Return-path: Received: from mail-ig0-f171.google.com ([209.85.213.171]:33839 "EHLO mail-ig0-f171.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752344AbbBWCFG (ORCPT ); Sun, 22 Feb 2015 21:05:06 -0500 Received: by mail-ig0-f171.google.com with SMTP id h15so14996804igd.4 for ; Sun, 22 Feb 2015 18:05:05 -0800 (PST) MIME-Version: 1.0 In-Reply-To: References: Date: Sun, 22 Feb 2015 18:05:05 -0800 Message-ID: (sfid-20150223_030511_161548_ED9CB4D5) Subject: Re: AR9462 problems connecting again.. From: Adrian Chadd To: Linus Torvalds Cc: "Luis R. Rodriguez" , Kalle Valo , QCA ath9k Development , "ath9k-devel@lists.ath9k.org" , Linux Wireless List Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: So the interesting part thus far: 14:07:23.513500: RTM_NEWLINK: operstate=1 ifi_flags=0x1003 ([UP]) 14:07:23.513662: RTM_NEWLINK, IFLA_IFNAME: Interface 'wlp1s0' added 14:07:23.513842: nl80211: if_removed already cleared - ignore event 14:07:23.536309: nl80211: Event message available 14:07:23.536454: nl80211: Delete station 20:9f:db:e7:80:80 14:07:23.542786: nl80211: Event message available 14:07:23.542882: nl80211: MLME event 39 14:07:23.542903: nl80211: Deauthenticate event 14:07:23.542927: wlp1s0: Event DEAUTH (12) received 14:07:23.542946: wlp1s0: Deauthentication notification 14:07:23.542964: wlp1s0: * reason 2 14:07:23.542982: wlp1s0: * address 20:9f:db:e7:80:80 14:07:23.542997: Deauthentication frame IE(s) - hexdump(len=0): [NULL] 14:07:23.543055: wlp1s0: CTRL-EVENT-DISCONNECTED bssid=20:9f:db:e7:80:80 reason=2 14:07:23.543096: wlp1s0: Auto connect enabled: try to reconnect (wps=0 wpa_state=9) 14:07:23.543115: wlp1s0: Setting scan request: 0 sec 100000 usec 14:07:23.543133: Added BSSID 20:9f:db:e7:80:80 into blacklist You've associated, you do the 4-way handshake, that works fine; then one second later you do it again; then one second later it fails and it sends a disconnect. Whenever I've seen this in the past, it's been because some part of the EAPOL exchange is being transmitted at the wrong rate or something about the sequence number/etc is incorrect. So, hm, I'd want to verify that the whole exchange is working correctly and peek into the header values for each of those. -adrian