Hi Dave,
There is the "Differential Analysis" feature ( https://docops.ca.com/ca-apm/10-5/en/administrating/manage-metric-data-by-using-management-modules/configure-differential-analysis ) that baselines frontend metrics and then monitors and detects, if a metric exceeds baselined value by some sigma analyses. If value exceeds boundaries, EM tries to set the baseline value as an auto trace threshold for that metric in agent in order to automatically trace all transactions that are above the baseline.
The WARN message means that setting the threshold for the metric in agent failed. Most likely, because agent did not find required metric in its metric pool or there was a connection problem between EM-Agent and therefore cannot set the threshold value.
The impact is that the auto tracing part of this feature does not work for the metric in question.
From agent perspective things are quite simple, it receives the request from EM with metric name and threshold. It looks it up and sets value. When transactions happen, value is checked and auto tracing is triggered.
Agent does have DEBUG and TRACE logs to track these requests to see what metric EM is setting, where agent is looking for it and, if and why it fails:
log4j.logger.IntroscopeAgent.IntelligentInstrumentationService=TRACE#com.wily.util.feedback.Log4JSeverityLevel
Suggestions:
1) Check for any possible Agent-EM connection issue, check the Agent logs, use the Agent supportability metrics
2) Enable TRACE for SI as indicated above, try to correlate the WARN messages with the agent logs and find out if the EM is trying to set threshold incorrectly.
3) Go to the SuperDomain|slzpsr000000082|JBoss|AS2Remediation_PRODSG01-server03, "traces" tab in the investigator and try to find the failing traces.
If you are unable to identify the root cause, open a support case and attach EM + agent TRACE logs + a backup of your traces db.
I hope this helps,
Regards,
Sergio