From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jouni Malinen Date: Thu, 24 Feb 2011 11:27:45 +0200 Subject: [ath9k-devel] ar9271 SIOCSIFFLAGS error and authentication timeout In-Reply-To: <20110224091510.GA24543@fly> References: <20110223204641.GA19782@fly> <20110223213502.4280.qmail@stuge.se> <20110224091510.GA24543@fly> Message-ID: <1298539665.25101.5.camel@jm-desktop> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ath9k-devel@lists.ath9k.org On Thu, 2011-02-24 at 01:15 -0800, Cedric Sodhi wrote: > PS, if that matters: When I observe the ERFKILL status with rfkill I get > a > > idx 0 type 1 op 2 soft 0 hard 0 > idx 0 type 1 op 2 soft 0 hard 0 > > every time I restart /etc/init.d/net.wlan1. Those lines are saying that WLAN (type 1) was changes (op 2) to having both soft and hard block off. There would need to be a case where either the soft or the hard block value would change to 1 for ERFKILL to be indicated. Could you please run "rfkill event" while injected the card and keep it running until you see the first ERFKILL? I would like to see at which point soft/hard block is enabled and what may have happened just before that.. > Attached is a log of dmesg. If you want me to enable certain DEBUG > options and grab another one, please say so. I've just read that certain > DEBUG options might cause problems so I disabled them all, to be > careful. This looks like a separate issue. For some reason, the card is either not transmitting or receiving frames when trying to authenticate. If you can reproduce this easily, it would be interesting to see a wireless capture log (if you have another device that could run in monitor mode) to confirm whether the authentication frames are indeed transmitted. - Jouni