TiDB 2.0.6 Release Notes
On August 6, 2018, TiDB 2.0.6 is released. Compared with TiDB 2.0.5, this release has great improvement in system compatibility and stability.
TiDB
- Improvements
- Make "set system variable" log shorter to save disk space #7031
- Record slow operations during the execution of
ADD INDEX
in the log, to make troubleshooting easier #7083 - Reduce transaction conflicts when updating statistics #7138
- Improve the accuracy of row count estimation when the values pending to be estimated exceeds the statistics range #7185
- Choose the table with a smaller estimated row count as the outer table for
Index Join
to improve its execution efficiency #7277 - Add the recover mechanism for panics occurred during the execution of
ANALYZE TABLE
, to avoid that the tidb-server is unavailable caused by abnormal behavior in the process of collecting statistics #7228 - Return
NULL
and the corresponding warning when the results ofRPAD
/LPAD
exceed the value of themax_allowed_packet
system variable, compatible with MySQL #7244 - Set the upper limit of placeholders count in the
PREPARE
statement to 65535, compatible with MySQL #7250
- Bug Fixes
- Fix the issue that the
DROP USER
statement is incompatible with MySQL behavior in some cases #7014 - Fix the issue that statements like
INSERT
/LOAD DATA
meet OOM after openingtidb_batch_insert
#7092 - Fix the issue that the statistics fail to automatically update when the data of a table keeps updating #7093
- Fix the issue that the firewall breaks inactive gPRC connections #7099
- Fix the issue that prefix index returns a wrong result in some scenarios #7126
- Fix the panic issue caused by outdated statistics in some scenarios #7155
- Fix the issue that one piece of index data is missed after the
ADD INDEX
operation in some scenarios #7156 - Fix the wrong result issue when querying
NULL
values using the unique index in some scenarios #7172 - Fix the messy code issue of the
DECIMAL
multiplication result in some scenarios #7212 - Fix the wrong result issue of
DECIMAL
modulo operation in some scenarios #7245 - Fix the issue that the
UPDATE
/DELETE
statement in a transaction returns a wrong result under some special sequence of statements #7219 - Fix the panic issue of the
UNION ALL
/UPDATE
statement during the process of building the execution plan in some scenarios #7225 - Fix the issue that the range of prefix index is calculated incorrectly in some scenarios #7231
- Fix the issue that the
LOAD DATA
statement fails to write the binlog in some scenarios #7242 - Fix the wrong result issue of
SHOW CREATE TABLE
during the execution process ofADD INDEX
in some scenarios #7243 - Fix the issue that panic occurs when
Index Join
does not initialize timestamps in some scenarios #7246 - Fix the false alarm issue when
ADMIN CHECK TABLE
mistakenly uses the timezone in the session #7258 - Fix the issue that
ADMIN CLEANUP INDEX
does not clean up the index in some scenarios #7265 - Disable the Read Committed isolation level #7282
- Fix the issue that the
TiKV
- Improvements
- Enlarge scheduler's default slots to reduce false conflicts
- Reduce continuous records of rollback transactions, to improve the Read performance when conflicts are extremely severe
- Limit the size and number of RocksDB log files, to reduce unnecessary disk usage in long-running condition
- Bug Fixes
- Fix the crash issue when converting the data type from string to decimal