Auto Handle Unconfigured objects

[peter.zanetti]

We were assigned a task to monitor Hosts which are submitting Passive check results utilizing NRDP API. Right now testing with a few devices but potentially there will be 2K +.
As right now it comes in as an “Unconfigured Object” and will not be processed by the monitoring engine until they manually configure in Nagios XI.
Given number of devices to monitor here is my thought process:
1. Pre-configure a rule that handles these hosts, correlate it with an existing host/service (e.g. where notifications should go, what host/servicegroup they should belong to). So When Nagios XI first sees a passive check comes in, it reconfigures based on that rule and eliminate human intervention.
2. When Nagios XI first sees a passive check that doesn't correlate with an existing host/service, it creates a new one and reconfigures the monitoring engine.
3. The original passive check is re-sent to Nagios XI (once it is reconfigured to handle it) and an alert is/can be sent out

Does this makes sense or is even possible?
Look forward to the feedback, thanks.

[tmcdonald]

It's certainly possible, but not currently built-in. It's been on the dev roadmap for a while, so that's a plus. As far as us architecting a solution for you however, that would fall under custom development or consulting and is something you would need to talk to Sales about. I can tell looking at your thought process that it's one way of going about the automation, but the implementation details are not something that we have readily available.

[Box293]

From a deployment perspective, if you are rolling out NRDP via a script or some automation method then you should be able to automate the Nagios bit.

With XI 5 there is a new backend API which would allow you to add the new host and then add it to the required hostgroup all via a curl command.

Have a look in the help menu on the left pane under "Nagios XI API Docs".

Does this help?

[peter.zanetti]

Is it possible that if we prepopulate Nagios with the host that is showing up as an unconfigured object that the next time it sends a passive check it will reflect on that host instead of in a unconfigured object?

[tmcdonald]

Only for the host check - if there are not matching services already configured then those will go into Unconfigured as usual.

[peter.zanetti]

Well we are using one service for all the devices, so if we were to preload the service as well and attach the service to the preloaded hosts, would that work?

[tmcdonald]

Should work, yea. Just as some background, when you send a passive result in to Nagios for a host or host+service combo that does not exist, it gets logged. The Unconfigured Objects cron job parses the logs for this sort of message and that's where it gets the info for new hosts/services. So as long as you have one service, attached to one host, then anything matching that coming in should go to that specific host+service combo. If either part is not already in the CCM, then it will show up in Unconfigured Objects as expected. But if you have already pre-configured the hosts and services, it kinda defeats the purpose of Unconfigured Objects :)

[peter.zanetti]

Adding the host and service and attaching the host to the service worked! Thanks for the help!

[tmcdonald]

No problem, glad it's working. I'll be closing this thread now, but feel free to open another if you need anything in the future!