This repository has been archived by the owner on Nov 14, 2024. It is now read-only.
0.235.2-rc1
Pre-release
Pre-release
·
2644 commits
to develop
since this release
33f7748
This commit was signed with the committer’s verified signature.
The key has expired.
| Type | Description | Link |
|---|---|---|
| Fix | Four small metrics cuts: - Only log the background sweep batch multiplier if <1. - Don't instrument TimestampManagementService endpoints (used only in timelock migration): saves 10 series.- Don't instrument CoordinationService endpoints (used only in transactions2 migration): saves 10 series.- Don't publish timelockSuccessfulRequest and timelockFailedRequest: saves 4 series.Please contact the AtlasDB team if the removal of these series will be an issue. |
#4867 |
| Improvement | Some Cassandra client pool metrics have been removed, to save costs: - approximatePoolSize: please use numIdle + numActive instead.- proportionDestroyedByEvictor: please use destroyedByEvictor / created instead.- destroyedByBorrower and proportionDestroyedByBorrower: these have been 0 everywhere in production for the past year, because we do not do any validation on objects returned from the pool (at least, within the object pooling framework). |
#4873 |
| Fix | TimeLock is now more resilient to single nodes being slow, even in the learners. | #4896 |