[SOLVED] TrueNAS VM crashing every 5 minutes

[Errol-Hassall]

Hey everyone.

I'm not sure if this counts as a Proxmox issue, but I wasn't previously having this issue running TrueNas baremetal so I think it might be a hypervisor issue. However, every 5 minutes with TrueNas simply idling it will crash sending a stop signal to Proxmox. It consistently will crash every 5 minutes, previously this happened with TrueNas core as well, so I swapped to Scale, I did a 300GB transfer and it was fine, then it started happening again.

I couldn't find a lot online that matched what I was facing however I have tried the following:
- tried other VMs, they don't crash
- LXCs are fine, no issues at all
- one thread suggested to change the CPU for the VM from host to kvm64, this didn't help
- I have tried increasing the base amount of ram from 8 to 12gb, that didn't do anything
- updated to the lastest version of truenas

I have attached the logs at the time of the crash, but I don't quite understand what the logs are really telling me. As far as I can tell, its calling the stop command on VM 301 which is the TrueNas vm, which then stops and reboots.

System specs:
- i5 2500k with an overclock (this has been stable for over 10 years) (2 cores for vm
- 16gb of ram (12gb for vm)
- 32gb passed in virtual drive with 2 2TB HDD passed in directly for ZFS to work on baremetal

Given that this is quite an old cpu with out IOMMU support, could this play any role? I'm not passing any PCI devices so I would suspect that is fine. I have been stuck on this for quite some time now and I'm at a bit of a loss on how to debug this, being quite new to Proxmox. So if there are any tips on how to debug these issues that would be greatly appreciated too.

Below are the settings for this VM



Any help would be greatly appreciated.

 

[leesteken]

The logs indeed shot a controlled shutdown of VM 301 and a start of VM 301; no crashes or problems.
Did you happen to install some third-party software to monitor the VM 301 but it does not see it running and restarts it automatically after 5 minutes, and forgot about it? This may sound silly but you'd be surprised how many threads there are about my VM is restarting all the time after 5 or 6 minutes and it's because of something like that.

 

[Errol-Hassall]

Hey leesteken

You were exactly right, I hadn't followed all the steps on https://tteck.github.io/Proxmox/ "Proxmox VE Monitor-All". For truenas I wasn't using the guest agent and thus it just rebooted.

If anyone finds this in the future, simply update this file `nano /etc/systemd/system/ping-instances.service`

adding the VM id to this line `ExecStart=/usr/local/bin/ping-instances.sh` for example

`ExecStart=/usr/local/bin/ping-instances.sh 301` would exclude 301 from the monitoring.

This is only required if you don't install the qemu guest agent to a VM

 

[leesteken]

Please edit your first post in this thread and select Solved from the pull-down menu, to help people who do use the search.