For newbies like me who ran into the same issue, this is what I did to get Immich working again: Stopped both the Immich and PostgreSQL_Immich containers. Changed the image of the PostgreSQL_Immich container to ghcr.io/immich-app/postgres:16-vectorchord0.3.0-pgvectors0.3.0, keeping the same container name. To do that, I edited the container in Unraid and pasted ghcr.io/immich-app/postgres:16-vectorchord0.3.0-pgvectors0.3.0 in the Repository field Then I clicked Apply and the container updated. I didn’t touch the container port or any of the volume paths (/mnt/user/appdata/PostgreSQL_Immich → /var/lib/postgresql/data ) In Immich, leave these as they are, as I don't use a Redis container but Redis inside Immich Container: DOCKER_MODS=imagegenius/mods:universal-redis REDIS_HOSTNAME=localhost After that: Start the PostgreSQL_Immich container first. Then start the Immich container. For a while the Immich container was busy reindexing all the vector indexes, and after a few minutes it was accessible again through the web UI. The Immich logs no longer show any errors and I’ve got access to all my photos and videos back. To avoid this happening again in the future, I’ve disabled automatic updates for both the Immich and PostgreSQL containers In the future I’ll look into doing updates via Docker Compose, like Immich officially recommends, but for now my setup is working again. For anyone running into the same issue: don’t panic. Your photo files are not lost; they just need to be reindexed, and you have to switch the Postgres image to a newer version that’s compatible with Immich 3.0 and above. Thanks a lot to everyone who replied in the forum and shared their input