Mailinglist Archive: opensuse-bugs (4060 mails)

< Previous Next >
[Bug 662317] DomU crashes under high load on OBS workers

https://bugzilla.novell.com/show_bug.cgi?id=662317

https://bugzilla.novell.com/show_bug.cgi?id=662317#c9


Jan Beulich <jbeulich@xxxxxxxxxx> changed:

What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
CC| |pmillett@xxxxxxxxxx
Found By|--- |Other
InfoProvider| |adrian@xxxxxxxxxx
AssignedTo|jdouglas@xxxxxxxxxx |jbeulich@xxxxxxxxxx

--- Comment #9 from Jan Beulich <jbeulich@xxxxxxxxxx> 2011-01-10 08:37:26 UTC
---
(In reply to comment #0)
I have no way to trigger it reproducable yet, but it seems to happen every ~
15
minutes on some host. What other information can I provide ?

SysRq-t output from affected guest(s) (assuming these are pv guests, "xm sysrq"
should be usable for this as long as the guest does not hang completely).

If the guest hangs completely or can be observed to spin on one or more CPUs
(observable through "xm vcpu-list"), xenctx run against the corresponding
vCPU-s would likely provide further insight.

If the guest hangs/spins inside Xen (which is unlikely, as you didn't say the
whole box got locked), 'd' sent from the serial console would be the most
likely thing to provide further information.

As to there not being unusual output from "xm dmesg" - are you running with
"loglvl=all guest_loglvl=all" on the Xen command line?

In any case, the most helpful thing of course would be to have a way to
reproduce this.

--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

< Previous Next >
References