Proxmox is unstable?

[jdcm]

Hi friends,

when I clone the VM, (sata or virtio, qemu or raw), sometimes the server up load average to 11... and crashes, the rest of VM fall.

The server is a Dell 1950 with 2 CPU and 16 GB RAM, proxmox 3, the machine is new.

any special configuration?


Regards!!

 

[tom]

provide details about your physical storage, I assume you are working with local storage.?

(raid controller, hard disks, cache-settings, file system, raid setup, ...)

and run our simple IO benchmark, pveperf (but only if the server is idle)

 

[jdcm]

Hi Tom,

the configuration is this:

Cluster with 2 nodes Proxmox 3. (Dell 1950 2CPU and 16GB RAM).

This a partition table in both servers:

Raid:
Raid 1 by Hardware (Perc Controller).

Storage:
Is a NFS shared with both servers:
root@proxmox2:/# df -h
Filesystem Size Used Avail Use% Mounted on
udev 10M 0 10M 0% /dev
tmpfs 1.6G 336K 1.6G 1% /run
/dev/mapper/pve-root 95G 2.5G 88G 3% /
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 3.2G 41M 3.1G 2% /run/shm
/dev/mapper/pve-data 791G 46G 746G 6% /var/lib/vz
/dev/sda1 495M 34M 436M 8% /boot
/dev/fuse 30M 24K 30M 1% /etc/pve
172.16.0.2:/var/lib/vz/storage-proxmox2 791G 46G 746G 6% /mnt/pve/storage-proxmox2
172.16.0.1:/var/lib/vz/storage-proxmox1 791G 156G 636G 20% /mnt/pve/storage-proxmox1

Connected directly with a Gb nic.


I have been tested with multiple cloning, migration, etc, the CPU load up to 3 or 4, and the VM are not aware. All perfect!! Proxmox is incredible, not fails!

But, sometimes, the average gets too high and crashes the VM and the CLI (https://IP:8006), with 11 or 12 load and never drops... The solution: Force reboot. <-- This is a problem... The node is unstable when I clone the VM (on the same node or another node, is indifferent)


Thanks!