TiDB 3.0.9 Release Notes

Release date: January 14, 2020

TiDB version: 3.0.9

TiDB Ansible version: 3.0.9

Warning:

Some known issues are found in this version, and these issues are fixed in new versions. It is recommended that you use the latest 3.0.x version.

TiDB

  • Executor
    • Fix the incorrect result when the aggregate function is applied to the ENUM column and the collection column #14364
  • Server
    • Support the auto_increment_increment and auto_increment_offset system variables #14396
    • Add the tidb_tikvclient_ttl_lifetime_reach_total monitoring metric to monitor the number of pessimistic transactions with a TTL of 10 minutes #14300
    • Output the SQL information in the log when the SQL query causes a panic during its execution #14322
    • Add the plan and plan_digest fields in the statement summary table to record the plan that is being executed and the plan signature #14285
    • Adjust the default value of the stmt-summary.max-stmt-count configuration item from 100 to 200 #14285
    • Add the plan_digest field in the slow query table to record the plan signature #14292
  • DDL
    • Fix the issue that the results of anonymous indexes created using alter table ... add index on the primary column is inconsistent with MySQL #14310
    • Fix the issue that VIEWs are mistakenly dropped by the drop table syntax #14052
  • Planner
    • Optimize the performance of statements such as select max(a), min(a) from t. If an index exists in the a column, the statement is optimized to select * from (select a from t order by a desc limit 1) as t1, (select a from t order by a limit 1) as t2 to avoid full table scan #14410

TiKV

  • Raftstore
    • Speed up the configuration change to speed up the Region scattering #6421
  • Transaction
    • Add the tikv_lock_manager_waiter_lifetime_duration, tikv_lock_manager_detect_duration, and tikv_lock_manager_detect_duration monitoring metrics to monitor waiters’ lifetime, the time cost of detecting deadlocks, and the status of Wait table #6392
    • Optimize the following configuration items to reduce transaction execution latency caused by changing Region leader or the leader of deadlock detector in extreme situations #6429
      • Change the default value of wait-for-lock-time from 3s to 1s
      • Change the default value of wake-up-delay-duration from 100ms to 20ms
    • Fix the issue that the leader of the deadlock detector might be incorrect during the Region Merge process #6431

PD

  • Support using backlash / in the location label name #2083
  • Fix the incorrect statistics because the tombstone store is mistakenly included by the label counter #2067

Tools

  • TiDB Binlog
    • Add the unique key information in the binlog protocol output by Drainer #862
    • Support using the encrypted password for database connection for Drainer #868

TiDB Ansible

  • Support automatically creating directories to optimize the deployment of TiDB Lightning #1105