problem with stats and WPA on prism

Paul Fox pgf at brightstareng.com
Tue Nov 21 14:42:56 EST 2006


i guess no one has any ideas on this, eh?

paul

i wrote:
 > once in a while (actually, fairly often) when i connect to our
 > linksys AP with my engenius prism card (1.8.2 or 1.8.4 firmware),
 > i get connectivity, but the link stats are incorrect.  this
 > is reminiscent of an older bug that in the 0.3.x drivers, but
 > i was under the impression it had been fixed.
 > 
 > kernel:  2.4.24, with patch for wireless extensions 18 applied
 > driver:  hostap-driver-0.4.9
 > wpa_supplicant:  v0.4.9
 > wireless tools: 
 > 	
 >     $ iwconfig --version
 >     iwconfig  Wireless-Tools version 29
 > 	      Compatible with Wireless Extension v11 to v21.
 > 
 >     Kernel    Currently compiled with Wireless Extension v18.
 > 
 >     wifi0     Recommend Wireless Extension v18 or later,
 > 	      Currently compiled with Wireless Extension v18.
 > 
 >     eth1      Recommend Wireless Extension v18 or later,
 > 	      Currently compiled with Wireless Extension v18.
 > 
 > 
 > when everything is working properly, i have this:
 > 
 > $ ifconfig eth1
 > eth1      Link encap:Ethernet  HWaddr 00:02:6F:37:7F:A6  
 >           inet addr:192.168.1.58  Bcast:192.168.1.255  Mask:255.255.255.0
 >           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
 >           RX packets:11225 errors:0 dropped:394 overruns:0 frame:0
 >           TX packets:67 errors:0 dropped:0 overruns:0 carrier:0
 >           collisions:0 txqueuelen:0 
 >           RX bytes:994445 (971.1 Kb)  TX bytes:7487 (7.3 Kb)
 >           Interrupt:55 
 > 
 > $ iwconfig eth1
 > eth1      IEEE 802.11b  ESSID:"linksys54g"  
 >           Mode:Managed  Frequency:2.432 GHz  Access Point: 00:14:BF:74:25:2A   
 >           Bit Rate:11 Mb/s   Sensitivity=1/3  
 >           Retry min limit:8   RTS thr:off   Fragment thr:off
 >           Encryption key:off
 >           Power Management:off
 >           Link Quality=38/70  Signal level=-55 dBm  Noise level=-93 dBm
 >           Rx invalid nwid:0  Rx invalid crypt:294  Rx invalid frag:0
 >           Tx excessive retries:0  Invalid misc:10271   Missed beacon:0
 > 
 > 
 > when the problem occurs, i have this:
 > 
 > $ ifconfig eth1
 > eth1      Link encap:Ethernet  HWaddr 00:02:6F:37:7F:A6  
 >           inet addr:192.168.1.58  Bcast:192.168.1.255  Mask:255.255.255.0
 >           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
 >           RX packets:11346 errors:0 dropped:445 overruns:0 frame:0
 >           TX packets:83 errors:0 dropped:0 overruns:0 carrier:0
 >           collisions:0 txqueuelen:0 
 >           RX bytes:1005990 (982.4 Kb)  TX bytes:9382 (9.1 Kb)
 >           Interrupt:55 
 > 
 > $ iwconfig eth1
 > eth1      IEEE 802.11b  ESSID:"linksys54g"  
 >           Mode:Managed  Access Point: 00:14:BF:74:25:2A   Bit Rate:11 Mb/s   
 >           Sensitivity=1/3  
 >           Retry min limit:8   RTS thr:off   Fragment thr:off
 >           Encryption key:off
 >           Power Management:off
 >           Link Quality=0/70  Signal level=-100 dBm  Noise level=-93 dBm
 >           Rx invalid nwid:0  Rx invalid crypt:298  Rx invalid frag:0
 >           Tx excessive retries:0  Invalid misc:10291   Missed beacon:0
 > 
 > 
 > note the Link Quality values.  note also the missing channel information.
 > connectivity is fine -- it's just the stats that are bad (as far as
 > i can tell.)
 > 
 > 
 > i can fairly easily reproduce this, with a loop that downs the interface,
 > kills off wpa_supplicant, then restarts it, and ups the interface again.
 > wpa_supplicant's config looks like:
 > 
 >     network={
 >             proto=WPA
 >             key_mgmt=WPA-PSK
 >             ssid="linksys54g"
 >             psk="this_is_some_passphrase"
 >             scan_ssid=1
 >     }
 > 
 > 
 > the behavior is independent of whether the SSID is hidden or not.
 > 
 > i can supply more debug or config info as needed, and am happy
 > to insert whatever extra debug tracing that might be required.
 > 
 > btw, here's an exchange i had last year with tony espy about the
 > similar (but more consistent) bug that i/we observed in the 0.3.X
 > driver:
 >     http://lists.shmoo.com/pipermail/hostap/2005-December/012080.html
 > 
 > 
 > paul
 > =---------------------
 >  paul fox, pgf at brightstareng.com
 > _______________________________________________
 > HostAP mailing list
 > HostAP at shmoo.com
 > http://lists.shmoo.com/mailman/listinfo/hostap

=---------------------
 paul fox, pgf at brightstareng.com



More information about the HostAP mailing list