Running services, network usage, memory usage, bandwidth, disk I/O, successful logins, whether the thing is even alive, etc…
So far my only method has been “hope and pray”.
You must log in or # to comment.
Ultime Kuma, Beszel and Ntfy
I am using prometheus-nodeexporter which is scraped by Prometheus and then visualized in Grafana
You guys are monitoring your servers?
“can’t access this thing, server must be down”
I plan to perhaps, just maybe, I am not sure, to try self-hosting an email server. But I’ll need a good uptime.
Although I am really scared about security.Been running one for many years, mostly issue-free. Had to request whitelists way in the beginning, but things turned out ok.
Check out this super great guide. https://workaround.org/ispmail-trixie
Well, it’s a good learning experience. But…… don’t. Good luck getting around the major email providers’ smart screen shit. It’s probably not gonna happen. I tried a few years ago, and even microslop couldn’t figure out why their shit screen crap wouldn’t allow my emails. It’s not in there best interest to fix it. So they don’t. But it’s worth doing it in the short term to learn how to do it. Teaches you about email, spf, dkim etc. which actually does translate to real world skills. I used what I learnt setting it up in my real life job.
Yeah, email is one of the more annoying things to self host. Google rejects mine because my ISP has all their non static IPs on block lists (even though my IP hasn’t changed once since I’ve been with them it’s still not “static”, can’t change the reverse DNS) and I’m not paying 150% more a month for a business account to get a static IP. I ended up just using a free email relay. Still had to setup SPF, DKIM, DMARC, and all that jazz still.
I notice something stopped working, or someone in the household notices.
Then I go fix it.
And the best time to notice that the backups weren’t working is when you need access to the backup. 😅
Backups? In this economy?
But I need to get something working. I am about to move to VaultWarden and that makes me very nervous.
I run a vaultwarden container since forever and it’s really smooth. All updates went through just fine, never had an issue with it.
My 3-node Proxmox cluster has a dashboard, my NAS has a dashboard, and Home Assistant has a dashboard. Hell, my gaming PC has a dashboard, too (Webmin and Cockpit).
I go the easy route with munin and wazuh. It comes with a ton of stuff preconfigured
Uptime Kuma for monitoring web service status (just like if the jellyfin, etc. website is up), and Zabbix for monitoring system metrics like CPU/Memory/Disk I/O, etc. Zabbix can do a whole lot, but can also get a little complicated. There are plenty of guides out there for configuring it though.
I just look at the console.
Monit, simple to setup, gets the job done, runs on my router too.
Netdata, looking at dashboards when something breaks :)
Acronyms, initialisms, abbreviations, contractions, and other phrases which expand to something larger, that I’ve seen in this thread:
Fewer Letters More Letters AP WiFi Access Point DNS Domain Name Service/System ISP Internet Service Provider LXC Linux Containers NAS Network-Attached Storage
5 acronyms in this thread; the most compressed thread commented on today has 6 acronyms.
[Thread #122 for this comm, first seen 9th Oct 2026, 13:00] [FAQ] [Full list] [Contact] [Source code]
It’s running the private DNS for my phone. If it goes down, I realise quite quickly.
My primary monitoring system is Beszel, but since I just migrated from VMs to LXCs it can’t properly see the memory/CPU usage, so it just checks for disk usage and liveliness. Home assistant with Proxmox & Dockhand integrations gives me the rest of my monitoring needs.
I am also using Beszel after using checkmk. For me checkmk was way too complicated (but is also way more powerfull) to setup and maintain propperly. So I switcheed to beszel because it is much easier and good enough for my use case which is:
- Is the thing running at all?
- Disk full?
- CPU, network, etc. on full throttle for some strange reason?
- send me a mail if somethings wrong
Besides that I don’t need anything. The only thing missing is SNMP support to check switches, routers and AP’s.
Uptime Kuma for outages, Prometheus for metrics, rsyslog to Loki backend for logs. Grafana for ingesting Prometheus and Loki data.
That’s the neat part, I don’t.
