Devices Being Re-discovered after moving them to another poller

Question

We have 3 pollers and have moved some devices to even elements out among the 3 pollers.

Now, some devices pop up under our periodic device discoveries that still exist on one of the other pollers.

Obviously, when we "import" these devices, they are not added and the import process indicates that the device is already monitored by a different engine and not added to the database.

We are on 11.5.2. Is this something that may have been fixed in a later release? Does the discovery process not look at the database before reporting that it has "found" a new node that is not really new?

If this has been asked before, my apologies.

silverbacksays · Answer

Hmm, even if they are ICMP, Orion should be able to work out they're already in the database! Have you noticed anything else odd with your instance? Are both your primary polling engine and APE running identical versions of all you licensed products? Worth checking, perhaps version mismatches could cause these type of issues.

lhoyle · Answer

Not in this case. I would say 90% of these are ICMP only nodes, anyway.

I can't even tell you what kind of devices they are, they are gasoline dispensing devices that are IP-networked.

Thanks!

silverbacksays · Answer

Are you sure it's the devices that are being discovered, and not new resources on the existing nodes?

For example, I have a discovery job running every day on one client's instance, they are running 11.5.2 of NPM. The discovery initially discovered lots of 'changes' in some of the physical machines, but this was due to the disco job finding the interfaces I hadn't selected initially.

What I needed to do was select each of the newly discovered resources within the discovery results, and click 'add to ignore list'. This then stopped them from prompting to be imported on the following day's disco.

Not sure if this is happening for you, but I thought it was worth mentioning just in case.