The prototype in the Java agent appears to be uni-directional, where it receives a notification that a span started or ended, and then it starts/stops a transaction/segment using the internal New Relic data model.
It doesn't appear that this internal state is reflected back into the OTel API.
Maybe this uni-directional approach is good enough to start out with?
Maybe we can have a better offering if we can support reflecting some of our internal context back via the OTel APIs?
workato-integration[bot]
commented
1 month ago
The prototype in the Java agent appears to be uni-directional, where it receives a notification that a span started or ended, and then it starts/stops a transaction/segment using the internal New Relic data model.
It doesn't appear that this internal state is reflected back into the OTel API.
Maybe this uni-directional approach is good enough to start out with?
Maybe we can have a better offering if we can support reflecting some of our internal context back via the OTel APIs?
See the results described here for an example of some of the issues that need to be addressed: https://github.com/newrelic/newrelic-java-agent/issues/2060#issuecomment-2433868991