar errors, but of them refers to PCI boards, not PCMCIA. Any help is greatly appreciated! - -- This is Linux country. On a quiet night, you can hear Windows reboot. Leonardo Pereira Santos Engenheiro de Projetos PD3 Tecnologia av. Par? 330/202 (51) 3337 1237 Today Fortune tells us: Fry: "You know what I like best about you, Umbrielle? You find me fascinating, even when I'm not claiming to be a jewel thief or a lion tamer." -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.7 (GNU/Linux) iD8DBQE+KGJm8uAMCTjfaxIRAl+CAJ9of7cHVEgYhB6RZxuD+9OS6+bqLwCff4OT WMUX032FSEy+OqXK0XmzB44= =PvV7 -----END PGP SIGNATURE----- From raghu at sraamerica.com Fri Jan 17 15:31:54 2003 From: raghu at sraamerica.com (Raghu) Date: Fri, 17 Jan 2003 15:31:54 -0500 Subject: wlansniff setup Message-ID: <20030117203154.SQPK2474.imta01a2.registeredsite.com@smtp.registeredsite.com> Guillaume, Thankyou very much. Your suggestion worked great. -Raghu > > From: "Bichot Guillaume (Princeton)" > Date: 2003/01/17 Fri PM 02:40:12 EST > To: "'Raghu'" , hostap at shmoo.com > Subject: RE: wlansniff setup > > Raghu, > > Basically : iwconfig wlan0 mode "Monitor" > > And to select the type of header included in the monitored frames: > Prism2_param wlan0 monitor_type (0: 802.11 / 1: prism2 + 802.11) > > If it does not work (wireless tools < v25) then > > Iwpriv wlan0 monitor ((0: off /2: 802.11 / 3: prism2 + 802.11) > > Guillaume > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030117/d24a0817/attachment.htm From jdl at jdl.com Fri Jan 17 16:03:30 2003 From: jdl at jdl.com (Jon Loeliger) Date: Fri, 17 Jan 2003 13:03:30 -0800 Subject: hostap and compaq wl-200 In-Reply-To: Your message of "Sat, 18 Jan 2003 04:18:08 +1100." <20030117171808.GQ27132@taz.net.au> References: <20030117163800.GP27132@taz.net.au> <002501c2be49$1f03d4b0$010aa8c0@vasa> <20030117171808.GQ27132@taz.net.au> Message-ID: In message <20030117171808.GQ27132 at taz.net.au>, Craig Sanders mumbled: > On Fri, Jan 17, 2003 at 06:54:35PM +0200, Joachim Wickman wrote: > > I had some familiar problems, don't remember exactly how it did > > behave. But.. putting PCIC_OPTS="irq_mode=1" in rc.pcmcia did the > > trick. Maybe samething with yours. > > thanks, looks like that did the trick. > > although in debian, the file to edit is /etc/default/pcmcia rather than > rc.pcmcia > And for those of you who, like myself, compiled the yenta socket driver into the kernel rather than use the i82365, you have to set the PCIC="yenta_socket" in that same /etc/default/pcmcia file. This solves a whole bunch of "timeout" and interrupt delivery issues! :-) jdl From wireless at douglasheld.net Fri Jan 17 18:46:29 2003 From: wireless at douglasheld.net (wireless) Date: Fri, 17 Jan 2003 15:46:29 -0800 Subject: Trouble with enGenius card (no internal antenna!) References: <156AA90517398D479A3473C35CE99EDD0157A6CE@pvdexc99.jwu.edu> Message-ID: <000b01c2be82$a9f707c0$ec64000a@dheld2> That is correct. There is NO internal antenna in the enGenius. This place sells reasonably priced pigtails and antennas for that card: http://www.netgate.com/NL2511.html --Doug ----- Original Message ----- From: "Joshua Wright" To: ; Sent: Friday, January 17, 2003 8:21 AM Subject: RE: Trouble with enGenius card > I was using it in managed mode (STA). After a little more analysis, I'm thinking that the enGenius card doesn't actually have a built-in antenna at all - just the MMCX connectors. That may be my problem. :) > > -Joshua Wright > Team Leader, Networks and Systems > Johnson & Wales University > Joshua.Wright at jwu.edu > http://home.jwu.edu/jwright/ > > pgpkey: http://home.jwu.edu/jwright/pgpkey.htm > fingerprint: FDA5 12FC F391 3740 E0AE BDB6 8FE2 FC0A D44B 4A73 > > > > -----Original Message----- > > From: wireless at douglasheld.net [mailto:wireless at douglasheld.net] > > Sent: Friday, January 17, 2003 1:23 PM > > To: Joshua Wright; hostap at shmoo.com > > Subject: Re: Trouble with enGenius card > > > > > > Darn, I just bought one of those... > > > > Joshua, can you clarify: Is your enGenius functioning as as > > station or as > > the AP? > > > > --Doug > > > > ----- Original Message ----- > > From: "Joshua Wright" > > To: > > Sent: Friday, January 17, 2003 6:20 AM > > Subject: Trouble with enGenius card > > > > > > I recently ordered a 100mW enGenius PrismII card and am > > having trouble using > > it with the HostAP drivers. > > > > The drivers modprobe without a problem and I can configure > > the card with the > > standard iwconfig and prism2_param tools, but dmesg keeps reporting: > > > > wlan0: LinkStatus=1 (Connected) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0xa287 ScanReason=2 > > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > > wlan0: LinkStatus=1 (Connected) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0xa32b ScanReason=2 > > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > > wlan0: LinkStatus=1 (Connected) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0xa7f1 ScanReason=2 > > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > > wlan0: LinkStatus=1 (Connected) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0xa9e8 ScanReason=2 > > wlan0: LinkStatus=5 (Access point in range) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: JoinRequest: BSSID=00:e0:63:82:19:c6 channel=11 > > wlan0: LinkStatus=6 (Association failed) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0xa88f ScanReason=2 > > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > > wlan0: LinkStatus=1 (Connected) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0xa89d ScanReason=2 > > wlan0: LinkStatus=4 (Access point out of range) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > > wlan0: LinkStatus=6 (Association failed) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0x0000 ScanReason=2 > > wlan0: LinkStatus=5 (Access point in range) > > wlan0: LinkStatus: BSSID=44:44:44:44:44:44 > > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > > wlan0: LinkStatus=1 (Connected) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: ScanResults: Reserved=0x5c47 ScanReason=2 > > wlan0: LinkStatus=4 (Access point out of range) > > wlan0: LinkStatus: BSSID=00:e0:63:82:0d:27 > > wlan0: JoinRequest: BSSID=00:e0:63:82:0d:27 channel=3 > > wlan0: LinkStatus=1 (Connected) > > > > > > "iwconfig wlan0" seems to indicate I am associated with my AP > > (which is very > > close to my desk) but I keep seeing the LED on the enGenius > > card flashing > > which I presume means it is diassociating and scanning for a > > new connection. > > > > Any thoughts on making this card work? I am using the > > current CVS HostAP > > drivers with a 2.4.19 kernel, wireless tools 25 and wireless > > extensions 14. > > > > Thanks. > > > > -Joshua Wright > > Team Leader, Networks and Systems > > Johnson & Wales University > > Joshua.Wright at jwu.edu > > http://home.jwu.edu/jwright/ > > > > pgpkey: http://home.jwu.edu/jwright/pgpkey.htm > > fingerprint: FDA5 12FC F391 3740 E0AE BDB6 8FE2 FC0A D44B 4A73 > > > > > > > > > > iwconfig: > > > > wlan0 IEEE 802.11b ESSID:"JWU" > > Mode:Managed Frequency:2.422GHz Access Point: > > 00:E0:63:82:0D:27 > > Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 > > Retry min limit:8 RTS thr:off Fragment thr:off > > Power Management:off > > Link Quality:0/92 Signal level:-90 dBm Noise > > level:-100 dBm > > Rx invalid nwid:0 Rx invalid crypt:39 Rx invalid frag:0 > > Tx excessive retries:23 Invalid misc:3912 Missed beacon:0 > > > > wlan0ap IEEE 802.11b ESSID:"JWU" > > Mode:Managed Frequency:2.422GHz Access Point: > > 00:E0:63:82:0D:27 > > Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 > > Retry min limit:8 RTS thr:off Fragment thr:off > > Power Management:off > > > > wlan0sta IEEE 802.11b ESSID:"JWU" > > Mode:Managed Frequency:2.422GHz Access Point: > > 00:00:00:00:00:00 > > Bit Rate:11Mb/s Tx-Power=200 dBm Sensitivity=1/3 > > Retry min limit:8 RTS thr:off Fragment thr:off > > Power Management:off > > > > > > _______________________________________________ > > HostAP mailing list > > HostAP at shmoo.com > > http://lists.shmoo.com/mailman/listinfo/hostap > > > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From hozturk at nps.navy.mil Fri Jan 17 21:06:22 2003 From: hozturk at nps.navy.mil (Selcuk OZTURK) Date: Fri, 17 Jan 2003 18:06:22 -0800 Subject: Help authenticated WINXP 802.1x client Message-ID: Even though I got the initial accetance from the Freeradius , server doesnt authenticate, Could certicates cause this problem? Thanks ... SECO + LD_LIBRARY_PATH=/usr/local/openssl/lib + LD_PRELOAD=/usr/local/openssl/lib/libcrypto.so + export LD_LIBRARY_PATH LD_PRELOAD + /usr/local/sbin/radiusd -X -A Starting - reading configuration files ... reread_config: reading radiusd.conf Config: including file: /etc/raddb/proxy.conf Config: including file: /etc/raddb/clients.conf Config: including file: /etc/raddb/snmp.conf Config: including file: /etc/raddb/sql.conf main: prefix = "/usr/local" main: localstatedir = "/usr/local/var" main: logdir = "/usr/local/var/log/radius" main: libdir = "/usr/local/lib" main: radacctdir = "/usr/local/var/log/radius/radacct" main: hostname_lookups = no main: max_request_time = 30 main: cleanup_delay = 5 main: max_requests = 1024 main: delete_blocked_requests = 0 main: port = 0 main: allow_core_dumps = no main: log_stripped_names = no main: log_file = "/usr/local/var/log/radius/radius.log" main: log_auth = no main: log_auth_badpass = no main: log_auth_goodpass = no main: pidfile = "/usr/local/var/run/radiusd/radiusd.pid" main: user = "(null)" main: group = "(null)" main: usercollide = no main: lower_user = "no" main: lower_pass = "no" main: nospace_user = "no" main: nospace_pass = "no" main: checkrad = "/usr/local/sbin/checkrad" main: proxy_requests = yes proxy: retry_delay = 5 proxy: retry_count = 3 proxy: synchronous = no proxy: default_fallback = yes proxy: dead_time = 120 proxy: servers_per_realm = 15 security: max_attributes = 200 security: reject_delay = 1 security: status_server = no main: debug_level = 0 read_config_files: reading dictionary read_config_files: reading naslist read_config_files: reading clients read_config_files: reading realms radiusd: entering modules setup Module: Library search path is /usr/local/lib Module: Loaded expr Module: Instantiated expr (expr) Module: Loaded System unix: cache = yes unix: passwd = "/etc/passwd" unix: shadow = "/etc/shadow" unix: group = "/etc/group" unix: radwtmp = "/usr/local/var/log/radius/radwtmp" unix: usegroup = no unix: cache_reload = 600 HASH: Reinitializing hash structures and lists for caching... HASH: user root found in hashtable bucket 11726 HASH: user bin found in hashtable bucket 86651 HASH: user daemon found in hashtable bucket 11668 HASH: user adm found in hashtable bucket 26466 HASH: user lp found in hashtable bucket 54068 HASH: user sync found in hashtable bucket 42895 HASH: user shutdown found in hashtable bucket 71746 HASH: user halt found in hashtable bucket 7481 HASH: user mail found in hashtable bucket 79471 HASH: user news found in hashtable bucket 5375 HASH: user uucp found in hashtable bucket 38541 HASH: user operator found in hashtable bucket 21748 HASH: user games found in hashtable bucket 47657 HASH: user gopher found in hashtable bucket 47357 HASH: user ftp found in hashtable bucket 56226 HASH: user nobody found in hashtable bucket 99723 HASH: user mailnull found in hashtable bucket 78086 HASH: user rpm found in hashtable bucket 72383 HASH: user xfs found in hashtable bucket 17213 HASH: user ntp found in hashtable bucket 21418 HASH: user rpc found in hashtable bucket 72373 HASH: user rpcuser found in hashtable bucket 552 HASH: user nfsnobody found in hashtable bucket 51830 HASH: user nscd found in hashtable bucket 36306 HASH: user ident found in hashtable bucket 40304 HASH: user radvd found in hashtable bucket 66743 HASH: user pcap found in hashtable bucket 55326 HASH: user hozturk found in hashtable bucket 95961 HASH: Stored 28 entries from /etc/passwd HASH: Stored 38 entries from /etc/group Module: Instantiated unix (unix) Module: Loaded eap eap: default_eap_type = "tls" eap: timer_expire = 60 tls: rsa_key_exchange = no tls: dh_key_exchange = yes tls: rsa_key_length = 512 tls: dh_key_length = 512 tls: verify_depth = 0 tls: CA_path = "(null)" tls: pem_file_type = yes tls: private_key_file = "/etc/1x/srv.pem" tls: certificate_file = "/etc/1x/srv.pem" tls: CA_file = "/etc/1x/root.pem" tls: private_key_password = "XXXXX" tls: dh_file = "/etc/1x/dh" tls: random_file = "/etc/1x/random" tls: fragment_size = 1024 tls: include_length = yes rlm_eap_tls: conf N ctx stored rlm_eap: Loaded and initialized the type tls Module: Instantiated eap (eap) Module: Loaded preprocess preprocess: huntgroups = "/etc/raddb/huntgroups" preprocess: hints = "/etc/raddb/hints" preprocess: with_ascend_hack = no preprocess: ascend_channels_per_line = 23 preprocess: with_ntdomain_hack = no preprocess: with_specialix_jetstream_hack = no preprocess: with_cisco_vsa_hack = no Module: Instantiated preprocess (preprocess) Module: Loaded realm realm: format = "suffix" realm: delimiter = "@" Module: Instantiated realm (suffix) Module: Loaded files files: usersfile = "/etc/raddb/users" files: acctusersfile = "/etc/raddb/acct_users" files: preproxy_usersfile = "/etc/raddb/preproxy_users" files: compat = "no" Module: Instantiated files (files) Module: Loaded detail detail: detailfile = "/usr/local/var/log/radius/radacct/%{Client-IP-Address}/de tail-%Y%m%d" detail: detailperm = 384 detail: dirperm = 493 detail: locking = no Module: Instantiated detail (detail) Module: Loaded radutmp radutmp: filename = "/usr/local/var/log/radius/radutmp" radutmp: username = "%{User-Name}" radutmp: perm = 384 radutmp: callerid = yes Module: Instantiated radutmp (radutmp) Listening on IP address *, ports 1812/udp and 1813/udp, with proxy on 1814/udp. Ready to process requests. rad_recv: Access-Request packet from host 131.120.8.136:32776, id=194, length=15 2 User-Name = "hozturk" NAS-IP-Address = 131.xx NAS-Port = 1 Called-Station-Id = "00-05-5D-D9-55-A5:test" Calling-Station-Id = "00-05-5D-D9-57-59" Framed-MTU = 2304 NAS-Port-Type = Wireless-802.11 Connect-Info = "CONNECT 11Mbps 802.11b" EAP-Message = "\002\003\000\014\001hozturk" Message-Authenticator = 0xd11ebc27d5da3a88633c794226dd4722 modcall: entering group authorize modcall[authorize]: module "preprocess" returns ok modcall[authorize]: module "eap" returns updated rlm_realm: No '@' in User-Name = "hozturk", looking up realm NULL rlm_realm: No such realm NULL modcall[authorize]: module "suffix" returns noop users: Matched hozturk at 99 modcall[authorize]: module "files" returns ok modcall: group authorize returns updated rad_check_password: Found Auth-Type EAP auth: type "EAP" modcall: entering group authenticate rlm_eap: processing type tls modcall[authenticate]: module "eap" returns ok modcall: group authenticate returns ok Sending Access-Challenge of id 194 to 131.xxxxxx136:32776 EAP-Message = "\001\004\000\006\r " Message-Authenticator = 0x00000000000000000000000000000000 State = 0x7d3bb6cacfcf9085e25f7952651f672a66b5283ea47a593f7fe46a43b7cb0b 84060bbbea Finished request 0 Going to the next request --- Walking the entire request list --- Waking up in 6 seconds... rad_recv: Access-Request packet from host 131.xxxxx:32776, id=195, length=25 8 User-Name = "hozturk" NAS-IP-Address = 131.1xxxx NAS-Port = 1 Called-Station-Id = "00-05-5D-D9-55-A5:test" Calling-Station-Id = "00-05-5D-D9-57-59" Framed-MTU = 2304 NAS-Port-Type = Wireless-802.11 Connect-Info = "CONNECT 11Mbps 802.11b" EAP-Message = "\002\004\000P\r\200\000\000\000F\026\003\001\000A\001\000 \000=\003\001>(\264\357\276`Q\206\216\261=\032g|s\364\226\321M\370*\016\317\352\ 325\206\243uy\311\350I\000\000\026\000\004\000\005\000\n\000\t\000d\000b\000\003 \000\006\000\023\000\022\000c\001" State = 0x7d3bb6cacfcf9085e25f7952651f672a66b5283ea47a593f7fe46a43b7cb0b 84060bbbea Message-Authenticator = 0x216ed4c5c4f8ae16098352a125e27516 modcall: entering group authorize modcall[authorize]: module "preprocess" returns ok modcall[authorize]: module "eap" returns updated rlm_realm: No '@' in User-Name = "hozturk", looking up realm NULL rlm_realm: No such realm NULL modcall[authorize]: module "suffix" returns noop users: Matched hozturk at 99 modcall[authorize]: module "files" returns ok modcall: group authorize returns updated rad_check_password: Found Auth-Type EAP auth: type "EAP" modcall: entering group authenticate rlm_eap: Request found, released from the list rlm_eap: EAP_TYPE - tls rlm_eap: processing type tls rlm_eap_tls: Length Included undefined: before/accept initialization TLS_accept: before/accept initialization <<< TLS 1.0 Handshake [length 0041], ClientHello TLS_accept: SSLv3 read client hello A >>> TLS 1.0 Handshake [length 004a], ServerHello TLS_accept: SSLv3 write server hello A >>> TLS 1.0 Handshake [length 0641], Certificate TLS_accept: SSLv3 write certificate A >>> TLS 1.0 Handshake [length 00a1], CertificateRequest TLS_accept: SSLv3 write certificate request A TLS_accept: SSLv3 flush data TLS_accept:error in SSLv3 read client certificate A rlm_eap_tls: SSL_read Error Error code is ..... 2 SSL Error ..... 2 modcall[authenticate]: module "eap" returns ok modcall: group authenticate returns ok Sending Access-Challenge of id 195 to 131.xxxxxxx32776 EAP-Message = "\001\005\004\n\r\300\000\000\007;\026\003\001\000J\002\00 0\000F\003\001>(\265fA\276\\\260\235,aLxO\256\242\005f\310f{%t\350\231\230q\227\ 327*\275v \221\314$\205\364-\0304\317\371\311\213\224\001\235\023B!Im\010\353L / \364\366\273\nZ\352-\000\004\000\026\003\001\006A\013\000\006=\000\006:\000\002\ 2520\202\002\2460\202\002\017\240\003\002\001\002\002\001\0010\r\006\t*\206H\206 \367\r\001\001\004\005\0000\201\2171\0130\t\006\003U\004\006\023\002US1\0230\021 \006\003U\004\010\023\nCalifornia1\0210\017\006\003U\004" EAP-Message = "H\206\367\r\001\t\001\026\024hozturk at nps.navy.mil0\036\02 7\r030118001144Z\027\r040118001144Z0\201\2111\0130\t\006\003U\004\006\023\002US1 \0230\021\006\003U\004\010\023\nCalifornia1\0210\017\006\003U\004\007\023\010Mon terey1\r0\013\006\003U\004\n\023\004NPGS1\r0\013\006\003U\004\013\023\004SAAM1\0 170\r\006\003U\004\003\023\006radius1#0!\006\t*\206H\206\367\r\001\t\001\026\024 hozturk at nps.navy.mil0\201\2370\r\006\t*\206H\206\367\r\001\001\001\005\000\003\2 01\215\0000\201\211\002\201\201\000\315\226\201\027\332" EAP-Message = ":\245\344]\216,\270yFQ\375F\271\022MR\266\217\003\304\243 &\204xi_\371_\001n\354\221z\r\317\026\357\241N\t\354\\\341b\246\366au\346D\332\3 67\365\343\t"\353Y\315\350t\025\266\032\241\343\2175\337\023:\304i\004g\217#\244 \334\344\366i\235&r\255\004A\366\262C=g\t\245^:\377\202\332\204\333i\2761o\002\0 03\001\000\001\243\0260\0240\022\006\003U\035%\004\0130\t\006\007+\006\001\005\0 05\007\0030\r\006\t*\206H\206\367\r\001\001\004\005\000\003\201\201\000\203}\302 \366\013\310\354\211\212\311\242Vd\262\327\354\331/\214" EAP-Message = "\016\271u\245n[\367\005\241\343\3545H^"mo\n^\021\334e\227 N\351\302\256\355\221"\000\003\2120\202\003\2060\202\002\357\240\003\002\001\002 \002\001\0000\r\006\t*\206H\206\367\r\001\001\004\005\0000\201\2171\0130\t\006\0 03U\004\006\023\002US1\0230\021\006\003U\004\010\023\nCalifornia1\0210\017\006\0 03U\004\007\023\010Monterey1\r0\013\006\003U\004\n\023\004NPGS1\r0\013\006\003U\ 004\013\023\004SAAM1\0250\023\006\003U\004\003\023\014WirelessSAAM1#0!\006\t*\20 6H\206\367\r\001\t\001\026\024hozturk at nps.navy.mil0\036" EAP-Message = "\004\006\023\002US1\0230\021\006\003U\004\010\023\nCalifo rni" Message-Authenticator = 0x00000000000000000000000000000000 State = 0xf59c9d7a61f019980b440077feb3cbaa66b5283e21366799975b8a83722798 0be401bf4a Finished request 1 Going to the next request Waking up in 6 seconds... rad_recv: Access-Request packet from host 131.xxxxxx:32776, id=196, length=18 4 User-Name = "hozturk" NAS-IP-Address = 131.xxxxxx NAS-Port = 1 Called-Station-Id = "00-05-5D-D9-55-A5:test" Calling-Station-Id = "00-05-5D-D9-57-59" Framed-MTU = 2304 NAS-Port-Type = Wireless-802.11 Connect-Info = "CONNECT 11Mbps 802.11b" EAP-Message = "\002\005\000\006\r" State = 0xf59c9d7a61f019980b440077feb3cbaa66b5283e21366799975b8a83722798 0be401bf4a Message-Authenticator = 0x7b9266018ffdc922ef0a980c7981cd1a modcall: entering group authorize modcall[authorize]: module "preprocess" returns ok modcall[authorize]: module "eap" returns updated rlm_realm: No '@' in User-Name = "hozturk", looking up realm NULL rlm_realm: No such realm NULL modcall[authorize]: module "suffix" returns noop users: Matched hozturk at 99 modcall[authorize]: module "files" returns ok modcall: group authorize returns updated rad_check_password: Found Auth-Type EAP auth: type "EAP" modcall: entering group authenticate rlm_eap: Request found, released from the list rlm_eap: EAP_TYPE - tls rlm_eap: processing type tls rlm_eap_tls: Received EAP-TLS ACK message modcall[authenticate]: module "eap" returns ok modcall: group authenticate returns ok Sending Access-Challenge of id 196 to 131.xxxxx6:32776 EAP-Message = "\001\006\003E\r\200\000\000\007;a1\0210\017\006\003U\004\ 007\023\010Monterey1\r0\013\006\003U\004\n\023\004NPGS1\r0\013\006\003U\004\013\ 023\004SAAM1\0250\023\006\003U\004\003\023\014WirelessSAAM1#0!\006\t*\206H\206\3 67\r\001\t\001\026\024hozturk at nps.navy.mil0\201\2370\r\006\t*\206H\206\367\r\001 \001\001\005\000\003\201\215\0000\201\211\002\201\201\000\264\351\213\001e\211\0 20~\357\304\002\010\264r\371>\004Y\3169_N}<\335|\\\256F\231\223U\312\247\037\006 {\243\267"C*\313\371\000e\246M\022\275EO\232: \222\234" EAP-Message = "\240\301\236\213w-\007f\361\251\267?I\376(\021\216*\372)\ 250\325\2330W\002\003\001\000\001\243\201\3570\201\3540\035\006\003U\035\016\004 \026\004\024\034f\351\322^\200x\251 \234ZM\334\234\316\334\027\321\214L0\201\274 \006\003U\035#\004\201\2640\201\261\200\024\034f\351\322^\200x\251 \234ZM\334\23 4\316\334\027\321\214L\241\201\225\244\201\2220\201\2171\0130\t\006\003U\004\006 \023\002US1\0230\021\006\003U\004\010\023\nCalifornia1\0210\017\006\003U\004\007 \023\010Monterey1\r0\013\006\003U\004\n\023\004NPGS1\r" EAP-Message = "mil\202\001\0000\014\006\003U\035\023\004\0050\003\001\00 1\3770\r\006\t*\206H\206\367\r\001\001\004\005\000\003\201\201\000Gd\017_J\241\3 334o\302\306wZ7[\332\275\376\001\321T\326?\207\331U\340\016\037v\335\333\312~)\3 32\264I-\351\3057)o\235\351\020\210\265\020d\017\000\230\250\200\311i\332\313\30 2`Qe\020\236X\273\211\240\241\341\267\253?\365\332\035\214\005N\304=Ky at A\204\246 \200\r\255\007Q\265t\024\313i9\340\217\265\350\020\3167W\313^4\310\346>L`\230\21 5\323\3367_\312p\0106\366,\026\003\001\000\241\r\000\000" EAP-Message = "GS1\r0\013\006\003U\004\013\023\004SAAM1\0250\023\006\003 U\004\003\023\014WirelessSAAM1#0!\006\t*\206H\206\367\r\001\t\001\026\024hozturk @nps.navy.mil\016\000\000" Message-Authenticator = 0x00000000000000000000000000000000 State = 0xb7e68e6c65db68f2098dffa18097d84267b5283e7cba368ffcade49aca3b2b 5851446db9 Finished request 2 Going to the next request --- Walking the entire request list --- Waking up in 5 seconds... rad_recv: Access-Request packet from host 131.xxxxxxx:32776, id=197, length=18 4 User-Name = "hozturk" NAS-IP-Address = 131.1xxxx NAS-Port = 1 Called-Station-Id = "00-05-5D-D9-55-A5:test" Calling-Station-Id = "00-05-5D-D9-57-59" Framed-MTU = 2304 NAS-Port-Type = Wireless-802.11 Connect-Info = "CONNECT 11Mbps 802.11b" EAP-Message = "\002\006\000\006\r" State = 0xb7e68e6c65db68f2098dffa18097d84267b5283e7cba368ffcade49aca3b2b 5851446db9 Message-Authenticator = 0x1d538eba5a69abc9c4b591f6cc9512cd modcall: entering group authorize modcall[authorize]: module "preprocess" returns ok modcall[authorize]: module "eap" returns updated rlm_realm: No '@' in User-Name = "hozturk", looking up realm NULL rlm_realm: No such realm NULL modcall[authorize]: module "suffix" returns noop users: Matched hozturk at 99 modcall[authorize]: module "files" returns ok modcall: group authorize returns updated rad_check_password: Found Auth-Type EAP auth: type "EAP" modcall: entering group authenticate rlm_eap: Request found, released from the list rlm_eap: EAP_TYPE - tls rlm_eap: processing type tls rlm_eap_tls: Received EAP-TLS ACK message rlm_eap_tls: Invalid ACK received modcall[authenticate]: module "eap" returns invalid modcall: group authenticate returns invalid auth: Failed to validate the user. Delaying request 3 for 1 seconds Finished request 3 Going to the next request Waking up in 5 seconds... --- Walking the entire request list --- Cleaning up request 0 ID 194 with timestamp 3e28b566 Cleaning up request 1 ID 195 with timestamp 3e28b566 Sending Access-Reject of id 197 to 131.120.8.136:32776 EAP-Message = "\004\006\000\004" Message-Authenticator = 0x00000000000000000000000000000000 Waking up in 1 seconds... --- Walking the entire request list --- Cleaning up request 2 ID 196 with timestamp 3e28b567 Cleaning up request 3 ID 197 with timestamp 3e28b567 Nothing to do. Sleeping until we see a request. From zimage at del.bg Sat Jan 18 03:23:57 2003 From: zimage at del.bg (Theodor Milkov) Date: Sat, 18 Jan 2003 10:23:57 +0200 Subject: about noise level, signal level and quality In-Reply-To: References: Message-ID: <20030118082357.GB7133@zimage.urban.delbg.com> On Mon, Jan 13, 2003 at 06:30:07PM -0500, Hairong Sun wrote: > Since "signal" and "noise" are converted by (1x100+v*100/255-100) and > "quality" is converted by (signal-noise-35) in "prism2_ap_get_sta_qual()" from their > original values (last_rx_signal and last_rx_silence) in the firmware, my > questions are > > 1. what do "signal","noise", last_rx_signal, and last_rx_silence > really mean in the driver AND firmware? > 2. how to explain the noise level reading in my experiments (i.e.,noise > level changes little with the interference from neighboring AP)? Our experience with 2002-10-12 version of hostap driver shows that signal/noise readings in Master mode are in some unknown value and signal level change of 2..3 points is actually change with 10's of dBm. In Managed mode it seems that noise level doesn't change, but signal level is actually function of signal level AND noise level. However readings from hostap_cs/prism2 card and orinoco/ornoco_cs do not match. Does anyone have more information? -- Theodor Milkov Network Administrator Davidov Net Phone: +359 (2) 730158 From jkmaline at cc.hut.fi Sat Jan 18 21:15:18 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sat, 18 Jan 2003 18:15:18 -0800 Subject: State of the sources? In-Reply-To: <87wul3zh1g.fsf@toy.pzz.msk.ru> References: <20030117033438.GC1096@jm.kir.nu> <87wul3zh1g.fsf@toy.pzz.msk.ru> Message-ID: <20030119021518.GE2022@jm.kir.nu> On Fri, Jan 17, 2003 at 10:56:43PM +0300, Alexander Pevzner wrote: > BTW, could you please incorporate a minor fix into a top-level > Makefile? My Linux distribution, which is more or less standard > RedHat 6.2, doesn't have cpp command. Please, replace cpp with > gcc -E -. It will not affect the users of newer distributions. I'm pretty sure all Linux distributions that include a compiler do include preprocessor.. But yes, cpp is apparently not in the default path on some older Red Hat versions. Anyway, since gcc can be specified with CC variable, it is quite reasonable to use it to run preprocessor instead of directly calling cpp. CVS version is now using $(CC) -E -. -- Jouni Malinen PGP id EFC895FA From ckchoi95 at hotmail.com Sun Jan 19 03:46:40 2003 From: ckchoi95 at hotmail.com (=?ks_c_5601-1987?B?w9Ygw6KxuQ==?=) Date: Sun, 19 Jan 2003 17:46:40 +0900 Subject: TX rate unchange problem. Message-ID: Hello everyone. I am Kris Choi. I'm programming my own host based AP driver with Samsung Magic Lan which is using HFA3841 chip. So far, everything looks working fine except the AP transmit always in 2Mbps. I think I tried all method in Intersil Programmers Manual, but TX Rate never changed. (Neither Current Tx Rate record). Also, the host AP source code from http://hostp.epitest.fi include a comment that Current Tx Rate is always 2 on HFA384X chip set. I want to know the comment means whether transmitting 11Mbps with HFA3841 is impossible or I should do something else to enable 11Mbps transmission. Thank you for your reply. _________________________________________________________________ ???? ?? ??? ???? ???... MSN ??? ??? http://vod.msn.co.kr From wireless at douglasheld.net Sun Jan 19 04:12:53 2003 From: wireless at douglasheld.net (wireless at douglasheld.net) Date: Sun, 19 Jan 2003 01:12:53 -0800 Subject: What is the max power for my card? Message-ID: <000901c2bf9a$f4154a60$0200a8c0@buttrump> I have bought an EnGenius "EL-2011CD PLUS EXT2" pcmcia card. The documentation online for this card is unclear; I wonder if I can use iwconfig to query the maximum power setting? Currently, iwconfig accepts ridiculously large values, like 300mW, without complaint. Any ideas, anyone? --Doug From ammirata at econointl.com Sun Jan 19 12:09:35 2003 From: ammirata at econointl.com (Sergio M. Ammirata) Date: Sun, 19 Jan 2003 12:09:35 -0500 Subject: hostapd issues and questions In-Reply-To: <20030119021518.GE2022@jm.kir.nu> Message-ID: <000a01c2bfdd$966d2c50$017ba8c0@ammirata.net> I am using hostapd with latest cvs code. I start it with "hostapd /etc/hostapd.conf" in daemon mode after I initialize the kernel modules. 1) Changing the mac address of the wlan0 interface using ifconfig breaks routing (you can link but the tcp/ip layer does not work). This happens only with hostapd enabled. I also noticed that it is firmware dependent. It works with 1.4.9. The strange thing is that firmware 1.3.6 works if hostapd is not used. 2) Without hostapd there is wlan0 and wlan0wdsx. Now we have wlan0ap and wlan0sta as well. How are we supposed to setup Ethernet bridging now? Do we include all three in the bridge (wlan0, wlan0ap and wlan0sta)? 3) Has the functionality of wlan0wdsx changed? When and why would I use wlan0sta? 4) With hostapd active iwspy no longer reports any info on wds neighbors. How can I found out the strength of the links? Thanks in advance, Sergio From alexisb at netcourrier.com Sun Jan 19 12:11:59 2003 From: alexisb at netcourrier.com (Alexis) Date: Sun, 19 Jan 2003 12:11:59 -0500 Subject: iwlist txpower -> no answer !? Message-ID: <20030119121159.A169@alexis.itd.umich.edu> Hi all ! I just subscribed two days ago, but I tried to dig a bit through the archive and I could not find anything related... Here is my problem : wlan0 IEEE 802.11b ESSID:"test" Mode:Master Frequency:2.422GHz Access Point: 00:02:6F:05:E8:9F Bit Rate:11Mb/s Tx-Power:-5 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:0 Signal level:0 Noise level:0 Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 root at alexis:~# iwlist wlan0 txpower wlan0 No transmit-powers ? Please update driver... ^^^^^^ what does this mean ? wlan0 is a Senao SL-2511 CD PLUS (200mW) pcmcia card wlan0: NIC: id=0x800c v1.0.0 wlan0: PRI: id=0x15 v1.0.7 wlan0: STA: id=0x1f v1.3.6 HostAP driver is latest CVS wireless-tools is v25 I have another question : I read on a list that for iwconfig 20dBm was the max tx power regardless of the real card capability. But when connected, doing : iwconfig wlan0 txpower 20dBm will set the power output to 14dBm iwconfig wlan0 txpower 200mW will set the power output to 24dBm do you know why there is such a difference ? Thanks in advance, and many thanks for that wonderful driver, -- Alexis From jkmaline at cc.hut.fi Sun Jan 19 12:27:48 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 19 Jan 2003 09:27:48 -0800 Subject: TX rate unchange problem. In-Reply-To: References: Message-ID: <20030119172748.GB1404@jm.kir.nu> On Sun, Jan 19, 2003 at 05:46:40PM +0900, ?? ???? wrote: > So far, everything looks working fine except the AP transmit always in > 2Mbps. How did you verify what rate was used? > I think I tried all method in Intersil Programmers Manual, but TX Rate > never changed. (Neither Current Tx Rate record). > > Also, the host AP source code from http://hostp.epitest.fi include a > comment > that Current Tx Rate is always 2 on HFA384X chip set. I would suggest you would re-read both the manual and the driver source. hostapd_ioctl.c notes that the _RID_ seems to always be 2 Mbps. This is not the same as real TX rate.. > I want to know the comment means whether transmitting 11Mbps with HFA3841 > is impossible or I should do something else to enable 11Mbps transmission. Of course it is possible to transmit at 11 Mbps with a card that supports 11 Mbps ;-). In Host AP mode you need to manually control TX rate; see hostap_handle_sta_tx() for an example (in hostap_ap.c). -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Sun Jan 19 12:30:52 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 19 Jan 2003 09:30:52 -0800 Subject: What is the max power for my card? In-Reply-To: <000901c2bf9a$f4154a60$0200a8c0@buttrump> References: <000901c2bf9a$f4154a60$0200a8c0@buttrump> Message-ID: <20030119173052.GC1404@jm.kir.nu> On Sun, Jan 19, 2003 at 01:12:53AM -0800, wireless at douglasheld.net wrote: > The documentation online for this card is unclear; I wonder if I can use > iwconfig to query the maximum power setting? No, Host AP driver has no way of knowing the real maximum transmit power. > Currently, iwconfig accepts ridiculously large values, like 300mW, without > complaint. Everything over 20 dBm / 100 mW is interpreted as the maximum value for the card. In other words, setting it to 300 mW or 200 mW configures the card exactly in the same way as using 100 mW. -- Jouni Malinen PGP id EFC895FA From sachin at liman.rutgers.edu Sun Jan 19 12:53:00 2003 From: sachin at liman.rutgers.edu (Sachin Ganu) Date: Sun, 19 Jan 2003 12:53:00 -0500 (EST) Subject: What is the max power for my card? In-Reply-To: <20030119173052.GC1404@jm.kir.nu> Message-ID: Hi Jouni, On similar lines, everything below -47dBm is clamped. My question is what is the actual transmit power of the card when the card is set at -47dBm? I think there is a routine in prism2_ioctl.c that does the calculation But I am not sure on what is the actual Tx power of the card? ----------------------------------------------------------- Sachin Ganu Ph.D Candidate WINLAB, Rutgers University 73 Brett Rd, Piscataway NJ 08854 On Sun, 19 Jan 2003, Jouni Malinen wrote: > On Sun, Jan 19, 2003 at 01:12:53AM -0800, wireless at douglasheld.net wrote: > > > The documentation online for this card is unclear; I wonder if I can use > > iwconfig to query the maximum power setting? > > No, Host AP driver has no way of knowing the real maximum transmit power. > > > Currently, iwconfig accepts ridiculously large values, like 300mW, without > > complaint. > > Everything over 20 dBm / 100 mW is interpreted as the maximum value for > the card. In other words, setting it to 300 mW or 200 mW configures the > card exactly in the same way as using 100 mW. > > -- > Jouni Malinen PGP id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From roxxor-mail at binaervarianz.de Sun Jan 19 13:40:38 2003 From: roxxor-mail at binaervarianz.de (Mario Mlynek) Date: Sun, 19 Jan 2003 19:40:38 +0100 Subject: select accesspoint with use of bssid Message-ID: <20030119194038.01df77b8.roxxor-mail@binaervarianz.de> hello .. is it possible to select an accesspoint with use of the bssid and not the essid.. i have the problem that the ap with the best signal is not usable becourse of too many clients.. greedings mario From wlan at zaragozawireless.org Sun Jan 19 21:11:47 2003 From: wlan at zaragozawireless.org (wlan) Date: 20 Jan 2003 02:11:47 +0000 Subject: What is the max power for my card? In-Reply-To: <20030119173052.GC1404@jm.kir.nu> References: <000901c2bf9a$f4154a60$0200a8c0@buttrump> <20030119173052.GC1404@jm.kir.nu> Message-ID: <1043028708.1853.20.camel@localhost.localdomain> On Sun, 2003-01-19 at 17:30, Jouni Malinen wrote: > On Sun, Jan 19, 2003 at 01:12:53AM -0800, wireless at douglasheld.net wrote: > > > The documentation online for this card is unclear; I wonder if I can use > > iwconfig to query the maximum power setting? > > No, Host AP driver has no way of knowing the real maximum transmit power. Hostap is the only driver that it is not able to know the maximum transmit power, or it is common missing feature of all linux drivers? > > Currently, iwconfig accepts ridiculously large values, like 300mW, without > > complaint. > > Everything over 20 dBm / 100 mW is interpreted as the maximum value for > the card. In other words, setting it to 300 mW or 200 mW configures the > card exactly in the same way as using 100 mW. So i understand that this behaviour is due to the reason than we don't know the maximum value, in such a case won't be better to use percentages and change iwconfig man page? iwconfig eth0 txpower 100% or iwconfig eth0 txpower 25% Regards From wlan at zaragozawireless.org Sun Jan 19 21:18:00 2003 From: wlan at zaragozawireless.org (wlan) Date: 20 Jan 2003 02:18:00 +0000 Subject: select accesspoint with use of bssid In-Reply-To: <20030119194038.01df77b8.roxxor-mail@binaervarianz.de> References: <20030119194038.01df77b8.roxxor-mail@binaervarianz.de> Message-ID: <1043029080.1853.26.camel@localhost.localdomain> On Sun, 2003-01-19 at 18:40, Mario Mlynek wrote: > hello .. > is it possible to select an accesspoint with use of the bssid and not the essid.. > i have the problem that the ap with the best signal is not usable becourse of too many clients.. > greedings mario man iwconfig You can force the card to connect to a preselected AP with the following command: iwconfig eth0 ap 00:60:1D:01:23:45 anyway man page said: "When the quality of the connection goes too low, the driver may revert back to automatic mode (the card finds the best Access Point in range)" it is possible to turn off this feature and force the driver to connect ONLY to the preselected AP? Regards From jkmaline at cc.hut.fi Sun Jan 19 20:42:20 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 19 Jan 2003 17:42:20 -0800 Subject: What is the max power for my card? In-Reply-To: References: <20030119173052.GC1404@jm.kir.nu> Message-ID: <20030120014220.GA1514@jm.kir.nu> On Sun, Jan 19, 2003 at 12:53:00PM -0500, Sachin Ganu wrote: > On similar lines, everything below -47dBm is clamped. > My question is what is the actual transmit power of the card when > the card is set at -47dBm? This depends on the card model. I do not know of any method that would give the actual transmit power for the driver; apart from measuring it with an external device. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Sun Jan 19 22:23:53 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 19 Jan 2003 19:23:53 -0800 Subject: select accesspoint with use of bssid In-Reply-To: <1043029080.1853.26.camel@localhost.localdomain> References: <20030119194038.01df77b8.roxxor-mail@binaervarianz.de> <1043029080.1853.26.camel@localhost.localdomain> Message-ID: <20030120032353.GA2677@jm.kir.nu> On Mon, Jan 20, 2003 at 02:18:00AM +0000, wlan wrote: > You can force the card to connect to a preselected AP with the following > command: > iwconfig eth0 ap 00:60:1D:01:23:45 Assuming the driver supports this command. Host AP driver needs to be configured to use host-based roaming (prism2_param wlan0 host_roaming 1) fr this command. > "When the quality of the connection goes too low, the driver may > revert back to automatic mode (the card finds the best Access Point in > range)" > > it is possible to turn off this feature and force the driver to connect > ONLY to the preselected AP? Do you mean that the card should associate only to the selected AP, even if that would mean losing connection? With Host AP driver this would require small change in hostap_info.c: prism2_host_roaming() function would need to be modified to not issue JoinRequest command if local->preferred_ap is set, but it is not found in local->last_scan_results. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Sun Jan 19 22:30:56 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 19 Jan 2003 19:30:56 -0800 Subject: What is the max power for my card? In-Reply-To: <1043028708.1853.20.camel@localhost.localdomain> References: <000901c2bf9a$f4154a60$0200a8c0@buttrump> <20030119173052.GC1404@jm.kir.nu> <1043028708.1853.20.camel@localhost.localdomain> Message-ID: <20030120033056.GB2677@jm.kir.nu> On Mon, Jan 20, 2003 at 02:11:47AM +0000, wlan wrote: > Hostap is the only driver that it is not able to know the maximum > transmit power, or it is common missing feature of all linux drivers? As far as I know, it is common to all Prism2 drivers, no matter which OS. Manufacturers could of course hard code the value if they have own driver versions, but I do not know any generic method of finding this value. > So i understand that this behaviour is due to the reason than we don't > know the maximum value, in such a case won't be better to use > percentages and change iwconfig man page? Yes, it would probably be less confusing to use some raw value or percentage etc. However, it is not enough to just change man pages; this would need to be implemented first.. ;-) The value I use in TX power configuration seems to be logarithmic with most cards, so using dBm has its benifits. Anyway, I would rather start by reimplementing TX power setting using two BBP registers (one to set power level and another to give some measurements from the sent frames). This is the recommended method for host-based TX power controlling, but I haven't so far tested whether it works as expected. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Sun Jan 19 22:41:40 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 19 Jan 2003 19:41:40 -0800 Subject: iwlist txpower -> no answer !? In-Reply-To: <20030119121159.A169@alexis.itd.umich.edu> References: <20030119121159.A169@alexis.itd.umich.edu> Message-ID: <20030120034140.GC2677@jm.kir.nu> On Sun, Jan 19, 2003 at 12:11:59PM -0500, Alexis wrote: > root at alexis:~# iwlist wlan0 txpower > wlan0 No transmit-powers ? Please update driver... > > ^^^^^^ what does this mean ? Which part of that error message is unclear? The one saying that driver did not provide any transmit power levels or the one saying that you would need to update the driver to get this list? ;-) Host AP driver does not report TX levels in that list since it cannot give correct list and there would actually be 128 different values in that list. > I have another question : I read on a list that for iwconfig > 20dBm was the max tx power regardless of the real card capability. > But when connected, doing : > > iwconfig wlan0 txpower 20dBm will set the power output to 14dBm This should not really happen.. Host AP driver should use a local value and not query the card if you have configured txpower. What were the exact commands you used to get this result? > iwconfig wlan0 txpower 200mW will set the power output to 24dBm iwconfig converts that 200mW into 24dBm before passing the value to the driver and the driver return whatever was configured as explained above. -- Jouni Malinen PGP id EFC895FA From wlan at zaragozawireless.org Mon Jan 20 00:05:34 2003 From: wlan at zaragozawireless.org (wlan) Date: 20 Jan 2003 05:05:34 +0000 Subject: select accesspoint with use of bssid In-Reply-To: <20030120032353.GA2677@jm.kir.nu> References: <20030119194038.01df77b8.roxxor-mail@binaervarianz.de> <1043029080.1853.26.camel@localhost.localdomain> <20030120032353.GA2677@jm.kir.nu> Message-ID: <1043039134.13294.9.camel@localhost.localdomain> On Mon, 2003-01-20 at 03:23, Jouni Malinen wrote: > On Mon, Jan 20, 2003 at 02:18:00AM +0000, wlan wrote: > > "When the quality of the connection goes too low, the driver may > > revert back to automatic mode (the card finds the best Access Point in > > range)" > > > > it is possible to turn off this feature and force the driver to connect > > ONLY to the preselected AP? > > Do you mean that the card should associate only to the selected AP, even > if that would mean losing connection? Yes, maybe it is not a really common requirement but could be useful in some situations, for example: avoiding fake Aps for "stealing" your connection, or to force to register to an AP that maybe signal is weaker but it has less clients. I guest should not be very difficult to implement if you think it is worth. Thanks Jouni for you great work! . From jkmaline at cc.hut.fi Mon Jan 20 00:25:57 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 19 Jan 2003 21:25:57 -0800 Subject: select accesspoint with use of bssid In-Reply-To: <1043039134.13294.9.camel@localhost.localdomain> References: <20030119194038.01df77b8.roxxor-mail@binaervarianz.de> <1043029080.1853.26.camel@localhost.localdomain> <20030120032353.GA2677@jm.kir.nu> <1043039134.13294.9.camel@localhost.localdomain> Message-ID: <20030120052557.GA4321@jm.kir.nu> On Mon, Jan 20, 2003 at 05:05:34AM +0000, wlan wrote: > Yes, maybe it is not a really common requirement but could be useful in > some situations, for example: avoiding fake Aps for "stealing" your > connection, or to force to register to an AP that maybe signal is weaker > but it has less clients. First one would require small change that would make it configurable not to use other than configured AP. However, a fake AP can of course fake its BSSID to catch your station.. Second should work already since that AP has to be in range, it will be in the scan results and will be selected (if it was configured to be preferred AP) even if other APs have better signal quality. -- Jouni Malinen PGP id EFC895FA From vda at port.imtp.ilyichevsk.odessa.ua Mon Jan 20 02:45:16 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Mon, 20 Jan 2003 09:45:16 +0200 Subject: DLink DWL-520 + hostap in STA mode In-Reply-To: <200301170808.h0H88gs20786@Port.imtp.ilyichevsk.odessa.ua> References: <200301170808.h0H88gs20786@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <200301200746.h0K7k0s05270@Port.imtp.ilyichevsk.odessa.ua> On 17 January 2003 10:08, Denis Vlasenko wrote: > DLink DWL-520 > 2.4.20-pre11, wireless extensions 14, hostap-2002-10-12 > > We (my father and me) made a fairly long helical aerial. > We are trying to communicate over ~15 km with a small wireless cell. > (~10 hosts, one AP). > > Mainline kernel's orinoco_pci module works but seems to lose > an association within ~30 secs and is not able to restore it without > ifconfig down/up cycle. Win98 driver fares better but dies too > after ~3 mins. > > I am trying hostap now. It loads without problems but does not work, > I am unable to switch the node into STA mode. > Looks like card's firmware is old. Where can I find newer one? > (I'll go googling to try and find it myself...) Found 1.4.9 firmware. Flashed it in. Makes lots of difference. Basically everything works with mainline kernel's orinoco_pci (stopped using hostap since I do not need AP now). Thanks guys for all your numerous responses ;) -- vda From vda at port.imtp.ilyichevsk.odessa.ua Mon Jan 20 02:58:56 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Mon, 20 Jan 2003 09:58:56 +0200 Subject: Setting MAC address Message-ID: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> Hi wireless people. I seem to be unable to set MAC address. It seems to work and I can see my frames with right MAC address with tcpdump. The problem is, other hosts do not like me anymore :( I see zero incoming packets. This happens even if I set MAC addr before bringing iface up. I.e. no host can possibly cache stale ARP data... I'd be grateful if anyone will do a quick check on their hardware/firmware/driver combo and tell me the results. -- vda From cijoml at volny.cz Mon Jan 20 05:28:04 2003 From: cijoml at volny.cz (Michal Semler) Date: Mon, 20 Jan 2003 11:28:04 +0100 Subject: DLink DWL-520 + hostap in STA mode In-Reply-To: <200301200746.h0K7k0s05270@Port.imtp.ilyichevsk.odessa.ua> References: <200301170808.h0H88gs20786@Port.imtp.ilyichevsk.odessa.ua> <200301200746.h0K7k0s05270@Port.imtp.ilyichevsk.odessa.ua> Message-ID: Dne po 20. leden 2003 08:45 jste napsal(a): > On 17 January 2003 10:08, Denis Vlasenko wrote: > > DLink DWL-520 > > 2.4.20-pre11, wireless extensions 14, hostap-2002-10-12 > > > > We (my father and me) made a fairly long helical aerial. > > We are trying to communicate over ~15 km with a small wireless cell. > > (~10 hosts, one AP). > > > > Mainline kernel's orinoco_pci module works but seems to lose > > an association within ~30 secs and is not able to restore it without > > ifconfig down/up cycle. Win98 driver fares better but dies too > > after ~3 mins. > > > > I am trying hostap now. It loads without problems but does not work, > > I am unable to switch the node into STA mode. > > Looks like card's firmware is old. Where can I find newer one? > > (I'll go googling to try and find it myself...) > > Found 1.4.9 firmware. Flashed it in. Makes lots of difference. > Basically everything works with mainline kernel's orinoco_pci > (stopped using hostap since I do not need AP now). > > Thanks guys for all your numerous responses ;) If u want firmware 1.5.6 - here it is... :) http://aldebaran.feld.cvut.cz/~xmyslik/wifi Michal From james.harper at bigpond.com Mon Jan 20 06:33:59 2003 From: james.harper at bigpond.com (James Harper) Date: Mon, 20 Jan 2003 22:33:59 +1100 Subject: prism2_srec & capabilities Message-ID: <3E2BDEA7.1000803@bigpond.com> when i try to use prism2_srec I get an access denied message, but not before the card is crashed (iwpriv wlan0 reset 1 fixes it though). looks like the capability checking is done in hostap_ioctl. i'm using kernel 2.5.58 any suggestions? i have an idea my kernel doesn't properly support the capability stuff and is failing. setcap claims capabilities are not supported. james From thed at intracom.gr Mon Jan 20 11:31:11 2003 From: thed at intracom.gr (Dimitris Theofilatos) Date: Mon, 20 Jan 2003 16:31:11 +0000 Subject: HostAP and xsupplicant In-Reply-To: <20030114030006.GA1456@jm.kir.nu> References: <200301131513.25081.thed@intracom.gr> <20030114030006.GA1456@jm.kir.nu> Message-ID: <200301201631.11502.thed@intracom.gr> On Tuesday 14 January 2003 03:00, Jouni Malinen wrote: > On Mon, Jan 13, 2003 at 03:13:25PM +0000, Dimitris Theofilatos wrote: > > Does anyone know if the HostAP code interworks with the xsupplicant code > > from open1x? I followed the instructions from the site > > www.missl.cs.umd.edu/wireless/eaptls/. I used 3 linux machines with > > RedHat 7.3 and kernel 2.4.18. I used hostap-2002-10-12 driver and the > > latest version of xsupplicant (xsupplicant-0.5) but when the xsupplicant > > sends the first EAPOL Start message the hostapd daemon unauthorize the > > station. > > I tested xsupplicant v0.5 today, but it segfaulted when I tried to > authenticate.. I didn't yet have time to check why. Anyway, older > version of xsupplicant work fine with the same certificates and > configuration using hostapd as Authenticator and FreeRADIUS as > Authentication Server. > > Authenticator state machines set the port into unauthorized state when a > new station is added so it is OK to get that messages after EAPOL-Start. > The port is authorized for the first time only after successful > authentication has been completed. I used an older version of xsupplicant and the last CVS version of hostap and it worked with the minimal authentication server of the hostap. But i had a problem when i used another machine for RADIUS. I enabled routing at the machine which was acting as an Access Point and i saw that the Access Point sends an EAP-Request/Identity at the supplicant, the supplicant responds with a EAP-Response/Identity, then the AP sends an Access-Request to the Radius server, the Radius responds with an Access-Challenge but the authenticator does not forward this message to the supplicant. Do you know why? Thanks in advance Dimitris From mpolak at gup.jku.at Mon Jan 20 07:02:35 2003 From: mpolak at gup.jku.at (Martin Polak) Date: Mon, 20 Jan 2003 13:02:35 +0100 Subject: Two host_aps in WDS Bridgeing mode, Channel selection? Message-ID: <3E2BE55B.4030807@gup.jku.at> Hi Everyone! I have set up two hostaps like in the README.prism2 file, using kernel bridging and a WDS-Link between them and everything works fine. The only difference between the described setup and mine is, that both APs allow stations to associate. Do I see it right, that these two APs HAVE to be on the same frequency and there share the same Medium? Martin From jasonb at edseek.com Mon Jan 20 10:37:34 2003 From: jasonb at edseek.com (Jason Boxman) Date: Mon, 20 Jan 2003 10:37:34 -0500 Subject: Setting MAC address In-Reply-To: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> References: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <200301201037.34253.jasonb@edseek.com> On Monday 20 January 2003 02:58 am, Denis Vlasenko wrote: > Hi wireless people. > > I seem to be unable to set MAC address. > It seems to work and I can see my frames with right > MAC address with tcpdump. The problem is, other hosts > do not like me anymore :( I see zero incoming packets. > This happens even if I set MAC addr before bringing > iface up. I.e. no host can possibly cache stale ARP > data... I have similiar results. Communications simply stop or never start. Most distressing. > I'd be grateful if anyone will do a quick check on their > hardware/firmware/driver combo and tell me the results. > -- > vda -- Jason Boxman Administrator TrekWeb.COM http://trekweb.com/~jasonb/articles/ - Linux Pages From magnus_k at gmx.de Mon Jan 20 18:19:23 2003 From: magnus_k at gmx.de (Magnus) Date: Tue, 21 Jan 2003 00:19:23 +0100 Subject: one way bad: dlink dwl-520, netgear me102 Message-ID: <000001c2c0da$63252f20$0200a8c0@zoidberg> We use a combination of netgear me102 accesspoint and dlink dwl-520 (11 Mbit, using hostap_pci-driver) in the router to connect a lan to a dsl-router (suse-linux 8.2, kernelversion 2.4). The problem is that in one direction (from router to lan) the transmission is very good, but in the other direction (from lan to router) the transmission is very worse. The result is that emails with attachment cannot be sent, filesharing only works in one direction and no files can be uploaded per ftp to the server (it starts transmitting but after a while, maybe 500 KB, it stopps). Normal internet works, downloads ok. Running iwconfig on router prints: ================================== wlan0 IEEE 802.11-b ESSID:"wlan" Nickname:"uchajewanije" Mode:Managed Frequency:2.472GHz Access Point: 00:30:AB:22:85:3B Bit Rate:11Mb/s Tx-Power:10 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:24/92 Signal level:-86 dBm Noise level:-100 dBm Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:320 Invalid misc:7559 Missed beacon:0 ================================== Running ifconfig on router prints: ================================== ... wlan0 Link encap:Ethernet HWaddr 00:40:05:AE:CE:B2 inet addr:192.168.0.11 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::240:5ff:feae:ceb2/10 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2047522 errors:0 dropped:6659 overruns:0 frame:0 TX packets:2159915 errors:638 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:208447513 (198.7 Mb) TX bytes:1985993069 (1893.9 Mb) Interrupt:9 Memory:c28de000-c28df000 ================================== Why are there 6659 received packets dropped, while on every else interface 0 are dropped. And also transmitted packets on every else inface 0 errors occured. Also pinging a host from the router over the wlan-connection produces strange output: ================================== uchajewanije:/etc # ping -s 1000 192.168.0.2 PING 192.168.0.2 (192.168.0.2) from 192.168.0.11 : 1000(1028) bytes of data. 1008 bytes from 192.168.0.2: icmp_seq=1 ttl=128 time=6.20 ms 1008 bytes from 192.168.0.2: icmp_seq=2 ttl=128 time=5.97 ms wrong data byte #200 should be 0xc8 but was 0xca #8 8 9 a b c d e f 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 #40 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 38 39 3a 3b 3c 3d 3e 3f 40 41 42 43 44 45 46 47 ... #968 c8 c9 ca cb cc cd ce cf d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 da db dc dd de df e0 e1 e2 e3 e4 e5 e6 e7 1008 bytes from 192.168.0.2: icmp_seq=3 ttl=128 time=5.63 ms ================================== The bigger the packetsize is and the longer you send, the more wrong data byte errors occure. We think it is most possible that some parameters are wrong or the hostap-driver is buggy or our connection is semidirectional bad. Thank you for reply. Sincerly, Magnus. From jkmaline at cc.hut.fi Mon Jan 20 22:54:59 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 20 Jan 2003 19:54:59 -0800 Subject: one way bad: dlink dwl-520, netgear me102 In-Reply-To: <000001c2c0da$63252f20$0200a8c0@zoidberg> References: <000001c2c0da$63252f20$0200a8c0@zoidberg> Message-ID: <20030121035459.GA1478@jm.kir.nu> On Tue, Jan 21, 2003 at 12:19:23AM +0100, Magnus wrote: > We use a combination of netgear me102 accesspoint and dlink dwl-520 (11 > Mbit, using hostap_pci-driver) in the router to connect a lan to a > dsl-router (suse-linux 8.2, kernelversion 2.4). Which primary and station firmware versions are you using in D-Link DWL-520? If PRI < 1.0.7 or STA < 1.3.5, you should upgrade them. Older versions are known to corrupt packet data when it is transfered over PCI bus.. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Mon Jan 20 23:09:54 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 20 Jan 2003 20:09:54 -0800 Subject: Setting MAC address In-Reply-To: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> References: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <20030121040954.GA1743@jm.kir.nu> On Mon, Jan 20, 2003 at 09:58:56AM +0200, Denis Vlasenko wrote: > I seem to be unable to set MAC address. > It seems to work and I can see my frames with right > MAC address with tcpdump. The problem is, other hosts > do not like me anymore :( I see zero incoming packets. This works in my tests, both as an AP and as a station. When running hostapd, one must restart hostapd after this change. Other than that, there should not be other limitations on this. Anyway, I would recommend changing the MAC address when the device is down. Please give some details about the system and configuration that behaves incorrectly. Are you changing the MAC address in an AP or a station? Which firmware versions are you using? I'm mostly interested in which command you run and in which order, what ifconfig shows as the MAC address of wlan0 device after the changed, do you see errors in 'dmesg' output, and what 'hostap_diag wlan0 -r | grep FC01' shows as the own address. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Mon Jan 20 23:16:45 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 20 Jan 2003 20:16:45 -0800 Subject: prism2_srec & capabilities In-Reply-To: <3E2BDEA7.1000803@bigpond.com> References: <3E2BDEA7.1000803@bigpond.com> Message-ID: <20030121041645.GB1743@jm.kir.nu> On Mon, Jan 20, 2003 at 10:33:59PM +1100, James Harper wrote: > when i try to use prism2_srec I get an access denied message, but not > before the card is crashed (iwpriv wlan0 reset 1 fixes it though). Some more details, please. Exactly what kind of access denied message? What was printed before this error message? Are there anything interesting in 'dmesg' output? What kind of prism2_srec command line did you use? What PRI and STA firmware version did the card have originally? > i'm using kernel 2.5.58 Have you tested with other versions? prism2_srec works on my laptop at least with 2.5.47; I haven't tested firmware downloading with 2.5.58. > any suggestions? i have an idea my kernel doesn't properly support the > capability stuff and is failing. setcap claims capabilities are not > supported. Firmware downloading should not be using anything special as far as capability verification is concerned. I have assumed that capable(CAP_NET_ADMIN) would work with any kernel (at least it works for me ;-). So, if you are able to set SSID with 'iwconfig wlan0 essid testing', you should have enough privileges for firmware downloading (assuming it is compiled into the driver). -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Mon Jan 20 23:18:39 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 20 Jan 2003 20:18:39 -0800 Subject: Two host_aps in WDS Bridgeing mode, Channel selection? In-Reply-To: <3E2BE55B.4030807@gup.jku.at> References: <3E2BE55B.4030807@gup.jku.at> Message-ID: <20030121041839.GC1743@jm.kir.nu> On Mon, Jan 20, 2003 at 01:02:35PM +0100, Martin Polak wrote: > I have set up two hostaps like in the README.prism2 file, using kernel > bridging and a WDS-Link between them and everything works fine. > Do I see it right, that these two APs HAVE to be on the same frequency > and there share the same Medium? Yes, if both APs have only one radio, you must use the same channel on both APs to make WDS link work. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Tue Jan 21 00:11:48 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 20 Jan 2003 21:11:48 -0800 Subject: HostAP and xsupplicant In-Reply-To: <200301201631.11502.thed@intracom.gr> References: <200301131513.25081.thed@intracom.gr> <20030114030006.GA1456@jm.kir.nu> <200301201631.11502.thed@intracom.gr> Message-ID: <20030121051148.GA2513@jm.kir.nu> On Mon, Jan 20, 2003 at 04:31:11PM +0000, Dimitris Theofilatos wrote: > But i had a problem when i used another machine for RADIUS. I enabled routing > at the machine which was acting as an Access Point and i saw that the Access > Point sends an EAP-Request/Identity at the supplicant, the supplicant > responds with a EAP-Response/Identity, then the AP sends an Access-Request to > the Radius server, the Radius responds with an Access-Challenge but the > authenticator does not forward this message to the supplicant. Do you know > why? Have you checked whether hostapd reports any errors when you run it with full debugging (add '-ddd' to the command line)? IP routing is not used for these frames and hostapd debug logs should show whether it received Access-Challenge (and whether it passed all verifications). In addition, logs should show if the frame was sent (or at least attempted) to the station. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Tue Jan 21 00:17:08 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 20 Jan 2003 21:17:08 -0800 Subject: hostapd issues and questions In-Reply-To: <000a01c2bfdd$966d2c50$017ba8c0@ammirata.net> References: <20030119021518.GE2022@jm.kir.nu> <000a01c2bfdd$966d2c50$017ba8c0@ammirata.net> Message-ID: <20030121051708.GB2513@jm.kir.nu> On Sun, Jan 19, 2003 at 12:09:35PM -0500, Sergio M. Ammirata wrote: > 1) Changing the mac address of the wlan0 interface using ifconfig breaks > routing (you can link but the tcp/ip layer does not work). > This happens only with hostapd enabled. Was hostapd running when you changed MAC address? You must restart hostapd after this change. > 2) Without hostapd there is wlan0 and wlan0wdsx. Now we have wlan0ap and > wlan0sta as well. > How are we supposed to setup Ethernet bridging now? Do we include all > three in the bridge (wlan0, wlan0ap and wlan0sta)? wlan0ap should not be included in bridge setup at all. It used only for management frames (802.11 auth/assoc, 802.1X EAPOL frames) within that one AP. wlan0sta is an experimental device used for station mode operations with another AP if assoc_ap_addr is set. It does not support layer-2 bridging (just like any other "normal" station mode card), so it should not be included in bridge setup. > 3) Has the functionality of wlan0wdsx changed? When and why would I use > wlan0sta? No, wlan0wds# work as before. You could potentially use wlan0sta as a wireless uplink with IP routing (and possible Proxy ARPing). > 4) With hostapd active iwspy no longer reports any info on wds > neighbors. How can I found out the strength of the links? That is not implemented yet. -- Jouni Malinen PGP id EFC895FA From cupid90 at samsung.co.kr Tue Jan 21 03:02:45 2003 From: cupid90 at samsung.co.kr (cupid90 at samsung.co.kr) Date: Tue, 21 Jan 2003 17:02:45 +0900 Subject: Somebody have the programmers guide of intersil chipset? Message-ID: Hi, this is Mr. roh I'd like to start debugging "Transmit timed out" symptom at pcmcia card. If somebody have the manuals and programer's guide, please send it to me. If I resolve the problems, I will upload the result. I'll wait your mail... Thanks Roh you-chang Application S/W Engineer System Solution Group Access Network SOC Team System LSI Business Device Solution Network SAMSUNG ELECTRONICS CO., LTD. phone:82-31-209-8314, fax:82-31-209-8309 e-mail:cupid90 at samsung.co.kr From Dave.Edwards at abicom-international.com Tue Jan 21 03:04:10 2003 From: Dave.Edwards at abicom-international.com (Dave Edwards) Date: Tue, 21 Jan 2003 08:04:10 +0000 Subject: Somebody have the programmers guide of intersil chipset? References: Message-ID: <3E2CFEFA.2020001@abicom-international.com> Contact Intersil, They will send you the latest manual once you sign an NDA. Dave cupid90 at samsung.co.kr wrote: >Hi, this is Mr. roh > >I'd like to start debugging "Transmit timed out" symptom at pcmcia card. > >If somebody have the manuals and programer's guide, please send it to me. > >If I resolve the problems, I will upload the result. > >I'll wait your mail... > >Thanks > > > > Roh you-chang > > Application S/W Engineer > System Solution Group > Access Network SOC Team > System LSI Business > Device Solution Network > SAMSUNG ELECTRONICS CO., LTD. > > phone:82-31-209-8314, fax:82-31-209-8309 > e-mail:cupid90 at samsung.co.kr > > >_______________________________________________ >HostAP mailing list >HostAP at shmoo.com >http://lists.shmoo.com/mailman/listinfo/hostap > > > From ammirata at econointl.com Tue Jan 21 04:31:29 2003 From: ammirata at econointl.com (Sergio Ammirata) Date: Tue, 21 Jan 2003 04:31:29 -0500 Subject: hostapd issues and questions In-Reply-To: <20030121051708.GB2513@jm.kir.nu> Message-ID: <000001c2c12f$e2b631d0$350aa8c0@sergioxp> > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] > On Behalf Of Jouni Malinen > Sent: Tuesday, January 21, 2003 12:17 AM > To: hostap at shmoo.com > Subject: Re: hostapd issues and questions > > > On Sun, Jan 19, 2003 at 12:09:35PM -0500, Sergio M. Ammirata wrote: > > > 1) Changing the mac address of the wlan0 interface using ifconfig > > breaks routing (you can link but the tcp/ip layer does not > work). This > > happens only with hostapd enabled. > > Was hostapd running when you changed MAC address? You must > restart hostapd after this change. Hostapd is not running when I change the MAC address. Here is the script I use: echo -n "Starting wlan0: " insmod hostap_crypt insmod hostap_crypt_wep insmod hostap # Start in infrastructure mode so that the old MAC Address does not get exposed. insmod hostap_cs essid=test iw_mode=2 channel=6 echo -n "Initializing pcmcia" /etc/rc.d/init.d/pcmcia init # Set the to the same mac as the ethernet interface HWADDR=`/sbin/ifconfig eth0 | sed -n 's/.*HWaddr\(.*\)/\1/p'` ifconfig wlan0 hw ether $HWADDR iwconfig wlan0 mode master # Configure Wireless parameters iwconfig wlan0 txpower 23 iwconfig wlan0 sens 3 iwconfig wlan0 nick $hostname # Disable bridging (use ethernet bridges instead) iwpriv wlan0 bridge_packets 0 iwpriv wlan0 host_roaming 1 # Now that everything is configured bring up the wlan0 interface and enabled the hostap daemon ifconfig wlan0 0.0.0.0 # Start user space daemon hostapd /etc/hostapd.conf echo "Done." > > > 2) Without hostapd there is wlan0 and wlan0wdsx. Now we > have wlan0ap > > and wlan0sta as well. How are we supposed to setup Ethernet > bridging > > now? Do we include all three in the bridge (wlan0, wlan0ap and > > wlan0sta)? > > wlan0ap should not be included in bridge setup at all. It > used only for management frames (802.11 auth/assoc, 802.1X > EAPOL frames) within that one AP. wlan0sta is an experimental > device used for station mode operations with another AP if > assoc_ap_addr is set. It does not support layer-2 bridging > (just like any other "normal" station mode card), so it > should not be included in bridge setup. > > 3) Has the functionality of wlan0wdsx changed? When and why would I > > use wlan0sta? > > No, wlan0wds# work as before. You could potentially use > wlan0sta as a wireless uplink with IP routing (and possible > Proxy ARPing). I noticed that wlan0sta does not have an independent essid. When you change it on wlan0 it also changes for wlan0sta. If the destination MAC for wlan0sta has a different essid than wlan0, will it still link? > > 4) With hostapd active iwspy no longer reports any info on wds > > neighbors. How can I found out the strength of the links? > > That is not implemented yet. Will it make it into this release? Is there an alternate way of measuring the link speed or to even found out that we are linked to neighbors other than pinging? > -- > Jouni Malinen PGP > id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap Thanks in advance, Sergio From kenw at keeper.org.uk Tue Jan 21 05:30:53 2003 From: kenw at keeper.org.uk (Ken Wolstencroft) Date: Tue, 21 Jan 2003 10:30:53 +0000 Subject: Dynamic WEP keying and Hostap Message-ID: <3E2D215D.7030709@keeper.org.uk> I have set up EAP-TLS using hostap/hostapd, but I am a structling to get dynamic WEP keying working. Can anyone give me some advice on how to set this up using hostap/hostapd. I am using Freeradius as the authenticator and WinXP supplicant. Thanks and all the best, Ken www.wnap.org.uk From kenw at keeper.org.uk Tue Jan 21 05:40:14 2003 From: kenw at keeper.org.uk (Ken Wolstencroft) Date: Tue, 21 Jan 2003 10:40:14 +0000 Subject: dynamic WEP keying Message-ID: <3E2D238E.5010201@keeper.org.uk> I have set up EAP-TLS using hostap/hostapd, but I am a structling (I mean struggling) to get dynamic WEP keying working. Can anyone give me some advice on how to set this up using hostap/hostapd. I am using Freeradius as the authenticator and WinXP supplicant. Thanks and all the best, Ken www.wnap.org.uk From mohit at mahindrabt.com Tue Jan 21 06:22:09 2003 From: mohit at mahindrabt.com (Mohit Bajpai) Date: Tue, 21 Jan 2003 16:52:09 +0530 Subject: IAPP using HostAP-2002-10-12 Message-ID: <002901c2c13f$56fb37c0$f009030a@mahindrabt.com> Hi Jouni, We are in an advanced stage of implementing IAPP on HostAP for which we are using HostAP 2002-10-12 release for our implementation. 802.11f Draft 3.1 has been taken as a reference for the development of IAPP stack. We have used a modular approach while implementing the same & done minimal changes to the main HostAP code. . IAPP specific code has been mostly implemented as an additional library of functions to be linked to the main HostAP .We are facing some problems while implementing this related to retrieval of STA information from HostAP. Could you please suggest a way to retrieve the STA related information from hostAP with respect to the following 1.. Is the STA already associated with HostAP. If yes, what is its association Id ? 2.. Where and how can I remove association of STA in the old AP after it moves to a New AP area. Please reply. Thanks and Regards Mohit Bajpai -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030121/511ab974/attachment.htm From magnus_k at gmx.de Mon Jan 20 19:10:12 2003 From: magnus_k at gmx.de (Magnus) Date: Tue, 21 Jan 2003 01:10:12 +0100 Subject: one way bad: dlink dwl-520, netgear me102 Message-ID: <000101c2c0e1$7ae94e00$0200a8c0@zoidberg> We use a combination of netgear me102 accesspoint and dlink dwl-520 (11 Mbit, using hostap_pci-driver) in the router to connect a lan to a dsl-router (suse-linux 8.2, kernelversion 2.4). The problem is that in one direction (from router to lan) the transmission is very good, but in the other direction (from lan to router) the transmission is very worse. The result is that emails with attachment cannot be sent, filesharing only works in one direction and no files can be uploaded per ftp to the server (it starts transmitting but after a while, maybe 500 KB, it stopps). Normal internet works, downloads ok. Running iwconfig on router prints: ================================== wlan0 IEEE 802.11-b ESSID:"wlan" Nickname:"uchajewanije" Mode:Managed Frequency:2.472GHz Access Point: 00:30:AB:22:85:3B Bit Rate:11Mb/s Tx-Power:10 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:24/92 Signal level:-86 dBm Noise level:-100 dBm Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:320 Invalid misc:7559 Missed beacon:0 ================================== Running ifconfig on router prints: ================================== ... wlan0 Link encap:Ethernet HWaddr 00:40:05:AE:CE:B2 inet addr:192.168.0.11 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::240:5ff:feae:ceb2/10 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2047522 errors:0 dropped:6659 overruns:0 frame:0 TX packets:2159915 errors:638 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:208447513 (198.7 Mb) TX bytes:1985993069 (1893.9 Mb) Interrupt:9 Memory:c28de000-c28df000 ================================== Why are there 6659 received packets dropped, while on every else interface 0 are dropped. And also transmitted packets on every else inface 0 errors occured. Also pinging a host from the router over the wlan-connection produces strange output: ================================== uchajewanije:/etc # ping -s 1000 192.168.0.2 PING 192.168.0.2 (192.168.0.2) from 192.168.0.11 : 1000(1028) bytes of data. 1008 bytes from 192.168.0.2: icmp_seq=1 ttl=128 time=6.20 ms 1008 bytes from 192.168.0.2: icmp_seq=2 ttl=128 time=5.97 ms wrong data byte #200 should be 0xc8 but was 0xca #8 8 9 a b c d e f 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e 1f 20 21 22 23 24 25 26 27 #40 28 29 2a 2b 2c 2d 2e 2f 30 31 32 33 34 35 36 37 38 39 3a 3b 3c 3d 3e 3f 40 41 42 43 44 45 46 47 ... #968 c8 c9 ca cb cc cd ce cf d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 da db dc dd de df e0 e1 e2 e3 e4 e5 e6 e7 1008 bytes from 192.168.0.2: icmp_seq=3 ttl=128 time=5.63 ms ================================== The bigger the packetsize is and the longer you send, the more wrong data byte errors occure. We think it is most possible that some parameters are wrong or the hostap-driver is buggy or our connection is semidirectional bad. Thank you for reply. Sincerly, Magnus. From matt at softec.uk.com Tue Jan 21 12:16:18 2003 From: matt at softec.uk.com (Matt Connew) Date: Tue, 21 Jan 2003 17:16:18 -0000 Subject: Error messages from hostap Message-ID: I'm running hostap in AP mode with a Netgear MA311 (A PCI card using the Prism2.5 chipset). Here's is the ID info from the HostAP start up: hostap_crypt: registered algorithm 'NULL' hostap_pci: hostap_pci.c 0.0.0 2002-10-12 (SSH Communications Security Corp, Jouni Malinen) hostap_pci: (c) Jouni Malinen kmaline at cc.hut.fi> PCI: Found IRQ 12 for device 00:0b.0 hostap_pci: Registered netdevice wlan0 prism2_hw_init() prism2_hw_config: initialized in 19798 iterations wlan0: NIC: id=0x8013 v1.0.0 wlan0: PRI: id=0x15 v1.0.7 wlan0: STA: id=0x1f v1.3.6 wlan0: defaulting to host-based encryption as a workaround for firmware bug in Host AP mode WEP wlan0: LinkStatus=2 (Disconnected) wlan0: Intersil Prism2.5 PCI: mem=0xdd102000, irq=12 Setting ALC off Setting TX power to 15 dBm I'm am using WDS to connect a wireless only AP to another AP wired to the network (both HostAP). The version of HostAP I'm using is 2002-10-12. The 2 AP's connect and forward packets to each other without any problems. The problem is that I get the following errors repeatedly in my syslog: AP: packet to non-associated STA 00:c0:df:02:9d:f0 wlan0: Could not find STA for this TX error The first error is complaining about the mac address of my workstation on the wired LAN. Why is the hostap driver complaining about this? The second error message just keeps on being emitted at a rate of several a second until I reboot, and then it starts all over again. I get this from both AP's, both the wired and wireless. I've tried changing the ap_bridge_packets mode and it still does it. Can anyone tell me what's going on and how to stop the error messages? Matt From maillists at sonictech.net Tue Jan 21 15:40:46 2003 From: maillists at sonictech.net (Ray) Date: Tue, 21 Jan 2003 13:40:46 -0700 Subject: Error messages from hostap In-Reply-To: References: Message-ID: <20030121204046.GA23163@sonictech.net> I don't have an answer to your question but there was some mention of a similar problem in the newsgroup comp.os.linux.hardware around Jan 7-8. I've got the pcmcia version of this card but I havn't had a chance to try it out yet (I'm using the Seno cards). -- Ray From magnus_k at gmx.de Tue Jan 21 16:09:47 2003 From: magnus_k at gmx.de (Magnus) Date: Tue, 21 Jan 2003 22:09:47 +0100 Subject: one way bad: dlink dwl-520, netgear me102 Message-ID: <000901c2c191$726add60$0200a8c0@zoidberg> >> We use a combination of netgear me102 accesspoint and dlink dwl-520 (11 >> Mbit, using hostap_pci-driver) in the router to connect a lan to a >> dsl-router (suse-linux 8.2, kernelversion 2.4). > > Which primary and station firmware versions are you using in D-Link > DWL-520? If PRI < 1.0.7 or STA < 1.3.5, you should upgrade them. Older > versions are known to corrupt packet data when it is transfered over PCI > bus.. > > -- > Jouni Malinen PGP id EFC895FA Since when use pci-networkadaptercards a firmware? On D-Link-Homepage no firmware for dwl-520 exists. Magnus From jkmaline at cc.hut.fi Tue Jan 21 21:47:22 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Tue, 21 Jan 2003 18:47:22 -0800 Subject: IAPP using HostAP-2002-10-12 In-Reply-To: <002901c2c13f$56fb37c0$f009030a@mahindrabt.com> References: <002901c2c13f$56fb37c0$f009030a@mahindrabt.com> Message-ID: <20030122024722.GA1488@jm.kir.nu> On Tue, Jan 21, 2003 at 04:52:09PM +0530, Mohit Bajpai wrote: > We are in an advanced stage of implementing IAPP on HostAP for which > we are using HostAP 2002-10-12 release for our implementation. Are you using hostapd for this or the IAPP functionality in a separate program? Please do note that hostapd in the CVS version of Host AP driver includes initial IAPP implementation that I'm going to use as the basis for full IAPP implementation.. > 802.11f Draft 3.1 has been taken as a reference for the development of > IAPP stack. Please note that a newer version (draft 5) is already available. > We have used a modular approach while implementing the same & done minimal changes to the main HostAP code. . IAPP specific code has been mostly implemented as an additional library of functions to be linked to the main HostAP .We are facing some problems while implementing this related to retrieval of STA information from HostAP. By "HostAP", do you refer to hostapd daemon or the kernel driver? Could you please describe what are the main design differences between your and my implementation of IAPP? > Could you please suggest a way to retrieve the STA related information from hostAP with respect to the following > > 1.. Is the STA already associated with HostAP. If yes, what is its association Id ? > 2.. Where and how can I remove association of STA in the old AP after it moves to a New AP area. I'm assuming you are linking your code with hostapd. In this case you can use ap_get_sta() function to fetch station info based on MAC address and sta->aid contains the association id allocated for that station (or 0 if AID has not yet been allocated). See CVS version of hostap/hostapd/iapp.c and function iapp_process_add_notify() for an example how to remove association on the old AP. -- Jouni Malinen PGP id EFC895FA From wireless at douglasheld.net Mon Jan 20 11:11:07 2003 From: wireless at douglasheld.net (wireless at douglasheld.net) Date: Mon, 20 Jan 2003 08:11:07 -0800 Subject: Setting MAC address References: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> <200301201037.34253.jasonb@edseek.com> Message-ID: <000001c2c1d9$5b470440$0200a8c0@buttrump> I didn't think every NIC card was able to reassign its MAC address. Is there a reason why the driver would neccesarily be able to do this? --Doug ----- Original Message ----- From: "Jason Boxman" To: Sent: Monday, January 20, 2003 7:37 AM Subject: Re: Setting MAC address On Monday 20 January 2003 02:58 am, Denis Vlasenko wrote: > Hi wireless people. > > I seem to be unable to set MAC address. > It seems to work and I can see my frames with right > MAC address with tcpdump. The problem is, other hosts > do not like me anymore :( I see zero incoming packets. > This happens even if I set MAC addr before bringing > iface up. I.e. no host can possibly cache stale ARP > data... I have similiar results. Communications simply stop or never start. Most distressing. > I'd be grateful if anyone will do a quick check on their > hardware/firmware/driver combo and tell me the results. > -- > vda -- Jason Boxman Administrator TrekWeb.COM http://trekweb.com/~jasonb/articles/ - Linux Pages _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap From wlan at zaragozawireless.org Wed Jan 22 08:07:02 2003 From: wlan at zaragozawireless.org (wlan) Date: 22 Jan 2003 13:07:02 +0000 Subject: IAPP using HostAP-2002-10-12 In-Reply-To: <20030122024722.GA1488@jm.kir.nu> References: <002901c2c13f$56fb37c0$f009030a@mahindrabt.com> <20030122024722.GA1488@jm.kir.nu> Message-ID: <1043240823.13188.30.camel@localhost.localdomain> > > Could you please suggest a way to retrieve the STA related information from hostAP with respect to the following > > > > 1.. Is the STA already associated with HostAP. If yes, what is its association Id ? > > 2.. Where and how can I remove association of STA in the old AP after it moves to a New AP area. > > I'm assuming you are linking your code with hostapd. In this case you > can use ap_get_sta() function to fetch station info based on MAC address > and sta->aid contains the association id allocated for that station (or > 0 if AID has not yet been allocated). Is there a external command to get stations associated with an hostap AP? Regards From bercab at bulma.net Wed Jan 22 08:03:38 2003 From: bercab at bulma.net (Bernardo Cabezas Serra) Date: Wed, 22 Jan 2003 14:03:38 +0100 Subject: Monitor mode with WE-16 Message-ID: <200301221403.38498.bercab@bulma.net> Hello all, I have problems while switching to Monitor mode since I upgraded Wireless Extensions API to version 16 (with 2.4.20 vanilla kernel). I'm using hostap from CVS, and my wireless card is a Conceptronic CON11CC. Ah, and wireless-tools' version is 26pre7. When I try "iwconfig wlan0 mode monitor", I get the following: -- begin error log --- [...] Jan 22 13:15:17 localhost kernel: prism2: wlan0: operating mode changed 2 -> 6 Jan 22 13:15:17 localhost kernel: Enabling monitor mode Jan 22 13:15:17 localhost kernel: wlan0: LinkStatus=2 (Disconnected) Jan 22 13:15:17 localhost kernel: wlan0: LinkStatus: BSSID=00:00:00:00:00:00 Jan 22 13:15:17 localhost kernel: wlan0: LinkStatus=2 (Disconnected) Jan 22 13:15:17 localhost kernel: wlan0: LinkStatus: BSSID=00:50:c2:01:9e:34 Jan 22 13:15:20 localhost kernel: wlan0: hfa384x_cmd: entry still in list? (entry=c48827e0, type=0, res=0) Jan 22 13:15:20 localhost kernel: wlan0: hfa384x_cmd: command was not completed (res=0, entry=c48827e0, type=0, cmd=0x0021, param0=0xfd43) Jan 22 13:15:20 localhost kernel: wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd43, len=6) Jan 22 13:15:20 localhost kernel: wlan0: hfa384x_cmd_issue: cmd reg was busy for 1000 usec Jan 22 13:15:20 localhost kernel: wlan0: hfa384x_cmd_issue - timeout - reg=0x8000 [...] -- end error log -- This continues until switching to Managed mode and reseting card (with iwpriv wlan0 reset 1) Am I doing anythig wrong? Another related issue is that almost all aplications have some trouble with the new WE API and hostap. For example, "wavemon" submits ioctls with ".u.data.length" field of struct iwreq to zero, and seems hostap don't like it. I've got it runnig by setting length field to the "sizeof" of the data pointed to by "iwreq.u.data.poiter" (and the same process for every ioctl in wavemon). Can be considered a bug in wavemon the fact of setting "length" fieds to 0 ? It seems hard to be "on the edge" ;-) Tank you very much (and sorry for my poor english) Regards, Bernardo -- Bernardo Cabezas From mohit at mahindrabt.com Wed Jan 22 08:38:33 2003 From: mohit at mahindrabt.com (Mohit Bajpai) Date: Wed, 22 Jan 2003 19:08:33 +0530 Subject: IAPP using HostAP-2002-10-12 Message-ID: <009a01c2c21b$8f364800$f009030a@mahindrabt.com> Thanks for replying. > We are in an advanced stage of implementing IAPP on HostAP for which > we are using HostAP 2002-10-12 release for our implementation. Are you using hostapd for this or the IAPP functionality in a separate program? Please do note that hostapd in the CVS version of Host AP driver includes initial IAPP implementation that I'm going to use as the basis for full IAPP implementation.. MOHIT : The CVS version of hostapd with IAPP was not available when we started implementation of IAPP, therefore we used the version of hostapd (2002-10-12) available then to implement IAPP. 802.11f Draft 3.1 has been taken as a reference for the development of > IAPP stack. Please note that a newer version (draft 5) is already available. MOHIT: Could you tell me the link from where I can download it. I could not find this draft on the Internet. We have used a modular approach while implementing the same & done minimal changes to the main HostAP code. IAPP specific code has been mostly implemented as an additional library of functions to be linked to the main HostAP .We are facing some problems while implementing this related to retrieval of STA information from HostAP. By "HostAP", do you refer to hostapd daemon or the kernel driver? Could you please describe what are the main design differences between your and my implementation of IAPP? MOHIT: We are referring to the hostapd daemon. The main difference between our design & yours is that we have not really changed the hostapd the main implementation of hostapd except for adding a few function calls. The main IAPP is implemented as a separate lib of functions, which is linked with hostapd. Could you please suggest a way to retrieve the STA related information from hostAP with respect to the following > > 1.. Is the STA already associated with HostAP. If yes, what is its association Id ? > 2.. Where and how can I remove association of STA in the old AP after it moves to a New AP area. I'm assuming you are linking your code with hostapd. In this case you can use ap_get_sta() function to fetch station info based on MAC address and sta->aid contains the association id allocated for that station (or 0 if AID has not yet been allocated). See CVS version of hostap/hostapd/iapp.c and function iapp_process_add_notify() for an example how to remove association on the old AP. MOHIT: Regarding your suggestion to access STA related data I would further like to ask that when STA moves to the new AP area. AP2 (new AP) sends move notify message to AP1 (old AP) then the function which is handling received Notify messages on AP1 does not have access to valid hostapd structure pointer which is a parameter in ap_get_sta () function to retrieve STA info. Could you please suggest a way to get this value. Any help will be highly appreciated. Thanks & Regards Mohit Bajpai -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030122/d76acdff/attachment.htm From Carsten at Tschach.de Wed Jan 22 01:40:01 2003 From: Carsten at Tschach.de (Carsten Tschach) Date: Wed, 22 Jan 2003 07:40:01 +0100 Subject: Which card has a most power?! Message-ID: <20030122064001.GB14530205@Tschach.com> Hi, I do have a Compaq WL200 card and was told, that this has 100 mW of power. But I now have heard, that there are other cards with more power - so I would like to know which one has the most.... Bye, Carsten From Jacques.Caron at IPsector.com Wed Jan 22 09:19:22 2003 From: Jacques.Caron at IPsector.com (Jacques Caron) Date: Wed, 22 Jan 2003 15:19:22 +0100 Subject: Which card has a most power?! In-Reply-To: <20030122064001.GB14530205@Tschach.com> Message-ID: <5.2.0.9.0.20030122151206.01a3d988@pop.mail.yahoo.com> Hi, I would be surprised if that card had 100mW, most PRISM-based cards only have 30mW. The most powerful cards around have 200mW, and Senao has one of those (and it is PRISM-based). Note however that in Europe, ETSI regulations prohibit going over 100mW EIRP (to avoid interference to other users of the 2.4GHz band), which means about 30mW output power with a standard dipole antenna, and less as you increase your antenna gain. Cards sold for Europe usually have this limit set in the card, but if you import cards made for the US market (where the limits are much higher in general, and even higher for ptp links), it is your responsibility to make sure you don't go over the limit. There have been known cases in ETSI-regulated countries (e.g. France) of people being fined and asked to tear down their installation because they were going over the limits. Which all means that unless you are willing to take the risk (and potentially annoy all your neighbors), there is no reason to use a card with more than 30mW output power. Hope that helps, Jacques. At 07:40 22/01/2003, Carsten Tschach wrote: >Hi, > >I do have a Compaq WL200 card and was told, that this has 100 mW of power. >But I now have heard, that there are other cards with more power - so I >would like to know which one has the most.... > >Bye, Carsten >_______________________________________________ >HostAP mailing list >HostAP at shmoo.com >http://lists.shmoo.com/mailman/listinfo/hostap -- Jacques Caron, IP Sector Technologies Join the discussion on public WLAN open global roaming: http://lists.ipsector.com/listinfo/openroaming From lsantos at pd3.com.br Wed Jan 22 09:29:52 2003 From: lsantos at pd3.com.br (Leonardo Pereira Santos) Date: Wed, 22 Jan 2003 12:29:52 -0200 Subject: Which card has a most power?! In-Reply-To: <5.2.0.9.0.20030122151206.01a3d988@pop.mail.yahoo.com> References: <5.2.0.9.0.20030122151206.01a3d988@pop.mail.yahoo.com> Message-ID: <200301221229.53421.lsantos@pd3.com.br> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 You must remember one thing: if you have one card with great tx power, but are trying to communicate with cards that don't have as much tx power, you problably end up with you "powerful" card not being able to listen to the other card over great distances. If you want a card to make a AP that is able to connect over longer distances, it'll be better to have a good antenna than a lot of power, because antennas increase rx sensibility as well. If you want to make a point-to-point link, then go ahead and use both cards with lots of tx power, but in this case a good directional antenna will not only increase the distance you'll be able to cover but might even avoid intercerence with neighboors. On Wednesday 22 January 2003 12:19, Jacques Caron wrote: > Hi, > > I would be surprised if that card had 100mW, most PRISM-based cards only > have 30mW. The most powerful cards around have 200mW, and Senao has one of > those (and it is PRISM-based). > > Note however that in Europe, ETSI regulations prohibit going over 100mW > EIRP (to avoid interference to other users of the 2.4GHz band), which means > about 30mW output power with a standard dipole antenna, and less as you > increase your antenna gain. Cards sold for Europe usually have this limit > set in the card, but if you import cards made for the US market (where the > limits are much higher in general, and even higher for ptp links), it is > your responsibility to make sure you don't go over the limit. There have > been known cases in ETSI-regulated countries (e.g. France) of people being > fined and asked to tear down their installation because they were going > over the limits. Which all means that unless you are willing to take the > risk (and potentially annoy all your neighbors), there is no reason to use > a card with more than 30mW output power. > > Hope that helps, > > Jacques. > > At 07:40 22/01/2003, Carsten Tschach wrote: > >Hi, > > > >I do have a Compaq WL200 card and was told, that this has 100 mW of power. > >But I now have heard, that there are other cards with more power - so I > >would like to know which one has the most.... > > > >Bye, Carsten > >_______________________________________________ > >HostAP mailing list > >HostAP at shmoo.com > >http://lists.shmoo.com/mailman/listinfo/hostap > > -- Jacques Caron, IP Sector Technologies > Join the discussion on public WLAN open global roaming: > http://lists.ipsector.com/listinfo/openroaming > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap - -- This is Linux country. On a quiet night, you can hear Windows reboot. Leonardo Pereira Santos Engenheiro de Projetos PD3 Tecnologia av. Par? 330/202 (51) 3337 1237 Today Fortune tells us: "I'd love to go out with you, but I did my own thing and now I've got to undo it." -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.7 (GNU/Linux) iD8DBQE+Lqrg8uAMCTjfaxIRAkNgAJ48PC1JTi8oY/KAE0wIo0dAreHq8ACeOgW2 Errw4x84e7a6pL/Neg+d9TI= =XGLM -----END PGP SIGNATURE----- From gbritton at doomcom.org Wed Jan 22 10:21:45 2003 From: gbritton at doomcom.org (Gerald Britton) Date: Wed, 22 Jan 2003 10:21:45 -0500 Subject: Monitor mode with WE-16 In-Reply-To: <200301221403.38498.bercab@bulma.net>; from bercab@bulma.net on Wed, Jan 22, 2003 at 02:03:38PM +0100 References: <200301221403.38498.bercab@bulma.net> Message-ID: <20030122102145.A8625@nevermore.toe.doomcom.org> On Wed, Jan 22, 2003 at 02:03:38PM +0100, Bernardo Cabezas Serra wrote: > I have problems while switching to Monitor mode since I upgraded Wireless > Extensions API to version 16 (with 2.4.20 vanilla kernel). > > I'm using hostap from CVS, and my wireless card is a Conceptronic > CON11CC. Ah, and wireless-tools' version is 26pre7. > > When I try "iwconfig wlan0 mode monitor", I get the following: Is your essid set to "" (aka. any)? This is invalid in master mode and I've found the hardware gets very sick if you have it "" in any mode other than managed. If you do an "iwconfig wlan0 essid test" and then go into monitor mode it should work fine. I suggested a while ago that some sort of safety be put into the driver to prevent you doing that. -- Gerald From kenw at keeper.org.uk Wed Jan 22 10:25:23 2003 From: kenw at keeper.org.uk (Ken Wolstencroft) Date: Wed, 22 Jan 2003 15:25:23 +0000 Subject: WEP rekeying Message-ID: <3E2EB7E3.3090109@keeper.org.uk> Which version of hostap do I need to use WEP rekeying, and where can I find details on how to get it working. Thanks and all the best, Ken From Richard.Docksey at liffe.com Wed Jan 22 10:58:04 2003 From: Richard.Docksey at liffe.com (Richard Docksey) Date: Wed, 22 Jan 2003 15:58:04 -0000 Subject: updating firmware on zoomair 4100(05) Message-ID: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> Hi, I have zoomair 4100(05) pci/pc cards with firmware version 1.4.2 and I need to update the pci card for it to work in host ap mode. I've not been able to find a way to do it so far; the prism2 testsuite (available from intersil) claims to have a winupdate program - but, after installing it (twice!), the exe isn't available. I raised a tech support questions with zoom and got the following response -- Hello, There are no firmware updates available for the 4100 Zoomair Cards. Thank you RobertO Zoom Technical Support -- after searching this mail list archive I did find a hex txt file for flashing the card with but don't know of a tool I can use to do it. does anybody have any experience of carrying out this update with the zoomair cards? any advice or help would be very welcome. thanks in advance, Richard. ----------------------------------------------------------------------- The information contained in this e-mail is confidential and solely for the intended addressee(s). Unauthorised reproduction, disclosure, modification, and/or distribution of this email may be unlawful. If you have received this email in error, please notify the sender immediately and delete it from your system. The views expressed in this message do not necessarily reflect those of LIFFE Holdings Plc or any of its subsidiary companies. ----------------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030122/9215719e/attachment.htm From lsantos at pd3.com.br Wed Jan 22 11:07:32 2003 From: lsantos at pd3.com.br (Leonardo Pereira Santos) Date: Wed, 22 Jan 2003 14:07:32 -0200 Subject: updating firmware on zoomair 4100(05) In-Reply-To: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> Message-ID: <200301221407.32735.lsantos@pd3.com.br> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 You'll need a tweaked upgrader, as the upgrader is hard-coded to the firmware version. I managed to find the correct app, if you want it, I can send it to you, but if you go to google and do a search with "prism upgrade 1.4.9 utility", you'll find what you need. And you can get the upgrader from linksys, they just bundleded it in an installer. On Wednesday 22 January 2003 13:58, Richard Docksey wrote: > Hi, > > I have zoomair 4100(05) pci/pc cards with firmware version 1.4.2 and I need > to update the pci card for it to work in host ap mode. I've not been able > to find a way to do it so far; the prism2 testsuite (available from > intersil) claims to have a winupdate program - but, after installing it > (twice!), the exe isn't available. I raised a tech support questions with > zoom and got the following response - -- This is Linux country. On a quiet night, you can hear Windows reboot. Leonardo Pereira Santos Engenheiro de Projetos PD3 Tecnologia av. Par? 330/202 (51) 3337 1237 Today Fortune tells us: Reality is for people who lack imagination. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.7 (GNU/Linux) iD8DBQE+LsHE8uAMCTjfaxIRAhIEAKCF1+ngJiLGIdGw0Ouice61lqk5tACeNMWc hHh87nv/hkQyy9N+FJ6Jqno= =Adoc -----END PGP SIGNATURE----- From ian.cass at mblox.com Wed Jan 22 11:11:26 2003 From: ian.cass at mblox.com (Ian Cass) Date: Wed, 22 Jan 2003 16:11:26 -0000 Subject: updating firmware on zoomair 4100(05) References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> Message-ID: <03a501c2c230$ea783380$4c28a8c0@cblan.mblox.com> > does anybody have any experience of carrying out this update with the > zoomair cards? any advice or help would be very welcome. Compile and install the latest HostAP (CVS). Compile and the prism2_srec from the utils directory. Copy prism2_srec to /usr/local/sbin Get the latest firmware hex files http://lists.shmoo.com/pipermail/hostap/2002-November/000295.html For me, the relevant file was r1010506.hex. Copy it to /etc/pcmcia/firmware.hex Insert the following in your /etc/pcmcia/wireless.. # Set all desired settings via iwconfig # Do a hard reset to make sure the card is initialised properly log $IWPATH/iwpriv $DEVICE reset 1 # Upgrade the firmware first log /usr/local/sbin/prism2_srec -r $DEVICE /etc/pcmcia/firmware.hex sleep 2 Works a treat for me. -- Ian Cass From cornelisse at bonbon.net Wed Jan 22 11:23:00 2003 From: cornelisse at bonbon.net (Cornelisse) Date: Wed, 22 Jan 2003 17:23:00 +0100 Subject: radius & traffic shaping Message-ID: <001401c2c232$8babca40$6912210a@vthoog.nl> Hello all, I've been playing around with a soekris net4521 for a couple of weeks now. So far I've tried Leaf Wisp-dist and the pebble release, but these two releases lack a couple of features I need. What I'm trying to build is a authentication gateway (802.1x) in combination with a Radius server (eg freeradius). So if a customer wants access on an AP a login box appears (windows), after that the gateway checks with the radius server what type of subscription the customer has (1024/512 or 512/64) and then sets the appropriate rate for that session (traffic shaping) I know that with tc and ipchains this can already be achieved, but I want to integrate this into a radius server. I this even possible ? Regards, Cornelisse -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030122/4384c352/attachment.htm From koma at lysator.liu.se Wed Jan 22 11:31:13 2003 From: koma at lysator.liu.se (Magnus Ekhall) Date: Wed, 22 Jan 2003 17:31:13 +0100 Subject: WEP rekeying References: <3E2EB7E3.3090109@keeper.org.uk> Message-ID: <000d01c2c233$adc8bdd0$451d72d5@salamander> I'm trying to change the WEP keys as well. I'm trying to use local->priv->ops->set_key(), but it does not work. Maybe it has to be changed in the prism chipset as well? Thanks, Magnus ----- Original Message ----- From: "Ken Wolstencroft" To: Sent: Wednesday, January 22, 2003 4:25 PM Subject: WEP rekeying > Which version of hostap do I need to use WEP rekeying, and where can I > find details on how to get it working. > > Thanks and all the best, > Ken > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From dsen at homemail.com Wed Jan 22 11:36:35 2003 From: dsen at homemail.com (D. Sen) Date: Wed, 22 Jan 2003 11:36:35 -0500 Subject: Linksys WMP11 problems (disconnects continuously) Message-ID: <3E2EC893.8050104@homemail.com> Bought a friend of mine a WMP11 PCI card. She is experiencing constant disconnects (even from a few feet away). I understand that a firmware/driver update might solve the problem? But the firmware update on linksys's ftp site doesnt work....(comes back saying "card not inserted"). Any ideas? DS From cijoml at volny.cz Wed Jan 22 12:08:01 2003 From: cijoml at volny.cz (Michal Semler) Date: Wed, 22 Jan 2003 18:08:01 +0100 Subject: write firmware into nonvolatile memory Message-ID: Hello, I have ZCOMAX XI-626 Prism2 based wifi card. It has fw from factory wlan0: NIC: id=0x8013 v1.0.0 wlan0: PRI: id=0x15 v1.0.7 wlan0: STA: id=0x1f v1.3.6 After upgrade to volatile memory I have this fw... wlan0: NIC: id=0x8013 v1.0.0 wlan0: PRI: id=0x15 v1.0.7 wlan0: STA: id=0x1f v1.5.6 Is there any possibility to write fw into non-volatile memory? Is there newer version of PRI and NIC fw? Where? Thanks Michal From hick at gink.org Wed Jan 22 11:36:37 2003 From: hick at gink.org (gARetH baBB) Date: Wed, 22 Jan 2003 16:36:37 +0000 (GMT) Subject: updating firmware on zoomair 4100(05) In-Reply-To: <03a501c2c230$ea783380$4c28a8c0@cblan.mblox.com> References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> <03a501c2c230$ea783380$4c28a8c0@cblan.mblox.com> Message-ID: On Wed, 22 Jan 2003, Ian Cass wrote: > For me, the relevant file was r1010506.hex. Copy it to > /etc/pcmcia/firmware.hex I haven't found a 1.5.6 for 8003, has anyone ? From ian.cass at mblox.com Wed Jan 22 12:27:33 2003 From: ian.cass at mblox.com (Ian Cass) Date: Wed, 22 Jan 2003 17:27:33 -0000 Subject: updating firmware on zoomair 4100(05) References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> <03a501c2c230$ea783380$4c28a8c0@cblan.mblox.com> Message-ID: <041501c2c23b$8cbe83b0$4c28a8c0@cblan.mblox.com> > I haven't found a 1.5.6 for 8003, has anyone ? bomberman:~/hostap/utils/srec_temp.uWgtFv# /usr/local/bin/prism2_srec -r wlan0 r1010506.hex srec summary for r1010506.hex Included file name: r1010506.hex Component: 0x001f 1.5.6 (station firmware) Supported platforms: 0x8002 1.0.0, 0x8002 1.0.1, 0x8003 1.0.0, 0x8003 1.0.1, 0x8004 1.0.0 0x8008 1.0.0 -- Ian Cass From panxer at b-online.gr Wed Jan 22 12:41:42 2003 From: panxer at b-online.gr (panxer) Date: Wed, 22 Jan 2003 19:41:42 +0200 Subject: dwl-520+ and linux (works!) Message-ID: So the + generation of dlink's products seems to finally work..? I think i might have made dwl-520+ work under linux.. Even though dlink announced in last december that they would publish drivers (be it binary) for their TI - 22mbps products, such thing never happened.. Some point in November as i recall someone published a binary driver on behalf of a eusso company for a pPCI devicethat has the same chipset as the + generation by dlink (and others), the acx100 by Texas Instruments.. gryppas:~# cat /proc/pci Bus 0, device 12, function 0: Network controller: PCI device 104c:8400 (Texas Instruments) (rev 0). IRQ 11. Master Capable. Latency=32. I/O at 0xdc00 [0xdc1f]. Non-prefetchable 32 bit memory at 0xdc010000 [0xdc010fff]. Non-prefetchable 32 bit memory at 0xdc000000 [0xdc00ffff]. The driver was setup for a mandrake kernel, a 2.4.18-16mdk, see what the module has to say; gryppas:/home/panxer# modinfo acx100_pci.o filename: acx100_pci.o description: "TI ACX100 WLAN 22Mbps driver" author: "Lancelot Wang" license: "GPL" Mentioning GPL makes me happy, maybe they finnaly _do_ plan to release the source code. I noticed in the lwan mailing list that none had succeded in loading the driver and making it work. I tried to load the driver in my 2.4.18 kernel yet it came oute with 5 or 6 unresolved symbols, plus it was annoying telling me that it was made for a -16mdk.. Using hexeditor i erased every mention of -16mdk, and it stopped picking me about it. After contacting some guys from lkml i realised that CONFIG_DEBUG_KERNEL should be on, so i recompiled the kernel with this oprion and the driver came out with only one unresolved symbol which as i suspected it had to do with the fact that the built tree of the kernel was in /usr/src/whatever instead of the /usr/src/linux..(?) So I mv /usr/src/whatever /usr/src/linux, placed the driver in /lib/modules and echo "alias wlan0 acx100_pci" >> /etc/modules.conf After a reboot the driver loaded properly! gryppas:~# lsmod |grep a Module Size Used by Tainted: GF acx100_pci 181952 0 I got the wlan0 up and gryppas:~# ifconfig wlan0 wlan0 Link encap:Ethernet HWaddr 00:08:05:BF:66:50 inet addr:192.168.0.3 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::208:5ff:febf:6650/10 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:4 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:0 (0.0 b) TX bytes:288 (288.0 b) Interrupt:11 iwconfig speaks: gryppas:~# iwconfig wlan0 wlan0 v0.9.0.3a 11b+ ESSID:"Wesley2" Nickname:"" Mode:Ad-Hoc Channel:1 Cell: 59:51:5C:E6:3F:09 Wesley2 is the default essid as seen in the driver... gryppas:/etc/wireless/acx100# iwconfig wlan0 essid panxer gryppas:/etc/wireless/acx100# iwconfig wlan0 wlan0 v0.9.0.3a 11b+ ESSID:"panxer" Nickname:"" Mode:Ad-Hoc Channel:1 Cell: 66:EE:E3:59:80:B6 nice! gryppas:~# uname -a &&lsmod Linux gryppas 2.4.18 #5 ??? ??? 20 18:51:53 EET 2003 i686 unknown Module Size Used by Tainted: GF acx100_pci 181952 1 Lets see if its functional gryppas:~# ping 192.168.0.3 PING 192.168.0.3 (192.168.0.3): 56 data bytes 64 bytes from 192.168.0.3: icmp_seq=0 ttl=255 time=0.3 ms 64 bytes from 192.168.0.3: icmp_seq=19 ttl=255 time=0.2 ms --- 192.168.0.3 ping statistics --- 20 packets transmitted, 20 packets received, 0% packet loss round-trip min/avg/max = 0.1/0.1/0.3 ms I borrowed a dw-900+ AP from a friend partner in the AWMN project ( www.athenswireless.net ) to do some testing.. It occured that the 900+ some how gets crazy when ping it hard, but it might be the driver or the 520+ --- 192.168.0.49 ping statistics --- 52 packets transmitted, 38 packets received, 26% packet loss round-trip min/avg/max = 1.9/8374.0/32007.4 ms The strange thing is that; gryppas:/etc/wireless/acx100# ping 192.168.0.49 PING 192.168.0.49 (192.168.0.49): 56 data bytes 64 bytes from 192.168.0.49: icmp_seq=13 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=14 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=15 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=16 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=17 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=18 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=19 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=20 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=21 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=22 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=23 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=24 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=25 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=26 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=27 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=28 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=29 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=30 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=31 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=32 ttl=127 time=4.3 ms 64 bytes from 192.168.0.49: icmp_seq=0 ttl=127 time=32007.4 ms 64 bytes from 192.168.0.49: icmp_seq=1 ttl=127 time=31008.6 ms 64 bytes from 192.168.0.49: icmp_seq=2 ttl=127 time=30010.6 ms 64 bytes from 192.168.0.49: icmp_seq=3 ttl=127 time=29011.3 ms 64 bytes from 192.168.0.49: icmp_seq=4 ttl=127 time=28012.2 ms 64 bytes from 192.168.0.49: icmp_seq=5 ttl=127 time=27013.3 ms 64 bytes from 192.168.0.49: icmp_seq=6 ttl=127 time=26014.0 ms 64 bytes from 192.168.0.49: icmp_seq=7 ttl=127 time=25015.1 ms 64 bytes from 192.168.0.49: icmp_seq=8 ttl=127 time=24016.4 ms 64 bytes from 192.168.0.49: icmp_seq=9 ttl=127 time=23017.2 ms 64 bytes from 192.168.0.49: icmp_seq=10 ttl=127 time=22018.0 ms 64 bytes from 192.168.0.49: icmp_seq=11 ttl=127 time=21018.6 ms 64 bytes from 192.168.0.49: icmp_seq=46 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=47 ttl=127 time=2.6 ms 64 bytes from 192.168.0.49: icmp_seq=48 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=49 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=50 ttl=127 time=1.9 ms 64 bytes from 192.168.0.49: icmp_seq=51 ttl=127 time=1.9 ms Some icmp packets are delayed somehow (low basic Tx rate maybe?).. I did some tweaks on the 900+ and it seems to be handling dwl-520+ better, I still get some packetloss (few) and plenty of DUPs. I have to add that i can switch channel on the 520+ only after switching the AP to the desired channel and setting the same essid.. gryppas:/etc# iwconfig wlan0 mode managed channel 6 essid awmn gryppas:/etc# iwconfig lo no wireless extensions. wlan0 v0.9.0.3a 11b+ ESSID:"awmn" Nickname:"" Mode:Managed Channel:6 Access Point: 00:40:05:CA:80:1E Encryption key:off eth0 no wireless extensions. So, let us see what happens.. gryppas:~# ping -s 1024 192.168.0.202 PING 192.168.0.202 (192.168.0.202): 1024 data bytes 1032 bytes from 192.168.0.202: icmp_seq=0 ttl=128 time=4.8 ms 1032 bytes from 192.168.0.202: icmp_seq=1 ttl=128 time=5.9 ms --- 192.168.0.202 ping statistics --- 76 packets transmitted, 73 packets received, +9 duplicates, 3% packet loss round-trip min/avg/max = 4.6/5.4/10.2 ms kinda nice i must say.. My logs are full of debugging stuff so maybe that suggests that the plan to release the driver someday.. Havent been able to get in contact with the guy who developed the driver.. In lwan-ng they imply that he used their code (which is under GPL) so we vould force him to publish the source.. Or else try to reverse eng. it. :P After playing with the 900+ again i get better responces but lots of DUPs... The fact is that dwl-520+ works under linux (at least in some point), i thing that all wifi NICs using acx100 might work also, check it out.. Regards, Panagiotis Moustafellos (aka panXer) www.athenswireless.net From jt at bougret.hpl.hp.com Wed Jan 22 12:49:48 2003 From: jt at bougret.hpl.hp.com (Jean Tourrilhes) Date: Wed, 22 Jan 2003 09:49:48 -0800 Subject: Monitor mode with WE-16 Message-ID: <20030122174948.GA22632@bougret.hpl.hp.com> Bernardo Cabezas Serra wrote : > > I have problems while switching to Monitor mode since I upgraded Wireless > Extensions API to version 16 (with 2.4.20 vanilla kernel). I have trouble seeing what could be the relation between WE16 and monitor mode. On the other hand, that might be related to the switch WE14 => WE15 (which change the way monitor mode is set). > Another related issue is that almost all aplications have some trouble with > the new WE API and hostap. For example, "wavemon" submits ioctls with > ".u.data.length" field of struct iwreq to zero, and seems hostap don't like > it. I've got it runnig by setting length field to the "sizeof" of the data > pointed to by "iwreq.u.data.poiter" (and the same process for every ioctl in > wavemon). > Can be considered a bug in wavemon the fact of setting "length" fieds to 0 ? That something that I introduced in WE13/WirelessTools23 and that is enforced strictly on WE15. It has nothing to do with hostap and is enforced in the kernel. At the time I released WE13 (over a year ago), I notified various authors of tools, including the author of wavemon. I sent him a friendly reminder the 4 Sep 2002. What more can I do ? Your fix is mostly correct, but lenght is (unfortunately) not always the sizeof(data). Please check in the Wireless Tools for each case. Good luck... Jean From panxer at b-online.gr Wed Jan 22 13:29:40 2003 From: panxer at b-online.gr (panxer) Date: Wed, 22 Jan 2003 20:29:40 +0200 Subject: [lwlan-devel] dwl-520+ and linux (works!) In-Reply-To: <20030122101112.B374@home.tig-grr.com> References: <20030122101112.B374@home.tig-grr.com> Message-ID: On Wednesday 22 January 2003 20:11, Tom Marshall wrote: > I don't see the driver listed on the download page. Did they remove it? > Does anyone have a copy they can post? > > http://www.eusso.com/Model2/GL2422-VP/GL2422-VP.htm > http://www.eusso.com/Model2/GL2422/GL2422_utility.htm > > On Wed, Jan 22, 2003 at 07:41:42PM +0200, panxer wrote: > > So the + generation of dlink's products seems to finally work..? > > I think i might have made dwl-520+ work under linux.. > > Even though dlink announced in last december that they would publish > > drivers (be it binary) for their TI - 22mbps products, such thing never > > happened.. Some point in November as i recall someone published a binary > > driver on behalf of a eusso company for a pPCI devicethat has the same > > chipset as the + generation by dlink (and others), the acx100 by Texas > > Instruments.. check out http://radio.isc.tuc.gr/vstoidis/forum/download.php?id=406 -- Panagiotis Moustafellos (aka panXer) From jasonb at edseek.com Wed Jan 22 13:51:03 2003 From: jasonb at edseek.com (Jason Boxman) Date: Wed, 22 Jan 2003 13:51:03 -0500 (EST) Subject: updating firmware on zoomair 4100(05) In-Reply-To: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.co m> References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.co m> Message-ID: <1786.209.208.68.56.1043261463.squirrel@rebecca.edseek.com> Richard Docksey said: > > Hi, > > after searching this mail list archive I did find a hex txt file for > flashing the card with but don't know of a tool I can use to do it. Yes, you can use HostAP itself to 'soft' update your card. http://trekweb.com/~jasonb/articles/hostap_20021012.shtml I have written a section on it in my HostAP article linked above. > does anybody have any experience of carrying out this update with the > zoomair cards? any advice or help would be very welcome. > > thanks in advance, > Richard. > > > From bercab at bulma.net Wed Jan 22 14:54:27 2003 From: bercab at bulma.net (Bernardo Cabezas Serra) Date: Wed, 22 Jan 2003 20:54:27 +0100 Subject: Monitor mode with WE-16 In-Reply-To: <20030122174948.GA22632@bougret.hpl.hp.com> References: <20030122174948.GA22632@bougret.hpl.hp.com> Message-ID: <200301222054.27876.bercab@bulma.net> El Mi?rcoles, 22 de Enero de 2003 18:49, Jean Tourrilhes writed: > At the time I released WE13 (over a year ago), I notified > various authors of tools, including the author of wavemon. I sent him > a friendly reminder the 4 Sep 2002. What more can I do ? > Your fix is mostly correct, but lenght is (unfortunately) not > always the sizeof(data). Please check in the Wireless Tools for each > case. Ok, it's clear. I will enjoy learning WE programming :) Thank you for your response (and for your grear work) Regards, Bernardo. From kenw at keeper.org.uk Wed Jan 22 15:46:51 2003 From: kenw at keeper.org.uk (Ken Wolstencroft) Date: Wed, 22 Jan 2003 20:46:51 -0000 Subject: Still having problems with WEP rekeying Message-ID: <008101c2c257$64323330$9864a8c0@ken> I still can't get WEP rekeying to work, EAP-TLS and EAP-TTLS work fine with static WEP keys, but when I switch on WEP rekeying my station can not authenticate. I am using a Dlink DWL-520+ in my Windows XP station, and a Prism 2 card in my Linux server running hostapd. Can anyone help, could it be firmware issues on one of the cards etc? Thanks, Ken Here is an extract from my hostapd.conf: ieee8021x=1 #minimal_eap=0 eap_message=hello wep_key_len_broadcast=13 #wep_key_len_unicast=13 wep_rekey_period=300 eapol_key_index_workaround=1 Here is an extract from my hostap log: [root at wnap-ram01 hostapd]# ./hostapd -d ./hostapd.conf Opening raw packet socket for ifindex 5 Using interface wlan0ap with hwaddr 00:80:c6:e3:5c:94 and ssid 'KEBAB' Default WEP key - hexdump(len=5): 73 1f 67 3a 82 Flushing old station entries Deauthenticate all stations mgmt::auth authentication: STA=00:40:05:c7:73:06 auth_alg=0 auth_transaction=1 status_code=0 New STA Station 00:40:05:c7:73:06 authentication OK (open system) MGMT (TX callback) ACK mgmt::auth cb Station 00:40:05:c7:73:06 authenticated mgmt::assoc_req association request: STA=00:40:05:c7:73:06 capab_info=0x01 listen_interval=12 new AID 1 Station 00:40:05:c7:73:06 association OK (aid 1) MGMT (TX callback) ACK mgmt::assoc_resp cb Station 00:40:05:c7:73:06 associated (aid 1) IEEE 802.1X: Start authentication for new station 00:40:05:c7:73:06 IEEE 802.1X: 00:40:05:c7:73:06 AUTH_PAE entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 BE_AUTH entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 AUTH_KEY_TX entering state NO_KEY_TRANSMIT IEEE 802.1X: 00:40:05:c7:73:06 AUTH_PAE entering state DISCONNECTED IEEE 802.1X: Unauthorizing station 00:40:05:c7:73:06 IEEE 802.1X: Sending canned EAP packet FAILURE to 00:40:05:c7:73:06 (identifier 0) IEEE 802.1X: 00:40:05:c7:73:06 BE_AUTH entering state IDLE IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 AUTH_PAE entering state CONNECTING IEEE 802.1X: Sending EAP Request-Identity to 00:40:05:c7:73:06 (identifier 1) IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE DATA (TX callback) ACK DATA (TX callback) ACK IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE From kenw at keeper.org.uk Wed Jan 22 16:08:54 2003 From: kenw at keeper.org.uk (Ken Wolstencroft) Date: Wed, 22 Jan 2003 21:08:54 -0000 Subject: Still having problems with WEP rekeying References: <008101c2c257$64323330$9864a8c0@ken> Message-ID: <001001c2c25a$78a01f00$9864a8c0@ken> It works.... I had to upgrade the DWL520+ driver to 2.3e.... Thanks, Ken ----- Original Message ----- From: "Ken Wolstencroft" To: "hostAP" Sent: Wednesday, January 22, 2003 8:46 PM Subject: Still having problems with WEP rekeying > I still can't get WEP rekeying to work, EAP-TLS and EAP-TTLS work fine with > static WEP keys, but when I switch on WEP rekeying my station can not > authenticate. I am using a Dlink DWL-520+ in my Windows XP station, and a > Prism 2 card in my Linux server running hostapd. > > Can anyone help, could it be firmware issues on one of the cards etc? > > Thanks, > Ken > > Here is an extract from my hostapd.conf: > > ieee8021x=1 > #minimal_eap=0 > eap_message=hello > wep_key_len_broadcast=13 > #wep_key_len_unicast=13 > wep_rekey_period=300 > eapol_key_index_workaround=1 > > Here is an extract from my hostap log: > > [root at wnap-ram01 hostapd]# ./hostapd -d ./hostapd.conf > Opening raw packet socket for ifindex 5 > Using interface wlan0ap with hwaddr 00:80:c6:e3:5c:94 and ssid 'KEBAB' > Default WEP key - hexdump(len=5): 73 1f 67 3a 82 > Flushing old station entries > Deauthenticate all stations > mgmt::auth > authentication: STA=00:40:05:c7:73:06 auth_alg=0 auth_transaction=1 > status_code=0 > New STA > Station 00:40:05:c7:73:06 authentication OK (open system) > MGMT (TX callback) ACK > mgmt::auth cb > Station 00:40:05:c7:73:06 authenticated > mgmt::assoc_req > association request: STA=00:40:05:c7:73:06 capab_info=0x01 > listen_interval=12 > new AID 1 > Station 00:40:05:c7:73:06 association OK (aid 1) > MGMT (TX callback) ACK > mgmt::assoc_resp cb > Station 00:40:05:c7:73:06 associated (aid 1) > IEEE 802.1X: Start authentication for new station 00:40:05:c7:73:06 > IEEE 802.1X: 00:40:05:c7:73:06 AUTH_PAE entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 BE_AUTH entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 AUTH_KEY_TX entering state NO_KEY_TRANSMIT > IEEE 802.1X: 00:40:05:c7:73:06 AUTH_PAE entering state DISCONNECTED > IEEE 802.1X: Unauthorizing station 00:40:05:c7:73:06 > IEEE 802.1X: Sending canned EAP packet FAILURE to 00:40:05:c7:73:06 > (identifier 0) > IEEE 802.1X: 00:40:05:c7:73:06 BE_AUTH entering state IDLE > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 AUTH_PAE entering state CONNECTING > IEEE 802.1X: Sending EAP Request-Identity to 00:40:05:c7:73:06 (identifier > 1) > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > DATA (TX callback) ACK > DATA (TX callback) ACK > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > IEEE 802.1X: 00:40:05:c7:73:06 REAUTH_TIMER entering state INITIALIZE > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap From jkmaline at cc.hut.fi Wed Jan 22 22:40:45 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 22 Jan 2003 19:40:45 -0800 Subject: IAPP using HostAP-2002-10-12 In-Reply-To: <1043240823.13188.30.camel@localhost.localdomain> References: <002901c2c13f$56fb37c0$f009030a@mahindrabt.com> <20030122024722.GA1488@jm.kir.nu> <1043240823.13188.30.camel@localhost.localdomain> Message-ID: <20030123034045.GC1177@jm.kir.nu> On Wed, Jan 22, 2003 at 01:07:02PM +0000, wlan wrote: > Is there a external command to get stations associated with an hostap > AP? You can either look at /proc/net/hostap/wlan0/ files or signal hostapd to dump its station data to a text file with USR1 signal. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Wed Jan 22 22:49:36 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 22 Jan 2003 19:49:36 -0800 Subject: IAPP using HostAP-2002-10-12 In-Reply-To: <009a01c2c21b$8f364800$f009030a@mahindrabt.com> References: <009a01c2c21b$8f364800$f009030a@mahindrabt.com> Message-ID: <20030123034936.GD1177@jm.kir.nu> On Wed, Jan 22, 2003 at 07:08:33PM +0530, Mohit Bajpai wrote: > MOHIT: Could you tell me the link from where I can download it. I could not find this draft on the Internet. Unfortunately, IEEE 802.11 drafts are not publicly available. > MOHIT: Regarding your suggestion to access STA related data I would further like to ask that when STA moves to the new AP area. AP2 (new AP) sends move notify message to AP1 (old AP) then the function which is handling received Notify messages on AP1 does not have access to valid hostapd structure pointer which is a parameter in ap_get_sta () function to retrieve STA info. Could you please suggest a way to get this value. If you are linking your code to the same binary, easiest way to get that pointer would be to use the same event loop implementation as the rest of hostapd is using. You can register a handler for a "read socket" with eloop_register_read_sock() function. Your handler will then be called with pointer to the main hostapd data as eloop_ctx (second argument) when new data packet (in case of UDP) or new incoming connection (TCP) is available. -- Jouni Malinen PGP id EFC895FA From vda at port.imtp.ilyichevsk.odessa.ua Thu Jan 23 01:50:33 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Thu, 23 Jan 2003 08:50:33 +0200 Subject: one way bad: dlink dwl-520, netgear me102 In-Reply-To: <000901c2c191$726add60$0200a8c0@zoidberg> References: <000901c2c191$726add60$0200a8c0@zoidberg> Message-ID: <200301230651.h0N6pRs21439@Port.imtp.ilyichevsk.odessa.ua> On 21 January 2003 23:09, Magnus wrote: > >> We use a combination of netgear me102 accesspoint and dlink > >> dwl-520 (11 > >> Mbit, using hostap_pci-driver) in the router to connect a lan to a > >> dsl-router (suse-linux 8.2, kernelversion 2.4). > > > > Which primary and station firmware versions are you using in D-Link > > DWL-520? If PRI < 1.0.7 or STA < 1.3.5, you should upgrade them. > > Older versions are known to corrupt packet data when it is > > transfered over PCI bus.. > > Since when use pci-networkadaptercards a firmware? On D-Link-Homepage > no firmware for dwl-520 exists. It exists. Install of flash utility for Windows: ftp://ftp.dlink.com/Wireless/DWL650/Firmware/dwl650_firmware_8c2.exe Firmwares: daemon wget -c http://www.netgate.com/support/prism_firmware/1.4.9.zip daemon wget -c http://www.netgate.com/support/prism_firmware/1.3.4.zip daemon wget -c http://www.netgate.com/support/prism_firmware/1.1.0.zip I used these and it helped a lot. BTW, I once wrote: >I seem to be unable to set MAC address. >It seems to work and I can see my frames with right >MAC address with tcpdump. The problem is, other hosts >do not like me anymore :( I see zero incoming packets. >This happens even if I set MAC addr before bringing >iface up. I.e. no host can possibly cache stale ARP >data... > >I'd be grateful if anyone will do a quick check on their >hardware/firmware/driver combo and tell me the results. It's a matter of simple # ifconfig ethN hw ether X:X:X:X:X:X or # ip link set dev ethN address X:X:X:X:X:X , upping iface and pinging Anybody? ;) -- vda From vda at port.imtp.ilyichevsk.odessa.ua Thu Jan 23 01:58:32 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Thu, 23 Jan 2003 08:58:32 +0200 Subject: Somebody have the programmers guide of intersil chipset? In-Reply-To: <3E2CFEFA.2020001@abicom-international.com> References: <3E2CFEFA.2020001@abicom-international.com> Message-ID: <200301230659.h0N6xQs21493@Port.imtp.ilyichevsk.odessa.ua> On 21 January 2003 10:04, Dave Edwards wrote: > Contact Intersil, > > They will send you the latest manual once you sign an NDA. Do you mean NIC's CPU (or whatever) docs? I wondered why we rely on someone else's code in the firware (and may be caught by fw bugs for years) instead of writing our own... Anyway, even 1.4.9 firmware files aren't that big (~64k). Do anybody know what CPU is used in Prism cards? Is it legal to try and disassemble firmware? -- vda From brianc at palaver.net Thu Jan 23 02:13:41 2003 From: brianc at palaver.net (Brian Capouch) Date: Thu, 23 Jan 2003 02:13:41 -0500 Subject: Somebody have the programmers guide of intersil chipset? References: <3E2CFEFA.2020001@abicom-international.com> <200301230659.h0N6xQs21493@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <3E2F9625.4010708@palaver.net> Denis Vlasenko wrote: > > I wondered why we rely on someone else's code in the firware > (and may be caught by fw bugs for years) instead of writing > our own... > > Anyway, even 1.4.9 firmware files aren't that big (~64k). > Do anybody know what CPU is used in Prism cards? > Is it legal to try and disassemble firmware? > -- Boy, those words sure got my excitement level raised!! Maybe if we could write our own firmware, we could adapt the technique regarding the "distributed" versus "short" interframe spaces put forth in the AT&T paper that would allow for Infrastructure mode connections to be solidly done from >~6000' from the access point. . . I'm all for it; hope it might be possible. B. From vda at port.imtp.ilyichevsk.odessa.ua Thu Jan 23 02:15:35 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Thu, 23 Jan 2003 09:15:35 +0200 Subject: DLink DWL-520 + hostap in STA mode In-Reply-To: References: <200301170808.h0H88gs20786@Port.imtp.ilyichevsk.odessa.ua> <200301200746.h0K7k0s05270@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <200301230716.h0N7GUs21596@Port.imtp.ilyichevsk.odessa.ua> On 20 January 2003 12:28, Michal Semler wrote: > Dne po 20. leden 2003 08:45 jste napsal(a): > > On 17 January 2003 10:08, Denis Vlasenko wrote: > > > DLink DWL-520 > > > 2.4.20-pre11, wireless extensions 14, hostap-2002-10-12 > > > > > > We (my father and me) made a fairly long helical aerial. > > > We are trying to communicate over ~15 km with a small wireless > > > cell. (~10 hosts, one AP). > > > > > > Mainline kernel's orinoco_pci module works but seems to lose > > > an association within ~30 secs and is not able to restore it > > > without ifconfig down/up cycle. Win98 driver fares better but > > > dies too after ~3 mins. > > > > > > I am trying hostap now. It loads without problems but does not > > > work, I am unable to switch the node into STA mode. > > > Looks like card's firmware is old. Where can I find newer one? > > > (I'll go googling to try and find it myself...) > > > > Found 1.4.9 firmware. Flashed it in. Makes lots of difference. > > Basically everything works with mainline kernel's orinoco_pci > > (stopped using hostap since I do not need AP now). > > > > Thanks guys for all your numerous responses ;) > > If u want firmware 1.5.6 - here it is... :) Wow! Where's my wget? > http://aldebaran.feld.cvut.cz/~xmyslik/wifi Following the link... -- vda From vda at port.imtp.ilyichevsk.odessa.ua Thu Jan 23 02:08:44 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Thu, 23 Jan 2003 09:08:44 +0200 Subject: Setting MAC address In-Reply-To: <20030121040954.GA1743@jm.kir.nu> References: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> <20030121040954.GA1743@jm.kir.nu> Message-ID: <200301230709.h0N79cs21553@Port.imtp.ilyichevsk.odessa.ua> On 21 January 2003 06:09, Jouni Malinen wrote: Ahh, I didn't see your post for a while 'coz you removed me from direct recipients... my mail sorting setup selects mails with me in To or CC and shows them first... > On Mon, Jan 20, 2003 at 09:58:56AM +0200, Denis Vlasenko wrote: > > I seem to be unable to set MAC address. > > It seems to work and I can see my frames with right > > MAC address with tcpdump. The problem is, other hosts > > do not like me anymore :( I see zero incoming packets. > > This works in my tests, both as an AP and as a station. When running > hostapd, one must restart hostapd after this change. Other than that, > there should not be other limitations on this. Anyway, I would > recommend changing the MAC address when the device is down. > > Please give some details about the system and configuration that > behaves incorrectly. Are you changing the MAC address in an AP or a > station? Which firmware versions are you using? I'm mostly interested > in which command you run and in which order, what ifconfig shows as > the MAC address of wlan0 device after the changed, do you see errors > in 'dmesg' output, and what 'hostap_diag wlan0 -r | grep FC01' shows > as the own address. DLink DWL-520, firmware 1.4.9, used in STA mode. kernel 2.4.20-pre11, wireless extensions 14. hostap-2002-10-12 is installed, but I use mainline kernel's orinoco_pci. Will try with hostap (and kernel 2.4.20) instead. Just curious: what "kernel: eth0: Station identity 001f:0006:0001:0003" means? It prints this at modprobe time. Gory details just in case you are interested: lspci -vvvxxx: ============== 00:0f.0 Network controller: Harris Semiconductor: Unknown device 3873 (rev 01) Subsystem: D-Link System Inc: Unknown device 3501 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- SERR- kernel: orinoco.c 0.11b (David Gibson and others) kernel: PCI: Found IRQ 11 for device 00:0f.0 kernel: PCI: Sharing IRQ 11 with 00:07.5 kernel: Detected Orinoco/Prism2 PCI device at 00:0f.0, mem:0xD8000000 to 0xD8000FFF -> 0xd0c66000, irq:11 kernel: Reset done.................................... .................................................. .................................................. .................................................. .................................................. ...........; kernel: Clear Reset................................... .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................................................. .................; kernel: pci_cor : reg = 0x0 - DD5 - DA3 kernel: eth0: Station identity 001f:0006:0001:0003 kernel: eth0: Looks like an Intersil firmware version 1.04 kernel: eth0: Ad-hoc demo mode supported kernel: eth0: IEEE standard IBSS ad-hoc mode supported kernel: eth0: WEP supported, 104-bit key kernel: eth0: MAC address 00:05:5D:FA:1A:DA kernel: eth0: Station name "Prism I" kernel: eth0: ready -- vda From jkmaline at cc.hut.fi Thu Jan 23 02:24:47 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Wed, 22 Jan 2003 23:24:47 -0800 Subject: Somebody have the programmers guide of intersil chipset? In-Reply-To: <200301230659.h0N6xQs21493@Port.imtp.ilyichevsk.odessa.ua> References: <3E2CFEFA.2020001@abicom-international.com> <200301230659.h0N6xQs21493@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <20030123072447.GA21122@jm.kir.nu> On Thu, Jan 23, 2003 at 08:58:32AM +0200, Denis Vlasenko wrote: > Do you mean NIC's CPU (or whatever) docs? I would assume that that is not available for anyone outside Intersil. One might be able to get driver programmer's documentation after signing NDA, but it does not cover MAC processor specific stuff. > I wondered why we rely on someone else's code in the firware > (and may be caught by fw bugs for years) instead of writing > our own... Because documentation or development environment for Intersil MAC is not available.. > Anyway, even 1.4.9 firmware files aren't that big (~64k). > Do anybody know what CPU is used in Prism cards? Intersil ;-). They don't even want to tell you what programming language was used for firmware coding (see Intersil knowledge base). > Is it legal to try and disassemble firmware? That probably depends on where you happen to live and local legislation, etc. Anyway, good luck trying to disassemble the code without even knowing the instruction codes or word length, etc. used in the MAC processor ;-). -- Jouni Malinen PGP id EFC895FA From vda at port.imtp.ilyichevsk.odessa.ua Thu Jan 23 04:02:21 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Thu, 23 Jan 2003 11:02:21 +0200 Subject: DLink DWL-520 + hostap in STA mode In-Reply-To: <200301230716.h0N7GUs21596@Port.imtp.ilyichevsk.odessa.ua> References: <200301170808.h0H88gs20786@Port.imtp.ilyichevsk.odessa.ua> <200301230716.h0N7GUs21596@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <200301230903.h0N93Fs22111@Port.imtp.ilyichevsk.odessa.ua> > > > > Looks like card's firmware is old. Where can I find newer one? > > > > (I'll go googling to try and find it myself...) > > > > > > Found 1.4.9 firmware. Flashed it in. Makes lots of difference. > > > Basically everything works with mainline kernel's orinoco_pci > > > (stopped using hostap since I do not need AP now). > > > > > > Thanks guys for all your numerous responses ;) > > > > If u want firmware 1.5.6 - here it is... :) > > Wow! Where's my wget? > > > http://aldebaran.feld.cvut.cz/~xmyslik/wifi > > Following the link... BTW, what for those copies in archive/ subdir? .bz2 seems to be corrupted. What tool did you use to obtain 'em? Other files are ok. Thank you. -- vda From hick.hostap at gink.org Thu Jan 23 04:42:32 2003 From: hick.hostap at gink.org (gARetH baBB) Date: Thu, 23 Jan 2003 09:42:32 +0000 (GMT) Subject: updating firmware on zoomair 4100(05) In-Reply-To: <041501c2c23b$8cbe83b0$4c28a8c0@cblan.mblox.com> References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> <03a501c2c230$ea783380$4c28a8c0@cblan.mblox.com> <041501c2c23b$8cbe83b0$4c28a8c0@cblan.mblox.com> Message-ID: On Wed, 22 Jan 2003, Ian Cass wrote: > bomberman:~/hostap/utils/srec_temp.uWgtFv# /usr/local/bin/prism2_srec -r > wlan0 r1010506.hex > srec summary for r1010506.hex > Included file name: r1010506.hex > Component: 0x001f 1.5.6 (station firmware) > Supported platforms: > 0x8002 1.0.0, 0x8002 1.0.1, 0x8003 1.0.0, 0x8003 1.0.1, 0x8004 1.0.0 > 0x8008 1.0.0 Hmmm, this is RAM based firmware though isn't it, not flash ... Anyone come across flash based 1.5.6 for 8003 ? Ditto for late(r|est) primary for 8003. From cijoml at volny.cz Thu Jan 23 04:45:15 2003 From: cijoml at volny.cz (Michal Semler) Date: Thu, 23 Jan 2003 10:45:15 +0100 Subject: updating firmware on zoomair 4100(05) In-Reply-To: References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> <041501c2c23b$8cbe83b0$4c28a8c0@cblan.mblox.com> Message-ID: Dne ?t 23. leden 2003 10:42 jste napsal(a): > On Wed, 22 Jan 2003, Ian Cass wrote: > > bomberman:~/hostap/utils/srec_temp.uWgtFv# /usr/local/bin/prism2_srec -r > > wlan0 r1010506.hex > > srec summary for r1010506.hex > > Included file name: r1010506.hex > > Component: 0x001f 1.5.6 (station firmware) > > Supported platforms: > > 0x8002 1.0.0, 0x8002 1.0.1, 0x8003 1.0.0, 0x8003 1.0.1, 0x8004 > > 1.0.0 0x8008 1.0.0 > > Hmmm, this is RAM based firmware though isn't it, not flash ... u have newer fw to flash into non-volatile memory? where can I download it? Michal From vda at port.imtp.ilyichevsk.odessa.ua Thu Jan 23 04:39:04 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Thu, 23 Jan 2003 11:39:04 +0200 Subject: hostapd does not detach from the terminal In-Reply-To: <20030117032823.GA1096@jm.kir.nu> References: <3E26E73C.7080109@keeper.org.uk> <20030117032823.GA1096@jm.kir.nu> Message-ID: <200301230940.h0N9dxs22332@Port.imtp.ilyichevsk.odessa.ua> On 17 January 2003 05:28, Jouni Malinen wrote: > On Thu, Jan 16, 2003 at 05:09:16PM +0000, kenw wrote: > > When I start hostapd it does not detach from the terminal. Is this > > normal behaviour, or should it fork giving me the terminal back. A > > couple of example command line I am using are below. > > It is normal behavior for the latest release, but CVS version has > support for forking into background. Not that I am against forking, but backgrounding code duplication in a unix utilities of all kinds is a waste of time. You can save your time by using already existing tools. Yes, there are tools to do it nicely. I use daemontools, they can start and daemonize programs, log their stdout with upper log size limit, *restart* crashed daemons (like infamous named) without losing logs etc etc etc. I am impressed by its simple, modular design. What a contrast with monstrosities I have to work with on my day job - Oracle under Windows 2000. That can be used to scare small and not-so-small children. -- vda From hick.hostap at gink.org Thu Jan 23 05:00:14 2003 From: hick.hostap at gink.org (gARetH baBB) Date: Thu, 23 Jan 2003 10:00:14 +0000 (GMT) Subject: updating firmware on zoomair 4100(05) In-Reply-To: References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> <041501c2c23b$8cbe83b0$4c28a8c0@cblan.mblox.com> Message-ID: On Thu, 23 Jan 2003, Michal Semler wrote: > u have newer fw to flash into non-volatile memory? > where can I download it? The list archives ... 1.4.9. Assuming we're talking (as per subject etc.) about firmware for 8003. From koma at lysator.liu.se Thu Jan 23 05:05:02 2003 From: koma at lysator.liu.se (Magnus Ekhall) Date: Thu, 23 Jan 2003 11:05:02 +0100 Subject: Setting of WEP keys from software Message-ID: <000901c2c2c6$e5026dd0$521d72d5@salamander> I'm trying to write code that changes a WEP key in the hostap driver. If I just call the local->crypt->ops->set_key functions it does not work, probably because the driver is set to use non-host based encryption. Which is the 'correct' way of changing a key so that both host based and non host-based encryption works? (Remember: I'm trying to do this from software...) Cheers, Magnus From cijoml at volny.cz Thu Jan 23 05:08:27 2003 From: cijoml at volny.cz (Michal Semler) Date: Thu, 23 Jan 2003 11:08:27 +0100 Subject: updating firmware on zoomair 4100(05) In-Reply-To: References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> Message-ID: Dne ?t 23. leden 2003 11:00 jste napsal(a): > On Thu, 23 Jan 2003, Michal Semler wrote: > > u have newer fw to flash into non-volatile memory? > > where can I download it? > > The list archives ... 1.4.9. and where can I download it? my card is in default wlan0: NIC: id=0x8013 v1.0.0 wlan0: PRI: id=0x15 v1.0.7 wlan0: STA: id=0x1f v1.3.6 i would like try to flash 1.4.9... Michal From ian.cass at mblox.com Thu Jan 23 05:23:04 2003 From: ian.cass at mblox.com (Ian Cass) Date: Thu, 23 Jan 2003 10:23:04 -0000 Subject: updating firmware on zoomair 4100(05) References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> <03a501c2c230$ea783380$4c28a8c0@cblan.mblox.com> <041501c2c23b$8cbe83b0$4c28a8c0@cblan.mblox.com> Message-ID: <052101c2c2c9$6a07db30$4c28a8c0@cblan.mblox.com> > Hmmm, this is RAM based firmware though isn't it, not flash ... > > Anyone come across flash based 1.5.6 for 8003 ? > > Ditto for late(r|est) primary for 8003. Yes it is, although you didn't specify that you wanted flash firmware in your original question ;) -- Ian Cass From hick.hostap at gink.org Thu Jan 23 05:28:14 2003 From: hick.hostap at gink.org (gARetH baBB) Date: Thu, 23 Jan 2003 10:28:14 +0000 (GMT) Subject: updating firmware on zoomair 4100(05) In-Reply-To: References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> Message-ID: On Thu, 23 Jan 2003, Michal Semler wrote: > and where can I download it? my card is in default > > wlan0: NIC: id=0x8013 v1.0.0 Is this really a ZoomAir 410x ? > i would like try to flash 1.4.9... > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap ... no doubt has a link to list archives ... From hick.hostap at gink.org Thu Jan 23 05:29:16 2003 From: hick.hostap at gink.org (gARetH baBB) Date: Thu, 23 Jan 2003 10:29:16 +0000 (GMT) Subject: updating firmware on zoomair 4100(05) In-Reply-To: <052101c2c2c9$6a07db30$4c28a8c0@cblan.mblox.com> References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> <03a501c2c230$ea783380$4c28a8c0@cblan.mblox.com> <041501c2c23b$8cbe83b0$4c28a8c0@cblan.mblox.com> <052101c2c2c9$6a07db30$4c28a8c0@cblan.mblox.com> Message-ID: On Thu, 23 Jan 2003, Ian Cass wrote: > Yes it is, I thought the RAM based images and the flash images were most certainly different. > although you didn't specify that you wanted flash firmware in your > original question ;) True ... From matt at softec.uk.com Thu Jan 23 05:48:53 2003 From: matt at softec.uk.com (Matt Connew) Date: Thu, 23 Jan 2003 10:48:53 -0000 Subject: Trying to track down problem Message-ID: As I mentioned in a previous message, I've got a problem with the HostAP driver spewing out thousands of "Could not find STA for TX error" messages. I've tracked down the error to hostap_handle_sta_tx_exc in the hostap_ap.c source file. The error is happening because txdesc->addr1 holds a mac address of 00:00:00:00:00:00. I've examined the values in addr2, addr3 and addr4 and they seem quite consistent as well when this error happens. Is there a manual/guide to the structure of the source code and data structures or can someone tell me what the significance of the various fields of the txdesc structure are? Here are the values of the addr fields when I get this error: addr1 : Always 00:00:00:00:00:00 addr2 : MAC address of AP addr3 : One of 01:80:c2:00:00:00 or ff:ff:ff:ff:ff:ff addr4 : Always 00:00:00:00:00:00 Thanks in advance Matt From tommy at home.tig-grr.com Wed Jan 22 13:11:12 2003 From: tommy at home.tig-grr.com (Tom Marshall) Date: Wed, 22 Jan 2003 10:11:12 -0800 Subject: [lwlan-devel] dwl-520+ and linux (works!) In-Reply-To: ; from panxer@b-online.gr on Wed, Jan 22, 2003 at 07:41:42PM +0200 References: Message-ID: <20030122101112.B374@home.tig-grr.com> I don't see the driver listed on the download page. Did they remove it? Does anyone have a copy they can post? http://www.eusso.com/Model2/GL2422-VP/GL2422-VP.htm http://www.eusso.com/Model2/GL2422/GL2422_utility.htm On Wed, Jan 22, 2003 at 07:41:42PM +0200, panxer wrote: > So the + generation of dlink's products seems to finally work..? > I think i might have made dwl-520+ work under linux.. > Even though dlink announced in last december that they would publish drivers > (be it binary) for their TI - 22mbps products, such thing never happened.. > Some point in November as i recall someone published a binary driver on > behalf of a eusso company for a pPCI devicethat has the same chipset as the + > generation by dlink (and others), the acx100 by Texas Instruments.. -- The little girl expects no declaration of tenderness from her doll. She loves it -- and that's all. It is thus that we should love. -- DeGourmont -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 240 bytes Desc: not available Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030122/3524e253/attachment.pgp From cijoml at volny.cz Thu Jan 23 07:30:34 2003 From: cijoml at volny.cz (Michal Semler) Date: Thu, 23 Jan 2003 13:30:34 +0100 Subject: updating firmware on zoomair 4100(05) In-Reply-To: References: <7306B0A474DDD41191D70002A52CA95A043C1343@NTPRDEXS053.admin.liffe.com> Message-ID: Dne ?t 23. leden 2003 11:28 jste napsal(a): > On Thu, 23 Jan 2003, Michal Semler wrote: > > and where can I download it? my card is in default > > > > wlan0: NIC: id=0x8013 v1.0.0 > > Is this really a ZoomAir 410x ? no, zcomax xi-626 pci version Michal From vda at port.imtp.ilyichevsk.odessa.ua Thu Jan 23 08:55:41 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Thu, 23 Jan 2003 15:55:41 +0200 Subject: Somebody have the programmers guide of intersil chipset? In-Reply-To: <3E2F9625.4010708@palaver.net> References: <200301230659.h0N6xQs21493@Port.imtp.ilyichevsk.odessa.ua> <3E2F9625.4010708@palaver.net> Message-ID: <200301231356.h0NDuas23569@Port.imtp.ilyichevsk.odessa.ua> On 23 January 2003 09:13, Brian Capouch wrote: > > Anyway, even 1.4.9 firmware files aren't that big (~64k). > > Do anybody know what CPU is used in Prism cards? > > Is it legal to try and disassemble firmware? > > -- > > Boy, those words sure got my excitement level raised!! > > Maybe if we could write our own firmware, we could adapt the > technique regarding the "distributed" versus "short" interframe > spaces put forth in the AT&T paper that would allow for > Infrastructure mode connections to be solidly done from >~6000' from > the access point. . . > > I'm all for it; hope it might be possible. It would be possible *if* we know which CPU is in there. And that's a big if. I didn't try to contact Intersil on this matter. -- vda From br1 at subnet.at Thu Jan 23 09:19:26 2003 From: br1 at subnet.at (bruno randolf) Date: Thu, 23 Jan 2003 15:19:26 +0100 Subject: Somebody have the programmers guide of intersil chipset? In-Reply-To: <200301231356.h0NDuas23569@Port.imtp.ilyichevsk.odessa.ua> References: <3E2F9625.4010708@palaver.net> <200301231356.h0NDuas23569@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <200301231519.26877.br1@subnet.at> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 > > Maybe if we could write our own firmware, we could adapt the > > technique regarding the "distributed" versus "short" interframe > > spaces put forth in the AT&T paper that would allow for > > Infrastructure mode connections to be solidly done from >~6000' from > > the access point. . . sorry, if i missed something, but could you please give us a link to this paper. thanks! br1 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.0 (GNU/Linux) iD8DBQE+L/nuqCK9CeX6cL8RAmpHAJ93VUe2BlPs4xEphJdx/HraA9N9XACdFcot /XuG467yUeEGadvwMgBXGLc= =eEhv -----END PGP SIGNATURE----- From location at snet.wit.ie Thu Jan 23 15:10:31 2003 From: location at snet.wit.ie (Location Proj) Date: Thu, 23 Jan 2003 20:10:31 +0000 (GMT) Subject: Mactime and sig strength Message-ID: In relation to wlansniff and the prism monitor mode: Can someone explain what mactime is? And My Linksys WPC11 cards seem to be reporting a signal strength from 0 to 100, is this actually what the card is reporting or is this a normalisation in the driver? thanks, -- Andrew From steve.trease at btinternet.com Thu Jan 23 17:31:34 2003 From: steve.trease at btinternet.com (Steve Trease) Date: Thu, 23 Jan 2003 22:31:34 -0000 Subject: Problems compiling HostAP on Red Hat 8.0 Message-ID: <000001c2c32f$44ea7310$0100a8c0@home> Hello, I'm trying to get HostAP driver 2002-10-12 working on a couple of Red Hat 8.0 systems running the stock Red Hat 2.4.18-19.8.0 kernel. The first system (Sagan) is a Duron with an SMC PLX card and I can compile the HostAP drivers and get the card working. No problem. The second system (Miller) is a PII also with an SMC PLX card. I can't compile the drivers on it because it is very low on disk space, so I'm trying to compile the modules for it on the first system. When I load the compiled drivers on the second machine depmod gives some "Unresolved symbols" errors listed at the bottom of this message. I've installed the kernel sources from the RedHat rpm. I then copy kernel-2.4.18-i686.config file from /usr/src/linux/configs to /usr/src/linux/.config. Then I'm doing a make dep in /usr/src/linux. After that I head on over to the HostAP src directory and do a "make clean" and "make plx". The modules seem to compile OK. Next I copy the four HostAP .o files over to /lib/modules/2.4.18-19.8.0/net on Miller and run depmod -ae. I'm almost certainly doing missing something pretty obvious. Any idea what it is? Thanks, Steve [root at miller steve]# depmod -ae depmod: *** Unresolved symbols in /lib/modules/2.4.18-19.8.0/net/hostap.o depmod: register_netdevice depmod: __netdev_watchdog_up depmod: __kfree_skb depmod: alloc_skb depmod: ether_setup depmod: unregister_netdevice depmod: create_proc_entry depmod: unregister_netdev depmod: request_module depmod: proc_mkdir depmod: register_netdev depmod: dev_queue_xmit depmod: remove_proc_entry depmod: skb_over_panic depmod: proc_net depmod: softnet_data depmod: irq_stat depmod: *** Unresolved symbols in /lib/modules/2.4.18-19.8.0/net/hostap_plx.o depmod: eth_type_trans depmod: __wake_up depmod: __kfree_skb depmod: alloc_skb depmod: create_proc_entry depmod: remove_wait_queue depmod: unregister_netdev depmod: register_netdev depmod: dev_queue_xmit depmod: ___pskb_trim depmod: remove_proc_entry depmod: netif_rx depmod: skb_over_panic depmod: add_wait_queue depmod: skb_clone depmod: softnet_data depmod: irq_stat [root at miller steve]# From vda at port.imtp.ilyichevsk.odessa.ua Fri Jan 24 06:25:04 2003 From: vda at port.imtp.ilyichevsk.odessa.ua (Denis Vlasenko) Date: Fri, 24 Jan 2003 13:25:04 +0200 Subject: Setting MAC address In-Reply-To: <200301230709.h0N79cs21553@Port.imtp.ilyichevsk.odessa.ua> References: <200301200759.h0K7xds05313@Port.imtp.ilyichevsk.odessa.ua> <20030121040954.GA1743@jm.kir.nu> <200301230709.h0N79cs21553@Port.imtp.ilyichevsk.odessa.ua> Message-ID: <200301241126.h0OBQ2s28212@Port.imtp.ilyichevsk.odessa.ua> On 23 January 2003 09:08, Denis Vlasenko wrote: > > This works in my tests, both as an AP and as a station. When > > running hostapd, one must restart hostapd after this change. Other > > than that, there should not be other limitations on this. Anyway, I > > would recommend changing the MAC address when the device is down. > > > > Please give some details about the system and configuration that > > behaves incorrectly.... [snip] > > DLink DWL-520, firmware 1.4.9, used in STA mode. > kernel 2.4.20-pre11, wireless extensions 14. > hostap-2002-10-12 is installed, but I use mainline kernel's > orinoco_pci. > > Will try with hostap (and kernel 2.4.20) instead. Tried. hostap_pci works even better that mainline's orinoco_pci. Specifically, with orinoco I get some dups when I ping and TCP is very slow. Hostap does not show dups, TCP is notably faster. I experience a bit of packet loss is due to weak signal, it's happening with both drivers. Thanks! -- vda From tanpaul at icr.a-star.edu.sg Fri Jan 24 12:24:25 2003 From: tanpaul at icr.a-star.edu.sg (Paul Tan) Date: Sat, 25 Jan 2003 01:24:25 +0800 Subject: Query Message-ID: <001b01c2c3cd$724bafb0$7e3f8489@PIGGY> Hi all, I am new to this mailing list -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030125/c9e979f6/attachment.htm From tanpaul at icr.a-star.edu.sg Fri Jan 24 12:30:50 2003 From: tanpaul at icr.a-star.edu.sg (Paul Tan) Date: Sat, 25 Jan 2003 01:30:50 +0800 Subject: Question of HostAP Message-ID: <002e01c2c3ce$57ecc810$7e3f8489@PIGGY> Hi all, I would like to know if the HostAP mode allows user-space program on the station or AP to add new 802.11 Information Element into the Beacon or Assoc frames. Does it provide the ability for deliver such information up to IP layer? Is there any 802.11 NIC vendors that allows such feature? I hope my questions are reasonable. 8-) Thank you in advance. Cheers, Paul -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030125/c353091f/attachment.htm From mrcool at stupidgamerz.com Fri Jan 24 13:40:24 2003 From: mrcool at stupidgamerz.com (Jerry) Date: Fri, 24 Jan 2003 12:40:24 -0600 Subject: Linksys WPC11, can load drivers (manually) but doesnt create device wlan0. Message-ID: <000801c2c3d8$0eaa76c0$5304000a@terraworld.net> SuSE 8.0 (fresh) Hostap 10/03.... The driver will compile (make pccard) and install (make install_pccard) fine... after installing the driver you can 'modprobe hostap_cs' and it will load the driver fine. But it wont create wlan0, the interface. if i reboot at this point, the machine finds the pcmcia adapter (pci to pcmcia adapter) and tries to initialize the prism2_cs driver (per dmesg), this doesnt work (you get the low freq beep), but it doesnt even TRY to load the hostap_cs driver... what am i doing wrong? I did the same thing with a plx adapter (same os/hostap) in a diff mobo and even though i had to modprobe to get it to load, at least it creates wlan0, and i can use it... Where am i going wrong and what info is needed to help troubleshoot this...? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030124/fa9e0bc0/attachment.htm From jlewis at netnitco.net Thu Jan 23 18:36:33 2003 From: jlewis at netnitco.net (James Lewis) Date: Thu, 23 Jan 2003 17:36:33 -600 Subject: HostAP Hard Lockup Message-ID: <3e307c819ffb17.71335272@netnitco.net> Hi Everyone, I'm using HostAP with a netgear MA311. I've observed a hard lock up on kernel versions 2.4.18 and 2.4.20. Whenever my client (unsure of the chipset) is transferring a large amount of data, the host PC will lock up enitrely. Before I post log information that I was able to collect (which isn't much) is there a known incompatiability with HostAP (oct-12-02) and a Netgera MA311? Thanks in advance, James From jsun at junsun.net Fri Jan 24 17:53:23 2003 From: jsun at junsun.net (Jun Sun) Date: Fri, 24 Jan 2003 14:53:23 -0800 Subject: driver warning with kismet Message-ID: <20030124145323.A8521@gateway.junsun.net> I am running kismet on top of recent CVS hostap driver. The driver normally works fine, but with kismet it emmits some annoying warnings like: wlan0: prism2_rx_80211: len(11023) > MAX(2304) Anybody knows what is wrong here? A driver bug? Or a configuration problem? Jun From jasonb at edseek.com Fri Jan 24 19:08:08 2003 From: jasonb at edseek.com (Jason Boxman) Date: Fri, 24 Jan 2003 19:08:08 -0500 (EST) Subject: driver warning with kismet In-Reply-To: <20030124145323.A8521@gateway.junsun.net> References: <20030124145323.A8521@gateway.junsun.net> Message-ID: <1126.209.208.68.56.1043453288.squirrel@rebecca.edseek.com> Jun Sun said: > > I am running kismet on top of recent CVS hostap driver. The driver > normally works fine, but with kismet it emmits some annoying warnings > like: > > wlan0: prism2_rx_80211: len(11023) > MAX(2304) > > Anybody knows what is wrong here? A driver bug? Or a configuration > problem? I get those too, a lot. I think they're just corrupt frames. dmesg -n 1 will get those off your console, at least. > > Jun > From sp0de at netzero.net Fri Jan 24 19:08:43 2003 From: sp0de at netzero.net (sp0de) Date: Fri, 24 Jan 2003 17:08:43 -0700 Subject: HostAP Hard Lockup References: <3e307c819ffb17.71335272@netnitco.net> Message-ID: <3E31D58B.3090908@netzero.net> I'm having a very similar problem. I can transfer files from the laptop to the AP computer, and from the AP computer to the laptop. I can also send files from the laptop through the AP computer to my desktop machine. The problem occurs when I try to send files to the laptop from the desktop. the AP will freeze up as soon as I start the transfer. I haven't been able to get any logs from it either. here's my setup: laptop: kernel 2.4.19, belkin F5D6020 pcmcia, HostAP configured as managed, id=0x8002 firmware PRI 0.3.0 STA 0.8.3 AP: kernel 2.4.20, belkin F5D6000 pci, HostAP configured as master, id=0x8013 firmware upgraded to PRI 1.1.0 STA 1.4.9 desktop: kernel 2.4.19, connected to AP via ethernet I am not running WEP on the network, and I am using sftp to transfer files between computers. Ohh.. it also works just fine for internet, only the high speed ethernet will freeze up the AP. Well, i guess you can add kernel 2.4.19 and belkin F5D6020 to that list. Sorry I don't have any answers for you *yet*, but now you know your not alone :) Sp0de James Lewis wrote: >Hi Everyone, > >I'm using HostAP with a netgear MA311. I've observed a hard lock up on kernel versions 2.4.18 and 2.4.20. Whenever my client (unsure of the chipset) is transferring a large amount of data, the host PC will lock up enitrely. > >Before I post log information that I was able to collect (which isn't much) is there a known incompatiability with HostAP (oct-12-02) and a Netgera MA311? > >Thanks in advance, >James > >_______________________________________________ >HostAP mailing list >HostAP at shmoo.com >http://lists.shmoo.com/mailman/listinfo/hostap > > > > From jasonb at edseek.com Fri Jan 24 19:21:37 2003 From: jasonb at edseek.com (Jason Boxman) Date: Fri, 24 Jan 2003 19:21:37 -0500 (EST) Subject: Linksys WPC11, can load drivers (manually) but doesnt create device wlan0. In-Reply-To: <000801c2c3d8$0eaa76c0$5304000a@terraworld.net> References: <000801c2c3d8$0eaa76c0$5304000a@terraworld.net> Message-ID: <1198.209.208.68.56.1043454097.squirrel@rebecca.edseek.com> Jerry said: > SuSE 8.0 (fresh) Hostap 10/03.... The driver will compile (make pccard) > and install (make install_pccard) fine... > after installing the driver you can 'modprobe hostap_cs' and it will load > the driver fine. But it wont create wlan0, the interface. if i reboot at > Where am i going wrong and what info is needed to help troubleshoot > this...? Well, what I do is in /etc/modules.conf (or in my case /etc/modutils/aliases on Debian) is add: alias wlan0 hostap_pci I always have my precious wlan0 now. From hueiping at yahoo.com Fri Jan 24 21:38:20 2003 From: hueiping at yahoo.com (Huei-Ping Chen) Date: Fri, 24 Jan 2003 18:38:20 -0800 (PST) Subject: one PC emulate lot of wireless client, possible ? Message-ID: <20030125023820.13589.qmail@web40703.mail.yahoo.com> Hi, I am wondering if it's possible, using one laptop with 2 wireless pcmcia card to emulate lots of wireless clients ? this is what I am thinking, using hostapd as example, I can modified it to send authentication/association request with made-up MAC address, so, the AP would think mutiple clients are trying to associate with it, and if I put the other card into monitor mode, then I can listen to all response from the AP. this way, I can transmit raw 802.11/802.1x packets, and can also listening to all packets, so I should be able to emualte lots of clients, right ? but there is one thing that's bothering me, that, I have seen Jouni mentioned a couple of times, that the client need to send back an ACK after receiving a packet, so, even though the card that's in monitor receives the packet, but it can't send ACK because it can't transmit in monitor, so, the above scheme would not work at all. is my understanding correct ? any suggestion/pointer is greatly appreciated, -- Ping __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com From jkmaline at cc.hut.fi Fri Jan 24 22:45:35 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Fri, 24 Jan 2003 19:45:35 -0800 Subject: Linksys WPC11, can load drivers (manually) but doesnt create device wlan0. In-Reply-To: <000801c2c3d8$0eaa76c0$5304000a@terraworld.net> References: <000801c2c3d8$0eaa76c0$5304000a@terraworld.net> Message-ID: <20030125034535.GD1351@jm.kir.nu> On Fri, Jan 24, 2003 at 12:40:24PM -0600, Jerry wrote: > the machine finds the pcmcia adapter (pci to pcmcia adapter) and tries to initialize the prism2_cs driver (per dmesg), this doesnt work (you get the low freq beep), but it doesnt even TRY to load the hostap_cs driver... That sounds like your pcmcia-cs (actually, cardmgr) is configured to use prism2_cs driver. If you want to use Host AP driver with that card, you would need to change pcmcia-cs configuration for that card. Host AP driver installation should have already copied hostap_cs.conf that describes hostap_cs.o as the driver for Linksys WPC11. However, some other configuration file can override this configuration. You would thus need to comment out that other part refering to prism2_cs driver. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Fri Jan 24 22:49:23 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Fri, 24 Jan 2003 19:49:23 -0800 Subject: driver warning with kismet In-Reply-To: <20030124145323.A8521@gateway.junsun.net> References: <20030124145323.A8521@gateway.junsun.net> Message-ID: <20030125034923.GE1351@jm.kir.nu> On Fri, Jan 24, 2003 at 02:53:23PM -0800, Jun Sun wrote: > I am running kismet on top of recent CVS hostap driver. The driver > normally works fine, but with kismet it emmits some annoying warnings > like: > > wlan0: prism2_rx_80211: len(11023) > MAX(2304) > > Anybody knows what is wrong here? A driver bug? Or a configuration problem? Those corrupted frames are coming from the wireless LAN card as far as I know. I'm not familiar with kismet operation details, but if it is changing channels often, this could be causing lot of corrupted RX'ed frames that are received while the channel is being changed. Monitor mode seems to pass such corrupted frames to the driver and the driver prints that warning. If you want to get rid of it, you can comment it out in prism2_rx_80211() function so that the driver just silently drops such frames. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Fri Jan 24 22:52:42 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Fri, 24 Jan 2003 19:52:42 -0800 Subject: Question of HostAP In-Reply-To: <002e01c2c3ce$57ecc810$7e3f8489@PIGGY> References: <002e01c2c3ce$57ecc810$7e3f8489@PIGGY> Message-ID: <20030125035242.GF1351@jm.kir.nu> On Sat, Jan 25, 2003 at 01:30:50AM +0800, Paul Tan wrote: > I would like to know if the HostAP mode allows user-space program on the station or AP to add new 802.11 Information Element into the Beacon or Assoc frames. You can do that for association reply (see handle_assoc() function), but Prism2 station firmware takes care of beacon generation within the wlan card and beacons cannot thus be changed in this way. -- Jouni Malinen PGP id EFC895FA From ammirata at econointl.com Sat Jan 25 00:30:59 2003 From: ammirata at econointl.com (Sergio M. Ammirata) Date: Sat, 25 Jan 2003 00:30:59 -0500 Subject: hostapd issues and questions Message-ID: <000301c2c432$f35e2f00$017ba8c0@ammirata.net> > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] > On Behalf Of Jouni Malinen > Sent: Tuesday, January 21, 2003 12:17 AM > To: hostap at shmoo.com > Subject: Re: hostapd issues and questions > > > On Sun, Jan 19, 2003 at 12:09:35PM -0500, Sergio M. Ammirata wrote: > > > 1) Changing the mac address of the wlan0 interface using ifconfig > > breaks routing (you can link but the tcp/ip layer does not > work). This > > happens only with hostapd enabled. > > Was hostapd running when you changed MAC address? You must > restart hostapd after this change. Hostapd is not running when I change the MAC address. Here is the script I use: echo -n "Starting wlan0: " insmod hostap_crypt insmod hostap_crypt_wep insmod hostap # Start in infrastructure mode so that the old MAC Address does not get exposed. insmod hostap_cs essid=test iw_mode=2 channel=6 echo -n "Initializing pcmcia" /etc/rc.d/init.d/pcmcia init # Set the to the same mac as the ethernet interface HWADDR=`/sbin/ifconfig eth0 | sed -n 's/.*HWaddr\(.*\)/\1/p'` ifconfig wlan0 hw ether $HWADDR iwconfig wlan0 mode master # Configure Wireless parameters iwconfig wlan0 txpower 23 iwconfig wlan0 sens 3 iwconfig wlan0 nick $hostname # Disable bridging (use ethernet bridges instead) iwpriv wlan0 bridge_packets 0 iwpriv wlan0 host_roaming 1 # Now that everything is configured bring up the wlan0 interface and enabled the hostap daemon ifconfig wlan0 0.0.0.0 # Start user space daemon hostapd /etc/hostapd.conf echo "Done." > > > 2) Without hostapd there is wlan0 and wlan0wdsx. Now we > have wlan0ap > > and wlan0sta as well. How are we supposed to setup Ethernet > bridging > > now? Do we include all three in the bridge (wlan0, wlan0ap and > > wlan0sta)? > > wlan0ap should not be included in bridge setup at all. It > used only for management frames (802.11 auth/assoc, 802.1X > EAPOL frames) within that one AP. wlan0sta is an experimental > device used for station mode operations with another AP if > assoc_ap_addr is set. It does not support layer-2 bridging > (just like any other "normal" station mode card), so it > should not be included in bridge setup. > > 3) Has the functionality of wlan0wdsx changed? When and why would I > > use wlan0sta? > > No, wlan0wds# work as before. You could potentially use > wlan0sta as a wireless uplink with IP routing (and possible > Proxy ARPing). I noticed that wlan0sta does not have an independent essid. When you change it on wlan0 it also changes for wlan0sta. If the destination MAC for wlan0sta has a different essid than wlan0, will it still link? > > 4) With hostapd active iwspy no longer reports any info on wds > > neighbors. How can I found out the strength of the links? > > That is not implemented yet. Will it make it into this release? Is there an alternate way of measuring the link speed or to even found out that we are linked to neighbors other than pinging? > -- > Jouni Malinen PGP > id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap Thanks in advance, Sergio _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030125/0628d81a/attachment.htm From jsun at junsun.net Sat Jan 25 00:40:36 2003 From: jsun at junsun.net (Jun Sun) Date: Fri, 24 Jan 2003 21:40:36 -0800 Subject: driver warning with kismet In-Reply-To: <20030125034923.GE1351@jm.kir.nu>; from jkmaline@cc.hut.fi on Fri, Jan 24, 2003 at 07:49:23PM -0800 References: <20030124145323.A8521@gateway.junsun.net> <20030125034923.GE1351@jm.kir.nu> Message-ID: <20030124214036.C8767@gateway.junsun.net> On Fri, Jan 24, 2003 at 07:49:23PM -0800, Jouni Malinen wrote: > On Fri, Jan 24, 2003 at 02:53:23PM -0800, Jun Sun wrote: > > > I am running kismet on top of recent CVS hostap driver. The driver > > normally works fine, but with kismet it emmits some annoying warnings > > like: > > > > wlan0: prism2_rx_80211: len(11023) > MAX(2304) > > > > Anybody knows what is wrong here? A driver bug? Or a configuration problem? > > Those corrupted frames are coming from the wireless LAN card as far as I > know. I'm not familiar with kismet operation details, but if it is > changing channels often, Indeed it is doing channel hopping a lot (three times a second, I think). That should explain why it works fine normaly but not with kismet. Jun From dave at cc0.net Sat Jan 25 04:04:08 2003 From: dave at cc0.net (Dave) Date: Sat, 25 Jan 2003 01:04:08 -0800 Subject: Automatic WDS problem Message-ID: <000501c2c450$b8a11f30$0300a8c0@ws7> Hello all, It is my understanding from reading (thoroughly) README.prism2 that WDS links can be created dynamically when another HostAP box is within range (3 address) or another standards compliant AP is within range (4 address, 1.5.x+ FW). I have three Soekris boxes each with all identical hardware/OS/configuration but I am working with two currently. I RAM upload 1.5.6 to the card on bootup (successfully) then perform the following commands: prism2_param wlan0 other_ap_policy 1 prism2_param wlan0 autom_ap_wds 1 prism2_param wlan0 wds_type 4 iwpriv wlan0 wds_add 00:00:00:00:00:00 iwpriv wlan0 wds_add 00:00:00:00:00:00 iwpriv wlan0 wds_add 00:00:00:00:00:00 brctl addbr br0 brctl addif br0 wlan0 brctl addif br0 wlan0wds0 brctl addif br0 wlan0wds1 brctl addif br0 wlan0wds2 ifconfig wlan0 0.0.0.0 up ifconfig wlan0wds0 0.0.0.0 up ifconfig wlan0wds1 0.0.0.0 up ifconfig wlan0wds2 0.0.0.0 up ifconfig br0 192.168.0.2 up At this point I see the bridge fire up and begin learning. When I do a cat /proc/net/hostap/wlan0/wds I see all the MACs for the preallocated WDS links as all 00:00:00:00:00:00. All is ok for now...about 3 or 4 minutes of waiting and with the other box up and running with same configuration (except alternate br0 ip address) and within a few feet of each other (no antennas) I again cat /proc/net/hostap/wlan0/wds and still do not see the MACs of the other box. I do see the MAC of the other box in /proc/net/hostap/wlan0/00:02:6f:01:f9:b6 and see stats in it...each box has the opposing MAC at that location... It looks to me like it is associating as a STA and not an AP. Is this assumption correct? I am using a 2.4.20 kernel with the kernel-patch and building the drivers in with the kernel. I opted to not enable hostapd functionality (do I need it?) and I also defined PRISM2_NO_STATION_MODES since I don't have any need for these boxes to be stations. Each box has an IP assigned to the br0 although neither can hit the other's IP. Anyone have any idea what is up? Thanks for any assistance you can provide! -Dave From czoffoli at xmerlin.org Fri Jan 24 05:45:59 2003 From: czoffoli at xmerlin.org (Christian Zoffoli) Date: Fri, 24 Jan 2003 11:45:59 +0100 Subject: Somebody have the programmers guide of intersil chipset? In-Reply-To: <20030123072447.GA21122@jm.kir.nu> References: <3E2CFEFA.2020001@abicom-international.com> <200301230659.h0N6xQs21493@Port.imtp.ilyichevsk.odessa.ua> <20030123072447.GA21122@jm.kir.nu> Message-ID: <3E311967.1070807@xmerlin.org> Jouni Malinen wrote: > On Thu, Jan 23, 2003 at 08:58:32AM +0200, Denis Vlasenko wrote: > > >>Do you mean NIC's CPU (or whatever) docs? > > > I would assume that that is not available for anyone outside Intersil. > One might be able to get driver programmer's documentation after signing > NDA, but it does not cover MAC processor specific stuff. The italian representative of Intersil says to me that I have to bought also one of the reference boards, if I want the programmer's documentation ...is it right? Thanks, Christian From paul at black-sun.demon.co.uk Fri Jan 24 19:38:47 2003 From: paul at black-sun.demon.co.uk (Paul Walker) Date: Sat, 25 Jan 2003 00:38:47 +0000 Subject: Strange effect with SparkLAN Message-ID: <20030125003847.GB7812@black-sun.demon.co.uk> Hi, I've got two SparkLAN cards, both using Prism2.5 (one PCI and one PCMCIA). The one in the desktop is using hostap to act as master. Running WinXP on the laptop with the PCMCIA card, it'll work fine for hours under heavy load. But.. running Linux on the laptop with various drivers (orinoco, prism2 etc.), things work for about 5 minutes at most before the card in the desktop dies to the extent I have to power-cycle it to work again. :/ Anyone have any idea what could cause this? Failing this, what debug info would help to narrow this down? Seems odd that a change in remote behaviour could kill the desktop, but there you go. Cheers, -- Paul And I raise my head and stare / Into the eyes of a stranger I've always known / That the mirror never lies -- Queensryche, "Eyes of a Stranger" From mrcool at stupidgamerz.com Sat Jan 25 10:32:39 2003 From: mrcool at stupidgamerz.com (Jerry) Date: Sat, 25 Jan 2003 09:32:39 -0600 Subject: Linksys WPC11, can load drivers (manually) but doesnt create device wlan0. References: <000801c2c3d8$0eaa76c0$5304000a@terraworld.net> <20030125034535.GD1351@jm.kir.nu> Message-ID: <005701c2c487$0187c7e0$f700a8c0@Home> Jason wrote: >Well, what I do is in /etc/modules.conf (or in my case >/etc/modutils/aliases on Debian) is add: >alias wlan0 hostap_pci >I always have my precious wlan0 now. Jouni wrote: > That sounds like your pcmcia-cs (actually, cardmgr) is configured to use > prism2_cs driver. If you want to use Host AP driver with that card, you > would need to change pcmcia-cs configuration for that card. Host AP > driver installation should have already copied hostap_cs.conf that > describes hostap_cs.o as the driver for Linksys WPC11. However, some > other configuration file can override this configuration. You would thus > need to comment out that other part refering to prism2_cs driver. > Thanks Jouni, Jason... Jason: ...had tried that already, its in /etc/modules.conf. doesnt make any difference... Jouni: I think you are right. But, Im a realtive newb to linux, and since i use SuSE (which has all kinds of wizards and such that work so well) im not familiar with ALL the little text files in etc that configure things... Does anyone know, perhaps, which config files i need to modify to make cardmanager load host_ap instead of prism2_cs? Oh, and i do have the hostap_cs.conf updated to what my linksys card has when you type 'cardctl ident', i have actually tried several different incarnations of this... Thanks Jerry From slimclity at hotmail.com Sat Jan 25 11:48:05 2003 From: slimclity at hotmail.com (Slim Clity) Date: Sat, 25 Jan 2003 16:48:05 +0000 Subject: HostAP & Prism GT Chipset? Message-ID: Will HostAP support the new Prism GT chipset? (D-Link AirPlus Xtreme G DWL-G520 PCI Adapter) _________________________________________________________________ MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus From panxer at b-online.gr Sat Jan 25 11:51:13 2003 From: panxer at b-online.gr (panxer) Date: Sat, 25 Jan 2003 18:51:13 +0200 Subject: Dwl-520+ under Linux HOWTO Message-ID: http://www.hack.gr/users/panxer/wireless/ -- Panagiotis Moustafellos (aka panXer) From jasonb at edseek.com Sat Jan 25 11:10:10 2003 From: jasonb at edseek.com (Jason Boxman) Date: Sat, 25 Jan 2003 11:10:10 -0500 Subject: driver warning with kismet In-Reply-To: <20030124214036.C8767@gateway.junsun.net> References: <20030124145323.A8521@gateway.junsun.net> <20030125034923.GE1351@jm.kir.nu> <20030124214036.C8767@gateway.junsun.net> Message-ID: <200301251110.10993.jasonb@edseek.com> On Saturday 25 January 2003 12:40 am, Jun Sun wrote: > On Fri, Jan 24, 2003 at 07:49:23PM -0800, Jouni Malinen wrote: > > On Fri, Jan 24, 2003 at 02:53:23PM -0800, Jun Sun wrote: > > > I am running kismet on top of recent CVS hostap driver. The driver > > > normally works fine, but with kismet it emmits some annoying warnings > > > like: > > > > > > wlan0: prism2_rx_80211: len(11023) > MAX(2304) > > > > > > Anybody knows what is wrong here? A driver bug? Or a configuration > > > problem? > > > > Those corrupted frames are coming from the wireless LAN card as far as I > > know. I'm not familiar with kismet operation details, but if it is > > changing channels often, > > Indeed it is doing channel hopping a lot (three times a second, I think). > That should explain why it works fine normaly but not with kismet. Yeah, that's one of the drawbacks of using a Prism based card for Kismet, it has to hop channels for you, a lot. If you want to listen on a specific channel you could always stop running kismet_hopper. :) > Jun -- Jason Boxman Administrator TrekWeb.COM http://trekweb.com/~jasonb/articles/ - Linux Pages From mojo.nichols at verizon.net Sat Jan 25 22:43:58 2003 From: mojo.nichols at verizon.net (Mojo B. Nichols) Date: 25 Jan 2003 22:43:58 -0500 Subject: range In-Reply-To: Message-ID: What is the "typical" range of 802.11b? How much do antennas help. I imagine that it greatly depends on the hardware as well. If so, What is a good pci card in terms of range that is hostap compatibilty tia mojo From sneeze at igreen.net Sat Jan 25 21:00:06 2003 From: sneeze at igreen.net (Ryan Abbenhuys) Date: Sun, 26 Jan 2003 13:00:06 +1100 Subject: Card initialisation problem under RH8 Message-ID: <002101c2c4de$a7ddf400$0200a8c0@silver> Occasionally my WL200 doesn't seem to initialise properley at bootup and sometimes requires that I restart the pcmcia service several times before it kicks in. The system is Redhat8 running kernel 2.4.20 The following is the output from dmesg. Does this shed any light on what might be happening? hostap_crypt: registered algorithm 'NULL' hostap_cs: 0.0.0 CVS (Jouni Malinen ) hostap_cs: setting Vcc=33 (constant) hostap_cs: CS_EVENT_CARD_INSERTION prism2_config() hostap_cs: setting Vcc=33 (from config) Checking CFTABLE_ENTRY 0x01 (default 0x01) Config has no IRQ info, but trying to enable IRQ anyway.. IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 io->flags = 0x0046, io.base=0x0000, len=64 hostap_cs: index 0x01: Vcc 3.3, irq 16, io 0x0180-0x01bf hostap_cs: Registered netdevice wlan0 wlan0: prism2_open prism2_hw_init() wlan0: init command completed too quickly - retrying prism2_hw_config: initialized in 9604 iterations wlan0: hfa384x_cmd: entry still in list? (entry=dfe699a0, type=0, res=0) wlan0: hfa384x_cmd: command was not completed (res=0, entry=dfe699a0, type=0, cmd=0x0001, param0=0x0000) wlan0: MAC port 0 enabling failed wlan0: could not enable MAC port wlan0: prism2_close wlan0: hfa384x_cmd: entry still in list? (entry=dfe7e320, type=0, res=0) wlan0: hfa384x_cmd: command was not completed (res=0, entry=dfe7e320, type=0, cmd=0x0021, param0=0xfd0b) wlan0: hfa384x_get_rid: CMDCODE_ACCESS failed (res=-110, rid=fd0b, len=8) Could not get RID for component NIC wlan0: PRI: id=0x15 v0.3.0 wlan0: STA: id=0x1f v1.4.9 wlan0: defaulting to bogus WDS frame as a workaround for firmware bug in Host AP mode WDS wlan0: LinkStatus=2 (Disconnected) -- Regards, Ryan Abbenhuys sneeze (at) alphalink.com.au rabbenhuys (at) mfbb.vic.gov.au ryana (at) igreen.net sneeze (at) igreen.net From matrix at spic.net Sun Jan 26 02:08:29 2003 From: matrix at spic.net (Francois Gurin) Date: Sun, 26 Jan 2003 02:08:29 -0500 Subject: [HostAP] Re: Linksys WPC11, can load drivers (manually) but doesnt create device wlan0. In-Reply-To: <005701c2c487$0187c7e0$f700a8c0@Home> References: <000801c2c3d8$0eaa76c0$5304000a@terraworld.net> <20030125034535.GD1351@jm.kir.nu> <005701c2c487$0187c7e0$f700a8c0@Home> Message-ID: <20030126070829.GS24359@shot.org> > > That sounds like your pcmcia-cs (actually, cardmgr) is configured to use > > prism2_cs driver. If you want to use Host AP driver with that card, you > > would need to change pcmcia-cs configuration for that card. Host AP > > driver installation should have already copied hostap_cs.conf that > > describes hostap_cs.o as the driver for Linksys WPC11. However, some > > other configuration file can override this configuration. You would thus > > need to comment out that other part refering to prism2_cs driver. > > Thanks Jouni, Jason... > > Jason: ...had tried that already, its in /etc/modules.conf. doesnt make any > difference... nope. it's actually in one of the configuration files in /etc/pcmcia/*. grep -l prism2_cs /etc/pcmcia/* will give you a list of files with the wrong setting. prism2_cs is what the module was called about a year ago. modifying the offending config files to hostap_cs should cardmgr loading the right driver. the erroneous entries could either be from a previously installed old version of hostap, or some wacky suse default. --francois From cijoml at volny.cz Sun Jan 26 10:10:06 2003 From: cijoml at volny.cz (Michal Semler) Date: Sun, 26 Jan 2003 16:10:06 +0100 Subject: flash with -f param Message-ID: Hello, I found 1.4.9 fw for my card and win utility - can I safe flash the hex with prism2srec??? firmware is here: http://prism.wz.cz/ Thanks for reply Michal From jkmaline at cc.hut.fi Sun Jan 26 11:48:27 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 26 Jan 2003 08:48:27 -0800 Subject: flash with -f param In-Reply-To: References: Message-ID: <20030126164827.GC1521@jm.kir.nu> On Sun, Jan 26, 2003 at 04:10:06PM +0100, Michal Semler wrote: > I found 1.4.9 fw for my card and win utility - can I safe flash the hex with > prism2srec??? prism2_srec supports '-f' option, but the current Host AP kernel driver does not include code for non-volatile downloads. So no, you cannot write the upgrade into flash using current Host AP driver. -- Jouni Malinen PGP id EFC895FA From harry at shoka.net Sun Jan 26 14:35:49 2003 From: harry at shoka.net (Harry Moyes) Date: Sun, 26 Jan 2003 19:35:49 +0000 Subject: Struggling with Pheenet card Message-ID: <200301261935.49239.harry@shoka.net> All Humble plea for assistance please. Trying to build a radio connected system. Hardware is IBM Pentium 200 box, with rtl8139 Ethernet and Pheenet WL-002 radio card. OS is stock RedHat 8.0, (make oldconfig) and rebuild custom Kernel. Hostap is last full release, not CVS. Obvious issues with symbol versioning etc overcome. /proc/pci is PCI devices found: Bus 0, device 0, function 0: Host bridge: Intel Corp. 430VX - 82437VX TVX [Triton VX] (rev 2). Master Capable. Latency=32. Bus 0, device 1, function 0: ISA bridge: Intel Corp. 82371SB PIIX3 ISA [Natoma/Triton II] (rev 1). Bus 0, device 1, function 1: IDE interface: Intel Corp. 82371SB PIIX3 IDE [Natoma/Triton II] (rev 0). Master Capable. Latency=32. I/O at 0xfff0 [0xffff]. Bus 0, device 1, function 2: USB Controller: Intel Corp. 82371SB PIIX3 USB [Natoma/Triton II] (rev 1). IRQ 15. Master Capable. Latency=64. I/O at 0x5400 [0x541f]. Bus 0, device 8, function 0: VGA compatible controller: Cirrus Logic GD 5446 (rev 0). Prefetchable 32 bit memory at 0x40000000 [0x40ffffff]. Bus 0, device 11, function 0: Network controller: PCI device 15e8:0131 (National Datacomm Corp) (rev 1). IRQ 15. I/O at 0x5424 [0x5433]. I/O at 0x1000 [0x103f]. Device 11 would appear to match PLXDEV(0x15e8, 0x0131, "National Datacomm NCP130 (TMD7160)"), in struct pci_device_id in hostap_plx.c Attempting to load the module gives ... [root at aichen shoka]# /sbin/insmod /lib/modules/2.4.18-14custom/net/hostap_crypt.o [root at aichen shoka]# /sbin/insmod /lib/modules/2.4.18-14custom/net/hostap.o [root at aichen shoka]# /sbin/insmod /lib/modules/2.4.18-14custom/net/hostap_plx.o /lib/modules/2.4.18-14custom/net/hostap_plx.o: init_module: No such device Hint: insmod errors can be caused by incorrect module parameters, including invalid IO or IRQ parameters. You may find more information in syslog or the output from dmesg Dmseg contains ... Linux version 2.4.18-14custom (root at aichen) (gcc version 3.2 20020903 (Red Hat Linux 8.0 3.2-7)) #10 Fri Jan 24 22:34:09 GMT 2003 BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 000000000009fc00 (usable) BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved) BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved) BIOS-e820: 0000000000100000 - 0000000004000000 (usable) BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved) BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved) BIOS-e820: 00000000fffe0000 - 0000000100000000 (reserved) 64MB LOWMEM available. On node 0 totalpages: 16384 zone(0): 4096 pages. zone(1): 12288 pages. zone(2): 0 pages. IBM machine detected. Enabling interrupts during APM calls. Kernel command line: ro root=LABEL=/ Initializing CPU#0 Detected 199.434 MHz processor. Speakup v-1.00 CVS: Tue Jun 11 14:22:53 EDT 2002 : initialized Console: colour VGA+ 80x25 Calibrating delay loop... 398.13 BogoMIPS Memory: 60972k/65536k available (1193k kernel code, 3664k reserved, 984k data, 200k init, 0k highmem) Dentry cache hash table entries: 8192 (order: 4, 65536 bytes) Inode cache hash table entries: 4096 (order: 3, 32768 bytes) Mount cache hash table entries: 1024 (order: 1, 8192 bytes) ramfs: mounted with options: ramfs: max_pages=7685 max_file_pages=0 max_inodes=0 max_dentries=7685 Buffer cache hash table entries: 4096 (order: 2, 16384 bytes) Page-cache hash table entries: 16384 (order: 4, 65536 bytes) CPU: Before vendor init, caps: 008001bf 00000000 00000000, vendor = 0 Intel Pentium with F0 0F bug - workaround enabled. CPU: After vendor init, caps: 008001bf 00000000 00000000 00000000 CPU: After generic, caps: 008001bf 00000000 00000000 00000000 CPU: Common caps: 008001bf 00000000 00000000 00000000 CPU: Intel Pentium MMX stepping 03 Checking 'hlt' instruction... OK. POSIX conformance testing by UNIFIX mtrr: v1.40 (20010327) Richard Gooch (rgooch at atnf.csiro.au) mtrr: detected mtrr type: none PCI: PCI BIOS revision 2.10 entry at 0xfd8bc, last bus=0 PCI: Using configuration type 1 PCI: Probing PCI hardware PCI: Cannot allocate resource region 2 of device 00:0b.0 Limiting direct PCI/PCI transfers. Activating ISA DMA hang workarounds. isapnp: Scanning for PnP cards... isapnp: No Plug & Play device found speakup: initialized device: /dev/synth, node (MAJOR 10, MINOR 25) Linux NET4.0 for Linux 2.4 Based upon Swansea University Computer Society NET3.039 Initializing RT netlink socket apm: BIOS version 1.2 Flags 0x03 (Driver version 1.16) Starting kswapd VFS: Diskquotas version dquot_6.5.0 initialized pty: 512 Unix98 ptys configured Serial driver version 5.05c (2001-07-08) with MANY_PORTS MULTIPORT SHARE_IRQ SERIAL_PCI ISAPNP enabled ttyS0 at 0x03f8 (irq = 4) is a 16550A ttyS1 at 0x02f8 (irq = 3) is a 16550A Real Time Clock Driver v1.10e block: 112 slots per queue, batch=28 Uniform Multi-Platform E-IDE driver Revision: 6.31 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx PIIX3: IDE controller on PCI bus 00 dev 09 PIIX3: chipset revision 0 PIIX3: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xfff0-0xfff7, BIOS settings: hda:DMA, hdb:pio hda: FUJITSU MPC3064AT, ATA DISK drive ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 blk: queue c0389c44, I/O limit 4095Mb (mask 0xffffffff) blk: queue c0389c44, I/O limit 4095Mb (mask 0xffffffff) hda: 12672450 sectors (6488 MB), CHS=788/255/63, (U)DMA ide-floppy driver 0.99.newide Partition check: hda: hda1 hda2 hda3 Floppy drive(s): fd0 is 1.44M FDC 0 is a National Semiconductor PC87306 NET4: Frame Diverter 0.46 RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize ide-floppy driver 0.99.newide md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27 md: Autodetecting RAID arrays. md: autorun ... md: ... autorun DONE. NET4: Linux TCP/IP 1.0 for NET4.0 IP Protocols: ICMP, UDP, TCP, IGMP IP: routing cache hash table of 512 buckets, 4Kbytes TCP: Hash tables configured (established 4096 bind 8192) Linux IP multicast router 0.06 plus PIM-SM NET4: Unix domain sockets 1.0/SMP for Linux NET4.0. RAMDISK: Compressed image found at block 0 Freeing initrd memory: 127k freed VFS: Mounted root (ext2 filesystem). Journalled Block Device driver loaded kjournald starting. Commit interval 5 seconds EXT3-fs: mounted filesystem with ordered data mode. Freeing unused kernel memory: 200k freed usb.c: registered new driver usbdevfs usb.c: registered new driver hub usb-uhci.c: $Revision: 1.275 $ time 00:41:16 Jan 24 2003 usb-uhci.c: High bandwidth mode enabled usb-uhci.c: USB UHCI at I/O 0x5400, IRQ 15 usb-uhci.c: Detected 2 ports usb.c: new USB bus registered, assigned bus number 1 hub.c: USB hub found hub.c: 2 ports detected usb-uhci.c: v1.275:USB Universal Host Controller Interface driver usb.c: registered new driver hiddev usb.c: registered new driver hid hid-core.c: v1.8.1 Andreas Gal, Vojtech Pavlik hid-core.c: USB HID support drivers mice: PS/2 mouse device common for all mice EXT3 FS 2.4-0.9.18, 14 May 2002 on ide0(3,3), internal journal Adding Swap: 514072k swap-space (priority -1) parport0: PC-style at 0x3bc [PCSPP] ohci1394: pci_module_init failed ip_tables: (C) 2000-2002 Netfilter core team 8139too Fast Ethernet driver 0.9.25 divert: allocating divert_blk for eth0 eth0: RealTek RTL8139 Fast Ethernet at 0xc48a1000, 00:e0:4c:39:87:fc, IRQ 11 eth0: Identified 8139 chip type 'RTL-8139C' eth0: Setting 100mbps full-duplex based on auto-negotiated partner ability 45e1.hostap_crypt: registered algorithm 'NULL' hostap_plx: hostap_plx.c 0.0.0 2002-10-12 (SSH Communications Security Corp, Jouni Malinen) hostap_plx: (c) Jouni Malinen TMD7160 PCI/PCMCIA adapter: io=0x5424, irq=15, pccard_io=0x1000 hostap_plx: Error setting COR (expected=0x45, was=0xff) hostap_plx: No devices found, driver not installed. Which would appear to be triggered here in hostap_plx.c .. if (tmd7160) { /* TMD7160 */ attr_mem = 0; /* no access to PC Card attribute memory */ printk(KERN_INFO "TMD7160 PCI/PCMCIA adapter: io=0x%x, " "irq=%d, pccard_io=0x%x\n", plx_ioaddr, pdev->irq, pccard_ioaddr); cor_offset = plx_ioaddr; cor_index = 0x04; outb(cor_index | COR_LEVLREQ | COR_ENABLE_FUNC, plx_ioaddr); mdelay(1); reg = inb(plx_ioaddr); if (reg != (cor_index | COR_LEVLREQ | COR_ENABLE_FUNC)) { printk(KERN_ERR "%s: Error setting COR (expected=" "0x%02x, was=0x%02x)\n", dev_info, cor_index | COR_LEVLREQ | COR_ENABLE_FUNC, reg); goto fail; } } else { Questions .. Has anybody experiance of specifically this variant prisam (??) card. Am I doing something obvious wrong.. (please presume I'm a complete dunce, we'll arrive at a reasonable assesment of my competancy much sooner that way ;-) ) Is this likely to be improved by moving to a current kernel and the CVS version of hostap, or does this imply an incompatability between hostap and this card? Cheers Harry Moyes From jkmaline at cc.hut.fi Sun Jan 26 13:54:34 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 26 Jan 2003 10:54:34 -0800 Subject: Struggling with Pheenet card In-Reply-To: <200301261935.49239.harry@shoka.net> References: <200301261935.49239.harry@shoka.net> Message-ID: <20030126185434.GA2974@jm.kir.nu> On Sun, Jan 26, 2003 at 07:35:49PM +0000, Harry Moyes wrote: > Pheenet WL-002 radio card. I have never used such card, but I hostap_plx has been reported to work with Pheecom WL-PCI, which has same PCI id as your card. > Bus 0, device 11, function 0: > Network controller: PCI device 15e8:0131 (National Datacomm Corp) (rev 1). > IRQ 15. > I/O at 0x5424 [0x5433]. > I/O at 0x1000 [0x103f]. Could you please send a bit more detailed info about this device ('lspci -vvd 15e8:0131')? > hostap_plx: (c) Jouni Malinen > TMD7160 PCI/PCMCIA adapter: io=0x5424, irq=15, pccard_io=0x1000 > hostap_plx: Error setting COR (expected=0x45, was=0xff) TMD7160 adapter was apparently found, but when the first control area was written to (COR register), the following read from the same register did not return the same value. I would assume that this indicates some problems with I/O, but one could test ignoring this by commenting out "goto fail;" line in the code snippet you quoted: > if (reg != (cor_index | COR_LEVLREQ | COR_ENABLE_FUNC)) { > printk(KERN_ERR "%s: Error setting COR (expected=" > "0x%02x, was=0x%02x)\n", dev_info, > cor_index | COR_LEVLREQ | COR_ENABLE_FUNC, > reg); > goto fail; > } > Am I doing something obvious wrong.. No. > Is this likely to be improved by moving to a current kernel and the CVS > version of hostap, or does this imply an incompatability between hostap and > this card? At least Host AP code has not changed on that part, so updating it should not produce any difference. -- Jouni Malinen PGP id EFC895FA From ammirata at econointl.com Sun Jan 26 17:18:21 2003 From: ammirata at econointl.com (Sergio M. Ammirata) Date: Sun, 26 Jan 2003 17:18:21 -0500 Subject: wds links Message-ID: <000301c2c588$d7138500$017ba8c0@ammirata.net> Does anyone have any suggestions on how to measure the strength of a wds link when using hostapd? This is feature is not yet implemented in the current cvs version. One idea is to use the wlan0sta to associate to each of the neighbors and record the results. Another one is to do a passive scan of all access points and filter out the ones that are not relevant. Before I embark in writing complex scripts I would like to know if anyone has other ideas. Thanks in advance, Sergio -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030126/023ae41f/attachment.htm From jkmaline at cc.hut.fi Sun Jan 26 23:02:40 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Sun, 26 Jan 2003 20:02:40 -0800 Subject: wds links In-Reply-To: <000301c2c588$d7138500$017ba8c0@ammirata.net> References: <000301c2c588$d7138500$017ba8c0@ammirata.net> Message-ID: <20030127040240.GA2461@jm.kir.nu> On Sun, Jan 26, 2003 at 05:18:21PM -0500, Sergio M. Ammirata wrote: > Does anyone have any suggestions on how to measure the strength of a wds > link when using hostapd? > > This is feature is not yet implemented in the current cvs version. As a temporary solution, you could try adding permanent station entry for each WDS link end-point with 'hostap_crypt_conf wlan0 none -p'. I haven't tested whether this really collects statistics, but it might. At least it will add /proc/net/hostap/wlan0/ entry. -- Jouni Malinen PGP id EFC895FA From dean-list-hostap at arctic.org Sun Jan 26 23:16:57 2003 From: dean-list-hostap at arctic.org (dean gaudet) Date: Sun, 26 Jan 2003 20:16:57 -0800 (PST) Subject: problems with EN2511CD / NL2511CD In-Reply-To: <20030113050625.GD28005@jm.kir.nu> References: <20021203062656.GA12480@jm.kir.nu> <20021204032243.GA1232@jm.kir.nu> <20030113050625.GD28005@jm.kir.nu> Message-ID: On Sun, 12 Jan 2003, Jouni Malinen wrote: > The capture log seemed to miss lot of messages. There was only one > beacon frame although there should have been about 10 beacons a second. > In addition, more or less all ACKs were missing. If these frames were ... so in the end i'm convinced this problem was related to the laptop i'm using as AP ... the netgate card has no troubles working as an AP in another laptop. the laptop in question has other odd pcmcia quirks as well, and i think i'm going to retire it as my gateway. thanks for your help though! -dean From dave at cc0.net Mon Jan 27 03:04:10 2003 From: dave at cc0.net (Dave) Date: Mon, 27 Jan 2003 00:04:10 -0800 Subject: WDS configuration issues Message-ID: <002401c2c5da$aeb84710$0500a8c0@ws7> Hi all, I have not successfully been able to get automatic WDS functioning no matter what version/configuration of hostap I use. I can get WDS to function by adding the MACs manually but I can only randomly see WDS links come up automatically using the instructions in README.prism2. All boxes are within the same room and within feet of each other. Could anyone help with the following questions so I can end this 4 day battle..;) 1) Which version of hostap works best for automated WDS? 2002-10-12 or CVS? and/or which version works best overall? 2) After adding wds links with iwpriv (iwpriv wlan0 wds_add 00:00:00:00:00:00), then adding the wds interfaces into a bridge (brctl addif br0 wlan0wds0, wlan0wds1, etc) and assigning 0.0.0.0 to each wds (ifconfig wlan0wds0 0.0.0.0 up) I will continually get syslog errors (Could not find STA for this TX error) I assume for the wds links that are not in use yet...how to avoid this and/or how to pre-allocate the wds links if I am doing it incorrectly? I recall a message on the mailing list where this problem was discussed but I cannot locate that message now or what the outcome was. 3) Will defining PRISM2_NO_STATION_MODES simply prevent the driver from ever acting as an STA to any other device or are there other ramifications of defining this? For the repeater AP's I don't see any reason why they would ever need to be a STA..unless this also affects wds links or something. My AP's only need to bridge between wlan0 and wlan0wds0/1/2 etc acting as an AP to clients connecting to the previous/next AP via wds. 4) Is it best to use the latest firmware (1.5.6) and to use 4 address WDS frames as opposed to the 3 address workaround? The current cards I use have 1.4.9 and I download 1.5.6 to the card on bootup. I do this right after the interface wlan0 is created...is there a better way to do it at a different time? 5) The following are the exact commands that are executed at boot time after the card is up and wlan0 is present. Does anyone see anything below that would prevent WDS or cause problems? This has been attempted using the current release (2002-10-12) or CVS snapshot, with and without hostapd, with and without PRISM2_NO_STATION_MODES. prism2_srec -r wlan0 /etc/pcmcia/rf010506.hex prism2_param wlan0 max_wds 16 prism2_param wlan0 wds_type 4 prism2_param wlan0 other_ap_policy 1 prism2_param wlan0 ap_bridge_packets 0 prism2_param wlan0 autom_ap_wds 1 prism2_param wlan0 ap_auth_algs 1 iwconfig wlan0 txpower 20 iwconfig wlan0 channel 5 iwpriv wlan0 wds_add 00:00:00:00:00:00 brctl addbr br0 brctl addif br0 wlan0 brctl addif br0 wlan0wds0 ifconfig wlan0 0.0.0.0 up ifconfig wlan0wds0 0.0.0.0 up ifconfig br0 10.10.1.5 netmask 255.255.255.0 broadcast 10.10.1.255 up Any help is GREATLY appreciated. The goal of this project is basically with one AP connected to a LAN via eth0 and providing wireless connectivity to the LAN via 20 or 30 repeating/bridging AP's thus (hopefully) extending coverage. Thanks! -Dave From yclin at iis.sinica.edu.tw Mon Jan 27 04:05:46 2003 From: yclin at iis.sinica.edu.tw (Lin Yin-chang) Date: Mon, 27 Jan 2003 17:05:46 +0800 Subject: Coexistence of tertiary firmware and 802.1X authentication Message-ID: <007501c2c5e3$6ad3d1f0$96126d8c@argay> Hi, I have successfully set up an AP with HostAP 2002-10-12 release and DWL-650. When tertiary firmware is downloaded to RAM on the wireless card, AP also works well. However, when I enable hostapd to support 802.1X authentication, the supplicant (running Odyssey client) cannot succesullty complete the process of authentication. Without downloading the firmware, 802.1X authentication works well. So far as I know, 802.1X authentication is processed at hostapd and subsequent per-session WEP encryption and decryption is done at hostap driver. Is it possilbe to move the job of per-session encryption to firmware? It not, does it mean that tertiary firmware cannot coexist with 802.1X authentication-supported hostapd? Any help will be highly appreciated. Thanks in advance. Best regards, Jack Lin From dave at cc0.net Mon Jan 27 05:29:35 2003 From: dave at cc0.net (Dave) Date: Mon, 27 Jan 2003 02:29:35 -0800 Subject: WDS configuration issues - Update (received WDS[4 addr] frame from unknown TA=00:02:6f:01:f9:b6) References: <002401c2c5da$aeb84710$0500a8c0@ws7> Message-ID: <000401c2c5ee$fd579920$0500a8c0@ws7> Still working on WDS... I am now seeing "wlan0: received WDS[4 addr] frame from unknown TA=00:02:6f:01:f9:b6" on one of the boxes (AP1) in dmesg as well as the expected "wlan0: Could not find STA for this TX error" The other box (AP2) promptly shows the following in dmesg: device wlan0 entered promiscuous mode wlan0: adding automatic WDS connection to AP 00:02:6f:01:fa:47 wlan0: using pre-allocated WDS netdevice wlan0wds0 device wlan0wds0 entered promiscuous mode and shows in /proc/net/hostap/wlan0/wds 00:02:6f:01:fa:47 so it appears to be working on AP2. I then took the EXACT same config from AP2 and placed on AP1 and still see the above WDS frame message and no wds shown in /proc Topology is as simple as could be. AP2 ~~ WDS (almost) ~~ AP1 -- eth0 Currently using hostap CVS with only PRISM2_DOWNLOAD_SUPPORT defined. (kernel is 2.4.20 with Wireless Tools version 26-pre7, WE-15 #6 and WE-16 #4) What in the world am I doing wrong? -Dave From kyler-keyword-hostap.67a4fa at lairds.org Sun Jan 26 14:49:32 2003 From: kyler-keyword-hostap.67a4fa at lairds.org (Kyler Laird) Date: Sun, 26 Jan 2003 14:49:32 -0500 Subject: Demarctech card locks machine with HostAP Message-ID: <20030126194932.GA1935@remote.lairds.org> I've got a few of the Demarctech ReliaWave 200mW cards that I'd hoped to use with HostAP. I finally got around to trying it yesterday and was surprised at how easy it was to install. Unfortunately, as soon as I insert one of these cards, my machine locks up hard. Panasonic Toughbook CF-71 Debian (unstable) Linux 2.4.20 kernel PCMCIA hostap-2002-10-12 It locks right after a message about registering wifi0. (I also tried last night's CVS snapshot, but it locked before even showing the registration message.) These cards worked (with some problems) using the Orinoco driver. Any advice on where to start poking? I've been trying to chew through all of the documentation I've found, but I'm not seeing any mention of lockup on insertion. Thank you. --kyler From mrcool at stupidgamerz.com Mon Jan 27 10:04:07 2003 From: mrcool at stupidgamerz.com (Jerry) Date: Mon, 27 Jan 2003 09:04:07 -0600 Subject: cant load hostap_cs automatically...an update. Message-ID: <001401c2c615$5728cde0$5304000a@terraworld.net> Well, thanks to Francois i can now load the drivers for Hostap automatically... but it still wont create wlan0. How can i do this manually (im not aginst doin that... ;) im gonna look on SuSE website for something on this, if i find something before you experts get around to helpin me, i will post how to do this... thanks everyone for your help. Jerryf -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030127/7bd3d54a/attachment.htm From alexey_23 at fromru.com Mon Jan 27 11:48:27 2003 From: alexey_23 at fromru.com (Alexey) Date: Mon, 27 Jan 2003 18:48:27 +0200 Subject: hostap in master mode and firmware 1.6.3 Message-ID: <20030127184827.28df9293.alexey_23@fromru.com> Hi all, We have PCMCIA PRISM 2.5 card with such ident: NIC ID=0x800c v1.0.0 PRI v1.1.0 STA v1.6.3 All work perfectly in Ad-hoc mode and in Managed mode with Lucent AP. But we can't get it worked in Master mode. When we turn on Master mode nothing happened. if we turn on debug with prism2_param debug 3 we can see only such lines in log: kernel: wlan0: TX status=0x0000 retry_count=0 tx_rate=0 tx_control=0x000c; jiffies=3047063 kernel: FC=0x0008 (type=2:0) dur=0x0000 seq=0x0000 data_len=36 kernel: A1=ff:ff:ff:ff:ff:ff A2=00:03:2f:00:99:e2 A3=02:03:9a:ed:99:e2 A4=00:00:00:00:00:00 kernel: dst=ff:ff:ff:ff:ff:ff src=00:03:2f:00:99:e2 len=36 ... Does it possible to run such type of card in Master mode? Any hints? Any information appricated. WBR, Alexey. From jasonb at edseek.com Mon Jan 27 14:12:05 2003 From: jasonb at edseek.com (Jason Boxman) Date: Mon, 27 Jan 2003 14:12:05 -0500 (EST) Subject: hostap in master mode and firmware 1.6.3 In-Reply-To: <20030127184827.28df9293.alexey_23@fromru.com> References: <20030127184827.28df9293.alexey_23@fromru.com> Message-ID: <1965.209.208.68.56.1043694725.squirrel@rebecca.edseek.com> Alexey said: > Hi all, Hi. > We have PCMCIA PRISM 2.5 card with such ident: > > NIC ID=0x800c v1.0.0 > PRI v1.1.0 > STA v1.6.3 While I don't have a solution, I am insatiably curious which card ships with STA 1.6.3? > All work perfectly in Ad-hoc mode and in Managed mode with Lucent AP. > But we can't get it worked in Master mode. When we turn on Master mode > nothing happened. > > Any information appricated. > > WBR, Alexey. > From linuxchuck at phoenixbox.org Mon Jan 27 20:57:49 2003 From: linuxchuck at phoenixbox.org (linuxchuck) Date: Mon, 27 Jan 2003 20:57:49 -0500 Subject: "original" WDT11/WPC11 combo for an AP? Message-ID: <200301280157.h0S1vnM2020252@phoenix.phoenixbox.org> Hello all. This is my first post, so I'll keep it as short as possible. At the risk of becoming flamebait for some post I probably missed in my searching... here's the problem: I've done a bit of research through the archives and haven't seen a reference to my hardware, so I thought I'd ask if there is anyone who has experience with it. I am using the first release of the WPC11 seated in a WDT11 and have had a "partial" success getting it to work as an AP with the Host_AP drivers. I have a laptop in the house with the same WPC11 pc card in the pcmcia slot. I can get good association, and signal strength, but cannot send any data. DHCP times out, and if I statically assign an IP, I still cannot ping to or past the AP. I have used the setup as described at http://trekweb.com/~jasonb/articles/hostap_20021012.shtml to the letter. I have two Ethernet connections, Eth0 is my internet connection, and Eth1 is my home network. I used the bridging technique from the same page. Here is the information from my ifconfig, iwconfig, and hostap_diag commands: *********hostap_diag wlan0********** Host AP driver diagnostics information for 'wlan0' NICID: id=0x8002 v1.0.0 (HWB3163-01,02,03,04 Rev A) PRIID: id=0x0015 v0.3.0 STAID: id=0x001f v0.8.0 (station firmware) **********iwconfig wlan0********** wlan0 IEEE 802.11-b ESSID:"test" Mode:Master Frequency:2.422GHz Access Point: 00:04:5A:CE:6D:C7 Bit Rate:11Mb/s Tx-Power:-4 dBm Sensitivity=1/3 Retry min limit:8 RTS thr:off Fragment thr:off Encryption key:off Power Management:off Link Quality:0 Signal level:0 Noise level:0 Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 **********ifconfig********** br0 Link encap:Ethernet HWaddr 00:04:5A:CE:6D:C7 inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:9723 errors:0 dropped:0 overruns:0 frame:0 TX packets:13716 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:654613 (639.2 Kb) TX bytes:15682965 (14.9 Mb) eth0 Link encap:Ethernet HWaddr 00:E0:29:24:9C:55 inet addr:68.47.118.231 Bcast:68.47.118.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:396352 errors:1 dropped:0 overruns:0 frame:0 TX packets:131138 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:46502816 (44.3 Mb) TX bytes:9530417 (9.0 Mb) Interrupt:10 Base address:0xd000 eth1 Link encap:Ethernet HWaddr 00:E0:29:24:9C:54 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:9727 errors:1 dropped:0 overruns:0 frame:0 TX packets:57955 errors:2 dropped:0 overruns:1 carrier:1 collisions:0 txqueuelen:100 RX bytes:790975 (772.4 Kb) TX bytes:18335851 (17.4 Mb) Interrupt:11 Base address:0xf000 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:986 errors:0 dropped:0 overruns:0 frame:0 TX packets:986 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:2734581 (2.6 Mb) TX bytes:2734581 (2.6 Mb) wlan0 Link encap:Ethernet HWaddr 00:04:5A:CE:6D:C7 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:45616 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:0 (0.0 b) TX bytes:3841942 (3.6 Mb) Interrupt:14 Base address:0xc000 My initial suspicion is that perhaps the age/firmware version I am running might be the problem. If that is the case, from my readings, I could not find a set of firmware that was "known good" and I could use with my specific card. If anyone has had any success with this particular setup, I'd appreciate any guidance you can provide. Thanks in advance... Charles -- For a copy of my public key, send an email with "send pgp key" in the subject. You really can't judge Windows until Microsoft comes out with a stable release. From jkmaline at cc.hut.fi Mon Jan 27 20:58:44 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 27 Jan 2003 17:58:44 -0800 Subject: hostap in master mode and firmware 1.6.3 In-Reply-To: <20030127184827.28df9293.alexey_23@fromru.com> References: <20030127184827.28df9293.alexey_23@fromru.com> Message-ID: <20030128015844.GB1320@jm.kir.nu> On Mon, Jan 27, 2003 at 06:48:27PM +0200, Alexey wrote: > STA v1.6.3 > > All work perfectly in Ad-hoc mode and in Managed mode with Lucent AP. > But we can't get it worked in Master mode. When we turn on Master mode nothing happened. I do not have v1.6.3 station firmware, so it is a bit difficult to guess what might have changed. It would be easier to debug any problems should I get the same version, and even more easier, if it would come with change log ;-). > if we turn on debug with prism2_param debug 3 > we can see only such lines in log: > > kernel: wlan0: TX status=0x0000 retry_count=0 tx_rate=0 tx_control=0x000c; jiffies=3047063 > kernel: FC=0x0008 (type=2:0) dur=0x0000 seq=0x0000 data_len=36 > kernel: A1=ff:ff:ff:ff:ff:ff A2=00:03:2f:00:99:e2 A3=02:03:9a:ed:99:e2 A4=00:00:00:00:00:00 > kernel: dst=ff:ff:ff:ff:ff:ff src=00:03:2f:00:99:e2 len=36 Is 02:03:9a:ed:99:e2 the MAC address of the card? If not, this looks odd. If it is, this might be OK. > Does it possible to run such type of card in Master mode? Any hints? Hard to say without debugging this a bit more. Could you please send me output of "utils/hostap_diag -ru wlan0"? If that looks OK, I would try to first check whether the card is sending beacon frames or not. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Mon Jan 27 21:01:49 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Mon, 27 Jan 2003 18:01:49 -0800 Subject: Coexistence of tertiary firmware and 802.1X authentication In-Reply-To: <007501c2c5e3$6ad3d1f0$96126d8c@argay> References: <007501c2c5e3$6ad3d1f0$96126d8c@argay> Message-ID: <20030128020149.GC1320@jm.kir.nu> On Mon, Jan 27, 2003 at 05:05:46PM +0800, Lin Yin-chang wrote: > So far as I know, 802.1X authentication is processed at hostapd and > subsequent per-session WEP encryption and decryption is done at hostap > driver. Is it possilbe to move the job of per-session encryption to > firmware? It not, does it mean that tertiary firmware cannot coexist with > 802.1X authentication-supported hostapd? I would guess that it is possible to use tertiary firmware with 802.1X, but I have not implemented this into Host AP driver or hostapd. After all, this is Host AP driver, not "tertiary AP driver" ;-). -- Jouni Malinen PGP id EFC895FA From imp at airlinktek.com Mon Jan 27 21:09:33 2003 From: imp at airlinktek.com (=?utf-7?B?K3ZCWFhaY1gwLQ==?=) Date: Tue, 28 Jan 2003 11:09:33 +0900 Subject: Where can I got the latest firmware ? Message-ID: Dear all... Could you tell me where can I got the latest firmware for PRISM 2, 2.5 chipset ? I'd very hard to find through internet searching... Thanks... From ammirata at econointl.com Tue Jan 28 00:31:06 2003 From: ammirata at econointl.com (Sergio M. Ammirata) Date: Tue, 28 Jan 2003 00:31:06 -0500 Subject: wds links In-Reply-To: <20030127040240.GA2461@jm.kir.nu> Message-ID: <000001c2c68e$7591a490$017ba8c0@ammirata.net> It created the entry in /proc/net/hostap/wlan0/ and it even shows up in iwspy ... but it does not collect any statistics :( Any other ideas? Sergio > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] On Behalf Of > Jouni Malinen > Sent: Sunday, January 26, 2003 11:03 PM > To: hostap at shmoo.com > Subject: Re: wds links > > On Sun, Jan 26, 2003 at 05:18:21PM -0500, Sergio M. Ammirata wrote: > > > Does anyone have any suggestions on how to measure the strength of a wds > > link when using hostapd? > > > > This is feature is not yet implemented in the current cvs version. > > As a temporary solution, you could try adding permanent station entry > for each WDS link end-point with 'hostap_crypt_conf wlan0 none > -p'. I haven't tested whether this really collects statistics, but it > might. At least it will add /proc/net/hostap/wlan0/ entry. > > -- > Jouni Malinen PGP id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap From greve at gnu.org Tue Jan 28 06:13:57 2003 From: greve at gnu.org (Georg C. F. Greve) Date: Tue, 28 Jan 2003 12:13:57 +0100 Subject: oddness/problem with hostap-2002-10-12 Message-ID: Hi, I have been trying kernel 2.4.21-pre3 with hostap-2002-10-12 on Debian with wireless-tools 25-2 (so v15 of the wireless extensions all around). The driver seems to load fine, but when I issue some iwconfig commands, like iwconfig wlan0 essid "TEST" I get a Error for wireless request "Set ESSID" (8B1A) : SET failed on device wlan0 ; Invalid argument. although it in fact *does* work, at least iwconfig wlan0 gives me wlan0 IEEE 802.11-DS ESSID:"TEST" Mode:Master Tx-Power:15 dBm Encryption key:off Link Quality:0 Signal level:0 Noise level:0 Rx invalid nwid:0 Rx invalid crypt:0 Rx invalid frag:0 Tx excessive retries:0 Invalid misc:0 Missed beacon:0 (works the same with any other ESSID). That seems odd and causes scripts to fail (obviously), so it is somewhat annoying. I couldn't find any reference to this on the web, so I thought I'd ask here. Regards, Georg P.S. The card is a Avaya/Lucent Gold PRISM card. P.P.S. Does anyone know a good "how to set up an AP" howto? I get things working, sometimes it would be useful to read a little about how it would be done best, though. But all I could find describes how to set up the driver, but doesn't talk about the best configurations... -- Georg C. F. Greve Free Software Foundation Europe (http://fsfeurope.org) Brave GNU World (http://brave-gnu-world.org) -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 258 bytes Desc: not available Url : http://lists.shmoo.com/pipermail/hostap/attachments/20030128/344e4fe0/attachment.pgp From jasonb at edseek.com Tue Jan 28 01:51:48 2003 From: jasonb at edseek.com (Jason Boxman) Date: Tue, 28 Jan 2003 01:51:48 -0500 Subject: "original" WDT11/WPC11 combo for an AP? In-Reply-To: <200301280157.h0S1vnM2020252@phoenix.phoenixbox.org> References: <200301280157.h0S1vnM2020252@phoenix.phoenixbox.org> Message-ID: <200301280151.48324.jasonb@edseek.com> On Monday 27 January 2003 08:57 pm, linuxchuck wrote: > Hello all. This is my first post, so I'll keep it as short as possible. > At the risk of becoming flamebait for some post I probably missed in my > searching... here's the problem: You've included plenty of useful details, I think, and that should help in resolving your issue. > I've done a bit of research through the archives and haven't seen a > reference to my hardware, so I thought I'd ask if there is anyone who has > experience with it. > > I am using the first release of the WPC11 seated in a WDT11 and have had a > "partial" success getting it to work as an AP with the Host_AP drivers. I First release, as in the WPC11v1? > have a laptop in the house with the same WPC11 pc card in the pcmcia slot. > I can get good association, and signal strength, but cannot send any data. > DHCP times out, and if I statically assign an IP, I still cannot ping to or > past the AP. I have used the setup as described at > http://trekweb.com/~jasonb/articles/hostap_20021012.shtml to the letter. I > have two Ethernet connections, Eth0 is my internet connection, and Eth1 is > my home network. I used the bridging technique from the same page. There are a few things you can do. First, I'd simplify the equation. Forget about getting 802.1d ethernet bridging working, for the moment. I would setup the WPC11 in the WDT11 in HostAP mode with a static address and setup the other station in managed mode with a static address in the same subnet. Then, see if you can ping yourself. If that works, see if you can ping the AP. If so, there's a bridging issue floating around. If not, then something else is going on. You said you were getting a good association and signal, so I'd imagine the AP is sending out beacon frames properly. > Here is the information from my ifconfig, iwconfig, and hostap_diag > commands: > > *********hostap_diag wlan0********** > Host AP driver diagnostics information for 'wlan0' > > NICID: id=0x8002 v1.0.0 (HWB3163-01,02,03,04 Rev A) > PRIID: id=0x0015 v0.3.0 > STAID: id=0x001f v0.8.0 (station firmware) As you mention later, the STA is suspiciously low. I don't recall if INFDROPs are fatal, however. Is there anything interesting happening in your syslog on the HostAP box during or after association with your wireless station? > **********iwconfig wlan0********** > wlan0 IEEE 802.11-b ESSID:"test" > **********ifconfig********** > br0 Link encap:Ethernet HWaddr 00:04:5A:CE:6D:C7 > inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:9723 errors:0 dropped:0 overruns:0 frame:0 > TX packets:13716 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:0 > RX bytes:654613 (639.2 Kb) TX bytes:15682965 (14.9 Mb) > > > eth1 Link encap:Ethernet HWaddr 00:E0:29:24:9C:54 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:9727 errors:1 dropped:0 overruns:0 frame:0 > TX packets:57955 errors:2 dropped:0 overruns:1 carrier:1 > collisions:0 txqueuelen:100 > RX bytes:790975 (772.4 Kb) TX bytes:18335851 (17.4 Mb) > Interrupt:11 Base address:0xf000 > > > wlan0 Link encap:Ethernet HWaddr 00:04:5A:CE:6D:C7 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:0 errors:0 dropped:0 overruns:0 frame:0 > TX packets:45616 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:100 > RX bytes:0 (0.0 b) TX bytes:3841942 (3.6 Mb) > Interrupt:14 Base address:0xc000 > > My initial suspicion is that perhaps the age/firmware version I am running > might be the problem. If that is the case, from my readings, I could not > find a set of firmware that was "known good" and I could use with my > specific card. If anyone has had any success with this particular setup, > I'd appreciate any guidance you can provide. Thanks in advance... First, I'd see if it works without bridging. If not, you could investigate your syslog (if you haven't already) for hints. You might also try doing an voliatile firmware update. You will have to reflash when the power cycles, but if it gets you up, it's worth it. Hopefully others will have suggestions as well. I don't think you mentioned what driver you're using on the client station. HostAP in managed mode I assume? What's the diagnostic output for that system? > Charles > -- > For a copy of my public key, send an email > with "send pgp key" in the subject. > -- Jason Boxman Administrator TrekWeb.COM http://trekweb.com/~jasonb/articles/ - Linux Pages From martin.whitlock at possio.com Tue Jan 28 11:27:09 2003 From: martin.whitlock at possio.com (Martin Whitlock) Date: Tue, 28 Jan 2003 17:27:09 +0100 Subject: hostap in master mode and firmware 1.6.3 In-Reply-To: <20030128015844.GB1320@jm.kir.nu> Message-ID: First of all I think that you should use PRI 1.1.1 together with STA 1.6.3. But still, it doesn't seem to work in HostAP mode. No stations can associate. I have verified with a spectrum analyzer that some kind of beacon is being sent periodically, and the transmitted frequency responds when I change channel with iwconfig. I suppose I should be able to capture the frames by putting another card in monitor mode, but I actually don't know exactly how to. Any hints? This is dmesg output from AP side: ******************************************** wlan0: RX status=0x0000 (port=0, type=0, fcserr=0) silence=10 signal=43 rate=10 rxflow=0; jiffies=97746 FC=0x0080 (type=0:8) dur=0x0000 seq=0xb200 data_len=40 A1=ff:ff:ff:ff:ff:ff A2=00:60:b3:6f:8b:56 A3=00:60:b3:6f:8b:56 A4=67:23:78:71 :74:aa dst=ff:ff:ff:ff:ff:ff src=00:50:ba:e7:75:67 len=40 wlan0: prism2_rx: dropped non-data frame (type=0x00, subtype=0x08) wlan0: RX status=0x0000 (port=0, type=0, fcserr=0) silence=14 signal=21 rate=10 rxflow=0; jiffies=97747 FC=0x0080 (type=0:8) dur=0x0000 seq=0xb850 data_len=37 A1=ff:ff:ff:ff:ff:ff A2=00:05:4e:81:04:6a A3=00:05:4e:81:04:6a A4=95:e3:ab:64 :6b:ac dst=ff:ff:ff:ff:ff:ff src=00:02:2d:27:4c:26 len=37 wlan0: prism2_rx: dropped non-data frame (type=0x00, subtype=0x08) wlan0: RX status=0x0000 (port=0, type=0, fcserr=0) silence=12 signal=26 rate=10 rxflow=0; jiffies=97748 FC=0x0080 (type=0:8) dur=0x0000 seq=0x94b0 data_len=35 A1=ff:ff:ff:ff:ff:ff A2=00:05:4e:81:02:a9 A3=00:05:4e:81:02:a9 A4=d1:c7:d6:b0 :64:42 dst=48:05:c2:a1:57:3a src=b0:46:9f:f6:b1:86 len=35 ******************************************** Funny thing is that none of these MAC-addresses seem familiar to me. Exactly what should these different address fields include? BTW, new features in STA 1.6.3 seems to be RIDs for Enhanced Security, i.e. hiding SSID in beacon and block responces to "ANY"-clients. "Frames to stations in promiscous mode will now be ACKed", but I havn't verified that yet. Otherwise it's mostly PS stuff. /Martin > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com]On Behalf Of > Jouni Malinen > Sent: den 28 januari 2003 02:59 > To: hostap at shmoo.com > Subject: Re: hostap in master mode and firmware 1.6.3 > > > On Mon, Jan 27, 2003 at 06:48:27PM +0200, Alexey wrote: > > > STA v1.6.3 > > > > All work perfectly in Ad-hoc mode and in Managed mode with Lucent AP. > > But we can't get it worked in Master mode. When we turn on > Master mode nothing happened. > > I do not have v1.6.3 station firmware, so it is a bit difficult to guess > what might have changed. It would be easier to debug any problems should > I get the same version, and even more easier, if it would come with > change log ;-). > > > if we turn on debug with prism2_param debug 3 > > we can see only such lines in log: > > > > kernel: wlan0: TX status=0x0000 retry_count=0 tx_rate=0 > tx_control=0x000c; jiffies=3047063 > > kernel: FC=0x0008 (type=2:0) dur=0x0000 seq=0x0000 data_len=36 > > kernel: A1=ff:ff:ff:ff:ff:ff A2=00:03:2f:00:99:e2 > A3=02:03:9a:ed:99:e2 A4=00:00:00:00:00:00 > > kernel: dst=ff:ff:ff:ff:ff:ff src=00:03:2f:00:99:e2 len=36 > > Is 02:03:9a:ed:99:e2 the MAC address of the card? If not, this looks > odd. If it is, this might be OK. > > > Does it possible to run such type of card in Master mode? Any hints? > > Hard to say without debugging this a bit more. Could you please send me > output of "utils/hostap_diag -ru wlan0"? If that looks OK, I would > try to first check whether the card is sending beacon frames or not. > > -- > Jouni Malinen PGP id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From jt at bougret.hpl.hp.com Tue Jan 28 12:55:26 2003 From: jt at bougret.hpl.hp.com (Jean Tourrilhes) Date: Tue, 28 Jan 2003 09:55:26 -0800 Subject: oddness/problem with hostap-2002-10-12 Message-ID: <20030128175526.GA16028@bougret.hpl.hp.com> Georg C. F. Greve wrote : > > P.S. The card is a Avaya/Lucent Gold PRISM card. This card won't work with the HostAP driver. Check Howto for details. Regards, Jean From jkim at ndosatech.com Tue Jan 28 19:54:17 2003 From: jkim at ndosatech.com (Jihong Kim) Date: Tue, 28 Jan 2003 19:54:17 -0500 Subject: Performance degradation with WEP enabled Message-ID: <000001c2c730$f396d020$0a01a8c0@Jihong> Hi, I'm using hostap-2002-10-12.tar.gz . I tested the throughput with the host AP (on Redhat Linux 7.2). With open system (no encryption), I got about 4.7 Mbps (GUI, daemons/processes are turned off). When they were on, the throughput was about 4 Mbps. With WEP (encryption) enabled, the highest I got was about 2.7 Mbps, when GUI and other daemons/processes were turned off. (With GUI and other daemons/processes running in the background, the throughput was about 600 Kbps.) I believe this is due to the fact that WEP encryption implemented in the device driver is used instead of the WEP encryption in PRISM MAC chip. Is there a way to improve the throughput of WLAN with WEP enabled? Do you think using a more powerful processor (CPU) will reduce the performance gap between open system and WEP? Thanks in advance. - Jihong -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030128/fc03aacc/attachment.htm From matthew at tedder.com Tue Jan 28 20:56:59 2003 From: matthew at tedder.com (Matthew Tedder) Date: Tue, 28 Jan 2003 17:56:59 -0800 Subject: Packages Needed Message-ID: <200301281756.59988.matthew@tedder.com> Hi, I would to setup 802.11B wireless access in my Open Source Cyber Cafe--what packages do I actually need? I have a D-Link Air DWL-520 card. I also DO NOT have a backup server--all the clients are X terminals. I've been running this way for a couple of months, with about 20 terminals, an 850Mhz AMD board, 512MB of RAM, and 20GB of disk space and CPU usage stays at about 1/4th capacity most of the time, during a full load. So I think adding this might not hurt much, right? -- Matthew C. Tedder SimpFlex Technologies, Inc. From travisr at rave.iinet.net.au Tue Jan 28 22:52:57 2003 From: travisr at rave.iinet.net.au (Travis Read) Date: Wed, 29 Jan 2003 11:52:57 +0800 (WST) Subject: prism2 routing table issues Message-ID: Greetings hostap'ers, I'm having a really weird issue with the hostap driver. I currently have a debian woody box with kernel 2.2.23 (custom compiled) + wavelan IEEE ISA cradle + orinoco silver pcmcica card. I also have a compaq wl200 with prism2 chipset. I'm finding the the hostap driver and the orinoco driver both work fine for a while. But after an unknown time period the route table goes away: cs: IO port probe 0x0a00-0x0aff: clean. cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff hostap_cs: index 0x01: Vcc 3.3, irq 9, io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 prism2_hw_init() prism2_hw_config: initialized in 9442 iterations wlan0: NIC: id=0x8002 v1.0.1 wlan0: PRI: id=0x15 v0.3.0 wlan0: STA: id=0x1f v0.8.0 Using data::nullfunc ACK workaround - firmware upgrade recommended wlan0: INFO - fid=0x01bc - len=2 type=0xf200 LinkStatus=2 (Disconnected) wlan0: INFO - fid=0x01bd - len=2 type=0xf200 LinkStatus=2 (Disconnected) wlan0: INFO - fid=0x01be - len=2 type=0xf200 LinkStatus=2 (Disconnected) Setting ALC off Setting TX power to 15 dBm wlan0: INFO - fid=0x01bf - len=2 type=0xf200 LinkStatus=2 (Disconnected) hermes.c: 5 Apr 2002 David Gibson orinoco.c 0.11b (David Gibson and others) orinoco_cs.c 0.11b (David Gibson and others) eth0: Initial media type MII. eth0: MII #24 status 786f, link partner capability 45e1, setting full-duplex. eth1: Station identity 001f:0001:0004:0034 eth1: Looks like a Lucent/Agere firmware version 4.52 eth1: Ad-hoc demo mode supported eth1: WEP supported, 104-bit key eth1: MAC address 00:60:1D:1B:F9:AD eth1: Station name "HERMES I" eth1: ready eth1: index 0x01: Vcc 5.0, irq 3, io 0x0180-0x01bf wlan0: prism2_open wlan0: dropped received packet from 00:60:1d:1b:f9:ad with no ToDS flag (type=0x02, subtype=0x00) wlan0: RX status=0x2000 (port=0, type=1, fcserr=0) silence=0 signal=81 rate=20 rxflow=0; jiffies=1715 FC=0x0008 (type=2:0) dur=0x0000 seq=0x0000 data_len=68 A1=ff:ff:ff:ff:ff:ff A2=00:60:1d:1b:f9:ad A3=00:00:00:00:00:00 A4=e2:1b:a9:af:05:20 dst=ff:ff:ff:ff:ff:ff src=00:60:1d:1b:f9:ad len=68 wlan0: prism2_close after it ways prism2_close the entire route table for pcmcica devices goes away and I'm left with: wavelan:~# route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.10.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 in /etc/init.d/pcmcia I have: PCIC_OPTS="irq_mode=1 fast_pci=1 extra_sockets=1 poll_interval=100" What am I missing? wavelan:~# cat /proc/interrupts CPU0 0: 34717 XT-PIC timer 1: 2 XT-PIC keyboard 2: 0 XT-PIC cascade 3: 15 XT-PIC orinoco_cs 9: 5 XT-PIC i82365, hostap_cs 11: 517 XT-PIC eth0 13: 1 XT-PIC fpu 14: 1613 XT-PIC ide0 NMI: 0 The beginning of dmesg is: Linux PCMCIA Card Services 3.2.3 kernel build: 2.2.23 #1 Tue Jan 28 20:33:54 HKT 2003 options: [pci] [cardbus] PCI routing table version 1.0 at 0xf7560 Intel ISA/PCI/CardBus PCIC probe: Cirrus PD6729 rev 00 PCI-to-PCMCIA at slot 00:0a, port 0xdff0 host opts [0]: [ring] [pci irq 9] [1/5/0] [1/20/0] host opts [1]: [ring] [pci irq 9] [1/5/0] [1/20/0] PCI card interrupts, PCI status changes Vadem VG-469 rev 00 ISA-to-PCMCIA at port 0x3e2 ofs 0x00 host opts [0]: none host opts [1]: none ISA irqs (scanned) = 3,4,7,15 polling interval = 1000 ms cs: memory probe 0x0d0000-0x0dffff: clean. hostap_cs: hostap_cs.c 0.0.0 2002-05-19 (SSH Communications Security Corp, Jouni Malinen) hostap_cs: (c) SSH Communications Security Corp hostap_cs: setting Vcc=33 (constant) hostap_cs: CS_EVENT_CARD_INSERTION prism2_config() hostap_cs: setting Vcc=33 (from config) CISTPL_MANFID: 0x0138, 0x0002 Checking CFTABLE_ENTRY 0x01 (default 0x01) Config has no IRQ info, but trying to enable IRQ anyway.. IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 io->flags = 0x0046, io.base=0x0000, len=64 cs: IO port probe 0x0100-0x04ff: excluding 0x170-0x177 0x370-0x37f 0x4d0-0x4d7 cs: IO port probe 0x0178-0x036f: clean. cs: IO port probe 0x0380-0x04cf: clean. cs: IO port probe 0x04d8-0x04ff: clean. cs: IO port probe 0x0800-0x08ff: clean. cs: IO port probe 0x0a00-0x0aff: clean. cs: IO port probe 0x0c00-0x0cff: excluding 0xcf8-0xcff hostap_cs: index 0x01: Vcc 3.3, irq 9, io 0x0100-0x013f hostap_cs: Registered netdevice wlan0 prism2_hw_init() prism2_hw_config: initialized in 9444 iterations wlan0: NIC: id=0x8002 v1.0.1 wlan0: PRI: id=0x15 v0.3.0 wlan0: STA: id=0x1f v0.8.0 Using data::nullfunc ACK workaround - firmware upgrade recommended I've tired the current version of hostap, the version before that, and the version before that. I've used pcmcia-cs version pcmcia-cs-3.2.3, pcmcia-cs-3.2.2 and pcmcia-cs-3.1.34. All do the same thing!?!?!? I'm not 100% sure this is a problem with the hostap driver but I'm still lost and can't figure out whats broken. What am I missing? Is it the firmware? Kind regards, ------------------------------------------------------------------------------ Travis Read iiNet Senior Support | Ph +61 8 9214 2222 Fx +61 8 9214 2211 travisr at staff.iinet.net.au | Level 6, Durack House, 263 Adelaide Terrace ------------------------------------------------------------------------------ " there is a war going on, it's not about who has the most bullets, it's about who controls the information " - SNEAKERS From jkmaline at cc.hut.fi Tue Jan 28 23:10:52 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Tue, 28 Jan 2003 20:10:52 -0800 Subject: Performance degradation with WEP enabled In-Reply-To: <000001c2c730$f396d020$0a01a8c0@Jihong> References: <000001c2c730$f396d020$0a01a8c0@Jihong> Message-ID: <20030129041052.GB1472@jm.kir.nu> On Tue, Jan 28, 2003 at 07:54:17PM -0500, Jihong Kim wrote: > With open system (no encryption), I got about 4.7 Mbps (GUI, > daemons/processes are turned off). When they were on, the throughput > was about 4 Mbps. > > With WEP (encryption) enabled, the highest I got was about 2.7 Mbps, > when GUI and other daemons/processes were turned off. (With GUI and > other daemons/processes running in the background, the throughput was > about 600 Kbps.) Few questions.. ;-) Which Prism2/2.5/3 and which station firmware version are you using? What about host CPU? Did you use host-based encryption and/or decryption? Which operation mode (Master?) were you using? Which card was used as the other part of the test and which driver? Have you seen the other card with good performance when using WEP? > I believe this is due to the fact that WEP encryption implemented in the > device driver is used instead of the WEP encryption in PRISM MAC chip. This can be the case, if the host system has very low-performance CPU. However, WEP is not very processor intensive at throughputs provided by IEEE 802.11b. You should also note that the Host AP driver defaults to hardware-based decryption (and also encryption, if STA firmware is version 1.4.9 or newer), so hardware WEP performance might also be an issue. At least some of the older Prism2 cards used to have poor WEP performance (I have seen results in 2-3 Mbps range when using hardware WEP). > Is there a way to improve the throughput of WLAN with WEP enabled? Do > you think using a more powerful processor (CPU) will reduce the > performance gap between open system and WEP? I have seen near 6 Mbps throughput with WEP, both using hardware encrypt/decrypt and doing encrypt/decrypt fully in the driver code. So yes, the throughput you are seeing can certainly be improved. The first thing would be to find out which component is the real bottleneck. If it is indeed host CPU, there are two options that could be used to improve situation. You could either use faster host CPU or try using hardware WEP by upgrading the wlan card firmware to at least STA f/w 1.4.9. -- Jouni Malinen PGP id EFC895FA From jkmaline at cc.hut.fi Tue Jan 28 23:31:38 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Tue, 28 Jan 2003 20:31:38 -0800 Subject: hostap in master mode and firmware 1.6.3 In-Reply-To: References: <20030128015844.GB1320@jm.kir.nu> Message-ID: <20030129043138.GC1472@jm.kir.nu> On Tue, Jan 28, 2003 at 05:27:09PM +0100, Martin Whitlock wrote: > First of all I think that you should use PRI 1.1.1 together with STA 1.6.3. > But still, it doesn't seem to work in HostAP mode. No stations can > associate. I have verified with a spectrum analyzer that some kind of beacon > is being sent periodically, and the transmitted frequency responds when I > change channel with iwconfig. I suppose I should be able to capture the > frames by putting another card in monitor mode, but I actually don't know > exactly how to. Any hints? Yes, getting a capture log should certainly help since I cannot test this myself at the moment. You can use Host AP driver in another host to do this capturing. Configure the sniffer on the same channel with the AP and set it to monitor mode with 'iwconfig wlan0 mode monitor' (or 'iwpriv wlan0 monitor 2' if your wireless tools do not understand new 'monitor' mode option). You will also need to set wlan0 device UP on the sniffer host. After this, you should be receiving the sniffed frames with 802.11 headers from wlan0 interface. You can make a capture log, e.g., using the latest version of libpcap with Ethereal or latest (CVS?) version of tcpdump. Alternatively, you could use wlansniff program in the Host AP package, but I prefer to get capture logs in some libpcap-compatible raw mode that I can load into Ethereal. > wlan0: RX status=0x0000 (port=0, type=0, fcserr=0) silence=10 signal=43 > rate=10 > rxflow=0; jiffies=97746 > FC=0x0080 (type=0:8) dur=0x0000 seq=0xb200 data_len=40 > A1=ff:ff:ff:ff:ff:ff A2=00:60:b3:6f:8b:56 A3=00:60:b3:6f:8b:56 > A4=67:23:78:71 > :74:aa > dst=ff:ff:ff:ff:ff:ff src=00:50:ba:e7:75:67 len=40 This is a beacon frame from 00:60:b3:6f:8b:56. type=0 (management) frames use addr1 as the destination, addr2 as the source, and addr3 as the BSSID. You can ignore A4, dst, and src for management frames. > wlan0: prism2_rx: dropped non-data frame (type=0x00, subtype=0x08) Aah.. Now this is interesting. The management frame was not catched in the RX path.. That would certainly break AP functionality. Now I would like to test that even more myself ;-). Which card manufacturer is using these new STA firmware images? As a semi-blind guess, could you please try the attached patch (to the current CVS version) with the new station firmware and report if there are any changes in Host AP mode behavior? > BTW, new features in STA 1.6.3 seems to be RIDs for Enhanced Security, i.e. > hiding SSID in beacon and block responces to "ANY"-clients. "Frames to > stations in promiscous mode will now be ACKed", but I havn't verified that > yet. Otherwise it's mostly PS stuff. Is that EnhSec RID also for Host AP mode now? It has been there long for tertiary firmware AP.. Fixing ACKs is always welcome, they are kind of needed in 802.11 ;-). -- Jouni Malinen PGP id EFC895FA -------------- next part -------------- Index: driver/modules/hostap_hw.c =================================================================== RCS file: /home/jm/cvsroot/hostap/driver/modules/hostap_hw.c,v retrieving revision 1.71 diff -u -p -r1.71 hostap_hw.c --- driver/modules/hostap_hw.c 11 Jan 2003 06:04:58 -0000 1.71 +++ driver/modules/hostap_hw.c 29 Jan 2003 04:29:37 -0000 @@ -2784,7 +2784,7 @@ static void hostap_rx_skb(local_info_t * if (hostap_rx_frame_invalid(dev, macport, status, len)) goto rx_dropped; - if (msg_type == HFA384X_RX_MSGTYPE_MGMT) { + if (msg_type == HFA384X_RX_MSGTYPE_MGMT || type == WLAN_FC_TYPE_MGMT) { if (hostap_rx_frame_mgmt(local, skb, type, stype)) goto rx_dropped; else From jkmaline at cc.hut.fi Wed Jan 29 00:02:15 2003 From: jkmaline at cc.hut.fi (Jouni Malinen) Date: Tue, 28 Jan 2003 21:02:15 -0800 Subject: prism2 routing table issues In-Reply-To: References: Message-ID: <20030129050215.GA1327@jm.kir.nu> On Wed, Jan 29, 2003 at 11:52:57AM +0800, Travis Read wrote: > I'm finding the the hostap driver and the orinoco driver both work fine > for a while. But after an unknown time period the route table goes away: > eth1: MAC address 00:60:1D:1B:F9:AD > wlan0: dropped received packet from 00:60:1d:1b:f9:ad with no ToDS flag > (type=0x02, subtype=0x00) > wlan0: RX status=0x2000 (port=0, type=1, fcserr=0) silence=0 signal=81 > rate=20 rxflow=0; jiffies=1715 > FC=0x0008 (type=2:0) dur=0x0000 seq=0x0000 data_len=68 > A1=ff:ff:ff:ff:ff:ff A2=00:60:1d:1b:f9:ad A3=00:00:00:00:00:00 > wlan0: prism2_close Your Orinoco card seemed to have sent a broadcast data frame that the Host AP driver dropped since it was adhoc frame. This should not cause such problems that you described. Something set wlan0 interface down after this. I don't think there is any code in Host AP driver that could do this, so I would assume that it was some unknown user space program. How do you configure IP addresses for wlan0 and eth1? Could you have, e.g., DHCP client in the background setting interfaces down after some timeout? > after it ways prism2_close the entire route table for pcmcica devices goes > away and I'm left with: Setting the interface down will remove all routing entries using that interface, so this is quite expected. > I'm not 100% sure this is a problem with the hostap driver but I'm still > lost and can't figure out whats broken. What am I missing? Is it the > firmware? I don't think this is Host AP driver, firmware, or hardware issue. It sounds more like something setting interfaces down for some odd reason. I can only come up with DHCP client as an example for now, but there might be other programs doing it, too. -- Jouni Malinen PGP id EFC895FA From travisr at rave.iinet.net.au Wed Jan 29 01:56:46 2003 From: travisr at rave.iinet.net.au (Travis Read) Date: Wed, 29 Jan 2003 14:56:46 +0800 (WST) Subject: prism2 routing table issues In-Reply-To: <20030129050215.GA1327@jm.kir.nu> Message-ID: I cracked it, Turns out an extry in cron had a script that was destroying the route table. I've since fixed it and hostap is working fine again :-) Kind regards, ------------------------------------------------------------------------------ Travis Read iiNet Senior Support | Ph +61 8 9214 2222 Fx +61 8 9214 2211 travisr at staff.iinet.net.au | Level 6, Durack House, 263 Adelaide Terrace ------------------------------------------------------------------------------ " there is a war going on, it's not about who has the most bullets, it's about who controls the information " - SNEAKERS On Tue, 28 Jan 2003, Jouni Malinen wrote: > On Wed, Jan 29, 2003 at 11:52:57AM +0800, Travis Read wrote: > > > I'm finding the the hostap driver and the orinoco driver both work fine > > for a while. But after an unknown time period the route table goes away: > > > > eth1: MAC address 00:60:1D:1B:F9:AD > > > wlan0: dropped received packet from 00:60:1d:1b:f9:ad with no ToDS flag > > (type=0x02, subtype=0x00) > > wlan0: RX status=0x2000 (port=0, type=1, fcserr=0) silence=0 signal=81 > > rate=20 rxflow=0; jiffies=1715 > > FC=0x0008 (type=2:0) dur=0x0000 seq=0x0000 data_len=68 > > A1=ff:ff:ff:ff:ff:ff A2=00:60:1d:1b:f9:ad A3=00:00:00:00:00:00 > > wlan0: prism2_close > > Your Orinoco card seemed to have sent a broadcast data frame that the > Host AP driver dropped since it was adhoc frame. This should not cause > such problems that you described. Something set wlan0 interface down > after this. I don't think there is any code in Host AP driver that could > do this, so I would assume that it was some unknown user space program. > > How do you configure IP addresses for wlan0 and eth1? Could you have, > e.g., DHCP client in the background setting interfaces down after some > timeout? > > > after it ways prism2_close the entire route table for pcmcica devices goes > > away and I'm left with: > > Setting the interface down will remove all routing entries using that > interface, so this is quite expected. > > > I'm not 100% sure this is a problem with the hostap driver but I'm still > > lost and can't figure out whats broken. What am I missing? Is it the > > firmware? > > I don't think this is Host AP driver, firmware, or hardware issue. It > sounds more like something setting interfaces down for some odd reason. > I can only come up with DHCP client as an example for now, but there > might be other programs doing it, too. > > -- > Jouni Malinen PGP id EFC895FA > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From jerryf at terraworld.net Tue Jan 28 15:59:38 2003 From: jerryf at terraworld.net (Jerry) Date: Tue, 28 Jan 2003 14:59:38 -0600 Subject: Hostap will load but WILL NOT create the wlan0 interface Message-ID: <000d01c2c710$2bb9db80$5304000a@terraworld.net> Hello, I have SuSE 80 with the stock kernel. I can compile and install the drivers without an error. I can also get the drivers to load at boot, or if i hotplug the pcmcia card (wpc11). My problem is that it wont create the "wlan0" interface. and I havent been able to figure out how to create it manually. The drivers seem to load fine, now, but no interface to give an IP to... any ideas? anyone?? I have been playing with 'ifup' but best i can get from it manually is 'ifup: Could not get a valid interface name: -> skipped'. Im kinda stuck... I have tried both the CVS and the October release, and both do the same thing. Thanks Jerryf -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030128/84637794/attachment.htm From mpolak at gup.jku.at Wed Jan 29 08:14:05 2003 From: mpolak at gup.jku.at (Martin Polak) Date: Wed, 29 Jan 2003 14:14:05 +0100 Subject: Hostap will load but WILL NOT create the wlan0 interface References: <000d01c2c710$2bb9db80$5304000a@terraworld.net> Message-ID: <3E37D39D.1020605@gup.jku.at> Jerry wrote: > Hello, > > I have SuSE 80 with the stock kernel. I can compile and install the > drivers without an error. I can also get the drivers to load at boot, > or if i hotplug the pcmcia card (wpc11). > > My problem is that it wont create the "wlan0" interface. and I havent > been able to figure out how to create it manually. The drivers seem > to load fine, now, but no interface to give an IP to... any ideas? anyone?? > > I have been playing with 'ifup' but best i can get from it manually is > 'ifup: Could not get a valid interface name: -> skipped'. Im kinda > stuck... I have tried both the CVS and the October release, and both do > the same thing. > You could do, what I ve done: change the name of the interface, the driver creates: go to: cd /driver/modules/ grep -n wlan hostap_hw.c //this should show you the line-numbers where the interface name is created: There are two lines: xxxx: sprintf(dev->name, "wlan%d", i++); yyyy: memcpy(dev->name, "wlan%d",7); open the file hostap_hw.c in an editor and replace this two occurrencies with the string "eth", recompile the host_ap driver, and your interface will now be called ethX. Martin From josh.auzins at mchp.siemens.de Wed Jan 29 10:41:41 2003 From: josh.auzins at mchp.siemens.de (Josh Auzins) Date: Wed, 29 Jan 2003 16:41:41 +0100 Subject: problems with Lucent cards associating with LinkSys AP Message-ID: <3E37F635.4F0C24FC@mchp.siemens.de> Hello, I have two different linux station running in Host AP mode, one with a D-Link650 and the other with a LinkSys v3. A station using Cisco Aironet 340 has no problems associating with either of these APs. However, a station using a Lucent card (I have two, WaveLAN Gold and Orinoco Gold) can associate with the only the D-Link650 AP, and not with the LinkSys AP. Trying to force an associate with the "iwconfig ap " results in the error "SIOCSIWAP Operation Not Supported". This command (to force the association) does work with the Cisco card. Any suggestions on how to get these Lucents cards to associate with the LinkSys AP? regards, Josh Auzins From martin.whitlock at possio.com Wed Jan 29 11:40:02 2003 From: martin.whitlock at possio.com (Martin Whitlock) Date: Wed, 29 Jan 2003 17:40:02 +0100 Subject: hostap in master mode and firmware 1.6.3 References: <20030128015844.GB1320@jm.kir.nu> <20030129043138.GC1472@jm.kir.nu> Message-ID: <3E3803E2.1040204@possio.com> Hi, Jouni Malinen wrote: >On Tue, Jan 28, 2003 at 05:27:09PM +0100, Martin Whitlock wrote: > >>wlan0: prism2_rx: dropped non-data frame (type=0x00, subtype=0x08) >> > >Aah.. Now this is interesting. The management frame was not catched in >the RX path.. That would certainly break AP functionality. Now I would >like to test that even more myself ;-). Which card manufacturer is using >these new STA firmware images? > >As a semi-blind guess, could you please try the attached patch (to the >current CVS version) with the new station firmware and report if there >are any changes in Host AP mode behavior? > As usual, a Jouni-labelled patch solves the problem! ;-) The STA 1.6.3 f/w now works like a charm, even in master mode. I also added support for enhanced security and it seems to work fine. A hackish patch towards cvs is attached where you can use 'iwpriv wlan0 prism2_param 30 [0..3]' to switch mode. 0 = normal mode (default) 1 = Show ESSID but deny "ANY"-clients 2 = Hide ESSID but accept "ANY" 3 = Hide ESSID and block "ANY" But actually mode 1, 2 and 3 seems to have the same effect... Have to test this further. Another issue is that using any other mode than 0 seems to shut out Win XP clients totally. Isn't there any way to force XP to connect to a ESSID, even if XP can't find the network? A HostAP client in managed mode with essid set to "ANY" behaves as expected. I.e. associates with an AP without enhanced security until you change from "ANY" to the correct ESSID. /Martin -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: STA163.patch Url: http://lists.shmoo.com/pipermail/hostap/attachments/20030129/b62a6929/attachment.txt From lsantos at pd3.com.br Wed Jan 29 11:16:26 2003 From: lsantos at pd3.com.br (Leonardo Pereira Santos) Date: Wed, 29 Jan 2003 14:16:26 -0200 Subject: Hostap will load but WILL NOT create the wlan0 interface In-Reply-To: <3E37D39D.1020605@gup.jku.at> References: <000d01c2c710$2bb9db80$5304000a@terraworld.net> <3E37D39D.1020605@gup.jku.at> Message-ID: <200301291416.26783.lsantos@pd3.com.br> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I guess this was not his problem. And the scripts ( /etc/pcmcia )use the interface name as a parameter. If the modules was loaded without problems, there should be at least an entry in /proc/net/dev, so do: #cat /proc/net/dev This should show all the net interfaces registered in tghe kernel, these being the ones you can bring up with ifconfig. One odd behaviour I noticed in my system was that if I configured /etc/pcmcia/config to load all three modules (hostap_crypt, hostap and hostap_cs ) upon card insertion, it wouldn't register the interface in the kernel. I have to load the first two modules with a insmod at /etc/rc.sysinit and just load hostap_cs upon card insertion. On Wednesday 29 January 2003 11:14, Martin Polak wrote: > Jerry wrote: > > Hello, > > > > I have SuSE 80 with the stock kernel. I can compile and install the > > drivers without an error. I can also get the drivers to load at boot, > > or if i hotplug the pcmcia card (wpc11). > > > > My problem is that it wont create the "wlan0" interface. and I havent > > been able to figure out how to create it manually. The drivers seem > > to load fine, now, but no interface to give an IP to... any ideas? > > anyone?? > > > > I have been playing with 'ifup' but best i can get from it manually is > > 'ifup: Could not get a valid interface name: -> skipped'. Im kinda > > stuck... I have tried both the CVS and the October release, and both do > > the same thing. > > You could do, what I ve done: > change the name of the interface, the driver creates: > > go to: > cd /driver/modules/ > grep -n wlan hostap_hw.c > > //this should show you the line-numbers where the interface name is > created: There are two lines: > xxxx: sprintf(dev->name, "wlan%d", i++); > yyyy: memcpy(dev->name, "wlan%d",7); > > open the file hostap_hw.c in an editor and replace this two occurrencies > with the string "eth", recompile the host_ap driver, and your interface > will now be called ethX. > > Martin > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap - -- This is Linux country. On a quiet night, you can hear Windows reboot. Leonardo Pereira Santos Engenheiro de Projetos PD3 Tecnologia av. Par? 330/202 (51) 3337 1237 Today Fortune tells us: Sorry, no fortune this time. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.7 (GNU/Linux) iD8DBQE+N/5a8uAMCTjfaxIRAgmPAKCYpSkWTKETzD7xxKstJ/QsgTUzPwCeMy1C 6ICXbH/969ktAbVX73aNtCM= =0RXP -----END PGP SIGNATURE----- From ycheng at cs.ucsd.edu Wed Jan 29 13:40:15 2003 From: ycheng at cs.ucsd.edu (Yuchung Cheng) Date: Wed, 29 Jan 2003 10:40:15 -0800 Subject: Sleep/Wake Time in Power Saving Mode Message-ID: <20030129104015.C22792@cs.ucsd.edu> Is it possible to control the sleep/wake time or ATIM in Power Saving Mode from the driver? Or I would appreciate somebody can point good reference for doing that. Thanks. Yuchung From ammirata at econointl.com Wed Jan 29 14:03:47 2003 From: ammirata at econointl.com (Sergio M. Ammirata) Date: Wed, 29 Jan 2003 14:03:47 -0500 Subject: hostap in master mode and firmware 1.6.3 In-Reply-To: <3E3803E2.1040204@possio.com> Message-ID: <000001c2c7c9$28111fc0$017ba8c0@ammirata.net> Will this security patch work with firmware version 1.3.6 or is it something specific to 1.6.3? Also, where can I download firmware 1.6.3? Thanks in advance, Sergio > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] On Behalf Of > Martin Whitlock > Sent: Wednesday, January 29, 2003 11:40 AM > To: hostap at shmoo.com > Subject: Re: hostap in master mode and firmware 1.6.3 > > Hi, > > Jouni Malinen wrote: > > >On Tue, Jan 28, 2003 at 05:27:09PM +0100, Martin Whitlock wrote: > > > >>wlan0: prism2_rx: dropped non-data frame (type=0x00, subtype=0x08) > >> > > > >Aah.. Now this is interesting. The management frame was not catched in > >the RX path.. That would certainly break AP functionality. Now I would > >like to test that even more myself ;-). Which card manufacturer is using > >these new STA firmware images? > > > >As a semi-blind guess, could you please try the attached patch (to the > >current CVS version) with the new station firmware and report if there > >are any changes in Host AP mode behavior? > > > As usual, a Jouni-labelled patch solves the problem! ;-) > > The STA 1.6.3 f/w now works like a charm, even in master mode. I also > added support for enhanced security and it seems to work fine. A hackish > patch towards cvs is attached where you can use 'iwpriv wlan0 > prism2_param 30 [0..3]' to switch mode. > 0 = normal mode (default) > 1 = Show ESSID but deny "ANY"-clients > 2 = Hide ESSID but accept "ANY" > 3 = Hide ESSID and block "ANY" > > But actually mode 1, 2 and 3 seems to have the same effect... Have to > test this further. Another issue is that using any other mode than 0 > seems to shut out Win XP clients totally. Isn't there any way to force > XP to connect to a ESSID, even if XP can't find the network? A HostAP > client in managed mode with essid set to "ANY" behaves as expected. I.e. > associates with an AP without enhanced security until you change from > "ANY" to the correct ESSID. > > /Martin From cijoml at volny.cz Wed Jan 29 14:14:00 2003 From: cijoml at volny.cz (Michal Semler) Date: Wed, 29 Jan 2003 20:14:00 +0100 Subject: hostap in master mode and firmware 1.6.3 In-Reply-To: <000001c2c7c9$28111fc0$017ba8c0@ammirata.net> References: <000001c2c7c9$28111fc0$017ba8c0@ammirata.net> Message-ID: Dne st 29. leden 2003 20:03 jste napsal(a): > Will this security patch work with firmware version 1.3.6 or is it > something specific to 1.6.3? > > Also, where can I download firmware 1.6.3? I look for 1.6.3 too, but u have older version then is available - download and flash 1.4.9 here. It's better then 1.3.6 :) http://prism.wz.cz/ Michal From martin.whitlock at possio.com Wed Jan 29 14:28:42 2003 From: martin.whitlock at possio.com (Martin Whitlock) Date: Wed, 29 Jan 2003 20:28:42 +0100 Subject: hostap in master mode and firmware 1.6.3 In-Reply-To: <000001c2c7c9$28111fc0$017ba8c0@ammirata.net> Message-ID: The patch would only work with STA 1.6.3 or newer. With older f/w the settings won't have any effect. /Martin > -----Original Message----- > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com]On Behalf Of > Sergio M. Ammirata > Sent: den 29 januari 2003 20:04 > To: 'Martin Whitlock'; hostap at shmoo.com > Subject: RE: hostap in master mode and firmware 1.6.3 > > > Will this security patch work with firmware version 1.3.6 or is it > something specific to 1.6.3? > > Also, where can I download firmware 1.6.3? > > Thanks in advance, > > Sergio > > > -----Original Message----- > > From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] On Behalf > Of > > Martin Whitlock > > Sent: Wednesday, January 29, 2003 11:40 AM > > To: hostap at shmoo.com > > Subject: Re: hostap in master mode and firmware 1.6.3 > > > > Hi, > > > > Jouni Malinen wrote: > > > > >On Tue, Jan 28, 2003 at 05:27:09PM +0100, Martin Whitlock wrote: > > > > > >>wlan0: prism2_rx: dropped non-data frame (type=0x00, subtype=0x08) > > >> > > > > > >Aah.. Now this is interesting. The management frame was not catched > in > > >the RX path.. That would certainly break AP functionality. Now I > would > > >like to test that even more myself ;-). Which card manufacturer is > using > > >these new STA firmware images? > > > > > >As a semi-blind guess, could you please try the attached patch (to > the > > >current CVS version) with the new station firmware and report if > there > > >are any changes in Host AP mode behavior? > > > > > As usual, a Jouni-labelled patch solves the problem! ;-) > > > > The STA 1.6.3 f/w now works like a charm, even in master mode. I also > > added support for enhanced security and it seems to work fine. A > hackish > > patch towards cvs is attached where you can use 'iwpriv wlan0 > > prism2_param 30 [0..3]' to switch mode. > > 0 = normal mode (default) > > 1 = Show ESSID but deny "ANY"-clients > > 2 = Hide ESSID but accept "ANY" > > 3 = Hide ESSID and block "ANY" > > > > But actually mode 1, 2 and 3 seems to have the same effect... Have to > > test this further. Another issue is that using any other mode than 0 > > seems to shut out Win XP clients totally. Isn't there any way to force > > XP to connect to a ESSID, even if XP can't find the network? A HostAP > > client in managed mode with essid set to "ANY" behaves as expected. > I.e. > > associates with an AP without enhanced security until you change from > > "ANY" to the correct ESSID. > > > > /Martin > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From mrcool at stupidgamerz.com Thu Jan 30 09:36:07 2003 From: mrcool at stupidgamerz.com (Jerry) Date: Thu, 30 Jan 2003 08:36:07 -0600 Subject: Hostap will load but WILL NOT create the wlan0 interface Message-ID: <000e01c2c86c$ece8d420$5304000a@terraworld.net> I still have not gotten hostap to create the interface wlan0. I have tried all of your suggestions, thanks for your help. But still no luck. Does anyone know which config files to modify? or what script needs to run to do the 'ifup' , or for that matter, what the '/etc/sysconfig/network/ifcfg-wlan0' file should look like? I know there HAS to be a way to manually create the interface and register it to the kernel... I realize this is not necessarily within the scope of the hostap mailing list, but this is frusterating, i can get the drivers to load, but cant do anything without the interface. Thanks. Jerryf -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030130/cd87504e/attachment.htm From jkim at ndosatech.com Thu Jan 30 11:09:27 2003 From: jkim at ndosatech.com (Jihong Kim) Date: Thu, 30 Jan 2003 11:09:27 -0500 Subject: Performance degradation with WEP enabled In-Reply-To: <20030129041052.GB1472@jm.kir.nu> Message-ID: <000201c2c879$f6c5ee80$899ffea9@Jihong> ** note - what I mean by host-based enc/decryption is WEP encryption/decryption performed solely on the host-CPU without using WEP enc/decryption feature on hardware or PRISM chip. The NIC that I used to build the host AP for this performance test is SMC2602W (PLX card that uses SMC2632W PCMCIA card.) I believe this card uses PRISM 2 chipset. The host CPU that was used to build a host AP (using the SMC2602W) is Pentium 4, 1.4GHz. The OS for host AP is Redhat Linux v7.2. I used the host-based encryption/decryption (which I believe is the default, when you enable WEP with 'iwconfig wlan0 key xxxx-xxxx-xx' command.). In the same test environment, I switched the host AP with Linksys AP (WAP 11 version 2.2), and I did the performance test again (Open system, WEP 64, WEP 128). In this test, there was no difference in throughput rate between different security modes. The rate was around 4.9 Mbps for open, WEP64, and WEP 128. I performed this test to see if anything other than host AP could have contributed to the performance gap between open mode and WEP 64 in the host AP performance test. However, with the findings from the Linksys AP performance test, it seems most likely that the host-based encryption/decryption (as opposed to hardware enc/decryption) would have contributed to the performance loss in host AP performance test. Question 1> Any opinion on this issue? Question 2> Jouni, you said you've seen near 6 Mbps throughput rate with host-based enc/decryption. Can you tell me how you've got 6 Mbps? What was its environment (host CPU, NIC used to build host AP, etc.)? I'm interested in how to boost performance with host-based enc/decryption, which, in my opinion, is likely to hurt the throughput rate. Thanks in advance. - Jihong -----Original Message----- From: hostap-admin at shmoo.com [mailto:hostap-admin at shmoo.com] On Behalf Of Jouni Malinen Sent: Tuesday, January 28, 2003 11:11 PM To: hostap at shmoo.com Subject: Re: Performance degradation with WEP enabled On Tue, Jan 28, 2003 at 07:54:17PM -0500, Jihong Kim wrote: > With open system (no encryption), I got about 4.7 Mbps (GUI, > daemons/processes are turned off). When they were on, the throughput > was about 4 Mbps. > > With WEP (encryption) enabled, the highest I got was about 2.7 Mbps, > when GUI and other daemons/processes were turned off. (With GUI and > other daemons/processes running in the background, the throughput was > about 600 Kbps.) Few questions.. ;-) Which Prism2/2.5/3 and which station firmware version are you using? What about host CPU? Did you use host-based encryption and/or decryption? Which operation mode (Master?) were you using? Which card was used as the other part of the test and which driver? Have you seen the other card with good performance when using WEP? > I believe this is due to the fact that WEP encryption implemented in the > device driver is used instead of the WEP encryption in PRISM MAC chip. This can be the case, if the host system has very low-performance CPU. However, WEP is not very processor intensive at throughputs provided by IEEE 802.11b. You should also note that the Host AP driver defaults to hardware-based decryption (and also encryption, if STA firmware is version 1.4.9 or newer), so hardware WEP performance might also be an issue. At least some of the older Prism2 cards used to have poor WEP performance (I have seen results in 2-3 Mbps range when using hardware WEP). > Is there a way to improve the throughput of WLAN with WEP enabled? Do > you think using a more powerful processor (CPU) will reduce the > performance gap between open system and WEP? I have seen near 6 Mbps throughput with WEP, both using hardware encrypt/decrypt and doing encrypt/decrypt fully in the driver code. So yes, the throughput you are seeing can certainly be improved. The first thing would be to find out which component is the real bottleneck. If it is indeed host CPU, there are two options that could be used to improve situation. You could either use faster host CPU or try using hardware WEP by upgrading the wlan card firmware to at least STA f/w 1.4.9. -- Jouni Malinen PGP id EFC895FA _______________________________________________ HostAP mailing list HostAP at shmoo.com http://lists.shmoo.com/mailman/listinfo/hostap From Dave.Edwards at abicom-international.com Thu Jan 30 11:41:34 2003 From: Dave.Edwards at abicom-international.com (Dave Edwards) Date: Thu, 30 Jan 2003 16:41:34 +0000 Subject: HostAP and Bridging performance Message-ID: <3E3955BE.5080601@abicom-international.com> Hi Folks, I'm looking for some pointers here. I have a system running RH8 operating as a Host based AP. I have remote systems that can connect and Authenticate without any problems, however when I enable bridging on the system the performance of the network goes through the floor. I'm relatively new to Linux and I guess that this issue is "operator error", but I can't figure out what is the problem. The performance that I'm seeing is as follows: Without bridging enabled I can get aound 650Kbytes /sec to the system When I enable bridging I get around 100Kbytes/sec! Does anyone have any pointers on where I can start looking to fix this? Thanks in advance Dave From raghu at sraamerica.com Thu Jan 30 13:26:37 2003 From: raghu at sraamerica.com (Raghu) Date: Thu, 30 Jan 2003 13:26:37 -0500 Subject: did not ACK authentication response Message-ID: <20030130182637.CGAY5165.imta02a2.registeredsite.com@smtp.registeredsite.com> Hi I get the following error and crash in hostapd, occasionally but immediately when I insert the card at the STA. When I reinsert the card (sometimes after n times), this problem goes away. Both STA & AP are using HOSTAP driver from cvs. I have the same dlink DWL 650 pc cards at both STA & AP. $ ./hostap_diag wlan0 Host AP driver diagnostics information for 'wlan0' NICID: id=0x800c v1.0.0 (PRISM II (2.5) PCMCIA (SST parallel flash)) PRIID: id=0x0015 v1.0.7 STAID: id=0x001f v1.3.5 (station firmware) Station 00:05:5d:ee:8a:f6 authentication OK (open system) STA 00:05:5d:ee:8a:f6 trying to associate before authentication sta: addr=00:05:5d:ee:8a:f6 aid=0 flags=0x0000 handle_auth_cb: STA 00:05:5d:ee:8a:f6 did not ACK authentication response Station 00:05:5d:ee:8a:f6 authentication OK (open system) STA 00:05:5d:ee:8a:f6 trying to associate before authentication sta: addr=00:05:5d:ee:8a:f6 aid=0 flags=0x0000 handle_auth_cb: STA 00:05:5d:ee:8a:f6 did not ACK authentication response Station 00:05:5d:ee:8a:f6 authentication OK (open system) STA 00:05:5d:ee:8a:f6 trying to associate before authentication sta: addr=00:05:5d:ee:8a:f6 aid=0 flags=0x0000 handle_auth_cb: STA 00:05:5d:ee:8a:f6 did not ACK authentication response Station 00:05:5d:ee:8a:f6 authentication OK (open system) handle_auth_cb: STA 00:05:5d:ee:8a:f6 did not ACK authentication response ./exe: line 1: 5515 Segmentation fault (core dumped) ./hostapd hostapd.conf $ gdb hostapd core.5515 GNU gdb Red Hat Linux (5.1.90CVS-5) Copyright 2002 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "i386-redhat-linux"... Core was generated by `./hostapd hostapd.conf'. Program terminated with signal 11, Segmentation fault. Reading symbols from /lib/i686/libc.so.6...done. Loaded symbols for /lib/i686/libc.so.6 Reading symbols from /lib/ld-linux.so.2...done. Loaded symbols for /lib/ld-linux.so.2 #0 hostapd_rotate_wep (eloop_ctx=0xbffff0e0, timeout_ctx=0x0) at hostapd.c:820 820 s->eapol_sm->keyAvailable = TRUE; (gdb) bt #0 hostapd_rotate_wep (eloop_ctx=0xbffff0e0, timeout_ctx=0x0) at hostapd.c:820 #1 0x0804adeb in eloop_run () at eloop.c:245 #2 0x0804a94d in main (argc=2, argv=0xbffff7e4) at hostapd.c:1120 #3 0x42017499 in __libc_start_main () from /lib/i686/libc.so.6 Please let me know if you need more info. -Raghu From kauer at biplane.com.au Thu Jan 30 12:48:02 2003 From: kauer at biplane.com.au (Karl Auer) Date: Thu, 30 Jan 2003 18:48:02 +0100 Subject: Hostap will load but WILL NOT create the wlan0 interface In-Reply-To: <000e01c2c86c$ece8d420$5304000a@terraworld.net> References: <000e01c2c86c$ece8d420$5304000a@terraworld.net> Message-ID: <20030130184802.20a58bcc.kauer@biplane.com.au> On Thu, 30 Jan 2003 08:36:07 -0600 "Jerry" wrote: > I still have not gotten hostap to create the interface wlan0. I'm no guru, but all I had to do to get SuSE 8.0 to use the interface properly (apart from compiling the software and putting the modules in the right place, obviously) was set up /etc/sysconfig/network/ifcfg-wlan0: BOOTPROTO="dhcp" WIRELESS="yes" STARTMODE="onboot" WIRELESS_MODE="managed" WIRELESS_ESSID="XXXXXXX" WIRELESS_KEY="XXXXXXXXXXXXX" Putting WIRELESS="yes" in there tells SuSE to run iwconfig to set the mode, essid etc after loading the modules, and it's that that actually "creates" the interface. You can hack the source to change the interface name if you don't like wlan0, or just rename the interface in /etc/modules.conf. These lines in /etc/modules.conf will let you use the hostap drivers and end up with an interface called eth1: alias eth1 hostap_pci post-install eth1 ip link set wlan0 name eth1 Remember you need to have ifcfg-eth1 in that case, not ifcfg-wlan0! If this doesn't work out of the box, you might have to install the latest versions of the wireless tools (which is a good idea annyway). To do all this manually (at least for a built in card, there are additional steps for PCMCIA I guess): - compile and install the hostap drivers - modify /etc/modules.conf appropriately - depmod -a - load the hostap modules - iwconfig wlan0 mode managed - iwconfig wlan0 essid XXXXXXX <-- if necessary - ifconfig wlan0 key XXXXXX <-- if necessary - ifconfig [insert appropriate IP setup here] - route [insert appropriate routing setup here] Regards, K. -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Karl Auer (kauer at biplane.com.au) +41-43-2660706 (h) http://www.biplane.com.au/~kauer/ +41- 1-6327531 (w) From linuxchuck at phoenixbox.org Thu Jan 30 17:46:44 2003 From: linuxchuck at phoenixbox.org (linuxchuck) Date: Thu, 30 Jan 2003 17:46:44 -0500 Subject: "original" WDT11/WPC11 combo for an AP? In-Reply-To: <200301280151.48324.jasonb@edseek.com> References: <200301280157.h0S1vnM2020252@phoenix.phoenixbox.org> <200301280151.48324.jasonb@edseek.com> Message-ID: <200301302246.h0UMkic8017923@phoenix.phoenixbox.org> Back in the day, Jason Boxman said: > > There are a few things you can do. First, I'd simplify the equation. Forget > about getting 802.1d ethernet bridging working, for the moment. I would > setup the WPC11 in the WDT11 in HostAP mode with a static address and setup > the other station in managed mode with a static address in the same subnet. > > Then, see if you can ping yourself. If that works, see if you can ping the > AP. If so, there's a bridging issue floating around. If not, then something > else is going on. You said you were getting a good association and signal, > so I'd imagine the AP is sending out beacon frames properly. snip-snip > First, I'd see if it works without bridging. If not, you could investigate > your syslog (if you haven't already) for hints. You might also try doing an > voliatile firmware update. You will have to reflash when the power cycles, > but if it gets you up, it's worth it. > snip-snip > > I don't think you mentioned what driver you're using on the client station. > HostAP in managed mode I assume? What's the diagnostic output for that > system? Good news. I have it fully functional now. It would seem the problem I had was transient. I removed the bridging as you suggested, and placed the system in ad-hoc mode. The laptop, (running win2k pro, by the way) refused to associate at all then. Then I found out that the drivers in 2k give 2 different modes of ad- hoc. One is written just plain ad-hoc, the other option is 802.11b ad-hoc. I set it to 802.11b ad-hoc, and it associated. I could also ping between them as well. Then, I placed my system in Master mode, and the laptop in infrastructure mode and again, had fully functional network connectivity between them. I finally set my bridging back up as I had it before, and everything worked. I must have had some small mistake in my previous setup that righted itself the second time around. Typical, huh? Either way, Thank you for your assistance. Now for another question: Are there any worthwhile benefits to flashing my WPC-11 firmware up to 1.4.9? As far as I can tell, it seems to be passing data without issues at my current version. I have not put any real load on it yet, as I only have one laptop at the moment that I can use for testing with it. I have found reference to windows based flash methods that will *supposedly* work with any prism2 based card. The specific reference is: http://lists.shmoo.com/pipermail/hostap/2002-October/000050.html Is it worth even considering? Again, my firmware versions are: Host AP driver diagnostics information for 'wlan0' NICID: id=0x8002 v1.0.0 (HWB3163-01,02,03,04 Rev A) PRIID: id=0x0015 v0.3.0 STAID: id=0x001f v0.8.0 (station firmware) From jasonb at edseek.com Thu Jan 30 18:24:08 2003 From: jasonb at edseek.com (Jason Boxman) Date: Thu, 30 Jan 2003 18:24:08 -0500 (EST) Subject: 'original' WDT11/WPC11 combo for an AP? In-Reply-To: <200301302246.h0UMkic8017923@phoenix.phoenixbox.org> References: <200301280157.h0S1vnM2020252@phoenix.phoenixbox.org> <200301280151.48324.jasonb@edseek.com> <200301302246.h0UMkic8017923@phoenix.phoenixbox.org> Message-ID: <2773.209.208.68.56.1043969048.squirrel@rebecca.edseek.com> linuxchuck said: > > Back in the day, Jason Boxman said: (: > Good news. I have it fully functional now. It would seem the problem I > had was > transient. I removed the bridging as you suggested, and placed the system > in > ad-hoc mode. The laptop, (running win2k pro, by the way) refused to > associate > at all then. Then I found out that the drivers in 2k give 2 different > modes of ad- > hoc. One is written just plain ad-hoc, the other option is 802.11b > ad-hoc. I set > it to 802.11b ad-hoc, and it associated. I could also ping between them > as well. Yeah, there's the standard Ad-Hoc mode and there's an old school variant that was used by older 2Mbps only cards I believe. I see you found the right one. > Are there any worthwhile benefits to flashing my WPC-11 firmware up to > 1.4.9? As > far as I can tell, it seems to be passing data without issues at my > current version. I > have not put any real load on it yet, as I only have one laptop at the > moment that I > can use for testing with it. I have found reference to windows based > flash methods > that will *supposedly* work with any prism2 based card. The specific > reference is: Don't Don't Don't flash your WPC card to STA 1.4.9 using the hacked utility. The site hosting it says he has received reports of several people frying their WPC11 cards doing so. If you really need to, you can ram upload STA FW. If the card is working okay with your existing firmware, you can leave it alone. I get the impression that newer revisions, like 1.5.6 have some cool features, but I haven't had time to try ram uploading those yet... I still don't even know which cards ship with this very new firmware. > http://lists.shmoo.com/pipermail/hostap/2002-October/000050.html > > Is it worth even considering? Again, my firmware versions are: > > Host AP driver diagnostics information for 'wlan0' > > NICID: id=0x8002 v1.0.0 (HWB3163-01,02,03,04 Rev A) > PRIID: id=0x0015 v0.3.0 > STAID: id=0x001f v0.8.0 (station firmware) > > > > From travisr at rave.iinet.net.au Thu Jan 30 20:37:59 2003 From: travisr at rave.iinet.net.au (Travis Read) Date: Fri, 31 Jan 2003 09:37:59 +0800 (WST) Subject: HostAP and Bridging performance In-Reply-To: <3E3955BE.5080601@abicom-international.com> Message-ID: Hey Dave, What kind of ethernet card are you brdiging with? If it's a Realtek 8139 based card I'd consider getting a card with a DEC Tulip chipset or a Intel (if you have the money). I've found that the realtek network cards do not work very well at all if you try and do anything involving large amounts of traffic. Maybe it's related? Kind regards, ------------------------------------------------------------------------------ Travis Read iiNet Senior Support | Ph +61 8 9214 2222 Fx +61 8 9214 2211 travisr at staff.iinet.net.au | Level 6, Durack House, 263 Adelaide Terrace ------------------------------------------------------------------------------ " there is a war going on, it's not about who has the most bullets, it's about who controls the information " - SNEAKERS On Thu, 30 Jan 2003, Dave Edwards wrote: > Hi Folks, > > I'm looking for some pointers here. I have a system running RH8 > operating as a Host based AP. > > I have remote systems that can connect and Authenticate without any > problems, however when I enable bridging on the system the performance > of the network goes through the floor. > > I'm relatively new to Linux and I guess that this issue is "operator > error", but I can't figure out what is the problem. > > The performance that I'm seeing is as follows: > > Without bridging enabled I can get aound 650Kbytes /sec to the system > > When I enable bridging I get around 100Kbytes/sec! > > Does anyone have any pointers on where I can start looking to fix this? > > Thanks in advance > > Dave > > > > _______________________________________________ > HostAP mailing list > HostAP at shmoo.com > http://lists.shmoo.com/mailman/listinfo/hostap > From Dave.Edwards at abicom-international.com Fri Jan 31 03:53:05 2003 From: Dave.Edwards at abicom-international.com (Dave Edwards) Date: Fri, 31 Jan 2003 08:53:05 +0000 Subject: HostAP and Bridging performance References: Message-ID: <3E3A3971.5000006@abicom-international.com> An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030131/7f2cbabd/attachment.htm From hostapd at cybercity.sk Fri Jan 31 05:00:53 2003 From: hostapd at cybercity.sk (hostapd at cybercity.sk) Date: Fri, 31 Jan 2003 11:00:53 +0100 Subject: Wireless links with HostAPd for 16 km. Message-ID: <16458.030131@cybercity.sk> Hello all, We are little ISP at small city at Slovakia. I would to thank you all for well job made on that super project. You can see your work in wild at http://www.netlars.sk/internet.php Thank you once again and I wish you good luck on that way. Best regards, Slavoj Hruska mailto:hostapd at cybercity.sk From jerryf at terraworld.net Thu Jan 30 14:24:35 2003 From: jerryf at terraworld.net (Jerry) Date: Thu, 30 Jan 2003 13:24:35 -0600 Subject: Hostap will load but WILL NOT create the wlan0 interface Message-ID: <001f01c2c895$39173620$5304000a@terraworld.net> This is a fresh SuSE 8.0 install (minimal with pcmcia, wireless tools, and a few other odds and ends). It uses the stock kernel and i have the kernel sources installed. 'lspci' shows the cardbus bridge (Ricoh RL5c475). Here is dmesg, after a boot... Linux version 2.4.18-4GB (root at Pentium.suse.de) (gcc version 2.95.3 20010315 (SuSE)) #1 Wed May 29 15:47:24 UTC 2002 BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 000000000009fc00 (usable) BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved) BIOS-e820: 00000000000e0000 - 0000000000100000 (reserved) BIOS-e820: 0000000000100000 - 0000000008000000 (usable) BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved) BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved) BIOS-e820: 00000000fffe0000 - 0000000100000000 (reserved) Scanning bios EBDA for MXT signature 128MB LOWMEM available. On node 0 totalpages: 32768 zone(0): 4096 pages. zone(1): 28672 pages. zone(2): 0 pages. Building zonelist for node : 0 Kernel command line: auto BOOT_IMAGE=linux ro root=303 apm=off acpi=off Initializing CPU#0 Detected 166.195 MHz processor. Console: colour VGA+ 80x25 Calibrating delay loop... 331.77 BogoMIPS Memory: 126284k/131072k available (1405k kernel code, 4400k reserved, 421k data, 120k init, 0k highmem) Dentry cache hash table entries: 16384 (order: 5, 131072 bytes) Inode cache hash table entries: 8192 (order: 4, 65536 bytes) Mount-cache hash table entries: 512 (order: 0, 4096 bytes) Buffer-cache hash table entries: 8192 (order: 3, 32768 bytes) Page-cache hash table entries: 32768 (order: 5, 131072 bytes) CPU: Before vendor init, caps: 000001bf 00000000 00000000, vendor = 0 Intel Pentium with F0 0F bug - workaround enabled. CPU: After vendor init, caps: 000001bf 00000000 00000000 00000000 CPU: After generic, caps: 000001bf 00000000 00000000 00000000 CPU: Common caps: 000001bf 00000000 00000000 00000000 CPU: Intel Pentium 75 - 200 stepping 0c Checking 'hlt' instruction... OK. POSIX conformance testing by UNIFIX mtrr: v1.40 (20010327) Richard Gooch (rgooch at atnf.csiro.au) mtrr: detected mtrr type: none PCI: PCI BIOS revision 2.10 entry at 0xfdb81, last bus=0 PCI: Using configuration type 1 PCI: Probing PCI hardware Limiting direct PCI/PCI transfers. Linux NET4.0 for Linux 2.4 Based upon Swansea University Computer Society NET3.039 Initializing RT netlink socket apm: BIOS not found. mxt_scan_bios: enter Starting kswapd kinoded started VFS: Diskquotas version dquot_6.5.0 initialized ACPI: disabled by cmdline, exiting Detected PS/2 Mouse Port. pty: 256 Unix98 ptys configured Serial driver version 5.05c (2001-07-08) with HUB-6 MANY_PORTS MULTIPORT SHARE_IRQ SERIAL_PCI enabled Real Time Clock Driver v1.10e Uniform Multi-Platform E-IDE driver Revision: 6.31 ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx PIIX: IDE controller on PCI bus 00 dev 38 PIIX: chipset revision 2 PIIX: not 100% native mode: will probe irqs later PIIX: neither IDE port enabled (BIOS) PIIX: IDE controller on PCI bus 00 dev 39 PIIX: chipset revision 2 PIIX: not 100% native mode: will probe irqs later ide0: BM-DMA at 0xffa0-0xffa7, BIOS settings: hda:DMA, hdb:pio ide1: BM-DMA at 0xffa8-0xffaf, BIOS settings: hdc:pio, hdd:pio hdb: C/H/S=0/0/0 from BIOS ignored hda: WDC AC23200L, ATA DISK drive ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 blk: queue c0322e64, I/O limit 4095Mb (mask 0xffffffff) hda: 6346368 sectors (3249 MB) w/256KiB Cache, CHS=6296/16/63, (U)DMA ide-floppy driver 0.99.newide Partition check: hda: hda1 hda2 hda3 Floppy drive(s): fd0 is 1.44M FDC 0 is a National Semiconductor PC87306 RAMDISK driver initialized: 16 RAM disks of 64000K size 1024 blocksize loop: loaded (max 16 devices) Cronyx Ltd, Synchronous PPP and CISCO HDLC (c) 1994 Linux port (c) 1998 Building Number Three Ltd & Jan "Yenya" Kasprzak. ide-floppy driver 0.99.newide SCSI subsystem driver Revision: 1.00 kmod: failed to exec /sbin/modprobe -s -k scsi_hostadapter, errno = 2 kmod: failed to exec /sbin/modprobe -s -k scsi_hostadapter, errno = 2 md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27 md: Autodetecting RAID arrays. md: autorun ... md: ... autorun DONE. NET4: Linux TCP/IP 1.0 for NET4.0 IP Protocols: ICMP, UDP, TCP, IGMP IP: routing cache hash table of 1024 buckets, 8Kbytes TCP: Hash tables configured (established 8192 bind 8192) Linux IP multicast router 0.06 plus PIM-SM NET4: Unix domain sockets 1.0/SMP for Linux NET4.0. RAMDISK: Compressed image found at block 0 Freeing initrd memory: 536k freed VFS: Mounted root (ext2 filesystem). reiserfs: checking transaction log (device 03:03) ... Using r5 hash to sort names ReiserFS version 3.6.25 VFS: Mounted root (reiserfs filesystem) readonly. Trying to move old root to /initrd ... failed Unmounting old root Trying to free ramdisk memory ... okay Freeing unused kernel memory: 120k freed md: Autodetecting RAID arrays. md: autorun ... md: ... autorun DONE. Adding Swap: 254008k swap-space (priority 42) 8139too Fast Ethernet driver 0.9.24 eth0: D-Link DFE-538TX (RealTek RTL8139) at 0xc88e3f00, 00:40:05:7a:b6:b3, IRQ 11 eth0: Identified 8139 chip type 'RTL-8139C' eth0: Setting half-duplex based on auto-negotiated partner ability 0000. Linux Tulip driver version 0.9.15-pre10 (Mar 8, 2002) tulip0: MII transceiver #1 config 3000 status 7829 advertising 01e1. eth1: Lite-On 82c168 PNIC rev 32 at 0xe800, 00:A0:CC:56:D0:8C, IRQ 10. Linux Kernel Card Services 3.1.22 options: [pci] [cardbus] [pm] ds: no socket drivers loaded! unloading Kernel Card Services hostap_crypt: registered algorithm 'NULL' isapnp: Scanning for PnP cards... isapnp: No Plug & Play device found Linux Kernel Card Services 3.1.22 options: [pci] [cardbus] [pm] PCI: Enabling device 00:11.0 (0000 -> 0002) PCI: No IRQ known for interrupt pin A of device 00:11.0. Please try using pci=biosirq. Yenta IRQ list 0000, PCI irq0 Socket status: 30000810 cs: IO port probe 0x0c00-0x0cff: excluding 0xc70-0xc77 cs: IO port probe 0x0816-0x08ff: excluding 0x86e-0x875 cs: IO port probe 0x0800-0x080f: clean. cs: IO port probe 0x0100-0x04ff: excluding 0x170-0x177 0x370-0x377 0x470-0x477 0x4d0-0x4d7 cs: IO port probe 0x0a00-0x0aff: clean. cs: memory probe 0xa0000000-0xa0ffffff: clean. hostap_cs: 0.0.0 CVS (Jouni Malinen ) hostap_cs: setting Vcc=33 (constant) hostap_cs: CS_EVENT_CARD_INSERTION prism2_config() hostap_cs: setting Vcc=33 (from config) Checking CFTABLE_ENTRY 0x01 (default 0x01) IO window settings: cfg->io.nwin=1 dflt.io.nwin=1 io->flags = 0x0046, io.base=0x0000, len=64 hostap_cs: RequestIRQ: Resource in use prism2_release : card already removed or not configured during shutdown release - done prism2_detach hostap_free_data: ap has not yet been initialized - skip resource freeing IPv6 v0.8 for NET4.0 IPv6 over IPv4 tunneling driver eth1: no IPv6 routers present "pci=biosirq" <- with this turned on, the machine locks when it loads the pci network cards... this setting works well with more modern hardware, but i just have a regular pentium here, and the older bios' dont like that setting. Does any of this help?? thanks Jerryf -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030130/60ddebcf/attachment.htm From msmith at cbnco.com Fri Jan 31 09:39:01 2003 From: msmith at cbnco.com (Michael Smith) Date: Fri, 31 Jan 2003 09:39:01 -0500 (EST) Subject: Getting out of PS mode Message-ID: Hi, A couple of days ago I ran into the problem where the access point gets confused and thinks the station has entered power saving mode. Both AP and STA are running HostAP 2002-10-12. I got it going again by resetting the card in the station (iwpriv wlan0 reset 0). What's the best way to prevent this from happening? Frequently resetting the card? Thanks, Mike From arthurnewt at yahoo.com Fri Jan 31 11:38:22 2003 From: arthurnewt at yahoo.com (a n) Date: Fri, 31 Jan 2003 08:38:22 -0800 (PST) Subject: Linksys WUSB12 Message-ID: <20030131163822.56903.qmail@web13304.mail.yahoo.com> Hi, have several questions on the new Linksys NIC called WUSB12 1. Does it use the Prism chipset ? If so which version ? 2. Does HostAp work with this NIC ? 3. It has an USB interface instead of PCI/PCMCIA. Does HostAP work with any similar NIC ? Arthur --------------------------------- Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030131/08874005/attachment.htm