Scheduled jobs defined in duplicati not running on fedora 29

Support
1

I set up the duplicati service to start at boot.

ps ax | grep duplicati
1041 ? SNsl 0:00 DuplicatiServer /usr/lib/duplicati/Duplicati.Server.exe
1216 ? SNl 0:01 /usr/bin/mono-sgen /usr/lib/duplicati/Duplicati.Server.exe

sudo systemctl status duplicati
● duplicati.service - Duplicati web-server
Loaded: loaded (/etc/systemd/system/duplicati.service; enabled; vendor prese>
Active: active (running) since Sun 2019-06-02 21:55:25 PDT; 1h 21min ago
Main PID: 1041 (mono)
Tasks: 14 (limit: 4915)
Memory: 66.2M
CGroup: /system.slice/duplicati.service
├─1041 DuplicatiServer /usr/lib/duplicati/Duplicati.Server.exe
└─1216 /usr/bin/mono-sgen /usr/lib/duplicati/Duplicati.Server.exe

But the tasks scheduled within duplicati do not run unless I start the duplicati gui with /usr/bin/duplicati or running duplicati from the menu. Then the overdue job runs.

In the backup job definition, automatically run backups is checked. Under allowed days, all days are checked.

What piece am I missing that is causing the job to not run automatically? I know that I could set up a cron job to run the backups but it seems like the duplicati scheduler was created to do this.

Duplicati looks like a great multi-platform backup system!

Thanks!

2

The systemd service is likely using configuration in /root/.config/duplicati while it starts using ~/.config/duplicati when you start duplicati manually.

This would cause two split configurations to be available, one for root and one for your user.

3

What would you suggest for a solution? Should I link the root duplicati config to my user config? Or should I run the service from cron or something as my user?

Thanks!

4

It sounds like you want it to run as your user. That way permissions will work and you’ll retain your current configuration.

You can add the “user=name” configuration to your systemd service.