From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:60760) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1W2m37-0001w8-00 for qemu-devel@nongnu.org; Mon, 13 Jan 2014 13:15:50 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1W2m31-0001PX-EY for qemu-devel@nongnu.org; Mon, 13 Jan 2014 13:15:44 -0500 Received: from indium.canonical.com ([91.189.90.7]:52630) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1W2m30-0001PS-Tq for qemu-devel@nongnu.org; Mon, 13 Jan 2014 13:15:39 -0500 Received: from loganberry.canonical.com ([91.189.90.37]) by indium.canonical.com with esmtp (Exim 4.71 #1 (Debian)) id 1W2m30-00063W-0f for ; Mon, 13 Jan 2014 18:15:38 +0000 Received: from loganberry.canonical.com (localhost [127.0.0.1]) by loganberry.canonical.com (Postfix) with ESMTP id 041922E8005 for ; Mon, 13 Jan 2014 18:15:38 +0000 (UTC) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Date: Mon, 13 Jan 2014 18:06:56 -0000 From: Wesley Holevinski Sender: bounces@canonical.com References: <20140113174654.7725.99941.malonedeb@soybean.canonical.com> Message-Id: <20140113180656.7362.75243.launchpad@soybean.canonical.com> Errors-To: bounces@canonical.com Subject: [Qemu-devel] [Bug 1268671] Re: CentOS guest crashing due to assertion failure in qemu-char.c Reply-To: Bug 1268671 <1268671@bugs.launchpad.net> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org ** Description changed: Here is the log in /var/log/libvirt/qemu/centos_heavy.log = qemu-kvm: /builddir/build/BUILD/qemu-kvm-0.12.1.2/qemu-char.c:630: io_wat= ch_poll_finalize: Assertion `iwp->src =3D=3D ((void *)0)' failed. 2014-01-13 16:50:31.576+0000: shutting down = The code it's failing the assertion on has an interesting comment: = =C2=A0=C2=A0=C2=A0=C2=A0static void io_watch_poll_finalize(GSource *sourc= e) =C2=A0=C2=A0=C2=A0=C2=A0{ =C2=A0=C2=A0=C2=A0=C2=A0/* Due to a glib bug, removing the last reference= to a source =C2=A0=C2=A0=C2=A0=C2=A0* inside a finalize callback causes recursive loc= king (and a =C2=A0=C2=A0=C2=A0=C2=A0* deadlock). This is not a problem inside other c= allbacks, =C2=A0=C2=A0=C2=A0=C2=A0* including dispatch callbacks, so we call io_rem= ove_watch_poll =C2=A0=C2=A0=C2=A0=C2=A0* to remove this source. A t this point, iwp->src= must =C2=A0=C2=A0=C2=A0=C2=A0* be NULL, or we would leak it. =C2=A0=C2=A0=C2=A0=C2=A0* =C2=A0=C2=A0=C2=A0=C2=A0* This would be solved much more elegantly by chi= ld sources, =C2=A0=C2=A0=C2=A0=C2=A0* but we support older glib versions that do not = have them. =C2=A0=C2=A0=C2=A0=C2=A0*/ =C2=A0=C2=A0=C2=A0=C2=A0IOWatchPoll *iwp =3D io_watch_poll_from_source(so= urce); =C2=A0=C2=A0=C2=A0=C2=A0assert(iwp->src =3D=3D NULL); =C2=A0=C2=A0=C2=A0=C2=A0} = ------ CPU Info: = http://pastebin.com/U7MrzFxK = -------- = Relevant RPM versions: = qemu-kvm-0.12.1.2-2.415.el6_5.3.x86_64 libvirt-0.10.2-29.el6_5.2.x86_64 = -------- = Domain config: = http://pastebin.com/Nf2VsER8 = (Note the use of the vmchannels; I believe this is playing a part in this crash) = - = --------- = uname -a: = Linux blizzard 2.6.32-431.3.1.el6.x86_64 #1 SMP Fri Jan 3 21:39:27 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux = --------- = CLI to start guest (included in attached dump): = http://pastebin.com/W01Xzyb0 = --------- + = + thread apply all bt: + = + http://pastebin.com/FTpUDU7A -- = You received this bug notification because you are a member of qemu- devel-ml, which is subscribed to QEMU. https://bugs.launchpad.net/bugs/1268671 Title: CentOS guest crashing due to assertion failure in qemu-char.c Status in QEMU: New Bug description: Here is the log in /var/log/libvirt/qemu/centos_heavy.log qemu-kvm: /builddir/build/BUILD/qemu-kvm-0.12.1.2/qemu-char.c:630: io_wat= ch_poll_finalize: Assertion `iwp->src =3D=3D ((void *)0)' failed. 2014-01-13 16:50:31.576+0000: shutting down The code it's failing the assertion on has an interesting comment: =C2=A0=C2=A0=C2=A0=C2=A0static void io_watch_poll_finalize(GSource *sourc= e) =C2=A0=C2=A0=C2=A0=C2=A0{ =C2=A0=C2=A0=C2=A0=C2=A0/* Due to a glib bug, removing the last reference= to a source =C2=A0=C2=A0=C2=A0=C2=A0* inside a finalize callback causes recursive loc= king (and a =C2=A0=C2=A0=C2=A0=C2=A0* deadlock). This is not a problem inside other c= allbacks, =C2=A0=C2=A0=C2=A0=C2=A0* including dispatch callbacks, so we call io_rem= ove_watch_poll =C2=A0=C2=A0=C2=A0=C2=A0* to remove this source. A t this point, iwp->src= must =C2=A0=C2=A0=C2=A0=C2=A0* be NULL, or we would leak it. =C2=A0=C2=A0=C2=A0=C2=A0* =C2=A0=C2=A0=C2=A0=C2=A0* This would be solved much more elegantly by chi= ld sources, =C2=A0=C2=A0=C2=A0=C2=A0* but we support older glib versions that do not = have them. =C2=A0=C2=A0=C2=A0=C2=A0*/ =C2=A0=C2=A0=C2=A0=C2=A0IOWatchPoll *iwp =3D io_watch_poll_from_source(so= urce); =C2=A0=C2=A0=C2=A0=C2=A0assert(iwp->src =3D=3D NULL); =C2=A0=C2=A0=C2=A0=C2=A0} ------ CPU Info: http://pastebin.com/U7MrzFxK -------- Relevant RPM versions: qemu-kvm-0.12.1.2-2.415.el6_5.3.x86_64 libvirt-0.10.2-29.el6_5.2.x86_64 -------- Domain config: http://pastebin.com/Nf2VsER8 (Note the use of the vmchannels; I believe this is playing a part in this crash) --------- uname -a: Linux blizzard 2.6.32-431.3.1.el6.x86_64 #1 SMP Fri Jan 3 21:39:27 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux --------- CLI to start guest (included in attached dump): http://pastebin.com/W01Xzyb0 --------- thread apply all bt: http://pastebin.com/FTpUDU7A To manage notifications about this bug go to: https://bugs.launchpad.net/qemu/+bug/1268671/+subscriptions