What's a reasonable size for the config volume?

[scubbo]

I'm running Jellyfin on Kubernetes via a homemade Helm chart. I recently switched from mounting /config directly from an NFS server, to using a size-constrained democratic-csi-provided volume, which prompted me to realize something surprising - I've repeatedly run into issues where the config volume has filled up and I've had to increase the size of the volume to get the service working again.

With 220 Movies and 116 TV Shows, my config volume is currently using 88% of a 5Gb mount, with 3.8G of that in /config/metadata. Is that a normal usage? If not, where should I look for possible issues?

[Efficient_Good_5784]

Please give the config storage as much space as you can afford. Running out of space for it while your server is running is a easy way to corrupt Jellyfin's database.

It's normal for it to keep balooning in size the more your server grows. The more media you scan into your server and the more metadata that's pulled into it, the more your config directory grows in size. There's metadata folders there. The DB also stores lots of that data too.