From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Jones Subject: iscsi related softlockup (3.0 rc3) Date: Mon, 20 Jun 2011 16:00:59 -0400 Message-ID: <20110620200059.GA21401@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: netdev@vger.kernel.org Return-path: Received: from mx1.redhat.com ([209.132.183.28]:33368 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751942Ab1FTUBD (ORCPT ); Mon, 20 Jun 2011 16:01:03 -0400 Received: from int-mx10.intmail.prod.int.phx2.redhat.com (int-mx10.intmail.prod.int.phx2.redhat.com [10.5.11.23]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id p5KK125w022100 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Mon, 20 Jun 2011 16:01:02 -0400 Received: from gelk.kernelslacker.org (ovpn-113-108.phx2.redhat.com [10.3.113.108]) by int-mx10.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id p5KK11Eo032022 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Mon, 20 Jun 2011 16:01:02 -0400 Received: from gelk.kernelslacker.org (gelk [127.0.0.1]) by gelk.kernelslacker.org (8.14.5/8.14.4) with ESMTP id p5KK10eL024203 for ; Mon, 20 Jun 2011 16:01:00 -0400 Received: (from davej@localhost) by gelk.kernelslacker.org (8.14.5/8.14.5/Submit) id p5KK0x7V024189 for netdev@vger.kernel.org; Mon, 20 Jun 2011 16:00:59 -0400 Content-Disposition: inline Sender: netdev-owner@vger.kernel.org List-ID: I made some changes to my syscall fuzzer as it hasn't been turning up anything interesting since the last lot of problems got fixed. I seem to have found a way to get iscsi really confused, to the point that the soft-lockup detector kicks in, and then the rcu stall detector. http://davej.fedorapeople.org/iscsi-lockup.txt is the dmesg from the box. sysrq was working, but I couldn't ssh into the box, or even log in on the console. I had ssh sessions already logged into the box which were completely wedged. I don't have any iscsi setup configured on this machine, the fuzzer just hit upon that protocol randomly. If it wasn't for the fact that the machine wedged I wouldn't be overly concerned, but it seems that something got into a state that just consumed all cpu time. Dave