dzr171712 opened a new pull request, #66752:
URL: https://github.com/apache/doris/pull/66752
### What problem does this PR solve?
Issue Number: close #xxx
Related PR: None
Problem Summary:
In Cloud mode, cumulative compaction requests the tablet global compaction
lock before increasing `_cumu_compaction_thread_pool_used_threads`.
The cleanup `Defer` is created before the lock request. If the global lock
request fails, the task returns before the thread counter is incremented, but
the cleanup logic still decrements the counter. As a result,
`_cumu_compaction_thread_pool_used_threads` can become negative.
A negative thread counter can cause the cumulative compaction scheduling
logic
to calculate the remaining thread capacity incorrectly and affect the
large-task delay decision.
This PR adds a guard to ensure that the thread counter is decremented only
when it has actually been incremented.
A regression unit test is added to simulate global compaction lock failure
and
verify that the counter remains balanced.
### Release note
None
### Check List (For Author)
- Test <!-- At least one of them must be included. -->
- [ ] Regression test
- [✅] Unit Test
- [ ] Manual test (add detailed scripts or steps below)
- [ ] No need to test or manual test. Explain why:
- [ ] This is a refactor/code format and no logic has been changed.
- [ ] Previous test can cover this change.
- [ ] No code files have been changed.
- [ ] Other reason <!-- Add your reason? -->
Test case:
```text
CloudCompactionTest.cumulative
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
