Ewan Mellor <ewan@xxxxxxxxxxxxx> writes:
> On Mon, Dec 04, 2006 at 08:51:38PM +0100, Markus Armbruster wrote:
>
>> Hi Steven,
>>
>> I had a look at your changes and started to backport them to versions
>> we use. I might have broken something, but before I go hunting for
>> that, could you please try two little things for me in your version,
>> to make sure its okay in your tree?
>
> Steven's 30000 feet over the Atlantic at the moment. Perhaps I can help?
Appreciated :)
>> 1. Does xm restore work for you? I got the following in xend.log:
>>
>> [2006-12-04 20:34:54 xend 3507] ERROR (XendDomain:268) Restore failed
>> Traceback (most recent call last):
>> File "/usr/lib/python2.4/site-packages/xen/xend/XendDomain.py", line 263,
>> in domain_restore_fd
>> return XendCheckpoint.restore(self, fd)
>> File "/usr/lib/python2.4/site-packages/xen/xend/XendCheckpoint.py", line
>> 134, in restore
>> dominfo = xd.restore_(vmconfig)
>> File "/usr/lib/python2.4/site-packages/xen/xend/XendDomain.py", line 288,
>> in restore_
>> dominfo = XendDomainInfo.restore(config)
>> File "/usr/lib/python2.4/site-packages/xen/xend/XendDomainInfo.py", line
>> 271, in restore
>> vm.createDevices()
>> File "/usr/lib/python2.4/site-packages/xen/xend/XendDomainInfo.py", line
>> 1515, in createDevices
>> self.createDevice(n, c)
>> File "/usr/lib/python2.4/site-packages/xen/xend/XendDomainInfo.py", line
>> 1086, in createDevice
>> return self.getDeviceController(deviceClass).createDevice(devconfig)
>> File "/usr/lib/python2.4/site-packages/xen/xend/server/vfbif.py", line 64,
>> in createDevice
>> raise VmError('Unknown vfb type %s (%s)' % (t, repr(config)))
>> VmError: Unknown vfb type None (['vfb', ['backend', '0']])
>
> It looks like your device configuration hasn't been saved properly, and so
> it's not present on restore, or it's not being parsed properly on restore.
> Try sticking some tracing in XendConfig.all_devices_sxpr to see whether it's
> getting saved properly, and in the if cfg_sxp bit of device_add to see whether
> it's being reparsed.
>
> This changed quite a lot on the same day as the xenfb patches got committed,
> so it wouldn't be a surprise if we had a merge conflict.
Oww. This looks quite different in 3.0.3. Does anybody remember how
device config was saved there? A quick pointer could save me some
digging...
>> 2. My xm shutdown spits this:
>>
>> [2006-12-04 20:33:37 xend.XendDomainInfo 3507] ERROR (XendDomainInfo:1382)
>> XendDomainInfo.cleanup: image.destroy() failed.
>> Traceback (most recent call last):
>> File "/usr/lib/python2.4/site-packages/xen/xend/XendDomainInfo.py", line
>> 1380, in cleanupDomain
>> self.image.destroy()
>> File "/usr/lib/python2.4/site-packages/xen/xend/image.py", line 210, in
>> destroy
>> if not self.pid:
>> AttributeError: LinuxImageHandler instance has no attribute 'pid'
>
> For me, the pid field is specific to the HVMImageHandler, and shouldn't be
> being referenced for LinuxImageHandler instances. Are you up to date? I
> don't have an "if not self.pid:" statement anywhere in image.py.
I missed this one:
changeset: 12683:fb0a586854c1d8a7b814a4b0d77388ee05bb5fe3
user: Steven Smith <ssmith@xxxxxxxxxxxxx>
date: Fri Dec 01 12:09:10 2006 +0000
files: tools/python/xen/xend/image.py
description:
[TOOLS] Remove some dead code.
The description is misleading: the removed code wasn't dead, it was
rendered incorrect by Steven's previous changes.
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
Home