Merge pull request #9660 from harness/PIPE-25753

Update documentation to clarify the behavior of EXPIRE_ALL status wrt to failure strategy
harness · Mar 10, 2025 · e5525a1 · e5525a1
2 parents 8b7e89f + 41c4ebc
commit e5525a1
Showing 1 changed file with 4 additions and 0 deletions.
diff --git a/docs/platform/pipelines/failure-handling/timeout-settings.md b/docs/platform/pipelines/failure-handling/timeout-settings.md
@@ -22,6 +22,10 @@ Stage timeouts mark stages as expired in the following scenarios:
 * The stage does not complete before the configured timeout.
 * The pipeline reaches its timeout limit when the stage is still running.
 
+:::info note
+When a stage time out is less than it's total step execution time, the failure strategy applied is `EXPIRE_ALL`, preventing rollback. This occurs because child nodes (steps) inherit the expired status from the stage, stopping further execution.
+:::
+
 ## Pipeline timeouts
 
 The pipeline timeout applies to the duration of the entire pipeline. It does not override stage/step timeouts, but stages and steps will timeout if the pipeline timeout limit is hit while a stage/step is running.