immich v1.124.0 released
(github.com)
from petsoi@discuss.tchncs.de to linux@lemmy.ml on 08 Jan 10:25
https://discuss.tchncs.de/post/28221895
from petsoi@discuss.tchncs.de to linux@lemmy.ml on 08 Jan 10:25
https://discuss.tchncs.de/post/28221895
High performance self-hosted photo and video management solution
threaded - newest
Pretty decent piece of software. Been using it for a while and it hasn’t really had many issues. Got it set up with my synology NAS and I like it.
I love it. I’m going to be deleting my Google Photos account soon.
Did you try Synology Photos? If so, how does it compare?
Synology photo is extremely slow and messy as hell. It keeps uploading photos as duplicates. It also kills the battery on my phone.
Thank you for mentioning what it is/does. Too many announcements I see don’t do that.
I love to see it evolving but I also wish it was stable enough to be auto-updatable
Apparently the stable release is planned for the first half of 2025, so you may not have to wait too much longer!
It is, I’ve been auto updating it for almost 2 years. Just needs a quick fix every now and then.
Obviously make sure your backups are working properly.
Still requires docker?
What else would you want to run it on?
It’s only ever gonna be incredibly niche if it requires some arcane programmer shit to run.
I want to use it but getting into docker is silly.
I would consider docker very easy to use and well worth exploring. The benefit of Docker is that it runs exactly the same on any system, no matter what the underlying OS is.
Here’s how difficult it was to set up Immich:
docker-compose.ymlfile, which describes the options for the Immich server.docker-compose up -dThat’s it. Immich is now installed and running.
Same here. I just installed Proxmox on an old machine, then Coolify, boom… Docker ready without the knowledge. There’s a ton of apps one click away, immich included.
lol docker ain’t that.
Lol docker is literally the easiest and most user-friendly server program administration method… It is literally one user-readable configuration file and everything is automatic.
Vm’s are more complicated and have you even tried managing many services on bare metal with conflicting libraries, database versions, etc…? That is truly arcane arts of programming scripts.
It would be the exact opposite if it wasn’t using docker, it requires specific minor versions of multiple software, people on Debian would always get versions too old while people on arch too new
I think you got docker mixed with something else, since docker does the exact opposite, i.e. allows you run services without all of the arcane shit involved. Just put the compose file in a folder and run
docker compose up -dand you’re done, whereas the alternative would be to install a database, configure it, install the immich service, connect it to the database, write a service file for both database and service to allow it to be auto-started, and face multiple issues due to missing dependencies or permissions.Not to mention there are so many more ways to fuck up security when configuring it all on your own outside a container.
Edit: of course one can also fuck up security with a container
Absolutely, and in most cases (while one shouldn’t rely on this since escaping a docker container is sometimes doable) even if there is a security breach the invader would be limited to the service image, which would get recreated upon restart.
I really need to take the time to
The first big import will always take time if you have a lot of photos, as will the associated machine learning processing. I’d say you don’t need a GPU like that for normal use though.
Does it mean I would do the first import on desktop, including ML, then I could switch the resulting data (via e.g Docker) to a RPi5 if I get <100 new photos per day?
You can also just do the initial import on a NAS. It might take a day perhaps but, honestly, does that matter?
I’m not sure I understand. Whatever machine is hosting the Immich server will be doing all the processing. Uploading to there will go as fast as your internet connection/storage speed can go, and ML processing will be based on the processing power of the server. You can also manage the ML tasks if you want (Administration > Jobs and/or Settings).
By default yes but it’s possible to do remote ML, cf immich.app/docs/guides/remote-machine-learning so my “ideal” setup would be all on RPi5 BUT initial ML tasks (because it’s going to be > 100GB of photos at first) either directly on my PC with GPU or remotely (cf guide).
Ah, yes that’d probably be a good idea.
Community still locked to proprietary MS GitHub + Discord?
No, they can quit whenever they want, they just don’t want to.