Answer by Rajesh G. ·
Thanks Andi, for the quick response.
It turned out that Agent instrumentation was disabled: "Average instrumentation time for Agent '...' with version 5.6.0.5802 was too slow (26 ms for 30 classes, max allowed = 10 ms). I incresed the wait time using environment var 'DT_TRANSFORMATIONMAXAVGWAIT' and it fixed the PurePath corruption issue.
BTW, Why agent log doesn't have additional logging? As mentioned in my previous post, I turned on DT_DEBUG=true, DT_LOGLEVELFILE=DEBUG but still it didn't produce additional log messages.
Answer by Andreas G. ·
Hi
In this case I think its not a timeout - but - I think that you your not sending the Exit Event for your submitJob method. At least thats my assumption. Something seems to be missing on that method. thats why it shows up with the clock warning symbol and not timestamp for that method. Please double check your code and make sure you are sending the Exit event which tells our PurePath that this method completed its execution at a certain timestamp
Andi
Answer by Rajesh G. ·
I'm using native ADK macros to create PurePath in C++ application. When executing long running command, PurePath is marked as corrupted but the command itself completed successfully. Looks like PurePath is timing out. If the corrupted path is due to some timeout settings in Dynatrace, is it possible to increase the timeout?
Dynatrace version is 5.6.
Also, I changed log level to debug but there is no messages in the agent log related to timeout or corruption.
agent log:
2015-04-10 17:50:25 [f77676d0] info [native] Eventsender channel connected successfully
2015-04-10 17:50:33 [dbffeb70] info [native] License = license ok;
DT_DEBUG=true
DT_LOGLEVELFILE=DEBUG
DT_LOGLEVELCON=DEBUG
DT_WAIT=120
DT_SOCKTIMEOUT=120
DT_DEBUGFLAGS=debugPurePathADKNative=true,debugExceptionNative=true
JANUARY 15, 3:00 PM GMT / 10:00 AM ET