Hello Marcelo, well, my report might have been a bit misleading, 2.6.31.2 has been running there for almost 3 weeks. So the problem didn't occur JUST after the upgrade. But it never occured before it, while we were using older kernels (varius 2.6.30.x, 2.6.29.x and older). I've updated machine to 2.6.31.5, and it's been running without problem since then, but it's been just few days, so we'll see. I'll report if the problem appears again. Thanks. regards nik On Tue, Nov 03, 2009 at 12:46:38PM -0200, Marcelo Tosatti wrote: > On Fri, Oct 30, 2009 at 12:15:34PM +0100, Nikola Ciprich wrote: > > Ouch, typo in subject, it's 2.6.31.1 of course. sorry about that. > > also CCing kvm. > > n. > > > > On Fri, Oct 30, 2009 at 12:06:32PM +0100, Nikola Ciprich wrote: > > > Hi, > > > some time ago, I updated my KVM hosting machine to 2.6.31.1 and it just died horribly: > > Nikola, > > Upgraded from what? Did you see experience the crash again? > > > > Oct 30 10:45:17 vbox [706369.133516] Kernel panic - not syncing: CRED: put_cred_rcu() sees ffff880204e58c00 with usage 82150912 > > > Oct 30 10:45:17 vbox [706369.133519] > > > Oct 30 10:45:17 vbox [706369.144990] Pid: 19, comm: ksoftirqd/5 Not tainted 2.6.31lb.02 #1 > > > Oct 30 10:45:17 vbox [706369.151554] Call Trace: > > > Oct 30 10:45:17 vbox [706369.154332] <IRQ> > > > Oct 30 10:45:17 vbox [<ffffffff8104c1fa>] panic+0xaa/0x180 > > > Oct 30 10:45:17 vbox [706369.160280] [<ffffffff81322b90>] ? _spin_unlock+0x30/0x60 > > > Oct 30 10:45:17 vbox [706369.166256] [<ffffffff810f5671>] ? add_partial+0x21/0x90 > > > Oct 30 10:45:17 vbox [706369.172155] [<ffffffff810f6a92>] ? __slab_free+0x92/0x3c0 > > > Oct 30 10:45:17 vbox [706369.178127] [<ffffffff81102317>] ? file_free_rcu+0x37/0x50 > > > Oct 30 10:45:17 vbox [706369.184198] [<ffffffff8106c655>] put_cred_rcu+0x75/0x80 > > > Oct 30 10:45:17 vbox [706369.190008] [<ffffffff810a2525>] __rcu_process_callbacks+0x125/0x250 > > > Oct 30 10:45:17 vbox [706369.197020] [<ffffffff810a2689>] rcu_process_callbacks+0x39/0x60 > > > Oct 30 10:45:17 vbox [706369.203624] [<ffffffff81052a61>] __do_softirq+0xc1/0x250 > > > Oct 30 10:45:17 vbox [706369.209506] [<ffffffff81053860>] ? ksoftirqd+0x0/0x1a0 > > > Oct 30 10:45:17 vbox [706369.215182] [<ffffffff8100c4dc>] call_softirq+0x1c/0x30 > > > Oct 30 10:45:17 vbox [706369.220986] [<ffffffff8100e46d>] do_softirq+0x3d/0x80 > > > Oct 30 10:45:17 vbox [706369.227317] [<ffffffff81053860>] ? ksoftirqd+0x0/0x1a0 > > > Oct 30 10:45:17 vbox [706369.233020] [<ffffffff810538e4>] ksoftirqd+0x84/0x1a0 > > > Oct 30 10:45:17 vbox [706369.238622] [<ffffffff81066686>] kthread+0xa6/0xb0 > > > Oct 30 10:45:17 vbox [706369.243956] [<ffffffff8100c3da>] child_rip+0xa/0x20 > > > Oct 30 10:45:17 vbox [706369.249390] [<ffffffff810665e0>] ? kthread+0x0/0xb0 > > > Oct 30 10:45:17 vbox [706369.254806] [<ffffffff8100c3d0>] ? child_rip+0x0/0x20 > > > Oct 30 10:45:17 vbox [706369.260454] Rebooting in 10 seconds.. > > > (trace is obtained from netconsole, so hopefully it's not mangled). > > > The machine was running ~30 KVM guests, it's 8CPU 16GB x86_64, when it crashed, it was only > > > moderately loaded. Never had this (or any other) kind of crash on it before. > > > I know there's 2.6.31.5 already out, but I'm not sure if some related problem has been > > > reported/fixed and I'm obviously not able to quicky test/reproduce it with latest kernel, > > > so I'm rather reporting. > > > Should more information/testing/etc be required, I'll be glad to help > > > regards > > > nik > -- ------------------------------------- Nikola CIPRICH LinuxBox.cz, s.r.o. 28. rijna 168, 709 01 Ostrava tel.: +420 596 603 142 fax: +420 596 621 273 mobil: +420 777 093 799 www.linuxbox.cz mobil servis: +420 737 238 656 email servis: servis@xxxxxxxxxxx -------------------------------------
Attachment:
pgpjW0ksxGRgG.pgp
Description: PGP signature
