TiDB 6.5.5 Release Notes
Release date: September 21, 2023
TiDB version: 6.5.5
Quick access: Quick start | Production deployment
Improvements
TiDB
- Add new optimizer hints, including NO_MERGE_JOIN(), NO_INDEX_JOIN(), NO_INDEX_MERGE_JOIN(), NO_HASH_JOIN(), and NO_INDEX_HASH_JOIN() #45520 @qw4990
- Add request source information related to the coprocessor #46514 @you06
TiKV
- Add the backoff mechanism for the PD client in the process of connection retries, which gradually increases retry intervals during error retries to reduce PD pressure #15428 @nolouch
- Add monitoring metrics for snapshots #15401 @SpadeA-Tang
- Improve stability of PITR checkpoint lag during leader transfers #13638 @YuJuncen
- Add logs and monitoring metrics related to
safe-ts
#15082 @ekexium - Provide more logs and monitoring metrics for
resolved-ts
#15082 @ekexium
Tools
Bug fixes
TiDB
- Fix the issue that Stale Read might select an unavailable replica #46198 @zyguan
- Fix the issue that additional overhead is incurred due to the incompatibility between Stale Read and Schema Cache #43481 @crazycs520
TiKV
- Fix the issue that TiKV fails to start when Titan is enabled and the
Blob file deleted twice
error occurs #15454 @Connor1996 - Fix the issue that Online Unsafe Recovery cannot handle merge abort #15580 @v01dstar
- Fix the issue that TiKV fails to start when Titan is enabled and the
PD
- Fix the issue that the scheduler takes a long time to start up #6920 @HuSharp
- Fix the issue that the logic for handling Leaders and Peers in Scatter Region is inconsistent #6962 @bufferflies
- Fix the issue that the
empty-region-count
monitoring metric is abnormal when the cluster is restarted or the PD Leader is switched #7008 @CabinfeverB
Tools
Backup & Restore (BR)
- Fix the issue that restoring implicit primary keys by PITR might lead to conflicts #46520 @3pointer
- Fix the issue that an error occurs when PITR recovers the meta-kv #46578 @Leavrth
- Fix an error in BR integration test cases #46561 @purelind
- Fix the issue that PITR fails to restore data from GCS #47022 @Leavrth
TiCDC
- Fix the issue of high TiCDC replication latency caused by network isolation of PD nodes #9565 @asddongmen
- Fix the issue that TiCDC incorrectly changes the
UPDATE
operation toINSERT
when using the CSV format #9658 @3AceShowHand - Fix the issue that user passwords are recorded in some logs #9690 @sdojjy
- Fix the issue that using the SASL authentication might cause TiCDC to panic #9669 @sdojjy
- Fix the issue that TiCDC replication tasks might fail in some corner cases #9685 #9697 #9695 #9736 @hicqu @CharlesCheung96
- Fix the issue that TiCDC cannot recover quickly from TiKV node failures when there are a lot of Regions upstream #9741 @sdojjy
TiDB Lightning
- Fix the issue that TiDB Lightning fails to start when TiCDC is deployed on the target server #41040 @lance6716
- Fix the issue that TiDB Lightning fails to start when PD topology is changed #46688 @lance6716
- Fix the issue that TiDB Lightning cannot continue importing data after PD switching Leaders #46540 @lance6716
- Fix the issue that precheck cannot accurately detect the presence of a running TiCDC in the target cluster #41040 @lance6716