Upgrade problem

[tgaillard]

Hi,

I have a Proxmox cluster with 4 nodes in it. When i want to upgrade one of my node, I dispatch my VMs of the node that I want to upgrade through the other node. Then I upgrade my node and I restart it, and when I restart it the VMs that are in the other node restarts for no reason. So i wanted to know why the VMs restart even if i don't want to. Can you explain me please ?

Have a nice day.

 

[mira]

Please check the journal (journalctl) or syslog (/var/log/syslog) for any errors or other valuable information right before the reboot (~5 minutes before until the reboot).

 

[tgaillard]

Thanks for your reply but I did not express myself well.
What I wanted to say is that I notice that when my node is upgraded an re joins the cluster he sends the signal to restart the VMs, I wanted to know if it's normal or not ?
I also wanted to know what are the good practices to upgrade a node, should I delete it from the cluster in order to upgrade it an rejoint it after or is it the method that I did which is dispatch the VMs through the cluste during the upgrade and rejoin the cluster

 

[mira]

Did you look into the syslog for any information that might explain such a behavior?

This is not something that PVE does, so there has to be something else at play here. But to narrow it down, the first step would be to look into the syslog on both the rebooted node and the node where the VMs are during the time of the reboot (slightly before the reboot until after the VMs are restarted).