[6.10.2] Server hangs (with high load)

You should post the diagnostics grabbed when the issue is occurring.

Just updated to 6.10.3 and in this version my problem seems to be resolved.
I tested by running simultaneous:
- duplicati backup to onedrive
- running Parity-Check
- streaming a video

- and some other high I/O processes.
 

Changed Status to Solved

Kind of unexpected that just updating to v6.10.3 helps with this, but good news anyway.

I thought everything was working as expected, but just a few minutes ago all load went up again and a there were a lot of  processes in uninterruptible sleep. After stopping Docker (/etc/rc.d/rc.docker stop) all load went down again.
I had just a moment to watch docker stats, but there were no dockers running in high CPU.
I was able to create a diagnostic which can be find underneath.

I downgraded back to 6.9.2

 

optimus-diagnostics-20220616-0035.zip

Edited June 15, 2022 by capino

Changed Status to Open

Server load average is very high, shfs seems to play a big part of that, unfortunately and AFAIK not quite clear what causes this for some users.

 

Depending on where you have the data for the worst offending dockers any chance you can change the mappings to use disk shares instead of user shares? E.g., lets say the data Duplicati is syncing is on a pool or a single disk, change the mappings form "/mnt/user/share" to "/mnt/pool_name/share" or "/mnt/disk#/share" and test.

I already use "/mnt/pool_name/share" for most configurations in Dockers.
Except for systems where the data is fragmented over the Disk array.

Last night I had stopped the Duplicati docker.
So this could not be the docker that creates this problem.

I looked like the problem last night started at the moment the "Auto Update Applications" app did a check on available updatable dockers. The auto update had not started yet.