VM resetting after every 6 minutes

[rbhalla]

Hi All,

Proxmox novice here. I have been running proxmox for some time without issues. I upgraded it to Virtual Environment 7.4-3 recently.

One particulat VM with ID 100 keeps resetting after every 6 minutes. IF i manually stop it, Proxmoc keeps starting it automatically as well. I am not sure where to look to stop this from happening. All other VM's and LXC containers dont have this issue. I have restored the VM from a backup but the issue still exists

 

[leesteken]

That's weird. Please share the VM configuration file and some information about the operating system inside the VM.

 

[rbhalla]

Hi @leesteken,
Please see the vm config file below.



The VM has Ubuntu 20.04 LTS running on it.

Here is the Task history for the VM:

 

[fiona]

Hi,
maybe some script is triggering the resets via API? You can check in /var/log/pveproxy/access.log to see where it comes from if it comes from outside, e.g.

::ffff:192.168.16.20 - root@pam [11/05/2023:09:30:23 +0200] "POST /api2/extjs/nodes/pve701/qemu/101/status/reset HTTP/1.1" 200 83

Otherwise, you should check your cron scripts and systemd units started by timers.

 

[rbhalla]

I checked the access.log. The reset happened at 10:28:32 but i dont see any any /status/reset POST logs. i only see GET ones, which should not be the cause, right?


In the Proxmox Syslog, i do see this though:
May 11 10:28:32 proxmox qm[3774994]: <root@pam> starting task UPID: proxmox:00399A17:014B7375:645CB540:qmreset:100:root@pam:
May 11 10:28:32 proxmox qm[3774994]: <root@pam> end task UPID: proxmox:00399A17:014B7375:645CB540:qmreset:100:root@pam: OK

 

[rbhalla]

I checked the access.log. The reset happened at 10:28:32 but i dont see any any /status/reset POST logs. i only see GET ones, which should not be the cause, right?
View attachment 50268

In the Proxmox Syslog, i do see this though:
May 11 10:28:32 proxmox qm[3774994]: <root@pam> starting task UPID: proxmox:00399A17:014B7375:645CB540:qmreset:100:root@pam:
May 11 10:28:32 proxmox qm[3774994]: <root@pam> end task UPID: proxmox:00399A17:014B7375:645CB540:qmreset:100:root@pam: OK

anybody hve any ideas ?

 

[TDavLinguist]

I had this problem as a result of installing the Proxmox VE Monitor-All helper script from tteck. The instruction I missed was that any machine that does not have QEMU guest agent installed must be excluded from the list. Guess what? None of my VMs had the guest agent installed and thus they were all seen as "non-responsive" by the script and restarted . So, the solution was to uninstall that script and, as predicted, the restarting stopped happening!

This may not be relevant to you, but someone frustrated and Googling like I was may run into this issue through the same mistake I made.

 

[rbhalla]

I had this problem as a result of installing the Proxmox VE Monitor-All helper script from tteck. The instruction I missed was that any machine that does not have QEMU guest agent installed must be excluded from the list. Guess what? None of my VMs had the guest agent installed and thus they were all seen as "non-responsive" by the script and restarted . So, the solution was to uninstall that script and, as predicted, the restarting stopped happening!

This may not be relevant to you, but someone frustrated and Googling like I was may run into this issue through the same mistake I made.

Spot on. Thanks @TDavLinguist - i completely forgot that i tested this feature out recently. removed Monitor-All helper script and no more resets. Phew!

 

[bonez]

I had this problem as a result of installing the Proxmox VE Monitor-All helper script from tteck. The instruction I missed was that any machine that does not have QEMU guest agent installed must be excluded from the list. Guess what? None of my VMs had the guest agent installed and thus they were all seen as "non-responsive" by the script and restarted . So, the solution was to uninstall that script and, as predicted, the restarting stopped happening!

This may not be relevant to you, but someone frustrated and Googling like I was may run into this issue through the same mistake I made.

I signed up to this forum just to say THANK YOU!
I recently upgraded from Proxmox 7.4 to 8 and ever since then, I've had 2 Debian VM's constantly in a boot loop cycle - it's been driving me insane! I started rebuilding one of them from scratch and noticed that even during setup, it would get almost to the end and then reboot.

All fixed now that I have removed the monitor script (which I installed just after upgrading to Proxmox 8).
Thanks again

 

[RustedShader]

I had this problem as a result of installing the Proxmox VE Monitor-All helper script from tteck. The instruction I missed was that any machine that does not have QEMU guest agent installed must be excluded from the list. Guess what? None of my VMs had the guest agent installed and thus they were all seen as "non-responsive" by the script and restarted . So, the solution was to uninstall that script and, as predicted, the restarting stopped happening!

This may not be relevant to you, but someone frustrated and Googling like I was may run into this issue through the same mistake I made.

Thanks it worked

 

[m4k1]

I had this problem as a result of installing the Proxmox VE Monitor-All helper script from tteck. The instruction I missed was that any machine that does not have QEMU guest agent installed must be excluded from the list. Guess what? None of my VMs had the guest agent installed and thus they were all seen as "non-responsive" by the script and restarted . So, the solution was to uninstall that script and, as predicted, the restarting stopped happening!

This may not be relevant to you, but someone frustrated and Googling like I was may run into this issue through the same mistake I made.

Thanks, I had a similar problem and the problem was the monitoring script.

 

[s0n1c]

I had this problem as a result of installing the Proxmox VE Monitor-All helper script from tteck. The instruction I missed was that any machine that does not have QEMU guest agent installed must be excluded from the list. Guess what? None of my VMs had the guest agent installed and thus they were all seen as "non-responsive" by the script and restarted . So, the solution was to uninstall that script and, as predicted, the restarting stopped happening!

This may not be relevant to you, but someone frustrated and Googling like I was may run into this issue through the same mistake I made.

OMG.. Thank you...

I was wondering why this one VM would keep restarting.. Was pulling my hair out..


Thank You..

 

[bogos]

Spot on. Thanks @TDavLinguist - i completely forgot that i tested this feature out recently. removed Monitor-All helper script and no more resets. Phew!

I had the exact same behaviour and this solved the problem as well.

 

[Fantasenf]

THY that helped a lot