CorrelationEngine- Could not pair node for event
Server log is flooded with these messages:
2012-02-06 15:15:14 WARNING [CorrelationEngine] Received wrong event as first node of a PurePath. Event (com.dynatrace.diagnostics.core.datacenter.events.InsertLinkEventRecord, type: 36, numOfAttachments:0, xxxxx:1576), expected event of type EntryEventRecord with hopCount 1. Suppressing this warning for another 60000 ms.
2012-02-06 15:15:44 WARNING [CorrelationEngine] Could not pair node for event (com.dynatrace.diagnostics.core.datacenter.events.ExitEventRecord, type: 2, numOfAttachments:1, xxxxxxxxxxx:1576, tagHopCount=3 serialNo:472976). Suppressing this warning for another 60 seconds.
I made a couple of configuration changes trying to make User Experience working. We restarted agents and the server. I am not sure if these event were before the changed and I am not sure why are they are coming and how to fix the problem..
-Stan
Hi Stan
I would create a support ticket for this so that our engineering team can have a look at this in more detail. When opening a Support Ticket you will be asked to attach a Support Archive which contains all relevant information (System Profile, Log Files, ...) to troubleshoot the problem.
- The Support Archive can be exportet through the System Information Dashlet.
- You can create a ticket through http://support.dynatrace.com
Andi
I've recieving the same messages from my server, the agent is located on a apache servicemix jvm.
2012-06-11 06:44:40 WARNING [com.dynatrace.diagnostics.a.bu] Could not pair node for event (com.dynatrace.diagnostics.core.datacenter.events.ExitEventRecord, type: 2, numOfAttachments:0, >>>agent<<<@>>>host<<<:8828, tagHopCount=4 serialNo:240366000) to Path (null). Suppressing this warning for another 60 seconds.
Is there any solution? - I'm running on version 4.1.0.2974, Platform: Linux 2.6.32.49-0.3-xen, amd64 64bit, Java HotSpot(TM) 64-Bit Server VM 1.6.0_27
Hi Günther
This message is caused when you loose events from your agent. dynaTrace sends so called entry and exit events when methods are executed. In case the Entry event is lost the exit event will cause this log message as we never received the entry event.
This can be caused by multiple things, e.g: you had a problem with your network connection between your agent and collector/server and that caused some events to be lost. it could however also be a specific problem with your environment where we have an issue generating the correct events.
Please check if you may loose events. Check out the Deployment Dashboards or the Agent Overview Dashlet to verify these values.
If you need additional help on this I would contact our support team
Andi
Hi Andi,
if i look at the "server health" dashboard the meassure "Skipped Events (Server)" shows a sum of around 1070220 skipped events for the last 6h.
When i look at the agents dashboard i see this:
Name |
Started |
Event Count |
Total Class Load Count |
Total PurePath Count |
Total Skipped Events |
Skipped PurePaths |
agent1 |
2012-06-07T22:36:21 |
245589606 |
123362 |
787986 |
0 |
0 |
agent2 |
2012-06-07T22:50:45 |
246725970 |
106230 |
786851 |
0 |
0 |
best regards
Oliver
Hi Oliver,
please create a ticket on the support portal for this issue. This is most likely related to a config issue, network problem or the server not being able to handle the load.
regards,
Günter
