Detect container with high activity command

[luison]

This might be more an LXC question but I was wondering what are the key forensic commands when the pve server becomes unstable or with a high activity.
When running 6-10 containers, how can we have a generic view of their activity together apart from the web interface. I understand there was a vztop command that is no longer active, what are the alternatives? Thanks.

 

[ness1602]

Unstable is really broad term, is it CPU,storage,RAM stability problem. Is it internal/external(eg NFS). Depending on your answers, there are various commands to monitor. Load is really not that useful,except when it mounts to 10000.

 

[luison]

Thanks @ness1602 . The question is in general but in this case we are having CPU issues. Some other times could be RAM. The question is how to isolate it per container to at least know which is causing the issue rather that the process itself as for example we have apache in many of them or php, etc.