On 11/9/05, <b class="gmail_sendername">Jouni Malinen</b> <<a href="mailto:jkmaline@cc.hut.fi" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">jkmaline@cc.hut.fi</a>> wrote:<span class="gmail_quote">
</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On Wed, Nov 09, 2005 at 12:09:35PM -0600, Tony Jarnigan wrote:<br><br>> My Linux box (A) is using wpa_supplicant(v4.4) to do WPA-PSK to a Proxim AP.<br>> It connects fine but when another machine(B) on the network tries to ping
<br>> it, there's no response; no ssh or vnc either although the daemons are<br>> running. It's not iptables or any other firewall software, nothing is<br>> running.<br>><br>> Oddly enough, if A pings B, B can connect no problem. If I discontinue the
<br>> ping, the problem seems to come up again within 2 minutes. If I turn off<br>> WPA-PSK and not run wpa_supplicant, everything works as expected.<br><br>That sounds like issues with multicast/broadcast receive on the host
<br>(A). Pinging A->B allows (B) to update its ARP table without (A) having<br>to receive broadcast arp who-has request and that 2 minutes would be a<br>timeout for the ARP cache.</blockquote><div><br> I tried compiling the new madwifi-ng drivers and also wpa_supplicant
0.4.6 but they didn't seem to work together (I didn't spend much time debugging). I went back to my previous madwifi drivers but used the newer wpa_supplicant and things seem to work now. I'll certainly want madwifi-ng working in the future but this gets me what I want.
<br><br>Tony<br></div><br>