24 Aug 2018 02:29 PM
In AppMon we had the 10K node limit or 10 minutes before a purepath was considered corrupt. Does that also apply in some way for Dynatrace?
We are looking at the possibility to instrument a batch server with a OneAgent. Maybe we need to define custom entry points, but that's not an issue. Typically batch processes run a long time and generate a lot of nodes. I would like to know about the limitations before we go further.
Thanks
Solved! Go to Solution.
27 Nov 2018 08:43 AM
it is not recommended but you can:
increase the 10 min limit
"To
prevent purepaths from timing out you can do the already mentioned 2 ways.
-Increase time out limit -->
-Dcom.dynatrace.diagnostics.completenessTimeout=XXXX
-Prevent omitting timed out pathes --> -Dcom.dynatrace.diagnostics.realtime.omitTimedOutPurePaths=false"
10k node limit
"In order to increase maximum node per PurePath go to the DT Server
and navigate to DT_HOME/server/conf/server.config.xml and edit this
file. In this file you will see:
<purepath maxnodes="100000" />"
27 Nov 2018 10:32 AM
Hey Kay,
I think you described the ways to go around it for AppMon. But my question was about the new Dynatrace. Does that still apply there?
27 Nov 2018 10:58 AM
According to my opinion, this is not possible in the new Dynatrace.
Collecting data works in a completely different way than in AppMon.
/ I also think that the above change configuration to AppMon /
28 Nov 2018 10:14 AM
Hi Bert,
I just checked some batch-jobs in a Dynatrace SaaS environment, and found jobs with complete purepaths that ran for more than 1 hour (Message Driven Bean). So, without knowing the exact max duration, it's more than 1 hour 🙂
Cheers!
Sjoerd
PS. I also hope for a response with the exact values....
28 Nov 2018 02:44 PM
Thanks Sjoerd. Makes sense indeed as we now have "service" level while appmon had purepath as deepest.