VMs randomly closing.

[Script20]

Hey guys, so I'm new to proxmox so if you require anything you'll need to let me know how to get the info you require.

So I have proxmox server 7.4-3, and I have a ubuntu Desktop VM with a Minecraft server running, and a TrueNAS Core Server and a Portainer Ubuntu Server.

Randomly through out the day all the vms will close, I do not tell it to do it and it's obviously very annoying as people tell me the MC server is down and I have to load it up every more then 3 times a day.

In the bottom tasks part it doesn't say anything about them turning off so I'm not sure.

 

[rbe]

What does your host RAM usage say? If the VMs are really stopped, it might relate to oomkiller events.
Check journalctl and see if you can find events around the time where the VM stopped.

 

[Dunuin]

What does your host RAM usage say? If the VMs are really stopped, it might relate to oomkiller events.
Check journalctl and see if you can find events around the time where the VM stopped.

If cat /var/log/syslog | grep oom is returning anything you should assign less RAM to your guests or buy more RAM. Or limit the ARC in case ZFS is used.

 

[Script20]

Hello, so nothing shows when I'm grepping oom, I have 64GB, and it's strange so when I look on the summary part of the VM it will be like 96 ish % but when I go in the vm and free -h it shows so much less ram usage then the summary.

 

[leesteken]

Hello, so nothing shows when I'm grepping oom, I have 64GB, and it's strange so when I look on the summary part of the VM it will be like 96 ish % but when I go in the vm and free -h it shows so much less ram usage then the summary.

That's probably filesystem cache. See the many, many threads about this on this forum.

Randomly through out the day all the vms will close, I do not tell it to do it and it's obviously very annoying as people tell me the MC server is down and I have to load it up every more then 3 times a day.

In the bottom tasks part it doesn't say anything about them turning off so I'm not sure.

Unless you see the actual shutdown tasks in the Proxmox GUI task list and the VMs are shutting down gracefully, something is indeed going wrong with the kvm processes (or inside the VMs, check their logs also). Please search journalctl about possible clues as to why the VMs are being killed or crashing.

 

[Script20]

That's probably filesystem cache. See the many, many threads about this on this forum.

Unless you see the actual shutdown tasks in the Proxmox GUI task list and the VMs are shutting down gracefully, something is indeed going wrong with the kvm processes (or inside the VMs, check their logs also). Please search journalctl about possible clues as to why the VMs are being killed or crashing.

Hello, I looked in the journactl and grepped stop, and I see

Jun 03 16:14:04 prox pve-guests[28342]: <root@pam> starting task UPIDrox:00006EB7:000300B0:647B58BC:stopall::root@pam:
Jun 03 16:14:04 prox pve-guests[28343]: all VMs and CTs stopped
Jun 03 16:14:04 prox pve-guests[28342]: <root@pam> end task UPIDrox:00006EB7:000300B0:647B58BC:stopall::root@pam: OK

What is that and how do I stop it, I think that's whats doing it.

 

[leesteken]

Jun 03 16:14:04 prox pve-guests[28342]: <root@pam> starting task UPIDrox:00006EB7:000300B0:647B58BC:stopall::root@pam:
Jun 03 16:14:04 prox pve-guests[28343]: all VMs and CTs stopped
Jun 03 16:14:04 prox pve-guests[28342]: <root@pam> end task UPIDrox:00006EB7:000300B0:647B58BC:stopall::root@pam: OK
What is that and how do I stop it, I think that's whats doing it.

You usually get that when the Proxmox host is shutting down (or rebooting). This is normal if you or someone else did that.

Hello, I looked in the journactl and grepped stop, and I see

Sorry but you really need to look a little more yourself (around the time that your VM(s) are stopping or crashing). Just grepping words won't really help because we don't know what is happening.