If you want to do this, what you probably want is to pump your logs into a log drain, something like betterstack is good. They then allow you to set up discrepancy thresholds and can send you emails when something seems to be out of the ordinary. There’s probably a self hosted thing that works the same way but I’ve never found a simple setup. You can do the whole Prometheus, influxdb, grafana setup but imo it’s too much work, and then you still have to set up email smtp separate from that.

Came to write basically this. I would try caddy but my compose file is 600 lines long now and half of that is traefik labels, I can’t be arsed with the migration.

I do have nightly off-site backups, that’s true. Still, having the git repo be on the same machine doesn’t seem right to me.

That would fill the same role as watchtower I guess? I’ve previously tried to have a look at having portainer manage the docker compose stack that it’s running inside but at least back then it seemed to be a dead end and not really what portainer is meant to do. I’m not interested in moving away from docker compose at this time.

I’d be a bit concerned with having the git repo also be hosted on the machine itself. If the drives break it’s all gone. I could of course have two remotes but then pushing changes still becomes a multi step procedure.

Oh for sure for sure. I just know that a lot of people use their homelab to learn skills that they can put on their resume when looking for a job. It’s totally fair to over engineer your self hosting setup if that’s your goal.

You should definitely figure out some infra as code system now while it’s manageable. Normally I’d recommend docker-compose as it’s very easy to learn and has a huge ecosystem, but since you’re using proxmox you might need to look at ansible like the other commenter said. Having IaC with git makes it so much easier to test new stuff, roll changes back, and all that good stuff, in addition to solving your original problem of forgetting what is running where.

Just find the simplest IaC solution possible. Unless you are gunning for a job in infrastructure you don’t need to go into kubernetes or terraform or anything like that, you just need something reproducible that you can easily understand and modify.

Is it still confusing after my new explanation? The posts are hidden because they’re marked as read for me, and I’m trying to clear the read posts for a community so they won’t be hidden anymore.

No, I didn’t say that. I’m unsure where the confusion arises so I’ll start from the beginning. And try to clarify my problem.

1. There’s a setting that makes it so posts you’ve already read are automatically hidden. I have this turned on.
2. There’s a setting that makes it so posts get marked as read when you scroll past them. I have this turned on as well.
3. There’s an option in the drop-down when viewing communities that says “clear read”. This button doesn’t seem to do anything.

Let’s take a specific community, doctor who @ lemmy.world. This community is the main reason I’m having a problem, because usually I’ll see a post on my feed from there before I watch an episode, so I want to read it later, but when I scroll past the post it gets marked as read. Now once I’ve gone and watched the episode, I’d like to go back to the community to read the post I skipped before. Of course that post is hidden, because it got marked as read when I scrolled past it. That’s when the button from point 3 seems like it should allow me to unhide the post, but it doesn’t work. I can verify that the post does still exist if I just visit that community in a web browser instead of in boost.

It’s a setting you can turn on where it doesn’t show posts you’ve already read

Ah, fooled by the title yet again!

Quick feedback: your css transitions are way too long, opening the hamburger menu should not make me feel like I’m waiting for it to open.

Also you’ve gone for the card layout on the app list, however cards create the expectation that they are actionable yet clicking them does nothing. At least make the app names clickable.

I started seeing this too. Interestingly pulling individual images works, it only does this when trying to pull all the images.

Finally! So happy to start using boost again now!

Same! Only reason why I’m still using Sync instead.

This is the one! This is why I wanted boost in the first place. For now I’ll just keep using Sync, which has this. I don’t need full feature parity with the old app, but this one is especially important since lemmy is so much slower paced than Reddit.

I’m using brave mixed with a network wide ad blocker, so while it’s nice that Firefox has UBO I’m fine without. Firefox has been presenting these issues every time I’ve tried switching, so for about a year now, so no not a recent issue.

This is great but I literally can’t use the base app on my S20. Like clicking on Google search hits causes the app to freeze. Trying to scroll up on a page triggers a reload 30% of the time. I want to use Firefox but it’s nowhere near good enough, and adding extensions on top of that state is not going to help.