Howdy folks,
As you may recall, we’ve been on a long journey to replace Nagios with Zabbix. A quick history:
- We had initial work from @dkirwan
- Last year I wrote a migration plan for moving it forward
- In January, the Infra team officially adopted Zabbix as the primary monitoring source
In the recent months, I’ve been porting all the remaining checks from Nagios to Zabbix, and as of this morning I have now completed that, and stopped Nagios on noc01 and noc02.
Farewell Nagios, you have served us well!
There’s a Flock talk!
If you want a more detailed overview of what’s changed, why, and how you can be involved, do check out the Zabbix talk on Tues - I was supposed to give this, but I can’t make it to Flock this year, so huge thanks to @zlopez for standing in for me. I’ll be on Matrix for questions though 
What’s next?
There’s still plenty to do. In particular:
- We have many checks that need refresheing / improving as they’ve been directly ported from Nagios
- Many of the new checks are quite noisy and need tuning
- We need to onboard more groups into consuming their own monitoring alerts
The first two land (mostly) on me, but the latter is a call to action. If you depend on Nagios in some way, let me know. I’m going to read through the notification config to see where things were going, but I may miss things, so get in touch. It’s also fine to ask about net-new monitoring that you’d like to have - I can’t promise everything will be possible, but we can discuss.
Also a shoutout to @frostyx who’s been helping me test the new notifications in the COPR Matrix rooms - this has been very useful for figuring out how to open this up, and I have @lecris from CI lined up to be next
You can check out the monitoring at https://zabbix.fedoraproject.org (FAS login required, use the SAML link). Feel free to ping me (gwmngilfen:fedora.im) in #noc or #admin if you have questions!
