Maybe it’s well known but I just came across journalctl-desktop-notification and I find it very useful so I thought I’d mention it. It’s basically a bash script that monitors systemd’s journal and pops up a notification when there are warnings or errors (or anything else you want to make it catch besides the default config).
What makes it so useful for the selfhoster is that it can monitor the journal on hosts your user has ssh access to with key authentication (set up in 2s with ‘ssh-copy-id’).
So case in point, this just popped up:
My reverse proxy can’t renew certs, that’s bad. For some reason netdata didn’t catch it, and the service didn’t trigger a system email that would have been forwarded to my smtp. Uptime kuma would have caught it when I would have had only a few days to fix it, but this caught it immediately, and I have 52 days to figure it out.
So you install that on your daily driver and you get these notifications on your desktop. They only have packages for Arch and Gentoo but the thing is just a batch script and a systemd unit. So to install anywhere you just download the “source”, extract it, cd to it, and run ‘sudo cp -r usr etc /’ which is exactly what the Arch package does (line 22).
Just a nifty little tool I wanted to share in case others haven’t heard of it.
Edit: I made .deb and .rpm packages so it’s a lot easier to install now :)
Holy shit, this is amazing. Thank you so, so much for sharing this. I had not heard of it, and I am often stymied by journal CTL, since I don’t really know how it works. So I will most definitely be using this on my desktop and possibly on my self-hosted stuff as well. Thank you again.
Ha ha ha.
I love how lennart’s cancer tries to replicate fucking syslog and it’s this bad. What a mess the kids worship.
Please transcribe any images with text
Wow! This sounds super handy! Thanks for sharing!
If you want the next level up, set up a log collector like graylog or something.
Gray lot requires a newer version of mongo. Mongo now requires a processor with the AVX instruction set; and my aging homelab is one gen before Sandy Bridge.
So basically no graylog for me because I ain’t got money to run that shit anymore let alone upgrade it
LGA1366? I’ve got a Xeon X5679 in mine. It was like an $80 upgrade from my i7-950 lol, and yeah no AVX
That seems like a flight of stairs up.
Hahaha nice. I have PTSD from teaching the interns how to search graylog and then locking up elastic with piped conditional searches.
Well yes, you take the stairs to the next level.
I’m using netdata for now but for some reason it didn’t notify me of this one.
This is great! Thank you for sharing
I just discovered it a few weeks ago. But I did not know about the ssh part. Thanks.
This is a good idea, thanks.
Assuming the uptime of your services are in any way important.
I’m not running a business here, I’ve got no big stakeholders. If something doesn’t work, at most me or someone close to me is affected. No one really cares if something is not available for a day.
I spent 0 minutes on monitoring and don’t intend to start now :D
Does your data matter ? There’s a data loss prevention risk and security.
If you don’t care about those either, then I guess your decision makes sense
This is very cool but all the machines I would use this on are headless with no GUI installed. Womp womp for me.
i think the idea is install this on your PC, and it displays notifications from your headless servers journals over ssh
Oh I didn’t catch that part, that’s even better than how I understood it, thanks so much for clarifying!
This is cool but for self-hosting you probably want a more robust monitoring system capable of alerting at all times. Prometheus is what I use. It also gathers data over time and can monitor many machines.
