Page 2 of 2

Re: Baselining

Posted: Thu Oct 08, 2015 10:57 am
by jdalrymple
Is this question somehow different from your earlier post/question?

https://support.nagios.com/forum/viewto ... =6&t=34775

If not I'd assume it's OK to just merge these 2 threads into 1? Did you try using the plugin that tmcdonald mentioned?

Re: Predictive baseline monitoring

Posted: Thu Oct 08, 2015 4:47 pm
by jdalrymple
Again, the answer remains the same today and probably will for eternity. Nagios Core is the core component of the monitoring platform. The actual collection and interpretation of metrics is handled by plugins and always will be. I'm speaking both as a support technician and a long time Nagios user saying that this is the right architecture and ongoing it's proper that metric interpretation needs to reside in the plugins.

The unfortunate reality is that plugins for this type of monitoring are scarce. Google around and you'll find nothing, not only for the Nagios platform, but I know of no good enterprise monitoring platforms with intelligent trend based monitoring. I have had personal experience where I needed to do trend based analysis (of temperature metrics) for my employer and this plugin had to be written from scratch.

Now the fortunate news is that if someone were to take it on to create such a plugin, this should be fairly doable. The plugin guidelines have semi-strict requirements on output format for performance data. Utilizing those constructs it would be possible to create a plugin that wraps around all existing plugins (that output proper data) to identify abnormal behavior based upon what you perceive as abnormal.

This is something you could create for yourself if you have the time and any knowledge of scripting in any language. To help you along I can share with you the nagios plugin builder guidelines (it seems like I've had to link these a lot lately). We could additionally put in a feature request for you, but it will be just that, a feature request. There are no guarantees on either the acceptance nor the timeliness of feature requests.

Re: Baselining

Posted: Fri Oct 09, 2015 2:33 am
by ranjitw
Hi,

Please merge the posts.

Re: Predictive baseline monitoring

Posted: Fri Oct 09, 2015 11:43 am
by jdalrymple
Merge complete...

Please let me know if any of the information I provided earlier is helpful. Hopefully it is because I probably won't be able to come up with anything better :D

Re: Predictive baseline monitoring

Posted: Mon Oct 12, 2015 1:22 am
by ranjitw
Hi ,

Is there a reference number for feature requests ?

Re: Predictive baseline monitoring

Posted: Mon Oct 12, 2015 3:22 pm
by jdalrymple
They're tracked in a couple of places. We have internal feature requests that we as support staff put in to make our lives as easier as well as to submit features that definitely meet the criteria required to be a good feature to be implemented for the masses. We can track those features internally using our own IDs.

Additionally you can submit your own feature requests at http://tracker.nagios.com - all of which are handled directly by the developers and do not have to be scrutinized by the support staff. Those feature requests can be tracked publicly by you.

Re: Predictive baseline monitoring

Posted: Wed Oct 14, 2015 1:50 am
by ranjitw
Thanks , Please close the post.