Services that appear and that seem related to OneAgent communication errors?

[AntonioSousa]

I have noticed that in some situations services with similar names appear, where there should clearly be only one. The following is an example:

The blocks in green and orange have exactly the same names in all cases I have observed. The only difference is that the one that has real traffic are always the ones that have the port numbers, like in this case ":80,443"

The other ones only have occasional traffic, and all the ones I've seen have in common Purepaths with errors, like the ones below:

• Some data could not be collected or transmitted. This is most likely due to a resource congestion on network, host or process level in your monitored environment (Diagnostic codes: C1, C2, A5)
• Some data could not be collected or transmitted. This is most likely due to a resource congestion on network, host or process level in your monitored environment (Diagnostic code: C2)

Now, this introduces some confusion when looking at the list of services. So, two questions:

1. Can someone confirm that this is a procedure of the Dynatrace platform, when it has to put Purepaths which it doesn't know where to put them?
2. Is it possible to deal with these Purepaths in some other way, through Settings?

[Julius_Loman · Accepted solution]

This happens typically for incomplete traces. As due to adaptive load reduction, oneagents do not send all data to the cluster, the cluster still tries to assembly PurePaths, but it may no have enough data. Similarly, it can show "Unexpected service" for a process group.

Previously (before Dynatrace version 1.164) those PurePaths were not visible in the UI at all. You cannot change the behaviour from UI.

[AntonioSousa]

Now that you've referenced the previous behavior, I remember that really some were lost...

I have seen at least on of these situations in the context of one vMotion that took about 3 minutes...