NRPE Packet Size

This support forum board is for support questions relating to Nagios XI, our flagship commercial network monitoring solution.
Fred Kroeger
Posts: 588
Joined: Wed Oct 19, 2011 11:36 pm
Location: Perth, Western Australia
Contact:

Re: NRPE Packet Size

Post by Fred Kroeger »

and so we come back to the question, how do we reliably monitor disks on servers without havinng to rely on "someone" remmbering to add/remove monitors when there is a chage in the disk envirnoment.
Having separate monitors also means you lose the ability to view all disks on one graph in Graph Explorer.
sreinhardt
-fno-stack-protector
Posts: 4366
Joined: Mon Nov 19, 2012 12:10 pm

Re: NRPE Packet Size

Post by sreinhardt »

The trouble here is not that nagios isn't reliably dynamically adding and removing disks. This is not the intent of the plugin or nagios as a whole. It can only do what the administrator informs it to check. Dynamically adding and removing things is a recipe for disaster as you could have variances in what you want checked, how you want it checked, and it may add or remove things in a way that you do not like. This is precisely why we do not do this. Additionally, in this particular case, it is a limit of both the nrpe packet size and and rrd track information. You cannot dynamically add and remove tracks, much less alter what device or mount point is on what track. It is simply not how rrd's are designed to be used. The custom development suggestion could work as you would define an amount of drives or mount points that could be filled in the future. Total and free space would then return 0 and should result in no warnings or critical alerts if the plugin is altered properly. Then as you alter the command run, to add a disk to the zeroed track, it would already be allocated and could be used. However this still would not resolve someone needing to add it unless you are also going to check mount points or something along those lines, which brings us back to issues of consistency and the plugin making judgement calls that may not be true for your network and devices.

As for your point regarding a single graph, I agree that having them on a single graph is nice, and probably something we can look into resolving. I think it would be a great feature request to go onto tracker.nagios.com as a suggestion to merge host or service group graphs onto a single graph from multiple checks. Please do post this there!
Nagios-Plugins maintainer exclusively, unless you have other C language bugs with open-source nagios projects, then I am happy to help! Please pm or use other communication to alert me to issues as I no longer track the forum.
Locked