Re: [Xen-users] Dom0 Crashes on High I/O

[Mark Williamson <mark.williamson@xxxxxxxxxxxx>]

> I upgraded in an effort to correct the problem - it was present on the
> older kernel as well

The older kernel being the XenSource XenLinux 2.6.18 kernel?

The reason this question gets asked by folks is that a 2.6.24 port of full Xen 
functionality is a long way removed from the upstream XenSource 2.6.18 and 
it's potentially possible (with all due respect to distro maintainers) that 
porting the patches forward 6 kernel releases may have introduced some 
interesting new bugs.

Cheers,
Mark

> Henri
>
> Ryan Burke wrote:
> >> Henri Cook wrote:
> >>> Hi all, I have a problem that's very similar to this one -
> >>> which doesn't appear to have ever been resolved:
> >>> http://lists.xensource.com/archives/html/xen-users/2007-03/msg00112.htm
> >>>l
> >>>
> >>> My dom0 will crash when I try to transfer large amounts of files in
> >>> (e.g. an rsync network backup), when I try to install Windows over HVM
> >>> (at the beginning of the installer when 'installation files are being
> >>> put into folders') and seemingly 'randomly' throughout the day -
> >>> although presumably this could be an instance of high I/O on
> >>> a DomU that
> >>> I would not be aware of.
> >>>
> >>> There are no errors, anywhere that I can see - i've checked
> >>> kern.log/syslog/messages/dmesg/xend.log/xend-debug.log
> >>>
> >>> Please; can anyone shed some light on this? At the moment I
> >>> can lose the machine up to five times a day and obviously
> >>> can't install windows or anything!
> >>>
> >>> System is running on Ubuntu Hardy, 2.6.24-14 - in case this is a bug
> >>> with their kernel port i've made a bug on launchpad:
> >>> https://bugs.launchpad.net/ubuntu/+source/xen-3.2/+bug/211751
> >>>
> >>> Has anyone had this before? How did you solve it?
> >>>
> >>> Swapping also causes a crash as observed in the initially referenced
> >>> post, but that's probably because it's a high i/o operation.
> >>> Turning my system's ability to swap off has helped the problem a
> >>> little bit (eliminating that point of failure at least)
> >>
> >> Have you tried upgrading your network and storage drivers to
> >> the latest versions?
> >>
> >> Have you performed a 'memtest' for several passes over your
> >> entire memory? IO is memory mapped and a bad byte can cause
> >> the server to abend.
> >>
> >> Have you tried a disk scan? Especially of your swap storage
> >> to rule out any bad sectors in swap. Swap doesn't have the
> >> bad sector mapping abilities that file systems have.
> >>
> >> Try doing all of these and see if it doesn't fix your
> >> problem.
> >>
> >> -Ross
> >
> > If possible could you downgrade your kernel to 2.6.18 so that it more
> > closely matches the officially supported Xen 3.2 kernel? You are at the
> > bleeding edge with the kernel you're using and to see if it is a problem
> > with Xen 3.2 or that version of the Kernel with Xen a downgrade would be
> > helpful.
> >
> > Ryan
> >
> >
> > _______________________________________________
> > Xen-users mailing list
> > Xen-users@xxxxxxxxxxxxxxxxxxx
> > http://lists.xensource.com/xen-users
>
> _______________________________________________
> Xen-users mailing list
> Xen-users@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-users



-- 
Push Me Pull You - Distributed SCM tool (http://www.cl.cam.ac.uk/~maw48/pmpu/)

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users