On Thu, 2005-09-15 at 10:22 -0400, Sean Dague wrote:
> On Thu, Sep 15, 2005 at 08:34:45AM -0400, Ted Kaczmarek wrote:
> > On Wed, 2005-09-14 at 23:38 +0100, Steven Hand wrote:
> > > >FC4, tyan 2462 SMP Athlon, 2 gig, appears that xend toasts, I noticed
> > > >that purging /var/lib/xenstored/store/domain/ appears to allow xend to
> > > >restart after this type pf error.
> > > >
> > > >The problem mostly occurs after issuing an xm shutdowm to a DomU.
> > >
> > > Can you check if it's been fixed by cset 6828?
> > >
> > > cheers,
> > >
> > > S.
> >
> > xm list is definetly behaving better, xm destroy is still having issues.
> >
> > The best test for xm destroy in my case is to try to start a domain that
> > is unable to mount root filesystem, once that happens the only way to
> > get rid of that domain is to kill xend and purge the files. Doing an xm
> > list shows the domain crashed. Will start a new thread on this.
>
> Dan Smith, can you create a test case for xm-test for that condition?
>
> -Sean
Just got a panic on Dom0 as well
Test Case
Bring up 4 DomU's
from terminal on 4th DomU issue a reboot
xm list is toast
And this leads to the panic on Dom0
xm create -c domain4
Using config file "/etc/xen/domain4".
Traceback (most recent call last):
File "/usr/sbin/xm", line 10, in ?
main.main(sys.argv)
File "/usr/lib/python/xen/xm/main.py", line 729, in main
handle_xend_error(argv[1], args[0], ex)
File "/usr/lib/python/xen/xm/main.py", line 163, in handle_xend_error
raise ex
xen.xend.XendProtocol.XendError: (3, 'No such process')
And the panic
kernel/workqueue.c:170
Unfortunately my only console cable was on the wrong machine :-(
If needed I may be able to reproduce, got hg down, must say the
mercurial cheat sheet is splendid.
Are you guys getting any mails from the list?
Starting to suspect my ISP, as my mail volume is very light today.
Ted
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home