Port-xen archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: Linux HVM domUs crashing



On Wed, 27 Apr 2011 20:57:29 +0200
Manuel Bouyer <bouyer%antioche.eu.org@localhost> wrote:

> On Wed, Apr 27, 2011 at 07:51:50PM +0100, Gerard Lally wrote:
> > I'm trying to get Debian and Scientific Linux HVM domUs going. I
> > have a Windows XP domU and a Slackware PV domU successfully
> > booting. I can't get Debian/SL working as either PV or HVM domUs.
> > 
> > For HVM I am using the example HVM config found
> > in /usr/pkg/share/examples/xen.
> > 
> > Here is my debian.cfg, for example, with comments stripped out.
> > 
> > 
> > ------------
> > kernel = "/usr/pkg/lib/xen/boot/hvmloader"
> > device_model = '/usr/pkg/libexec/qemu-dm'
> > builder='hvm'
> > name = "debian"
> > 
> > memory = 768
> > 
> > vif = [ 'mac=00:16:3e:00:00:13, bridge=bridge0, type=ioemu' ]
> > 
> > disk = [ 'file:/vnode/debian.img,ioemu:hda,w', \
> >     'file:/mnt/wd2a/debian.iso,ioemu:hdc:cdrom,r' ]
> > boot='d'
> > 
> > vnc = 1
> > vncunused = 1
> > 
> > usb=1
> > usbdevice='tablet'
> > ------------
> > 
> > I created the disk as follows:
> > 
> > dd if=/dev/zero of=/vnode/debian.img bs=1m count=12288
> > 
> > I start this domU as root from an xterm:
> > 
> > xm create /usr/pkg/etc/xen/debian.cfg
> > 
> > I use TightVNC viewer to bring up a screen which shows me the Debian
> > installer waiting for me to select Install or Graphical Install.
> > 
> > As soon as I click Install a ramdisk seems to load and then the
> > whole computer crashes - domU, dom0, the lot. The computer reboots.
> 
> Can you get something printed on console when this happens ?
> if you're running X11 on this system, maybe try in text mode so you
> have a chance to see something.

I get the following, but Googling this I see you consider these
harmless:

Failed to read /local/domain/0/device-model/1/script (No such file or
directory)

Failed to read /local/domain/0/backend/console/1/0/script (No such file
or directory)

I am just writing the error messages down at the moment because I
don't have a shared writable partition.

Another error I get is this, which perhaps isn't so harmless:

Attach vnd0d

vnd1 dos partition I/O error

Does this mean anything to you?


> Also set ddb.onpanic=1 in /etc/sysctl.conf to get more infos in case
> of panic.


Done, but I can't get back into NetBSD till tomorrow morning.
 
 
> > The same thing happens with Scientific Linux and Zenwalk Linux HVM
> > domUs.
> > 
> > The only Linux domU I am able to get working so far is a Slackware
> > PV domU. I tried Debian and Scientific Linux PV domUs as well but I
> > had no success.
> 
> Scientific Linux PV is running fine for me ...


I would of course much prefer SL 6 and Debian 6 as PV domUs. Do you mind
if I clarify the following?

Is it enough to extract the kernel and ramdisk from the DVD ISO and
put them in NetBSD / or do I need to download a separate Xen kernel and
ramdisk from the SL/Debian website? Will the same kernel work for domU
installation and daily use? I have even tried compiling a vanilla
kernel and ramdisk, which works for my Slackware domU but not for these
other distros.

-- 
Gerard Lally

Attachment: signature.asc
Description: PGP signature



Home | Main Index | Thread Index | Old Index