High CPU Usage since 11/1/22 or 10/29 Update

[GreatHolyCow]

Been getting higher then usual processor usage with crashplan doing nothing no new files and "logmonitor" process in crashplan keeps around 100-120% cpu usage. If I stop the container cpu goes back to idle.

[ddr2108]

Can confirm this as well. Since upgrading on 10/29, I see cpu high usage on idle and nothing is actually backing up

[GreatHolyCow]

I thought I upgraded on the 1st, now I think I also got this issue back on 10/29 after looking at my historical graphs It started then too. It just got worse after 11/1 it was less sporadic but now its more consistent

[jlesage]

Do you have a lot of activity in service.log.0? You can try to run tail -f /config/log/service.log.0 inside the container.

[GreatHolyCow]

I saw it kept talking about that log file but I tried to tail it and only saw log entries when I added/modified a file that it was monitoring/backing up. If there is no backup activity log is dead even with spikes of cpu usage.

[jlesage]

Can you provide more details about your setup? To which filesystem on the host /config is mapped to ?

[GreatHolyCow]

/config is mounted to the os nvme ssd ext4.

[cgbale]

I'm experiencing the same issue. I didn't have resource limitations enabled on the crashplan container and it brought my system to a halt. It's an older Synology 1812+ with 3 GB RAM. It's only been happening the past week after the newer updates. Swap utilization also seems to be very high (synolgoy kswadp0 process).

The only read-write file system access is "docker/crashplanpro" being mapped to "/config". (None of the read-only backup folder are mapped from synology Volume1)

[jlesage]

I think I found the problem, a fix is in progress.

[GreatHolyCow]

I think I found the problem, a fix is in progress.

Will continue to monitor but with the update seems the initial startup spike then idle back down to practically zero when nothing to backup.

[jlesage]

With the latest image, CPU usage should be the same as with older versions.