Another option you could consider is using NetPath to monitor the node from multiple polling stations. It's easy to add multiple different SolarWinds pollers into a NetPath monitor solution, and that can give you additional data from the point of view of each poller.
The only alternate solution I can think of, other than one I originally mentioned (above) is to have multiple IP addresses on a node, and poll each from a different APE.
I have seven polling engines monitoring nodes in three states, but I don't have a method of monitoring the same device from two different pollers unless I want to monitor a second IP address on that device.
But there's a work around, of sorts, for my needs.
I've found that sometimes my WAN provider occasionally has routing problems associated with one of his nodes that is undersized for the number of route tables required. When issues arise, one of my remote sites can't be pinged by my main instance poller that's local to my main location. Simultaneously, users in that remote site can't get to my data centers. But they CAN get to the site that hosts an additional NPM polling engine in it, and that APE can talk to my main instance.
The problem here is that Solarwinds reports the site is "up", even though users can't get to my data centers, and from their point of view everything is down. This is because the main instance collects data from all the Additional Polling Engines, and one of those APE's is monitoring this site's router, and that APE can see the router without any problem. But my MAIN Polling Engine CAN'T see that router. Yet Solarwinds doesn't show the site down because ONE of my APE's CAN see the site's router just fine. This creates a false positive for me, and was a thorn in my side. Yes, I was on the service provider's back to correct the issue, but until that could occur (it took three years to get them to replace that undersized BGP router), I needed to know whether the site was down from the perspective of my data centers, which the site's users needed to access.
The site has a router connected to a set of distribution switches, which support all the access switches in the site. (It's a bigger site--has about 180 access switches).
So my remote APE polls the site's router and reports the router's status as "up" to the main poller that I see in NPM's front page. And when the WAN provider's node has no more room for routes to that remote side, the remote site's router is "up" according to the remote APE, even though the site is "down" from my data centers' point of view.
My solution, that doesn't duplicate a node by polling a second IP address on the site's router: I left monitoring the same for the site's router, but removed the site's Distribution switch from being monitored via the remote APE, and began monitoring the site's Distribution switch with my local polling engine. That way I can see the symptom of the WAN provider's issue--the site is down from my point of view while monitoring the Distribution switch from my data centers, even when NPM sees the site's router is up from the point of view of the remote APE monitoring that router.
Even better, I can see the latency difference between my site and the remote site, and simultaneously see the latency between the remote site and the APE monitoring it. Now I have enough additional information to make some informed decisions about the site's true speed.
You don't need Distribution switches or multiple IP addresses on a node to do this if you have two or more APE's. Just monitor something other than the router at the site with a different APE. Maybe it's just a UPS, or an Access Switch. You'll see this node down and know there's some investigating to do.
I understand this is a unique use case, and perhaps it doesn't fill your need. But it's a way of thinking outside the box, and maybe it'll give you an idea about how you can accomplish your goals without having to duplicate a node in NPM.
Good luck, and let us know what you do, or how your specific use case differs.
If there is a way to monitor a single node from different pollers at the same time, I would be interested in that. I haven't found a way to do it yet without the node that we are monitoring having multiple IPs assigned to it and then adding multiple nodes for the same device and assigned to each poller.
SolarWinds solutions are rooted in our deep connection to our user base in the THWACK® online community. More than 150,000 members are here to solve problems, share technology and best practices, and directly contribute to our product development process. Learn more today by joining now.