monitoring or the new flashy term telemetry what do hardcore freebsd users use?

[azathoth]

I have used nagios zabbix cacti...

My pals showed me graphite.
I also briefly used munin.

So what is the latest greatest?

 

[Eric A. Borisch]

I’ve been happy with zabbix. A bit of an initial learning curve, but it works well, and custom discovery rules make it very powerful.

 

[azathoth]

PS I hated appdynamics and felt it was a way to ignore proper database perftuning and a java programmer bandaid megaystem.

 

[azathoth]

I’ve been happy with zabbix. A bit of an initial learning curve, but it works well, and custom discovery rules make it very powerful.

Were you able to make graphs with groups of servers easily? that is where I got slowed down.
What are custom discovery rules used for?
I remeber with linux the base module has 400 checks.
I disabled all but 7 or so and load went way down.
People loved the ability to view history to compare previous weeks to current day.
Postgresql 9 was a dream come true and great performance.

 

[Eric A. Borisch]

Yeah, that’s always a bit of an issue; I prefer to have a ‘dynamic display’ of a single system; you can set up one screen, and then use a drop down to select what machine is being reviewed.

I use custom discovery rules to auto detect zpools and zfs filesystems, and then automatically measure (and trigger alarms based on) status, capacity trends, snapshot usage, etc. Also for network interfaces, smart monitoring, etc.

 

[azathoth]

There are companies charging tons to other corps n gov for stuff thats crappier than zabbix.