This message was deleted.

Or for nodes in the pool that lived recently, made some connections and died out right? That's too indeed a good point. I will check that out 👍

Supposedly finished connections for old pods should be cleared out of the connections list which is rendered. If you have lots of churn of nodes and pods, always possible there could be something funny. If you see things that are stale that never go away, maybe there is a problem, but would need to know more about the behavior to try and reproduce and see if there is a bug.

@quiet-fountain-46593 Is there in fact an age out threshold for short lived pods and nodes in the network map? because we do have such a use case with AKS self-hosted github runner pods / nodes and it seems that the map is getting bigger daily because of this.

I was told they should definitely be removed soon after the pods stop. There are no enforcers down on any nodes, correct?

There weren't any on those runner nodes, but after your suggestion, I actually got them added by updating the tolerations.