The Homelab Show Episode 82: Server Monitoring Tools

I'm a big fan of LibreNMS, very mature and very active development, metrics, monitoring and alerting and a boatload of comms platforms hooks

I still use Nagios :D

I used Nagios at first then I switched to Zabbix. One thing I haven’t changed is how I get my alerts to my phone. I use the free tier of Opsgenie How I have it set up is Zabbix sends a email to a Opsgenie email which I setup using their free email integration. With this method Opsgenie gets an email from Zabbix Opsgenie then sends out a push notification to the app. I also have it setup to also send a text message both of these integrations are included in the free tier. Tbh you can’t beat it I get both push and text notifications for free.. Opsgenie also allows you to send notifications to multiple people and has an on-call system so it would work well in a team/business environment as well.

Uptime Kuma has replaced healthchecks and uptimerobot for me. It's nice that heartbeats can be push or pull.

No mention of twilio for notification options. Easy to integrate. Pay as you need. Supports voice calls or sms notifications.

If you like Nagios. Try Centreon. It uses the same fundamentals as Nagios but has better ui, integration, and Api. We have used for 12 years. We have a notification bot to GroupMe sms for notifications. Centreon can scale out to pollers.

Many cell providers have email to SMS gateways. You can send SMS messages via an email address for free.

Do you know of any tools that send data like cpu, ram, disk space etc over MQTT so I can manage it in Node-Red?

In terms of notifications; IFTTT web hooks work well for mobile device notifications as well.

I am not going to say the Grafana monitoring stack is easier to setup than nagios or zabbix but it feels like it is just as hard to to setup zabbix but gives you metrics and alongs along with alerts. Alerting has gotten way better in the past year with grafana managed alerts. Netdata is easy you can't self host a centralized panel for all hosts.
I will agree their docs need work.

Not very graceful, but for notifications I've set up both Telegram and Signal just in case one or the other goes out

I'm 77 and old fashioned, so I still use Conky for monitoring my desktop, laptop, backup-server, all Linux VMs and OpenZFS on all three PCs. By the way Virtualbox offers a reasonable graphical overview of the loads (CPU, Memory; Disks; Network and VM-Exits) on the Host and for each running VM. Unfortunately the erroneous CPU VMM load spoils the CPU graph completely.

Didn't get to watch the whole thing live but it's starting to sound like a I have a few projects ahead of me :)

Also, do y'all still do spicy food videos?

I just started watching and I wanted to mention that I'm thinking of implementing graylog at my current job. We have a lot of machines that we have to remote into to just check logs and it would be nice to just have a central location for logs.