Aias00 commented on PR #7709:
URL: https://github.com/apache/shenyu/pull/7709#issuecomment-6307589737

   CI status on `43a4c029`, with the root cause rather than just "red":
   
   **`pr_build` fails, and it is a test failure, not a compile error:**
   
   ```
   [shenyu-plugin-tracing] [ERROR] 
TraceContinuityTest.testTheGatewayStartsTheTraceWithoutAnIncomingContext:134->onlyFinishedSpan:203
 expected: <1> but was: <0>
   [shenyu-plugin-tracing] [ERROR] Tests run: 71, Failures: 1, Errors: 0, 
Skipped: 0
   [ERROR] Failed to execute goal ...maven-surefire-plugin:3.0.0-M4:test 
(default-test) on project shenyu-plugin-tracing
   ```
   
   **`build` (the `checking job status` aggregator) fails in 3 s** as a direct 
consequence — its only non-setup step is `checking job status`, so it has no 
independent signal. One root cause, two red checks.
   
   Where to look: `shenyu-plugin-tracing` is not a module this PR modifies, so 
either
   
   1. the new bootstrap distribution changes what is on the reactor classpath 
for that module (most likely if the new module declares a tracing-related 
dependency or shades one), or
   2. the test is timing-sensitive and this is flake — `onlyFinishedSpan` 
waiting for a span to be exported with no incoming context is exactly the shape 
that fails on a loaded runner.
   
   Please determine which before re-running, because a plain re-run only 
distinguishes them by luck. Running `./mvnw -pl shenyu-plugin-tracing test 
-Dtest=TraceContinuityTest` on the branch head locally will settle it in a 
minute: if it passes repeatedly on a quiet machine and the module's 
dependencies are unchanged by your PR, it is flake and a re-run is the right 
answer.
   
   Separately, a scope note: this PR is +3526/−2 across 41 files, which is past 
the size where a reviewer can meaningfully audit it. It would review much 
better split into (a) the bootstrap assembly/distribution and (b) the code 
changes it needs — the distribution change can then be judged against the 
existing `shenyu-bootstrap` module instead of being bundled with it. Not a 
blocker, just the reason the review here stops at CI.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to