tmcdonald wrote:but since it would be a pretty significant component I want to keep this open for comment, get some discussion and other views going
This is a feature I've thought about for a while so here are some of those thoughts. I'm gonna refer to the component as the IMC.
Firstly, the IMC should have data stored in it that is separate from the Normal XI database.
Inventory management also sort of falls into the category of asset tracking.
Asset tracking and reporting on assets is something that you may want to keep doing long after the asset exists. So there needs to be a way that the objects in the IMC cannot be easily deleted. Preferably a hide/archive option.
So lets say we add the IMC to an existing Nagios XI system. The IMC has an empty database. It should query all the hosts in Nagios to populate it's database. Now there should be an option to link a Nagios Host to a IMC Host Object. When an object in Nagios is deleted, it should not be deleted from IMC, IMC will detect that it no longer exists.
Inventory Management and Asset Tracking is about different sorts of information. Lets use Operating System version as an example.
So in IMC you create an Operating System Version field.
Nagios already has an Operating System service for a lot of hosts.
So IMC should be able to link an existing Nagios Service to a field. This way IMC will retrieve from the last service check the Service Output and populate the field.
There are also instances where the Linux service checks might be named differently to the Windows service checks. So there should be some mechanism to say for this group of hosts link xxx service object to this field and for this other group link yyy service object.
There could also be a way to manually override this field value for a specific host and manually enter data.
The general reason for my example I provided is that Nagios monitors different objects. It's hard to build a product that goes out and collects data and becomes a one solution that fits all. Some people like SNMP, others NSClient++, nrpe, nsca etc etc.
There's also data that needs to be manually entered like purchase date, warrantly expiry day.
Needs to be an API to perform reports, also possibly to be able to accept data from an event handler perhaps.
Tools menu sound like a great place to manage/configure/report.
A tab per host/service that can report and be updated would be beneficial.
Perhaps a dashlet that could give summary details as well.
Thats just some of the ideas I've had about such a product.
My last comment is that this should be an add-on to Nagios XI which costs $x. Customers who want the add-on will be willing to pay for it and it ensures it gets developed as a separate product. It doesn't have to be only for XI, it could be added to core as well and still be a paid product.