HappenLee opened a new pull request, #68637:
URL: https://github.com/apache/doris/pull/68637
### What problem does this PR solve?
Issue Number: N/A
Related PR: N/A
Problem Summary:
Creating the workload group that reaches `workload_group_max_num` is
incorrectly rejected. With the default limit of 15, a compute group containing
`normal` and 13 other workload groups cannot create its 15th group.
`checkGlobalUnlock` already includes the candidate group in its count, so
reject only counts greater than the configured maximum. Keep the existing
per-compute-group accounting and create-only check.
Add parameterized manager tests for limits 1, 2, and 15, including the
`normal` group. Verify that two compute groups can independently reach the
limit and that rejected creations leave both metadata maps and the edit log
unchanged.
### Release note
Fix an off-by-one check that prevented each compute group from reaching the
configured `workload_group_max_num` limit.
### Check List (For Author)
- Test:
- [x] Unit Test: `MAVEN_ARGS='-T48' ./run-fe-ut.sh --run
org.apache.doris.resource.workloadgroup.WorkloadGroupMgrTest` — 10 tests passed.
- Confirmed all 3 new parameterized boundary cases fail against the
original implementation.
- `mvn -T48 checkstyle:check -pl fe-core` — passed with 0 violations.
- `git diff --check` — passed.
- Behavior changed:
- [x] Yes. Allow exactly the configured number of workload groups per
compute group; continue rejecting one more than the limit.
- Does this need documentation?
- [x] No. Restore the existing maximum-count contract.
--
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]
