I'm not familiar with Weblogic EM/SOA, but it looks like Flow Trace here is very specific to Weblogic. Does it behaves more like a message-driven framework (like Kafka/JMS), or more like a request/response lifecycle where each would be instrumented as a transaction ?
We don't have automatic support for this, but you could still use APM agent + our Public API to create transactions programatically by using any hooks that Weblogic could provide (if there is something similar to Filter in Servlet API, that's a good candidate).
How can I configure the agent to commit a transaction based on a specific ID in the metadata? Any doc? weblogic has a flow ID, and I want to make a timelilne based on this ID, and not by classes and methods
So, while the trace itself will remain "as it was executed and captured by the agent", you can build correlation between the Flow ID and the Transaction/Span ID.
Also, if using log correlation feature, you can inject the Trace ID into the application logs. I need to check if we can also use tags here to allow injecting both Transaction ID and the Flow ID directly into logs.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.