From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Lezcano Subject: Re: Kernel oops on 2.6.38 on destroying a container Date: Wed, 07 Sep 2011 23:40:01 +0200 Message-ID: <4E67E4B1.6050101@free.fr> References: <88648D2CBB3136EB9A6EFE7B@nimrod.local> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <88648D2CBB3136EB9A6EFE7B-PdXK6nMiYlGE+EvaaNYduQ@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org Errors-To: containers-bounces-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org To: Alex Bligh Cc: containers-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org List-Id: containers.vger.kernel.org On 09/07/2011 09:21 PM, Alex Bligh wrote: > We are seeing oops like this from 2.6.38-11 Ubuntu on deleting containers. > > This one was somewhat complicated (but we grabbed the oops). > The trace below happened destroying a container. I believe the iptables > rules including conntrack rules would have been deleted immediately > before the container process exited; devices evrrXXXX and evrlXXXX > are veth pairs that would also just have been destroyed) > > My colleague can replicate this on different hardware (his desktop) using > lxc-create -n foo > lxc-start -n foo > Send a ^C - entire system crashes. No exciting container activity > at all. > > I haven't seen this on 2.6.32. Any ideas what it might be? We are > trying to get this to work on a slightly more modern kernel with longterm > support, so "fixed in Linus's current tree" is not quite what I need! IMO you should fill a bug at : https://bugs.launchpad.net/ubuntu/+source/linux/ and/or ask kernel-team-nLRlyDuq1AZFpShjVBNYrg@public.gmane.org They will be able to identify the fix in the mainstream, backport it and release a new kernel. An oops is critical enough to have them to pay attention to this problem. Hope that helps -- Daniel