Monitoring Frequency

Support forum for Nagios Core, Nagios Plugins, NCPA, NRPE, NSCA, NDOUtils and more. Engage with the community of users including those using the open source solutions.
Locked
joe1871
Posts: 28
Joined: Tue Feb 01, 2011 3:36 pm

Monitoring Frequency

Post by joe1871 »

I am trying to set up a policy for server monitoring. I need to specify howe often we need to monitor wspecific serversbased on their level of use, their importance etc. I was wondering if anyone had either any standards that you created or if there was a list of industry standards for monitoring frequency. How often do you ping for up/down? How often do you check disk levels? Memory usage? etc. I know this is highly subjective and difficult to answer, but I am looking for any guidelines anyone may have used in their environment or simply any feedback. I have about 650 machines that I am doing a ping check on now, but the company wants me to expand Nagios monitoring to disk, memroy, cpu, and swap.

Thanks in advance for any help you may be able to send my way.
mguthrie
Posts: 4380
Joined: Mon Jun 14, 2010 10:21 am

Re: Monitoring Frequency

Post by mguthrie »

I would suggest looking at it from the perspective of: "What is your maximum acceptable response time for this host or service?" 650 machines isn't actually a whole lot in the world of monitoring, so unless your using a lot of interpreted checks that eat a lot of CPU, there isn't a whole lot of reason why you couldn't keep everything at a default 5mn check_interval. Unless your total checks within 5mn are exceeding 3000-5000 performance is barely a consideration. Hope that helps!
Locked