Windows Server 2008R2 randomly crashes with stop code 0x00000101

[PeterS]

Hi,

I'm running a Windows SBS 2011 (Windows Server 2008R2 64 Bit) as a KVM-VM on a Proxmox 1.9 server.
Mainly the system works fine but from time to time the VM runns into a Bluescreen with stop code 0x00000101.
The Bluesreen usually happens during the daily vzdump backup job.

After implementing the reg. hack as described in http://support.microsoft.com/kb/975530
the system becomes much more stable, but I still get the bluescreen at least once a week.

Any hints?

Thanks + regards,
Peter

 

[tom]

provide also the following details:

1. pveversion -v
2. cat /etc/qemu-server/VMID.conf
3. run 'pveperf' when the server is not under load and also details about your physical hardware
4. detail about your backup job, including backup log

 

[PeterS]

Hi Tom, please find the details below:


> 1. pveversion -v

pve1:~# pveversion -v
pve-manager: 1.9-24 (pve-manager/1.9/6542)
running kernel: 2.6.32-6-pve
proxmox-ve-2.6.32: 1.9-43
pve-kernel-2.6.32-6-pve: 2.6.32-43
qemu-server: 1.1-32
pve-firmware: 1.0-13
libpve-storage-perl: 1.0-19
vncterm: 0.9-2
vzctl: 3.0.28-1pve5
vzdump: 1.2-15
vzprocps: 2.0.11-2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.15.0-1
ksm-control-daemon: 1.0-6


> 2. cat /etc/qemu-server/VMID.conf

pve1:~# cat /etc/qemu-server/201.conf
name: sbs
ide2: none,media=cdrom
vlan0: virtio=1A:6A:A9:74:26:8E
bootdisk: ide0
ostype: w2k8
ide0: local:201/vm-201-disk-1.raw
memory: 9216
sockets: 2
onboot: 1
description: MS Windows Small Business Server 2011 PROD
cores: 4


> 3. run 'pveperf' when the server is not under load and also details about your physical hardware

pve1:~# pveperf
CPU BOGOMIPS: 76798.88
REGEX/SECOND: 882006
HD SIZE: 94.49 GB (/dev/mapper/pve-root)
BUFFERED READS: 246.20 MB/sec
AVERAGE SEEK TIME: 7.36 ms
FSYNCS/SECOND: 3655.55
DNS EXT: 195.25 ms
DNS INT: 0.82 ms



> 4. detail about your backup job, including backup log

pve1:~# cat /mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_26-20_20_02.log
Mar 26 20:20:03 INFO: Starting Backup of VM 201 (qemu)
Mar 26 20:20:03 INFO: running
Mar 26 20:20:03 INFO: status = running
Mar 26 20:20:04 INFO: backup mode: snapshot
Mar 26 20:20:04 INFO: ionice priority: 7
Mar 26 20:20:04 INFO: Logical volume "vzsnap-pve1-0" created
Mar 26 20:20:06 INFO: creating archive '/mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_26-20_20_02.tgz'
Mar 26 20:20:06 INFO: adding '/mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_26-20_20_02.tmp/qemu-server.conf' to archive ('qemu-server.conf')
Mar 26 20:20:06 INFO: adding '/mnt/vzsnap0/images/201/vm-201-disk-1.raw' to archive ('vm-disk-ide0.raw')
Mar 27 03:54:36 INFO: Total bytes written: 102265923584 (3.58 MiB/s)
Mar 27 03:55:33 INFO: archive file size: 40.47GB
Mar 27 03:55:34 INFO: delete old backup '/mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_19-20_20_01.tgz'
Mar 27 03:55:52 INFO: Logical volume "vzsnap-pve1-0" successfully removed
Mar 27 03:55:52 INFO: Finished Backup of VM 201 (07:35:50)


The backup directory /mnt/pve/pve-backup-daily/ is mounted via nfs on a Synology NAS
The original gzip is replaced by a wrapper script for pigz in order to use more then once CPU core
while compressing the archive. Without pigz the compression takes ages...

Thanks a lot for your assistance!
Regards
Peter

 

[tom]

Hi Tom, please find the details below:


> 1. pveversion -v

pve1:~# pveversion -v
pve-manager: 1.9-24 (pve-manager/1.9/6542)
running kernel: 2.6.32-6-pve
proxmox-ve-2.6.32: 1.9-43
pve-kernel-2.6.32-6-pve: 2.6.32-43
qemu-server: 1.1-32
pve-firmware: 1.0-13
libpve-storage-perl: 1.0-19
vncterm: 0.9-2
vzctl: 3.0.28-1pve5
vzdump: 1.2-15
vzprocps: 2.0.11-2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.15.0-1
ksm-control-daemon: 1.0-6

you have old packages, update to latest stable. see http://pve.proxmox.com/wiki/Downloads#Update_a_running_Proxmox_Virtual_Environment_to_1.9

> 2. cat /etc/qemu-server/VMID.conf

pve1:~# cat /etc/qemu-server/201.conf
name: sbs
ide2: none,media=cdrom
vlan0: virtio=1A:6A:A9:74:26:8E
bootdisk: ide0
ostype: w2k8
ide0: local:201/vm-201-disk-1.raw
memory: 9216
sockets: 2
onboot: 1
description: MS Windows Small Business Server 2011 PROD
cores: 4

use virtio for the disk. see http://pve.proxmox.com/wiki/Paravirtualized_Block_Drivers_for_Windows#Adding_a_temporary_drive

> 3. run 'pveperf' when the server is not under load and also details about your physical hardware

pve1:~# pveperf
CPU BOGOMIPS: 76798.88
REGEX/SECOND: 882006
HD SIZE: 94.49 GB (/dev/mapper/pve-root)
BUFFERED READS: 246.20 MB/sec
AVERAGE SEEK TIME: 7.36 ms
FSYNCS/SECOND: 3655.55
DNS EXT: 195.25 ms
DNS INT: 0.82 ms

looks fine.

> 4. detail about your backup job, including backup log

pve1:~# cat /mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_26-20_20_02.log
Mar 26 20:20:03 INFO: Starting Backup of VM 201 (qemu)
Mar 26 20:20:03 INFO: running
Mar 26 20:20:03 INFO: status = running
Mar 26 20:20:04 INFO: backup mode: snapshot
Mar 26 20:20:04 INFO: ionice priority: 7
Mar 26 20:20:04 INFO: Logical volume "vzsnap-pve1-0" created
Mar 26 20:20:06 INFO: creating archive '/mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_26-20_20_02.tgz'
Mar 26 20:20:06 INFO: adding '/mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_26-20_20_02.tmp/qemu-server.conf' to archive ('qemu-server.conf')
Mar 26 20:20:06 INFO: adding '/mnt/vzsnap0/images/201/vm-201-disk-1.raw' to archive ('vm-disk-ide0.raw')
Mar 27 03:54:36 INFO: Total bytes written: 102265923584 (3.58 MiB/s)
Mar 27 03:55:33 INFO: archive file size: 40.47GB
Mar 27 03:55:34 INFO: delete old backup '/mnt/pve/pve-backup-daily/vzdump-qemu-201-2012_03_19-20_20_01.tgz'
Mar 27 03:55:52 INFO: Logical volume "vzsnap-pve1-0" successfully removed
Mar 27 03:55:52 INFO: Finished Backup of VM 201 (07:35:50)


The backup directory /mnt/pve/pve-backup-daily/ is mounted via nfs on a Synology NAS
The original gzip is replaced by a wrapper script for pigz in order to use more then once CPU core
while compressing the archive. Without pigz the compression takes ages...

Thanks a lot for your assistance!
Regards
Peter

the upcoming 2.0 use lzo compression, faster than gzip. 2.0 will be stable soon so probably you will take a look on the upgrade guide which will be published together with the release.

 

[PeterS]

you have old packages, update to latest stable. see http://pve.proxmox.com/wiki/Downloads#Update_a_running_Proxmox_Virtual_Environment_to_1.9

Ok, I will try an update during any weekend.

use virtio for the disk. see http://pve.proxmox.com/wiki/Paravirtualized_Block_Drivers_for_Windows#Adding_a_temporary_drive

OK, I see... This may reduce the CPU usage, but will not fix the problem with the Nahelem CPU bug...

I will be back as soon as I have any news.

Thanks + regards,
Peter

 

[mythicalbox]

Hello, I've been experiencing this too.. Would simply disabling ACPI solve the problem?