

As far as I know, the Jellyfin server flatpak is not made by the Jellyfin team, but by one from the community.
So I don’t see what would be stopping you from doing the same with Immich.
Linux enthusiast, family man and nerd
As far as I know, the Jellyfin server flatpak is not made by the Jellyfin team, but by one from the community.
So I don’t see what would be stopping you from doing the same with Immich.
Apple’s own suggested hardware should work, no?
Same here. My N100 box is running multiple websites/blogs, nextcloud, jellyfin, home assistant and a bunch of other small things.
If you can live without the dedicated GPU for the media center, you can get an Intel CPU instead, which have QuickSync for media transcoding. It will use a fraction of the power a dedicated GPU does with the same workload.
That’s likely the place you can gain most power effeciency.
Any caldav web app should be able to do it.
Any reason it needs to be a webapp and not a native app for your OS?
Can the peertube user actually read the mapped /dev/dri path?
You know what is running on your Pi, right? Just check the configs of those services and see if any are still using old.home.lab.
The hardware in your server should be able to handle 50-60 degrees for a long period of time, so going to 35 ambiant shouldn’t be a problem.
I tried Rcognize on my Nextcloud install, but appearently I have too few photos fot it to matter. It never started any clustering. Not even from the CLI commands. Had it running for about 6 months, then uninstalled it again as I was getting no real use from it.
Ah. It was started by a community member, but seems that a team member now vouches for it. CHange made 5 moths ago.