Database corruption after cleaning up music filesystem

Over the past few months I have been cleaning up duplicate and missing files in my music library (Stored on a 20TB Synology drive). I have noticed a pattern where if I change around a hundred files or more, the next time the roon server scans the filesystem it will not throw an error, but it will become unresponsive and then start presenting a ‘restore from backup’ message to clients attempting to connect.

I restored from backup the first couple of times this happened. Then I discovered that just restarting the roon server process causes it to unf*k the database and start serving up music again. The Roon Server is running on an intel NUC with 32GB of RAM and 8+ cores. The Roon server rarely consumes more that 15% of the CPU (even during startup/rescans) and barely shows up in network traces between the NUC and the Synology NAS (over a dual 2 Gbit Ethernet).

Any ideas on what the issue might be? Tweaks, fixes or workarounds?

Are you performing these cleanups while the Roon Server is running?

1 Like

Sadly it does not. It only allows you temporary access to your f*ked-up database. You need to actually troubleshoot and fix the issue. There is a guide that might explain a few things:

“Troubleshooting” = Reboot and/or reinstall? I feel like I am using a Microsoft product…

I’ll give it a shot…

2 Likes