Backup Problem

[SaNkA]

I'm back, this morning the server was frozen. It occurs on saturday morning during the backup of the 2nd qemu vm.
I do my backups on a mount of a windows share folder on another server with a big storage. I have nothing on the log, no reboot, the server just freeze and i can't do anything. I can just push the power button to switch it off. On my Nagios graphs i can see that occurs once a week.

Sorry for my bad english

 

[martin]

I'm back, this morning the server was frozen. It occurs on saturday morning during the backup of the 2nd qemu vm.
I do my backups on a mount of a windows share folder on another server with a big storage. I have nothing on the log, no reboot, the server just freeze and i can't do anything. I can just push the power button to switch it off. On my Nagios graphs i can see that occurs once a week.

Sorry for my bad english

vzdump logs, syslog?

 

[SaNkA]

nothing on the log! like if the server switched off but still powered on! So i don't know why it appends...

Vzdump log :

Nov 29 05:12:36 INFO: Starting Backup of VM 102 (qemu)
Nov 29 05:12:37 INFO: status = running
Nov 29 05:12:37 INFO: creating lvm snapshot of /dev/mapper/pve-data ('/dev/pve/vzsnap')
Nov 29 05:12:38 INFO:   Logical volume "vzsnap" created
Nov 29 05:12:38 INFO: mounting lvm snapshot
Nov 29 05:12:39 INFO: creating archive '/mnt/backup/vzdump-102.dat' (/mnt/vzsnap/images/102)

Syslog :

Nov 29 05:13:48 localhost snmpd[3042]: Connection from UDP: [10.1.4.13]:50105 
Nov 29 05:13:49 localhost last message repeated 43 times
Nov 29 05:13:56 localhost snmpd[3042]: Connection from UDP: [10.1.4.13]:46448 
Nov 29 05:13:57 localhost last message repeated 11 times
Nov 29 05:14:08 localhost snmpd[3042]: Connection from UDP: [10.1.4.13]:50166 
Nov 29 05:14:09 localhost last message repeated 123 times
Nov 29 05:15:01 localhost /USR/SBIN/CRON[7283]: (root) CMD (/usr/share/vzctl/scripts/vpsnetclean)
Nov 29 05:15:01 localhost /USR/SBIN/CRON[7284]: (root) CMD (/usr/share/vzctl/scripts/vpsreboot)
Dec  1 08:25:05 localhost kernel: Initializing cgroup subsys cpuset
Dec  1 08:25:05 localhost kernel: Linux version 2.6.24-1-pve (root@oahu) (gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)) #1 SMP PREEMPT Fri Oct 24 11:34:13 CEST 2008 (Ubuntu 2.6.24-4.6-server)
Dec  1 08:25:05 localhost kernel: Command line: root=/dev/mapper/pve-root ro

 

[dietmar]

What kind of disk subysystem do you have - RAID?

Whats the output of:

# pveperf

 

[SaNkA]

yes RAID 5

proxmox:~# pveperf
CPU BOGOMIPS:      17027.26
REGEX/SECOND:      762523
HD SIZE:           33.47 GB (/dev/pve/root)
BUFFERED READS:    66.37 MB/sec
AVERAGE SEEK TIME: 6.10 ms
FSYNCS/SECOND:     390.38
DNS EXT:           167.79 ms
DNS INT:           9.57 ms (____.com)

 

[dietmar]

FSYNCS/SECOND is a bit slow (do you have battery backed cache and enabled write back mode?)

Anyways, dont think thats a problem. Maybe the snapshot itself run out of space - try vzdump with '--size 2048' (allocate 2GB for the snapshot).

But usually such behaviour is an indication that there is something wrong with the hardware, but thats just a guess.

- Dietmar

 

[SaNkA]

thanks Dietmar, i will try with '--size 2048' if it doesn't work, i will check my disk but the server has 2 months...

 

[dietmar]

Also do some memory (RAM) tests.