[LeoPanthera]

Perhaps I can hijack this post to ask some advice on how to monitor servers.

I don't do this professionally. I have a small homelab that is mostly one router running opnsense, one fileserver running TrueNAS, and one container host running Proxmox.

Proxmox does have about 10-15 containers though, almost all Debian, and I feel like I should be doing more to keep an eye on both them and the physical servers themselves. Any suggestions?

[YZF]

- Prometheus/Grafana/Alertmanager

- Zabbix

What are your goals though? I.e. why are you monitoring? Do you really care that much about getting alerted when something is down? Are you going to be looking at dashboards? Or do you just want to experiment/learn the technologies?

[pnw]

I run Uptime Kuma, it's a simple but extensible set of monitors but perfect for a home network.

https://github.com/louislam/uptime-kuma

[eddyg]

https://github.com/netdata/netdata

[oriettaxx]

nope please

all the free stuff almost disappeared, plus you'll become stupid when you then want to remove their "client" (a huge bunch of stuff spread out everywhere)... I am talking about linux, of course.

Before installing everything assure you read the "uninstall" page, if any

[oriettaxx]

or yes, install it only as a container separated, as somebody suggested, but again: it costs a lot

[SamPlgmsTooth]

snmp v3 (v3, because anything lower then this is unencrypted and 4) therefore not advised unless it’s on a trusted network)

For most Linux distros and network equipment, this will get you CPU/Disk/RAM/Net statistics. Use LibreNMS for this which runs in a VM or CT which can alert you if metrics go out of spec.