fix: Fixed undici/fetch instrumentation to properly assign the parent-id portion of the traceparent header on outgoing requests to the active http external span id
#2951
Conversation
…-id portion of the `traceparent` header on outgoing requests to the active http external span id
| createExternalSegment({ shim, request, segment }) | ||
| addDTHeaders({ transaction, config, request }) |
There was a problem hiding this comment.
So there's some side effects in these function such that ordering of their invocation matters in order to get the correct parent id attached to the new request?
There was a problem hiding this comment.
yes. in createExternalSegment we create the segment and set it as active, then assign the outgoing w3c headers. Before we were setting outgoing w3c headers so it was on the parent span of the undici/fetch request. The versioned test proves this, if you swap the order the test fails
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #2951 +/- ##
==========================================
- Coverage 97.35% 97.30% -0.05%
==========================================
Files 317 317
Lines 48648 48688 +40
==========================================
+ Hits 47360 47378 +18
- Misses 1288 1310 +22
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
Description
See details in #2950 as to why. This PR adds more tests to the relevant places that inject
traceparentto ensure the parts are the expected values.Related Issues
Closes #2950