I haven’t looked into this, but in step 3 can you call the editor’s undo function?

You mean like this?

What about simply shelling out to ripgrep?

These devices have been recommended in the past, and it looks like they can run OpenWRT

https://www.amazon.com/GL-iNet-GL-SFT1200-Secure-Travel-Router/dp/B09N72FMH5

https://openwrt.org/toh/gl.inet/start

Bash and a dedicated user should work with very little effort. Basically, create a user on your VM (maybe called git), set up passwordless (and keyless) ssh for this user but force the command to be the git-shell. Next a simple bash script which iterates directories in this user’s home directory and runs git fetch —all. Set cron to run this script periodically (every hour?). To add a new repository, just ssh as your regular user and su to the git user, then clone the new repository into the home directory. To change the upstream, do the same but simply update the remote.

This could probably be packaged as a dockerfile pretty easily, if you don’t mind either needing to specify the port, or losing the machine’s port 22.

EDIT: I found this after posting, might be the easiest way to serve the repositories, in combination with the update script. There’s a bunch more info in the Git Book too, the next section covers setting up HTTP…

I would probably use ntfy.sh for this purpose. It doesn’t quite meet all your requirements, but you could use a random channel name and get some amount of security…

You can self host it, or use the hosted version. (I know it’s technically not chat, but it works on a series of messages, it just happens to call them notifications.)

Yes, I have. I should probsbly test them again though, as it’s been a while, and Immich at least has had many potentially significant changes.

LVM snapshots are virtually instant, and there is no merge operation, so deleting the snapshot is also virtually instant. The way it works is by creating a new space where the difference from the main volume are written, so each time the application writes to the main volume the old block will be copied to the snapshot first. This does mean that disk performance will be somewhat lower than without snapshots, however I’ve not really noticed any practical implications. (I believe LVM typically creates my snapshots on a different physical disk from where the main volume lives though.)

You can my backup script here.

I don’t bother stopping services during backup, each service is contained to a single LVM volume, so snapshotting is exactly the same as yanking the plug. I haven’t had any issues yet, either with actual power failures or data restores.

I know it’s not ideal, but if you can afford it, you could rent a VPS in a cloud provider for a week or two, and do the download from Google Takeout on that, and then use sync or similar to copy the files to your own server.