Make nodes go into an "Unknown" state when Additional Polling engine is down

We recently had an incident where we lost a link to an additional polling engine. We noticed all the nodes being monitored by the additional polling engine remained in their existing state. I opened a ticket with SolarWinds to find out if this is normal behavior which they confirmed it is. I believe if an additional polling engine is no longer able to update the SolarWinds database, the nodes monitored by that polling engine should go into an "Unknown" state. With the nodes showing as up, even though there has been no contact with the node for a period of time, it may provide a false sense of security.

Thank you,

Mike

Find more posts tagged with

polling engines

remote office

scalability engine

remote polling

Status: None

Comments

rschroeder

I haven't tested this in a long time, but I think I remember seeing the desired Unknown status displayed for notes & interfaces when a remote poller went unavailable some years ago. I'm surprised to learn my memory is faulty, or something has changed in Orion and that Unknown information is no longer the case.

jhandberg

I have seen this happen here when an additional poller had a service failure so it wasn't polling, but all the nodes it was polling were still showing as UP.

This was happening weekly, usually over a weekend for a while, but has not happened recently. I think I really need a special alert when the additional polling engine services fail. Trouble is they didn't show as failed or down, but they were locked up and not polling anything. Restarting the Orion services fixed the issue.

charlesjsmith

I'm on 2020.2.6 and I can confirm the behavior if an APE goes unreachable via ping (monitored via ping from our main polling engine) and does not update the database (showing red in polling engines page) everything pretty much freezes that was monitored by the APE It makes sense to me to have items monitored by the APE have an Unknown status.