Hi,
We are creating BTs to export data to Splunk and we added Agent measure as one of the splitting measure.
However we dont see the Agent information in the Splunk side. What could be the reason?
We were hoping to create a load distribution chart in Splunk suing the Agent splitting.
-Sreerag
Answer by Sreerag M. ·
I think you can chart any measure split by agent as long as the measure splitting is defined to have values for each agent in the Details tab of the measure.
That said the issue is that there is no effect in adding Agent measure in the BT splitting. Either way you can get the measure split by agent.
Now from a business transaction feed perspective having Agent splitting will help in understanding the load distribution and also relate any other metrics to the entry point app instance.
Say if we are getting many 500 http status from one app instance (agent); with current data export to Splunk we cannot identify this.
-Sreerag
Answer by Graeme W. ·
Sreerag,
'Agent' is a special splitting, like 'Application', except that the agent is not included in the business transaction information. It's only there to indicate that the result measures should be split by agent – so you can chart them split by agent.
-- Graeme
Answer by Sreerag M. ·
I think DT identifies agent name for each purepath and if we chart purepath measure and split by agent name we get only those agents in chart where the purepaths are started.
One of the characteristics of purepath is Agent name(agent where purepath is started) and it is shown in purepaths dashlet Agent column. I was expecting the same value when used in BT splitting.
-Sreerag
Answer by Reinhard W. ·
Sreerag,
you can't really categorize a PurePath that potentially crosses multiple agents into a exact bucket (or split) by agent name because you don't know where to look for the 'correct' agent (entry point, frontend tier, backend tier).
that's the reason why, i guess
JANUARY 15, 3:00 PM GMT / 10:00 AM ET