clusternode freeze during vm-load (openvz postgresql)

U

udo

Distinguished Member
Proxmox Subscriber
Apr 22, 2009
5,981
203
163
Ahrensburg; Germany
Hi all,
i had just an freezed clusternode during an postgresql-db-import in an openvz-vm.

The first import breaks, and the second gives a lot of duplicate key value (that's normal because i don't drop the database before).
Suddenly the error-messages stop and a top on the pve-host show an high load:
Code: 
top - 13:38:25 up 7 days, 21:11,  1 user,  load average: 6.02, 1.57, 0.54
Tasks: 213 total,  13 running, 200 sleeping,   0 stopped,   0 zombie
Cpu(s):  0.1%us, 75.2%sy,  0.0%ni, 24.5%id,  0.1%wa,  0.1%hi,  0.1%si,  0.0%st
Mem:   7914084k total,  3173840k used,  4740244k free,    85948k buffers
Swap:  7340024k total,    25892k used,  7314132k free,  2470876k cached

    PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                                                
   1820 root      20   0  158m  23m 3908 R  100  0.3  14:49.97 pvestatd                                                                               
 661045 root      20   0  185m  34m 5472 R  100  0.5   0:26.23 pvedaemon                                                                              
 665985 root      20   0  183m  32m 5304 R  100  0.4   0:50.48 pvedaemon                                                                              
    901 root      20   0     0    0    0 S    1  0.0   3:16.65 kjournald
The load rise up to 7.33 (nine second later) and the whole node freezed. After reset there was nothing in the logs...

Version:
Code: 
pve-manager: 2.0-7 (pve-manager/2.0/de5d8ab1)
running kernel: 2.6.32-6-pve
proxmox-ve-2.6.32: 2.0-46
pve-kernel-2.6.32-6-pve: 2.6.32-46
lvm2: 2.02.86-1pve1
clvm: 2.02.86-1pve1
corosync-pve: 1.4.1-1
openais-pve: 1.1.4-1
libqb: 0.5.1-1
redhat-cluster-pve: 3.1.7-1
pve-cluster: 1.0-9
qemu-server: 2.0-2
pve-firmware: 1.0-13
libpve-common-perl: 1.0-6
libpve-access-control: 1.0-1
libpve-storage-perl: 2.0-4
vncterm: 1.0-2
vzctl: 3.0.29-3pve2
vzdump: 1.2.6-1
vzprocps: 2.0.11-2
vzquota: 3.0.12-3
pve-qemu-kvm: 0.15.0-1
ksm-control-daemon: 1.1-1
After reboot i do the same - but now the database was imported without trouble.
How can it happens that pvestatd and pvedaemon block the system?
I guess it's not an io-problem, because the processes has an high cpu-usage - if they wait for io it should look different.

Udo
 
can you test with the latest kernel (pve-kernel-2.6.32-6-pve: 2.6.32-52)? just apt-get update/upgrade
 
oh, just noticed that its not yet in the repo. we try to push this to the public soon.
 
You must log in or register to reply here.
Top Bottom