Node next discovery stuck in the past

tbelanger

Hello Everyone,

I recently started to notice an issue with my NPM deployment involving nodes that wont rediscover. The next discovery field time just stays the same for that particular node and manually rediscovering does not appear to do anything. However, polling of interfaces and node status continue to function normally.

The problem that this is causing for me is that data such as the model number and running software version to not get updated if the hardware is replaced or rebooted to run new software.

The only way I've found to correct this is to manually re-add the node individually. This can be a tedious process and I would prefer to not have to manually readd every node experiencing this issue. I also figure that I can just delete all the affected nodes and rediscover them but I would prefer to retain historical information.

Has anyone experienced a similar issue ?

tbelanger

One thing that I've discovered is that most of the nodes experiencing the issue where set to the Solarwinds native node details poller. Awhile back I had made some custom pollers for various devices since I did not like how the model number was displayed, for example I would prefer to see 2960S-48TD instead of just saying 29xx stack.

Somehow a lot of my nodes were assigned back to the native poller. Upon changing them back to the one I created and doing a rediscover, most of the affected nodes started to discover properly again. I wonder if there is something in the built in one that is messed up causing the rediscovery to fail. I wish I could just disable the built in one permanently