Vista - System backup restarts Hyper-v host and pauses VMs

Hello,

I'm trying to figure out the cause of a VM being paused (some earlier post
o'mine).
Here is what I've found...

SBS Premium, MS scenario hyper-v setup (w2008-64bit as host, 2 vms: SBS
64-bit and W2008 32-bit).
Host with 5 volumes (C - host system, D - sbs vhdfixed, E - W2008 32-bit
vhdfixed, F - W2008 data vhdfixed, S - host swap).

Full system backups (C, D, E, F) are scheduled to run
every day at 9 p.m. (to a host external hard drive). Registers on the host
are patched to do such backups properly. For a few days everything was fine
but then a backup failed and I saw a weird issue in logs. Physical server
restarted at the beginnig of the backup. The first message after that was:
1101 - "Inspection events were dropped during transport. The realtime backup
file is corrupt by incorrect closing" [translated from polish - sorry for
mistakes].
and 6008 - unexpected restart
Then come warnings:
16050 (hyperv) - volume D almost out of space
16050 (hyperv) - volume F almost out of space (this one every minute)
a few hours later 2013 (srv) - D runs out of space
and a few hours later error
16060 (hyperv) - 'sbs' has been paused because it has run out of disk space
on 'D:\'. (Virtual machine ID 53EC00D5-1A97-47E1-8E09-C9C733FA30DE)
SBS machine is being paused with 16060 error - vol D runs out of disk space.
Volume D contains only sbs vhd file and empty trash (actually switched off
for this vol). Oryginally there is 1,08 GB free space on it. But when the
machine is paused the free space is 199MB or even 99MB (but no additional
files on it).

What I noticed is that
16050 (hyperv) - volume F almost out of space (this one every minute)
starts after every backup (even succeded) and
16050 (hyperv) - volume D almost out of space
2013 (srv) - D runs out of space
16060 (hyperv) - 'sbs' has been paused because it has run out of disk space
on 'D:\'. (Virtual machine ID 53EC00D5-1A97-47E1-8E09-C9C733FA30DE)
start only after restart of the phisical machine during backup.

F contains scsi-fixed vhd (data for W2008 32-bit) and oryginally has 2,5 GB
free space - this volume is not a problem here since it doesn't stop a VM

I stopped the system backups and there are NO 16050 and related
warnings/errors or restarts.

Can you give me a solution so I could run backups again?

Best regards,
--
Krzysztof Wozniak, solidea.pl

Got some additional info that may be helpful to resolve the issue.
The server is HP ML350 G5 - I checked ILO2 logs and there are messages
showing pre-reset situation.
Standard ILO log says: POWER REMOVED
IML gives some more info:

Critical OS 07/28/2009 21:30 07/28/2009 21:30 1 Abnormal Program Termination
(BugCheck, STOP: 0x0000007E (0xFFFFFFFFC0000047, 0xFFFFF800018BDE38,
0xFFFFFA6001DE05C0, 0xFFFFFA6001DE0660))

Critical OS 07/26/2009 21:35 07/26/2009 21:35 1 Abnormal Program Termination
(BugCheck, STOP: 0x0000007E (0xFFFFFFFFC0000047, 0xFFFFF8000186FE38,
0xFFFFFA6001DD95C0, 0xFFFFFA6001DD9660))

Those two are BSODs

Any ideas?

Regards,
--
Krzysztof Wozniak, solidea.pl

BSODs around 0x7e indicate a hardware or hardware driver issue, generally.
I'd start by making sure the BIOS is up to the latest, and that all the
latest drivers are installed for the ML350. Also, from your original
description, you _could_ be running critically short of disk space during
the backups. You might need more space generally, or your space divided up
into fewer volumes. But I'd start with drivers.

--
Charlie.
http://msmvps.com/blogs/xperts64
http://mvp.support.microsoft.com/profile/charlie.russel


"Solidea" <Solidea@newsgroup> wrote in message
news:2BC4384D-EF35-4A8E-81CE-8876F47C183B@newsgroup