tikv-server v6.1.4

Bug fixes

• Fix the issue that TiDB fails to start on Gitpod when there are multiple cgroup and mountinfo records #13660 @tabokie
• Fix the issue that tikv-ctl is terminated unexpectedly when executing the reset-to-version command #13829 @tabokie
• Fix the issue that TiKV mistakenly reports a PessimisticLockNotFound error #13425 @sticnarf
• Fix the issue that TiKV might panic when the size of one single write exceeds 2 GiB #13848 @YuJuncen
• Fix the data inconsistency issue caused by network failure between TiDB and TiKV during the execution of a DML after a failed pessimistic DML #14038 @MyonKeminta
• Fix the issue that _ in the LIKE operator cannot match non-ASCII characters when new collation is not enabled #13769 @YangKeao @tonyxuqqi

tikv-server v6.5.0

For new features, improvements, and bug fixes released in 6.5.0 for tikv, see TiDB 6.5.0 release notes.

tikv-server v5.1.5

Bug fixes

• Fix the issue of time parsing error that occurs when the DATETIME values contain a fraction and Z #12739
• Fix a bug that replica reads might violate the linearizability #12109
• Fix a bug that Regions might be overlapped if Raftstore is busy #13160
• Fix the TiKV panic issue that occurs when applying snapshot is aborted #11618
• Fix a bug that TiKV might panic if it has been running for 2 years or more #11940
• Fix the panic issue that might occur when the source peer catches up logs by snapshot in the Region merge process #12663
• Fix the issue that TiKV panics when performing type conversion for an empty string #12673
• Fix a bug that stale messages cause TiKV to panic #12023
• Fix the panic issue that might occur when a peer is being split and destroyed at the same time #12825
• Fix the TiKV panic issue that occurs when the target peer is replaced with the peer that is destroyed without being initialized when merging a Region #12048
• Fix the issue that TiKV reports the invalid store ID 0 error when using Follower Read #12478
• Fix the possible duplicate commit records in pessimistic transactions when async commit is enabled #12615
• Support configuring the unreachable_backoff item to avoid Raftstore broadcasting too many messages after one peer becomes unreachable #13054
• Fix the issue that successfully committed optimistic transactions may report the Write Conflict error when the network is poor #34066
• Fix the wrong expression of Unified Read Pool CPU in dashboard #13086

tikv-server v6.1.3

Bug fixes

• Fix abnormal Region competition caused by expired lease during snapshot acquisition #13553 @SpadeA-Tang

tikv-server v5.3.4

Improvements

• Reload TLS certificate automatically for each update to improve availability #12546

tikv-server v6.4.0

Improvements

• Add a new configuration item apply-yield-write-size to control the maximum number of bytes that the Apply thread can write for one Finite-state Machine in one round of poll, and relieve Raftstore congestion when the Apply thread writes a large volume of data #13313 @glorv
• Warm up the entry cache before migrating the leader of Region to avoid QPS jitter during the leader transfer process #13060 @cosven
• Support pushing down the json_constains operator to Coprocessor #13592 @lizhenhuan
• Add the asynchronous function for CausalTsProvider to improve the flush performance in some scenarios #13428 @zeminzhou

Bug fixes

• Fix the issue that TiDB fails to start on Gitpod when there are multiple cgroup and mountinfo records #13660 @tabokie
• Fix the wrong expression of a TiKV metric tikv_gc_compaction_filtered #13537 @Defined2014
• Fix the performance issue caused by the abnormal delete_files_in_range #13534 @tabokie
• Fix abnormal Region competition caused by expired lease during snapshot acquisition #13553 @SpadeA-Tang
• Fix errors occurred when FLASHBACK fails in the first batch #13672 #13704 #13723 @HuSharp

tikv-server v6.1.2

Improvements

• Support configuring the unreachable_backoff item to avoid Raftstore broadcasting too many messages after one peer becomes unreachable #13054 @5kbpers
• Support configuring the RocksDB write stall settings to a value smaller than the flow control threshold #13467 @tabokie

Bug fixes

• Fix the issue that the snapshot data might be incomplete caused by batch snapshot across Regions #13553 @SpadeA-Tang
• Fix the issue of QPS drop when flow control is enabled and level0_slowdown_trigger is set explicitly #11424 @Connor1996
• Fix the issue that causes permission denied error when TiKV gets an error from the web identity provider and fails back to the default provider #13122 @3pointer
• Fix the issue that the TiKV service is unavailable for several minutes when a TiKV instance is in an isolated network environment #12966 @cosven

tikv-server v5.4.3

Improvements

• Support configuring the RocksDB write stall settings to a value smaller than the flow control threshold #13467
• Support configuring the unreachable_backoff item to avoid Raftstore broadcasting too many messages after one peer becomes unreachable #13054

Bug fixes

• Fix the issue of continuous SQL execution errors in the cluster after the PD leader is switched or PD is restarted #12934
• Cause: This issue is caused by a TiKV bug that TiKV does not retry sending heartbeat information to PD client after heartbeat requests fail, until TiKV reconnects to PD client. As a result, the Region information on the failed TiKV node becomes outdated, and TiDB cannot get the latest Region information, which causes SQL execution errors.
• Affected versions: v5.3.2 and v5.4.2. This issue has been fixed in v5.3.3 and v5.4.3. If you are using v5.4.2, you can upgrade your cluster to v5.4.3.
• Workaround: In addition to upgrade, you can also restart the TiKV nodes that cannot send Region heartbeat to PD, until there is no Region heartbeat to send.
• Fix the issue that causes permission denied error when TiKV gets an error from the web identity provider and fails back to the default provider #13122
• Fix the issue that the PD client might cause deadlocks #13191
• Fix the issue that Regions might be overlapped if Raftstore is busy #13160

tikv-server v6.3.0

Improvements

• Support configuring the unreachable_backoff item to avoid Raftstore broadcasting too many messages after one peer becomes unreachable #13054 @5kbpers
• Improve the fault tolerance of TSO service #12794 @pingyu
• Support dynamically modifying the number of sub-compaction operations performed concurrently in RocksDB (rocksdb.max-sub-compactions) #13145 @ethercflow
• Optimize the performance of merging empty Regions #12421 @tabokie
• Support more regular expression functions #13483 @gengliqi
• Support automatically adjusting the thread pool size based on the CPU usage #13313 @glorv

Bug fixes

• Fix the issue that PD does not reconnect to TiKV after the Region heartbeat is interrupted #12934 @bufferflies
• Fix the issue that Regions might be overlapped if Raftstore is busy #13160 @5kbpers
• Fix the issue that the PD client might cause deadlocks #13191 @bufferflies #12933 @BurtonQin
• Fix the issue that TiKV might panic when encryption is disabled #13081 @jiayang-zheng
• Fix the wrong expression of Unified Read Pool CPU in Dashboard #13086 @glorv
• Fix the issue that the TiKV service is unavailable for several minutes when a TiKV instance is in an isolated network environment #12966 @cosven
• Fix the issue that TiKV mistakenly reports a PessimisticLockNotFound error #13425 @sticnarf
• Fix the issue that PITR might cause data loss in some situations #13281 @YuJuncen
• Fix the issue that causes checkpoint not advanced when there are some long pessimistic transactions #13304 @YuJuncen
• Fix the issue that TiKV does not distinguish the datetime type (DATETIME, DATE, TIMESTAMP and TIME) and STRING type in JSON #13417 @YangKeao
• Fix incompatibility with MySQL of comparison between JSON bool and other JSON value #13386 #37481 @YangKeao

tikv-server v5.3.3

Bug fix

• Fix the issue of continuous SQL execution errors in the cluster after the PD leader is switched or PD is restarted.