Monitoring Frequency

[EnvBroker1]

I know this question is kind of site specific and also depends on the priority of the machines.
But in general, for the monitoring frequency, how often do most people out there run their checks?


Monitoring Settings
Specify the parameters that determine how the service should be monitored.

Under normal circumstances...

Monitor the service every XXX minutes.

When a potential problem is first detected ...
Re-check the service every XXX minutes up to XXX times before generating an alert.

[sreinhardt]

Most people, will stick with the defaults of 5 minute check intervals, 5 retries at 1 minute intervals. There are definitely people that have other defaults, or select systems that need to be monitored more or less frequently, but 5-1-5 is about the norm.

[Box293]

Some checks might not need to be run every 5 minutes. A disk space check might be OK every 20 minutes.

Ask yourself, how important is the check. This will help determine the frequency.

[WillemDH]

I've created templates for hosts and services with prio1, prio2 and prio3 suffixes
interval - retries - retry interval
*prio1 = 05 - 02 - 02
*prio2 = 10 - 02 - 05
*prio3 = 15 - 02 - 05

Some checks run only once each hour because they take so long to run and consume too much resources. If you have performance data I noticed 1 hour interval is the maximum, otherwise the graphs stop working sometimes...

Got an old thread somewhere where Andy said they were working on a solution for that issue. Anyone know if this issue is still there?

Grtz

Willem

[lmiltchev]

@EnvBroker1

Let us know if you have any more questions.

[EnvBroker1]

Thanks

[tmcdonald]

I think at this point a month after the topic started we can close it. Over time you'll get a feel for what your environment needs.