VM's not booting up

[kaupaproxmox]

Hello,

We have been using proxmox for over a few years, it went fine up until now.
There is a dell server on linux with about 8 VM's running on linux and windows. Proxmox 4.1-1
There was no recent changes in SW or HW.

Now all of the sudden random machines started to give this error:
/bin/nc -l -p 5900 -w 10 -c '/usr/sbin/qm vncproxy 2>/dev/null'

or when trying to shutdown : VM quit/powerdown failed - got timeout

Tried to restore a backup on one, it started to work(for now at least), but did the same for the other - no effect.
Sometimes VM boots, but freezes.
I atached few errors i picked up from machines.
Tried to restore backup to VM, restart proxmox server, update proxmox server's OS.

If I run VM with 1x core 1x socket it doesn't give an error in proxmox console, VM fully started once but froze soon, but now gives errors at boot in VM's console.
If I run VM with normal hardware setup - like 4x cores 4x sockets, it always gives proxmox console errors and in VM.

There's a hunch that problem is in hardware, in all of those errors i've seen another one about CPU problems

Any advice would be appreciated

Thank you

 

[LnxBil]

Most of the errors you show are inside your guest. Proxmox is (most probably) not responsible for that. The Apache und Lighttpd problem is simple: You installed two webservers and of course they cannot work on the same port (but totally not a Proxmox problem).

After updating to the newest Proxmox packages, did you restart all your VMs and Proxmox Server?

 

[kaupaproxmox]

The problem is that all VM's worked fine, and had all the ports and everything for them as needed, this is just an outcome from multiple VM's console during boot after something happend and crashed everything. All separate VM in proxmox started to do so, even after backup, which should have fixed any config problem in VM.

 

[LnxBil]

The problem is that all VM's worked fine, and had all the ports and everything for them as needed, this is just an outcome from multiple VM's console during boot after something happend and crashed everything. All separate VM in proxmox started to do so, even after backup, which should have fixed any config problem in VM.

"Port in use" is not a Proxmox problem, it's a guest problem.

Please answer my question about upgrade.

 

[kaupaproxmox]

Server was restarted, and therfore all of the VMs

 

[LnxBil]

Next question:

Did you update to the newest packages? If not, please upgrade and report back if the problems still exist.

 

[kaupaproxmox]

dist-upgrade was done
I moved few backups to other fresh proxmox machine, now some backups of same VM shows more or less the same errors and some of them boots, but shows proxmox console error /bin/nc -l -p 5900 -w 10 -c '/usr/sbin/qm vncproxy 2>/dev/null'

Could this be corrupted data in hard disks and therefore corrupted backup files?

tried backups of a few days or a month on fresh proxmox machine and most of them doesn't work, even though there was no problems up until all crashed.

 

[LnxBil]

Could be, please check /var/log/messages for any hints.

 

[fireon]

Do also an Filesystemcheck if you are using ext4 this works: In Host first then VMs.

touch /forcefsck
reboot