Ouh! I have a checklist of things I need to add/update too, that I never check. Maybe we could mutualize! ;)

The theory is I use Docmost. The reality is I don’t, and I hope my backups are solid.

Just as a reference, my NUC with 40+ containers, runs at around 3-4 W, not counting the 2,5" 5400 rpm HDD attached.

I know SQL and I can pull some faraway memories of PHP to create a web interface (or Python, but I’ve only ever used it for data analysis)

But it’s a far cry between using something off the shelf and developing something from scratch, especially when on company time when it’s not supposed to be your job (and you’ll be told to use the shitty existing tools instead of taking the time necessary to create something custom).

As for Excel, sure it’s an option, but there’s also data conflicts and overwritten version and all those pesky things.

It’s not as bad as others, but not having access to the survey view is a downer for usability.

…

Plus the fact that I can’t seem to successfully deploy the v2.1.6 docker image…

I have no VPS but a home server and Uptime Kuma can’t help me if it’s down. So I just set up healthchecks.io + Pushover via a cron job that runs on the server and I can confirm that it works! Tested yesterday after a citywide blackout 😵‍💫

Simply the fact of having 70+ users. There’s also the number of records, but that may be a non issue (for now).

I was hoping it wouldn’t come to this, especially for advanced search, everything else is quite basic, indeed.

I love the fact that your example of a “small-ish” homelab is way more overpowered than mine. Because my entire 40+ containers homelab runs off of a 13 yo NUC with a 10+ yo HDD and a (new!) SSD plugged into it. I regularly want to replace it by a more recent NUC but… there is not good reason, yet.

I only set up auto-update since the stable release, but I don’t think I’ve ever had an issue since I set it up 2 years ago. Every time I read the release notes they said “go for it”, and… it worked. I guess trauma holds much longer for those who were there early ;)

Now that it’s in stable release, is it really still the case?

Well, this explains that, I’m stuck with Yattee 1.0, which seems to work for me, or at least has the same bugs as Invidious/Piped.

I used to use Yattee + Piped and it was great, until Piped stopped working on iOS devices (I tried all the fixes I found, none seems to work). And I just switched to Yattee + Invidious but it seems to be limited to 360p for me (whereas Invidious works normally elsewhere). But it’s the first time I hear about Yattee-server. What’s the point of it? What does it add to Yattee + Invidious?

I don’t know what up with your Immich instance but this seems like a lot of RAM usage for me. I have 45 docker containers, including Immich, running on a 10 yo NUC with 8 GB of RAM and that’s not an issue… I use 4,7 GB total in this instant.

I can’t seem to prove I’m not a robot on the selfhosted docker install… shame. I’m stuck at “verifying…”

Yeah I don’t think any non public instance has any chance of being blocked. It simply looks like one user among millions to them.

I still use Pinchflat. It doesn’t exactly do as I like, but I’m mostly just downloading two channels, so it’s okay enough for me.

I haven’t found how to connect Filen’s S3 to BackRest… the only thing keeping me from using the 110 Go I have with them to double backup my local backup on another drive.

Pinchflat does just that…

Never tried Yunohost, but I use the docker container Transmission-OpenVPN to have a torrent client, and only a torrent client, connected to a VPN. I don’t like my IP appearing anywhere and everywhere either.