[4.1.0] [Docker] [Unraid] Auto Resume not triggering

[Eiscreme]

Hello,

I just switched from nzbget to sabnzbd and am in the middle of setting it up to my liking.
My setup is as following:

Code: Select all

Minimum Free Space for Temporary Download Folder = 10G
Minimum Free Space for Completed Download Folder = 100G
Temporary Download Folder = /downloads/inc
Completed Download Folder = /downloads/done

The Temporary Download location is on a separate SSD to spread out R/W operations.



Pausing works fine, Resume sadly not



I just came back to it once more to find everything paused, while it was stopped 7hrs ago. The Temp folder has 110GB free according to the little banner on top, hence sabnzbd knows there's enough free space, but the cron for the auto-resume doesn't seem to run successfully (or at all)?



I checked the log file, but only see the

Code: Select all

Disk full! Forcing Pause

line, nothing about resume. Only messages like

Code: Select all

Resetting bad trylist for job 

but no errors or similar.

I tried unticking, saving, reticking, but that didn't fix it. I also tried restarting sabnzbd from within the Web UI (which resumed the downloads, but only because of the restart)

System Information:
Unraid 6.12.6
LinuxServer sabnzbd Docker Container
Version: 4.1.0 [b7e3401]

[safihre]

Have you enabled +Info or +Debug logging?
In the log you should see after the Disk full a line that says:
"Will resume when X has more than X GB free space"
Do you see that?

[Eiscreme]

I've changed the Log level to +Debug (default was on +Info), I also tried moving the complete and incomplete location to the same disk.

Code: Select all

2023-12-23 20:10:18,005::ERROR::[assembler:105] Disk full! Forcing Pause

Nothing regarding resume or similar. You can find the full Debug log here:
privatebin dot net/?4e0f1225dae599a3#9gYW7h1X2zCMoEJqiSe3nhWemd5c4dbEZxDVxnzvvVyF

complete_dir = /downloads/inc
download_free = ""
complete_dir = /downloads/done
complete_free = 10G
fulldisk_autoresume = 1

Looks correct to me, this is where complete_dir and complete_dir are on the same disk.

Edit:
I'm currently testing a fresh config with the Binhex Docker version

[safihre]

Ah now I understand.
The error you are seeing is when we really ran into a fatal crash because there was 0 space left. It doesn't even allow us to open a new file for writing.
We only support auto-resume for when we stop a bit before we hit the absolute limit.
I would suggest you set `Minimum Free Space for Temporary Download Folder` to something like 250M.