|
|
|
|
|
|
|
|
|
|
xen-devel
Re: [Xen-devel] Xen-Unstable : strange power cycling problem
On Sun, Jun 27, 2004 at 08:58:16AM +0100, Keir Fraser wrote:
> >
> > Found a old serial cable, but no useful dump at
> > console. Just "(XEN) Domain 0 shutdown: rebooting
> > machine!"
> >
> > That's how it looks...
>
> Add 'console=ttyS0' to your Linux command line /before/
> 'console=tty0'. Linux will then send boot messages and crash dumps to
> the serial line.
>
I've also seen behaviour similar to this recently, although the
child domains survive when they are just doing NFS to domain0.
What seems to reliably trigger it is running nmap on the new domain
to domain0.
In the other direction I managed to get it to Oops but not crash
nmaping from domain0 to a child domain. It didn't seem to matter
if I have the bridging configured or go via the vif interface.
Only kernel modules loaded in domain0 at crash time were ipv6, sis900 and
crc32.
bill
--
Bill Boughton <murble(at)@yuri.org.uk>
<bill@(at)xencat.murble.org.uk>
Down not across
xen-crash
Description: Text document
config-2.4.26-xen0
Description: Text document
|
|
|
|
|