Container CPU/Mem Usage Stats Not Reporting On Summary Page

[MrJester888]

New Proxmox install. Latest, fully patched. I have 5 containers, 3 are showing stats correctly. Two are showing zero for CPU and memory usage on the Summary tabs. This keeps happening and the only fix I can find so far, is to reboot the host. Restarting pvestatd doesn't help.

Is there a way to fix this without a reboot? Known issue? What would you need to identify the cause?

 

[hashime]

Have you found a solution for this? I have the same problem, reboot is not an option and i need those stats.

 

[oguz]

Have you found a solution for this? I have the same problem, reboot is not an option and i need those stats.

Have you tried restarting pveproxy and pvedaemon? Or restarting the containers themselves?

 

[hashime]

Have you tried restarting pveproxy and pvedaemon? Or restarting the containers themselves?

Does not help. Network in/out is still shown.
Interesting also: When this happens proxmox does not/can not export cpu/mem stats to influxdb for the affected container either

 

[oguz]

When did this start happening? There was a known issue when upgrading from PVE 4 to 5 where the pvestatd had problems with the cgroups and couldn't show the statistics because of that, but a reboot solves that problem. Could that be your situation as well?

 

[hashime]

When did this start happening? There was a known issue when upgrading from PVE 4 to 5 where the pvestatd had problems with the cgroups and couldn't show the statistics because of that, but a reboot solves that problem. Could that be your situation as well?

Never had PVE 4 running, it's a fairly new installation. Second time that happened now. The first time was fixed via a reboot. The second time now happened after migrating a container from one node to the other. Another Symptom is i can't change swap size for example, no matter how much swap i give the container its always 8G inside the Container, rebooting the container das not change that. Same for the CPU Core count.

 

[oguz]

Another Symptom is i can't change swap size for example, no matter how much swap i give the container its always 8G inside the Container, rebooting the container das not change that. Same for the CPU Core count.

How are you trying to make this change? GUI or config file? If config file, you should use /etc/pve/lxc/CTID.conf (and not /var/lib/lxc/CTID/config)

Never had PVE 4 running, it's a fairly new installation. Second time that happened now. The first time was fixed via a reboot. The second time now happened after migrating a container from one node to the other.

You would need to provide more verbose information for us to be able to make any comment on this. Screenshots, log files while this is happening, anything related. If it's reproducible, then please also post how to reproduce it.

 

[hashime]

How are you trying to make this change? GUI or config file? If config file, you should use /etc/pve/lxc/CTID.conf (and not /var/lib/lxc/CTID/config)



You would need to provide more verbose information for us to be able to make any comment on this. Screenshots, log files while this is happening, anything related. If it's reproducible, then please also post how to reproduce it.

I made the edits via the GUI and double checked the config files if the changes are there, and they are.
Could not find anything abnormal in logs whatsoever. I was not able to reproduce it, a reboot will fix this though. I don't know where to start to debug this tbh.
Here are 2 Screenshots, the first showing 0% cpu usage and 1.35MiB Memory usage (thats the number showing up all the time, no matter what and is obviously wrong) also you can see the swap space being different than the actual swap given, swap inside the container stays the same, always.

~# free -h
              total        used        free      shared  buff/cache   available
Mem:            21G        2.0G        961M         22M         19G         19G
Swap:          8.0G         22M        8.0G

The Second showing the graphs not showing anything: