2

I am monitoring certain FreeIPA servers that are normally forking <= 460 process (roughly). This generates the "Too many processes on {HOST.NAME}" as part of the "Template OS Linux" template.

The Expression that does the check is looking for processes greater than 300:

{Template OS Linux:proc.num[].avg(5m)}>300

What is the best way to over ride the value in another template I created particularly for these class of servers that is also inheriting "Template OS Linux" and "Template IPA Servers?"

Richlv
  • 3,954
  • 1
  • 17
  • 21
farhany
  • 1,243
  • 2
  • 17
  • 29

2 Answers2

11

Probably user macros. You would add a user macro in the original template:

{$TRIGGER_THRESHOLD_PROCESSES_RUNNING}

And you would modify the trigger expression like this :

{Template OS Linux:proc.num[].avg(5m)}>{$TRIGGER_THRESHOLD_PROCESSES_RUNNING}

Then you could define a user macro with the same name on the lower level template - or even individual hosts - with a different value.

The user macro name is up to you, as long as it follows the syntax rules.

Richlv
  • 3,954
  • 1
  • 17
  • 21
3

If you consistently have more than 300 processes on all the systems that you monitor, you can also just edit the original template. As an example, 450 is a better value for my specific use case since I always have at least 160+ processes enclosed in brackets that don't reflect what I'm actually trying to monitor - user space. You may want to use a number more appropriate to your environment than my example.

I changed this value by going to the original template, which you can find from Configuration --> Templates --> Template OS Linux --> Triggers --> Too many processes on {HOST.NAME}, and then changing the value from 300 to a more reasonable value for your specific need.

One thing to note is that not all servers in an environment are the same, and if your environment has multiple different types of servers, the user macros solution is going to be far superior.

Vladinatrix
  • 31
  • 5