As per title, suddendly after a variable x amount a time i found my ram full and a reeeeally slow access to webgui and always time-out access to docker app.
When i finally access the webgui i found the cpu almost at 100% and ram 97% full ( 64gb ) with >40 gb from docker ( usually i got like half ram free ).
Unfortunately the docker page times out so i have no way to see what's causing this -  same for "docker stats" command on cli. 
Note that i found strange: when run btop on ssh-terminal my cpu is at 20-30%, completely different from what the webgui it's showing? It's not the first time these diverge, but not by so much. ( actually they've never been comparable even with similar refresh timing )

So how can i solve this? what's causing it? -->I made a diagnostic log(unraid-diagnostics-20250319-1543.zip) via cli and i added the docker logs that looks incredibly suspicious(docker.log.1.txt). If you need something else let me know.

PS: Do you know why the cpus utilization diverge? Always been curious about that.

unraid-diagnostics-20250319-1543.zip

docker.log.1.txt

Edited March 19, 20251 yr by lory995
Added docker log

58 minutes ago, lory995 said:

when run btop on ssh-terminal my cpu is at 20-30%, completely different from what the webgui it's showing?

GUI includes i/o wait.

 

Failing a better option, you can try to find the problem container by running the server with only half of them enabled, if no issues, try the other half, then keep drilling down.

Jorge i'm grateful for your answer but... Is there really no other way? i got 50 docker, even if i got it in the first half, then i have to split it in half again and again, and considering it takes between 1 and 2 weeks on average for the problem to show itself, it becomes a life quest.... 
And that strange docker log is normal? 

Since there's nothing abnormal logged in the syslog don't know of a better way, maybe someone else will, and the docker log looks normal to me.

Is there anything looking really out to lunch if you look at Docker, Container Size