The AWS failures in Virginia last night really did a job on my Stardog server. My volume with Stardog got knocked out so I had to restore from a snapshot. And I hadn't been that focused on Stardog backups yet so I only had snapshots from times where Stardog was actually running, which isn't the best time to backup any database.
So I just restored a snapshot and restarted it. It seemed to start all right but then I had lots of weird issues: the one user my apps were using to log in would not let them in. The Stardog user list command said it existed but if I tried to reset the password I was told it didn't exist. Also all my databases were gone. The existed on the drive but Stardog didn't recognize them. So I recreated everything and it worked a little bit more. Then I realized I didn't recreate my one user as a superuser, so I dropped it and then tried to re-create it but kept being told that (despite being successfully dropped) it still existed. So I gave up and just rebuilt everything from scratch (I had a few trig files to recover from).
I know the moral of this story as it's one I'm used to from MSSQL. I need to set up regular backups of the databases. In this case just schedule a regular export of each database to a trig file. But then I started thinking, what about backing up the Stardog configuration? The users, the databases, etc? Could there be a way to backup and restore all that? Similar to backing up and restore the master and msdb databases in the MSSQL world.