The same for me: without the 'usleep' patch I can't connect to WPA2 AP at all. With the usleep (used 100000us), it works every time.
Ubuntu 9.10 (Karmic), wpasupplicant_0.6.9-3ubuntu1. Card: TL-WN821N (Atheros Otus chip, ID 0cf3:9170), ndiswrapper 1.54-2ubuntu1 uses windows driver for Netgear WNDA3100 (http://ubuntuforums.org/showthread.php?t=885520 ).
Patch position slightly differs, so here is the context diff:
*** wpa.c.ORIGINAL 2010-01-09 17:23:12.000000000 +0300
--- wpa.c 2010-01-09 23:52:26.000000000 +0300
***************
*** 993,998 ****
--- 993,999 ----
if (wpa_supplicant_send_4_of_4(sm, sm->bssid, key, ver, key_info,
NULL, 0, &sm->ptk))
return;
+ usleep(100000); // pause needed, because sending is in progress
/* SNonce was successfully used in msg 3/4, so mark it to be renewed
* for the next 4-Way Handshake. If msg 3 is received again, the old
The same for me: without the 'usleep' patch I can't connect to WPA2 AP at all. With the usleep (used 100000us), it works every time. 0.6.9-3ubuntu1. Card: TL-WN821N (Atheros Otus chip, ID 0cf3:9170), ndiswrapper 1.54-2ubuntu1 uses windows driver for Netgear WNDA3100 (http:// ubuntuforums. org/showthread. php?t=885520 ).
Ubuntu 9.10 (Karmic), wpasupplicant_
Patch position slightly differs, so here is the context diff:
*** wpa.c.ORIGINAL 2010-01-09 17:23:12.000000000 +0300 _send_4_ of_4(sm, sm->bssid, key, ver, key_info,
--- wpa.c 2010-01-09 23:52:26.000000000 +0300
***************
*** 993,998 ****
--- 993,999 ----
if (wpa_supplicant
NULL, 0, &sm->ptk))
return;
+ usleep(100000); // pause needed, because sending is in progress
/* SNonce was successfully used in msg 3/4, so mark it to be renewed
* for the next 4-Way Handshake. If msg 3 is received again, the old