High cpu usage of node_exporter?

Support
1

NethServer Version: 8
Module: your_module

Hello altogether,

I noticed using top in the console of one of my NethServer 8 servers, that the process node_exporter uses lots of CPU. There is always a percentage of between 70 and 75%. That leads to a load average of 40+ while having an 8 core CPU in that server.

I tried to:

  • disable the node_exporter service with systemctl, it’s “inactive (dead)”.
  • podman ps -a showed my a container with the status “stopping”… which I killed with podman kill…
  • then tried to kill the process directly (without and with -9).

Nothing helped, the process is still there… I’m hesitating to reboot the machine…

Is there another way to get rid of that? Do I need it really?

Thanks for any hints or where to read… tried to look in the NS8 documentation, but didn’t really find a good place to start regarding that process.

Thank you in advance and best regards,

Frank

2

Solved by reboot

1 Like
3

Sorry for the late reply. Some commands to troubleshoot a case like this:

journalctl -u node_exporter
podman stats
systemctl status node_exporter
systemctl restart node_exporter

Restart may take some time. When the operation times out, it should send SIGKILL by itself to the right processes.