-
Type:
Improvement
-
Resolution: Unresolved
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
StorEng - Defined Pipeline
-
5
Summary
When a rollback failure occurs in the python tests, there are retries of the test (via WT-9063), that should limit the number of failed runs. However, when we get a rollback error, there's nothing in the output log about this. There is info in results.txt, but we need to download artifacts to see it, and does anybody know what to search for? And how many times we are retrying, and did they really all fail? In fact, if we're running with an old version (< 3.7) of Python for whatever reason, the retry mechanism will not work (the first one will be a hard fail). A good log message will remove any doubt.
Inspired by WT-9575.
Motivation
- Does this affect any team outside of WT? No
- How likely is it that this use case or problem will occur? Certain
- If the problem does occur, what are the consequences and how severe are they? Minor
- Is this issue urgent? No