I have run into a problem several times now where one of my backups will stall. When this happens none of my other scheduled backup jobs will run until I either kill the hung job or just restart the Duplicati service.
This is a situation that requires me to be extremely vigilant about checking if backups were successful and manually correcting if one is hung. I’d like to see the software handle this more gracefully so I have 2 questions.
Is there a setting I’m not seeing to set a maximum execution time on a job so it is automatically killed?
Is there a setting to enable multiple jobs to run concurrently?
I wish I had meaningful logs to send but the job logs aren’t interesting and I don’t see anything for the date this hung up in the server logs.
There should be a timeout on the transfers, but I think this is set to 5 minutes. With 5 retries, this gives around 25min of stall before the backup fails.
No, there are no options for these two. The first should be do-able, the second requires some work to display multiple progress bars.
You can perhaps do this is you forgo the UI and schedule jobs via the CLI and crontab.
If you go to the “Log” and choose “Live” and “Profiling” you should be able to see what exactly is happening during the stall.
I’m experiencing the same situation on Duplicati 2. Since not a lot has changed since then I’m continuing this conversation. I have many cases that my backup end point is going offline during backups and this causes the backup to stall (not fail as would be preferred). I’ve captured a screenshot of the “Live” logs to give you an idea. I took this screenshot at 9 AM on the next day. This backup stalled on the 6 PM the previous day already.
The stalling occurs in what I can see mostly when my storage destination goes offline or becomes unreachable. This happens ONLY during the Verify stage and during the Deleting Old Backups stage.
In the cases where the storage destination goes offline, during the backup phase, it fails in a natural way with an error that the “destination is unreachable” error message.
I use an FTP destination over a Point to Point VPN connection. The connection goes down regularly, so this is the point that causes Duplicati to struggle during these phases.
It would seem as if the code that handle the connection during the Verify and Old Backup Delete phase are not so robust (good at handling connection issues) as the code during the backup phase.
