Releases · tikv/pd

Improvements

Support managing the global memory threshold to alleviate the OOM problem (experimental) #5827 @hnes
Add the GC Tuner to alleviate the GC pressure (experimental) #5827 @hnes
Add the balance-witness-scheduler scheduler to schedule witness #5763 @ethercflow
Add the evict-slow-trend-scheduler scheduler to detect and schedule abnormal nodes #5808 @innerr
Add the keyspace manager to manage keyspace #5293 @AmoebaProtozoa

Fix the issue that the Region Scatter task generates redundant replicas unexpectedly #5909 @HundunDM
Fix the issue that the Online Unsafe Recovery feature would get stuck and time out in auto-detect mode #5753 @Connor1996
Fix the issue that the execution replace-down-peer slows down under certain conditions #5788 @HundunDM
Fix the PD OOM issue that occurs when the calls of ReportMinResolvedTS are too frequent #5965 @HundunDM

Fix the issue that a removed tombstone store appears again after the PD leader transfer #4941
Fix the issue that scheduling cannot start immediately after the PD leader transfer #4769
Fix the wrong status code of not leader #4797
Fix the issue that PD cannot correctly handle dashboard proxy requests #5321
Fix a bug of TSO fallback in some corner cases #4884
Fix the issue that the TiFlash learner replica might not be created in specific scenarios #5401
Fix the issue that the label distribution has residual labels in the metrics #4825
Fix the issue that when there exists a Store with large capacity (2T for example), fully allocated small Stores cannot be detected, which results in no balance operator being generated #4805
Fix the issue that schedulers do not work when SchedulerMaxWaitingOperator is set to 1 #4946