Similarly, make your methods short - really, really short. If you have 10 lines of code and put them each in their own method, great! Maybe it isnt the fastest code to read - but readable method names will make it obvious what each line does, and combining independent methods together is far easier than breaking one massive method into chunks.

I agree with a lot you said and this reads like advice straight from Robert Martins “Clean Code”, but I’ve recently read a discussion between him and John Ousterhout, where John makes compelling arguments for longer, “deeper” functions. I found the discussion very interesting and actually started reading “A Philosophy of Software Design” shortly after. Would recommend!

I’ve always had a weird feeling about FUTO, but I thought them giving money to open source projects can’t be that bad. Now I hope they don’t fuck up Immich.

Allegedly, backups simply couldn’t be kept, due to the G-Drive system’s massive capacity.

X Doubt. Things like S3 can also store massive amounts of data and still support backups or at least geo replication. It’s probably just a matter of cost.

But it gets worse. It turns out that before the fire, the Ministry of the Interior and Safety had apparently instructed government employees to store everything in the G-Drive cloud and not on their office PCs.

Which is totally fine and reasonable? The problem isn’t the order to use the centralized cloud system, but that the system hasn’t been sufficiently secured against possible data loss.

Maybe try this: https://docs.docker.com/compose/how-tos/multiple-compose-files/merge/

The question about the single most favorite self hosted software is impossible to answer.

No, I also got the mail and it stated pretty clearly that they only increase the price of the previously free plesk license to 5€. If you don’t use plesk (aka don’t have a license as you described), the VPS will still cost 1€/mon.

Oof sorry, that sucks.

What are the reviewers of these PRs doing? Out-of-scope PRs can be rejected and closed, no?

Thank you! I’d wish Radicale would include a dedicated sharing feature, at least there’s a workaround.

How do you handle shared calendars with radicale?

Check the permissions/owner of the authorized_keys file. I’m not at home right now to give you the path to it, but I have had a similar problem after I add a new ssh key to my gitea/forgejo account. It turned out that in doing so, sometimes the permissions change and gitea/forgejo then refuses to use the file. You should see warnings about this in the logs.

In my case the problem is probably rooted within the uid/gid thats used inside the container and/or the nfs mount I use for the container volume. I never bothered to get to the bottom of it though.

$10/month just for a static website is a lot, especially with free alternatives out there.

There’s a crowdsourced list of compatible banking apps which could give you a hint. My banking app kept working.

What doesn’t work is face unlock and contactless payment.l, e.g. Google Wallet. For the latter there may be workarounds, but I didn’t care.

I did the exact same thing, ignored the AI agent update and moved to Graphene. It really isn’t hard, the Web Installer is extremely handy.

For the migration I just made a list of my apps, created backups of those that needed one, installed GOS and reinstalled my apps. Nothing special.

jfc, I really want F-Droid to succeed because we urgently need alternatives to the Play Store, but amateur errors like this and the recent story about their ancient build servers does not exactly inspire confidence in the project.

I’m thinking about just doing something outside kubernetes that just copies the data from the directory that NFS provides to another storage.

This is what I’m doing for the most part. A TrueNAS server provides the NFS shares and periodically backs them up with restic.

Some apps don’t like NFS very much, especially those that require SQLite. If you’re running Jellyfin over NFS you probably know what I mean. For those apps I use Ceph instead, which is highly available and a lot faster but also more complicated. Those PVCs I backup from within kubernetes to S3 storage with velero.

🤌🤌🤌

Uhhh, git add -p?

I use Promtail + Loki + Grafana to monitor application logs. Promtail scrapes logs, Loki stores and indexes them and Grafana can query Loki with LogQL and also send alerts.

Apparently Promtail is superseded by Grafana Alloy, which I don’t have experience with.

Anyway, I set this up mostly for fun and to preserve logs of terminated pods in my kubernetes cluster. I don’t have any alerts in place, but I probably could.

https://grafana.com/docs/loki/latest/

No, it’s permanent. They call it “VPS XS”, here (in german). Sadly a initial one-time payment of 10€ required, I forgot about that.

On ionos.com the same VPS costs $2/month. No one-time payment though.

The unique selling point of this VPS for me was the low price combined with unlimited traffic. Sometimes my nebula lighthouse needs to proxy traffic for peers that can’t talk to each other directly. It’s nice not to worry about traffic then.