From ed4906078058813dae360ef4ffca6628be9122f0 Mon Sep 17 00:00:00 2001 From: ti-srebot <66930949+ti-srebot@users.noreply.github.com> Date: Mon, 20 Jun 2022 16:59:59 +0800 Subject: [PATCH 01/38] releases: add TiDB 5.3.2-16-59-57 release notes * releases: add TiDB 5.3.2-16-59-57 release notes * Update releases/release-5.3.2-16-59-57.md --- releases/release-5.3.2-16-59-57.md | 178 +++++++++++++++++++++++++++++ 1 file changed, 178 insertions(+) create mode 100644 releases/release-5.3.2-16-59-57.md diff --git a/releases/release-5.3.2-16-59-57.md b/releases/release-5.3.2-16-59-57.md new file mode 100644 index 000000000000..94a05015fe01 --- /dev/null +++ b/releases/release-5.3.2-16-59-57.md @@ -0,0 +1,178 @@ +--- +title: TiDB 5.3.2 Release Notes +category: Releases +--- + + + +# TiDB 5.3.2 Release Notes + +Release Date: June 20, 2022 + +TiDB version: 5.3.2 + +## __unsorted + ++ PingCAP/TiDB + + - ```release-note [#35498](https://github.com/pingcap/tidb/pull/35498) + - ```release-note [#35405](https://github.com/pingcap/tidb/pull/35405) + - ```release-note [#35398](https://github.com/pingcap/tidb/pull/35398) + - ```release-note [#35384](https://github.com/pingcap/tidb/pull/35384) + - ```release-note [#35371](https://github.com/pingcap/tidb/pull/35371) + - Fixed an issue where extra datums may break binlog. [#35165](https://github.com/pingcap/tidb/pull/35165) + - ```release-note [#34618](https://github.com/pingcap/tidb/pull/34618) + - ```release-note [#34516](https://github.com/pingcap/tidb/pull/34516) + - ```release-note [#34474](https://github.com/pingcap/tidb/pull/34474) + - ```release-note [#34432](https://github.com/pingcap/tidb/pull/34432) + - ```release-note [#34377](https://github.com/pingcap/tidb/pull/34377) + - ```release-note [#34334](https://github.com/pingcap/tidb/pull/34334) + - lightning: split and scatter regions in batches [#34257](https://github.com/pingcap/tidb/pull/34257) + - ```release-note [#34225](https://github.com/pingcap/tidb/pull/34225) + - Fix the issue that the table attributes don't support index and won't be updated when the partition changes [#34024](https://github.com/pingcap/tidb/pull/34024) + - Fix a bug of duplicate primary key when insert record into table after incremental restoration. [#33987](https://github.com/pingcap/tidb/pull/33987) + - ```release-note [#33960](https://github.com/pingcap/tidb/pull/33960) + - ```release-note [#33942](https://github.com/pingcap/tidb/pull/33942) + - ```release-note [#33913](https://github.com/pingcap/tidb/pull/33913) + - Fix the issue that the schedulers won't be resumed after BR/Lightning exits abnormally. [#33815](https://github.com/pingcap/tidb/pull/33815) + - Fix the issue that privilege-related operations may fail for upgraded clusters. [#33604](https://github.com/pingcap/tidb/pull/33604) + - Fix the issue that NewCollationEnable config not checked during restoration. [#33533](https://github.com/pingcap/tidb/pull/33533) + - Fix a bug that BR incremental restore return error by mistake caused by ddl job with empty query. [#33516](https://github.com/pingcap/tidb/pull/33516) + - Fix the issue that BR not retry enough when region not consistency during restoration. [#33469](https://github.com/pingcap/tidb/pull/33469) + - ```release-note [#33339](https://github.com/pingcap/tidb/pull/33339) + - Fix a bug that caused BR get stuck when restore meets some unrecoverable error. [#33267](https://github.com/pingcap/tidb/pull/33267) + - executor: fix wrong result of delete multiple tables using left join [#33122](https://github.com/pingcap/tidb/pull/33122) + - planner: Fix the issue that TiDB may dispatch duplicated tasks to TiFlash [#32840](https://github.com/pingcap/tidb/pull/32840) + - lightning: fix checksum encountered “GC life time is shorter than transaction duration” error [#32798](https://github.com/pingcap/tidb/pull/32798) + - Fix BR failure on backup rawkv. [#32792](https://github.com/pingcap/tidb/pull/32792) + - Fix some connections and goroutines leak caused by not closed HTTP response body [#30600](https://github.com/pingcap/tidb/pull/30600) + + ++ TiKV/TiKV + + - Fix bug which causes frequent pd client reconnection [#12832](https://github.com/tikv/tikv/pull/12832) + - Report bad health status if raftstore stops working. [#12817](https://github.com/tikv/tikv/pull/12817) + - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12745](https://github.com/tikv/tikv/pull/12745) + - Fix tikv crash when conv empty string [#12692](https://github.com/tikv/tikv/pull/12692) + - Fix possible duplicate commit record in async-commit pessimistic transactions. [#12653](https://github.com/tikv/tikv/pull/12653) + - Fix a bug that sometimes generates a message with zero store id when doing follower read [#12529](https://github.com/tikv/tikv/pull/12529) + - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#12457](https://github.com/tikv/tikv/pull/12457) + - Report bad health status if raftstore stops working. [#12447](https://github.com/tikv/tikv/pull/12447) + - fix race between split check and destroy [#12405](https://github.com/tikv/tikv/pull/12405) + - Fixes that successfully committed optimistic transactions may report false WriteConflict on network errors. [#12377](https://github.com/tikv/tikv/pull/12377) + - fix tikv panic and peer unexpected destroy due to fake merge target [#12296](https://github.com/tikv/tikv/pull/12296) + - Fix stale message cause panic [#12287](https://github.com/tikv/tikv/pull/12287) + - Solve the problem of raft msg memory metrics overflow, which will cause intermittent packet loss and oom. [#12226](https://github.com/tikv/tikv/pull/12226) + - Fix crash when profiling in Ubuntu 18.04. [#12213](https://github.com/tikv/tikv/pull/12213) + - Fix logic of error string match in `bad-ssts`. [#12149](https://github.com/tikv/tikv/pull/12149) + - Pass leader transferee to cdc observer to reduce TiCDC latency spike. [#12136](https://github.com/tikv/tikv/pull/12136) + - Fix potential linearizability violation in replica reads. [#12119](https://github.com/tikv/tikv/pull/12119) + - fix panic when target peer is replaced with an destroyed uninitialized peer during merge [#12085](https://github.com/tikv/tikv/pull/12085) + - Fixes the bug that TiKV keep running over 2 years may panic. [#12041](https://github.com/tikv/tikv/pull/12041) + - None. [#11438](https://github.com/tikv/tikv/pull/11438) + - None. [#11380](https://github.com/tikv/tikv/pull/11380) + + ++ PingCAP/TiFlash + + - Fix potential query error when select on a table with many delete operations [#4759](https://github.com/pingcap/tiflash/pull/4759) + - Fix bug that TiFlash query will meet keepalive timeout error randomly. [#4731](https://github.com/pingcap/tiflash/pull/4731) + - Avoid leaving data on tiflash node which doesn't corresponding to any region range [#4716](https://github.com/pingcap/tiflash/pull/4716) + - Fix a bug that MPP tasks may leak threads forever [#4647](https://github.com/pingcap/tiflash/pull/4647) + - Fix the problem that empty segments cannot be merged after gc [#4523](https://github.com/pingcap/tiflash/pull/4523) + - Fix wrong result of cast(float as decimal) when overflow happens [#4387](https://github.com/pingcap/tiflash/pull/4387) + - Fix the potential crash issue that occurs when TLS is enabled [#4369](https://github.com/pingcap/tiflash/pull/4369) + - fix the problem that expired data was not recycled timely due to slow gc speed [#4254](https://github.com/pingcap/tiflash/pull/4254) + - Fix the bug that canceled MPP query may cause tasks hang forever when local tunnel is enabled. [#4236](https://github.com/pingcap/tiflash/pull/4236) + - Fix cast datetime to decimal wrong result bug [#4159](https://github.com/pingcap/tiflash/pull/4159) + - Fix the bug that invalid storage dir configurations lead to unexpected behavior [#4134](https://github.com/pingcap/tiflash/pull/4134) + - Fix the bug that some exceptions are not handled properly [#4114](https://github.com/pingcap/tiflash/pull/4114) + - Fix potential query error after add column under heavy read workload [#4027](https://github.com/pingcap/tiflash/pull/4027) + - Fix str_to_date() function incorrectly handles leading zeros when parsing Microseconds [#3936](https://github.com/pingcap/tiflash/pull/3936) + - Fix the bug that castStringAsReal has different behaivor between tiflash and tikv/tidb. [#3757](https://github.com/pingcap/tiflash/pull/3757) + - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3740](https://github.com/pingcap/tiflash/pull/3740) + - Fix random `EstablishMPPConnection` fail after TiFlash server restart. [#3708](https://github.com/pingcap/tiflash/pull/3708) + - Fix the bug that results of `where ` is wrong because it will be converted to int type. [#3480](https://github.com/pingcap/tiflash/pull/3480) + + ++ PD + + - server: disable swagger server [#5177](https://github.com/tikv/pd/pull/5177) + - Fix the issue that the hot region may cause panic due to no leader [#5040](https://github.com/tikv/pd/pull/5040) + - None. [#4999](https://github.com/tikv/pd/pull/4999) + - Fix the issue that scheduling cannot immediately start after PD leader transfers [#4969](https://github.com/tikv/pd/pull/4969) + - Fix the issue that the removed tombstone store shows again after transferring the PD leader [#4958](https://github.com/tikv/pd/pull/4958) + - Fix the corner case that may cause TSO fallback. [#4891](https://github.com/tikv/pd/pull/4891) + - None. [#4847](https://github.com/tikv/pd/pull/4847) + - Fix the issue that the label distribution has residual labels [#4831](https://github.com/tikv/pd/pull/4831) + + ++ Tools + + + PingCAP/TiCDC + + - Fix TiCDC incorrectly display stale metrics data on dashboard. [#5894](https://github.com/pingcap/tiflow/pull/5894) + - Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. + +If you don't think this PR needs a release note then fill it with `None`. [#5678](https://github.com/pingcap/tiflow/pull/5678) + - Fix a bug that after auto resume, DM will use more disk space. [#5636](https://github.com/pingcap/tiflow/pull/5636) + - Fix a bug in redo log manager that flush log executed before writing logs [#5628](https://github.com/pingcap/tiflow/pull/5628) + - Fix a bug that resolved ts moves too fast when part of tables are not maintained redo writer. [#5617](https://github.com/pingcap/tiflow/pull/5617) + - Add uuid suffix to redo log file name to prevent name conflict, which may cause data loss. [#5613](https://github.com/pingcap/tiflow/pull/5613) + - `Fix a bug that mysql sink may save a wrong checkpointTs`. [#5436](https://github.com/pingcap/tiflow/pull/5436) + - Please add a release note. + +Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. + +If you don't think this PR needs a release note then fill it with `None`. [#5419](https://github.com/pingcap/tiflow/pull/5419) + - Fix DM can't replicate uppercase tables when the task has case-sensitive: false [#5308](https://github.com/pingcap/tiflow/pull/5308) + - Fix an issue where TiCDC clusters may panic after an upgrade [#5304](https://github.com/pingcap/tiflow/pull/5304) + - save table checkpoint after a DDL is filtered [#5292](https://github.com/pingcap/tiflow/pull/5292) + - `None`. [#5226](https://github.com/pingcap/tiflow/pull/5226) + - fix the issue that ticdc failed to start when connects to multiple pd endpoints with tls-enabled and the 1st endpoint is not available [#5209](https://github.com/pingcap/tiflow/pull/5209) + - fix tracker panic when pk of downstream table orders behind [#5178](https://github.com/pingcap/tiflow/pull/5178) + - `None` [#5171](https://github.com/pingcap/tiflow/pull/5171) + - fix tracker panic when pk of downstream table orders behind [#5165](https://github.com/pingcap/tiflow/pull/5165) + - `Fix a bug that openapi may be stuck when pd is abnormal` [#5111](https://github.com/pingcap/tiflow/pull/5111) + - send one heartbeat for successive skipped GTID when enable relay log [#5092](https://github.com/pingcap/tiflow/pull/5092) + - fix bug of relay log may turn off after master reboot in v5.3.1 [#4877](https://github.com/pingcap/tiflow/pull/4877) + - Syncer will use working directory of DM-worker rather than /tmp to write internal files, and clean it after task is stopped [#4804](https://github.com/pingcap/tiflow/pull/4804) + - Fix stability problem in workerpool, which is used by Unified Sorter. [#4682](https://github.com/pingcap/tiflow/pull/4682) + - Fix a bug that sequence should not be replicated even if force-replication is true. +Fix data loss when upstream transaction conflicts during cdc reconnection. [#4666](https://github.com/pingcap/tiflow/pull/4666) + - `None`. [#4650](https://github.com/pingcap/tiflow/pull/4650) + - Please add a release note. +`None`. [#4647](https://github.com/pingcap/tiflow/pull/4647) + - `None`. [#4345](https://github.com/pingcap/tiflow/pull/4345) + + +## Bug Fixes + ++ PingCAP/TiDB + + - fix the problem that dumpling can't dump with --compress and s3 output directory. [#35356](https://github.com/pingcap/tidb/pull/35356) + - Fix the data inconsistency caused by invalid usage of lazy existence check and untouch key optimization. [#30912](https://github.com/pingcap/tidb/pull/30912) + - Fix the bug that sql got cancel if including json column joins char column. [#30777](https://github.com/pingcap/tidb/pull/30777) + + ++ PingCAP/TiFlash + + - Fix potential wrong result after a lot of insert and delete operations [#4967](https://github.com/pingcap/tiflash/pull/4967) + - Fix the metadata corruption issue when `Prepare Merge` is triggered after a new election is finished but the isolated peer is not informed [#4495](https://github.com/pingcap/tiflash/pull/4495) + - Fix the issue that a query containing `JOIN` could be hung if an error was encountered [#4269](https://github.com/pingcap/tiflash/pull/4269) + + +## Compatibility Changes + ++ PingCAP/TiDB + + - Fix an issue that REPLACE statement changing other rows when the auto ID is out of range [#32324](https://github.com/pingcap/tidb/pull/32324) + + +## Improvements + ++ TiKV/TiKV + + - Delay in raft client to reduce syscalls and improve CPU efficiency [#11761](https://github.com/tikv/tikv/pull/11761) + + From c47862bea9861c52312bdc7631da962cda025bd4 Mon Sep 17 00:00:00 2001 From: Ran Date: Mon, 20 Jun 2022 17:05:27 +0800 Subject: [PATCH 02/38] update release notes Signed-off-by: Ran --- TOC.md | 1 + releases/{release-5.3.2-16-59-57.md => release-5.3.2.md} | 9 +++++++-- releases/release-notes.md | 1 + releases/release-timeline.md | 1 + 4 files changed, 10 insertions(+), 2 deletions(-) rename releases/{release-5.3.2-16-59-57.md => release-5.3.2.md} (98%) diff --git a/TOC.md b/TOC.md index af5b1fa58997..3bbd73ff21f4 100644 --- a/TOC.md +++ b/TOC.md @@ -771,6 +771,7 @@ - v5.4 - [5.4.0](/releases/release-5.4.0.md) - v5.3 + - [5.3.2](/releases/release-5.3.2.md) - [5.3.1](/releases/release-5.3.1.md) - [5.3.0](/releases/release-5.3.0.md) - v5.2 diff --git a/releases/release-5.3.2-16-59-57.md b/releases/release-5.3.2.md similarity index 98% rename from releases/release-5.3.2-16-59-57.md rename to releases/release-5.3.2.md index 94a05015fe01..b27201607087 100644 --- a/releases/release-5.3.2-16-59-57.md +++ b/releases/release-5.3.2.md @@ -1,16 +1,21 @@ --- title: TiDB 5.3.2 Release Notes -category: Releases --- # TiDB 5.3.2 Release Notes -Release Date: June 20, 2022 +Release Date: June xx, 2022 TiDB version: 5.3.2 +## Compatibility change(s) + +## Improvements + +## Bug fixes + ## __unsorted + PingCAP/TiDB diff --git a/releases/release-notes.md b/releases/release-notes.md index 960dd2d2ce13..ac41e71597d9 100644 --- a/releases/release-notes.md +++ b/releases/release-notes.md @@ -15,6 +15,7 @@ aliases: ['/docs/dev/releases/release-notes/','/docs/dev/releases/rn/'] ## 5.3 +- [5.3.2](/releases/release-5.3.2.md) - [5.3.1](/releases/release-5.3.1.md) - [5.3.0](/releases/release-5.3.0.md) diff --git a/releases/release-timeline.md b/releases/release-timeline.md index b7c99fe6fff4..6d8803aea9e6 100644 --- a/releases/release-timeline.md +++ b/releases/release-timeline.md @@ -9,6 +9,7 @@ This document shows all the released TiDB versions in reverse chronological orde | Version | Release Date | | :--- | :--- | +| [5.3.2](/releases/release-5.3.2.md) | 2022-06-xx | | [5.2.4](/releases/release-5.2.4.md) | 2022-04-26 | | [6.0.0-DMR](/releases/release-6.0.0-dmr.md) | 2022-04-07 | | [5.3.1](/releases/release-5.3.1.md) | 2022-03-03 | From 7a870fd50cbeeed2a60ae7727a4f8c58c08ce49b Mon Sep 17 00:00:00 2001 From: Ran Date: Mon, 20 Jun 2022 17:46:34 +0800 Subject: [PATCH 03/38] pull to issue Signed-off-by: Ran --- releases/release-5.3.2.md | 428 ++++++++++++++++++++++---------------- 1 file changed, 245 insertions(+), 183 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index b27201607087..1cb8233be17c 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -1,183 +1,245 @@ ---- -title: TiDB 5.3.2 Release Notes ---- - - - -# TiDB 5.3.2 Release Notes - -Release Date: June xx, 2022 - -TiDB version: 5.3.2 - -## Compatibility change(s) - -## Improvements - -## Bug fixes - -## __unsorted - -+ PingCAP/TiDB - - - ```release-note [#35498](https://github.com/pingcap/tidb/pull/35498) - - ```release-note [#35405](https://github.com/pingcap/tidb/pull/35405) - - ```release-note [#35398](https://github.com/pingcap/tidb/pull/35398) - - ```release-note [#35384](https://github.com/pingcap/tidb/pull/35384) - - ```release-note [#35371](https://github.com/pingcap/tidb/pull/35371) - - Fixed an issue where extra datums may break binlog. [#35165](https://github.com/pingcap/tidb/pull/35165) - - ```release-note [#34618](https://github.com/pingcap/tidb/pull/34618) - - ```release-note [#34516](https://github.com/pingcap/tidb/pull/34516) - - ```release-note [#34474](https://github.com/pingcap/tidb/pull/34474) - - ```release-note [#34432](https://github.com/pingcap/tidb/pull/34432) - - ```release-note [#34377](https://github.com/pingcap/tidb/pull/34377) - - ```release-note [#34334](https://github.com/pingcap/tidb/pull/34334) - - lightning: split and scatter regions in batches [#34257](https://github.com/pingcap/tidb/pull/34257) - - ```release-note [#34225](https://github.com/pingcap/tidb/pull/34225) - - Fix the issue that the table attributes don't support index and won't be updated when the partition changes [#34024](https://github.com/pingcap/tidb/pull/34024) - - Fix a bug of duplicate primary key when insert record into table after incremental restoration. [#33987](https://github.com/pingcap/tidb/pull/33987) - - ```release-note [#33960](https://github.com/pingcap/tidb/pull/33960) - - ```release-note [#33942](https://github.com/pingcap/tidb/pull/33942) - - ```release-note [#33913](https://github.com/pingcap/tidb/pull/33913) - - Fix the issue that the schedulers won't be resumed after BR/Lightning exits abnormally. [#33815](https://github.com/pingcap/tidb/pull/33815) - - Fix the issue that privilege-related operations may fail for upgraded clusters. [#33604](https://github.com/pingcap/tidb/pull/33604) - - Fix the issue that NewCollationEnable config not checked during restoration. [#33533](https://github.com/pingcap/tidb/pull/33533) - - Fix a bug that BR incremental restore return error by mistake caused by ddl job with empty query. [#33516](https://github.com/pingcap/tidb/pull/33516) - - Fix the issue that BR not retry enough when region not consistency during restoration. [#33469](https://github.com/pingcap/tidb/pull/33469) - - ```release-note [#33339](https://github.com/pingcap/tidb/pull/33339) - - Fix a bug that caused BR get stuck when restore meets some unrecoverable error. [#33267](https://github.com/pingcap/tidb/pull/33267) - - executor: fix wrong result of delete multiple tables using left join [#33122](https://github.com/pingcap/tidb/pull/33122) - - planner: Fix the issue that TiDB may dispatch duplicated tasks to TiFlash [#32840](https://github.com/pingcap/tidb/pull/32840) - - lightning: fix checksum encountered “GC life time is shorter than transaction duration” error [#32798](https://github.com/pingcap/tidb/pull/32798) - - Fix BR failure on backup rawkv. [#32792](https://github.com/pingcap/tidb/pull/32792) - - Fix some connections and goroutines leak caused by not closed HTTP response body [#30600](https://github.com/pingcap/tidb/pull/30600) - - -+ TiKV/TiKV - - - Fix bug which causes frequent pd client reconnection [#12832](https://github.com/tikv/tikv/pull/12832) - - Report bad health status if raftstore stops working. [#12817](https://github.com/tikv/tikv/pull/12817) - - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12745](https://github.com/tikv/tikv/pull/12745) - - Fix tikv crash when conv empty string [#12692](https://github.com/tikv/tikv/pull/12692) - - Fix possible duplicate commit record in async-commit pessimistic transactions. [#12653](https://github.com/tikv/tikv/pull/12653) - - Fix a bug that sometimes generates a message with zero store id when doing follower read [#12529](https://github.com/tikv/tikv/pull/12529) - - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#12457](https://github.com/tikv/tikv/pull/12457) - - Report bad health status if raftstore stops working. [#12447](https://github.com/tikv/tikv/pull/12447) - - fix race between split check and destroy [#12405](https://github.com/tikv/tikv/pull/12405) - - Fixes that successfully committed optimistic transactions may report false WriteConflict on network errors. [#12377](https://github.com/tikv/tikv/pull/12377) - - fix tikv panic and peer unexpected destroy due to fake merge target [#12296](https://github.com/tikv/tikv/pull/12296) - - Fix stale message cause panic [#12287](https://github.com/tikv/tikv/pull/12287) - - Solve the problem of raft msg memory metrics overflow, which will cause intermittent packet loss and oom. [#12226](https://github.com/tikv/tikv/pull/12226) - - Fix crash when profiling in Ubuntu 18.04. [#12213](https://github.com/tikv/tikv/pull/12213) - - Fix logic of error string match in `bad-ssts`. [#12149](https://github.com/tikv/tikv/pull/12149) - - Pass leader transferee to cdc observer to reduce TiCDC latency spike. [#12136](https://github.com/tikv/tikv/pull/12136) - - Fix potential linearizability violation in replica reads. [#12119](https://github.com/tikv/tikv/pull/12119) - - fix panic when target peer is replaced with an destroyed uninitialized peer during merge [#12085](https://github.com/tikv/tikv/pull/12085) - - Fixes the bug that TiKV keep running over 2 years may panic. [#12041](https://github.com/tikv/tikv/pull/12041) - - None. [#11438](https://github.com/tikv/tikv/pull/11438) - - None. [#11380](https://github.com/tikv/tikv/pull/11380) - - -+ PingCAP/TiFlash - - - Fix potential query error when select on a table with many delete operations [#4759](https://github.com/pingcap/tiflash/pull/4759) - - Fix bug that TiFlash query will meet keepalive timeout error randomly. [#4731](https://github.com/pingcap/tiflash/pull/4731) - - Avoid leaving data on tiflash node which doesn't corresponding to any region range [#4716](https://github.com/pingcap/tiflash/pull/4716) - - Fix a bug that MPP tasks may leak threads forever [#4647](https://github.com/pingcap/tiflash/pull/4647) - - Fix the problem that empty segments cannot be merged after gc [#4523](https://github.com/pingcap/tiflash/pull/4523) - - Fix wrong result of cast(float as decimal) when overflow happens [#4387](https://github.com/pingcap/tiflash/pull/4387) - - Fix the potential crash issue that occurs when TLS is enabled [#4369](https://github.com/pingcap/tiflash/pull/4369) - - fix the problem that expired data was not recycled timely due to slow gc speed [#4254](https://github.com/pingcap/tiflash/pull/4254) - - Fix the bug that canceled MPP query may cause tasks hang forever when local tunnel is enabled. [#4236](https://github.com/pingcap/tiflash/pull/4236) - - Fix cast datetime to decimal wrong result bug [#4159](https://github.com/pingcap/tiflash/pull/4159) - - Fix the bug that invalid storage dir configurations lead to unexpected behavior [#4134](https://github.com/pingcap/tiflash/pull/4134) - - Fix the bug that some exceptions are not handled properly [#4114](https://github.com/pingcap/tiflash/pull/4114) - - Fix potential query error after add column under heavy read workload [#4027](https://github.com/pingcap/tiflash/pull/4027) - - Fix str_to_date() function incorrectly handles leading zeros when parsing Microseconds [#3936](https://github.com/pingcap/tiflash/pull/3936) - - Fix the bug that castStringAsReal has different behaivor between tiflash and tikv/tidb. [#3757](https://github.com/pingcap/tiflash/pull/3757) - - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3740](https://github.com/pingcap/tiflash/pull/3740) - - Fix random `EstablishMPPConnection` fail after TiFlash server restart. [#3708](https://github.com/pingcap/tiflash/pull/3708) - - Fix the bug that results of `where ` is wrong because it will be converted to int type. [#3480](https://github.com/pingcap/tiflash/pull/3480) - - -+ PD - - - server: disable swagger server [#5177](https://github.com/tikv/pd/pull/5177) - - Fix the issue that the hot region may cause panic due to no leader [#5040](https://github.com/tikv/pd/pull/5040) - - None. [#4999](https://github.com/tikv/pd/pull/4999) - - Fix the issue that scheduling cannot immediately start after PD leader transfers [#4969](https://github.com/tikv/pd/pull/4969) - - Fix the issue that the removed tombstone store shows again after transferring the PD leader [#4958](https://github.com/tikv/pd/pull/4958) - - Fix the corner case that may cause TSO fallback. [#4891](https://github.com/tikv/pd/pull/4891) - - None. [#4847](https://github.com/tikv/pd/pull/4847) - - Fix the issue that the label distribution has residual labels [#4831](https://github.com/tikv/pd/pull/4831) - - -+ Tools - - + PingCAP/TiCDC - - - Fix TiCDC incorrectly display stale metrics data on dashboard. [#5894](https://github.com/pingcap/tiflow/pull/5894) - - Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. - -If you don't think this PR needs a release note then fill it with `None`. [#5678](https://github.com/pingcap/tiflow/pull/5678) - - Fix a bug that after auto resume, DM will use more disk space. [#5636](https://github.com/pingcap/tiflow/pull/5636) - - Fix a bug in redo log manager that flush log executed before writing logs [#5628](https://github.com/pingcap/tiflow/pull/5628) - - Fix a bug that resolved ts moves too fast when part of tables are not maintained redo writer. [#5617](https://github.com/pingcap/tiflow/pull/5617) - - Add uuid suffix to redo log file name to prevent name conflict, which may cause data loss. [#5613](https://github.com/pingcap/tiflow/pull/5613) - - `Fix a bug that mysql sink may save a wrong checkpointTs`. [#5436](https://github.com/pingcap/tiflow/pull/5436) - - Please add a release note. - -Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. - -If you don't think this PR needs a release note then fill it with `None`. [#5419](https://github.com/pingcap/tiflow/pull/5419) - - Fix DM can't replicate uppercase tables when the task has case-sensitive: false [#5308](https://github.com/pingcap/tiflow/pull/5308) - - Fix an issue where TiCDC clusters may panic after an upgrade [#5304](https://github.com/pingcap/tiflow/pull/5304) - - save table checkpoint after a DDL is filtered [#5292](https://github.com/pingcap/tiflow/pull/5292) - - `None`. [#5226](https://github.com/pingcap/tiflow/pull/5226) - - fix the issue that ticdc failed to start when connects to multiple pd endpoints with tls-enabled and the 1st endpoint is not available [#5209](https://github.com/pingcap/tiflow/pull/5209) - - fix tracker panic when pk of downstream table orders behind [#5178](https://github.com/pingcap/tiflow/pull/5178) - - `None` [#5171](https://github.com/pingcap/tiflow/pull/5171) - - fix tracker panic when pk of downstream table orders behind [#5165](https://github.com/pingcap/tiflow/pull/5165) - - `Fix a bug that openapi may be stuck when pd is abnormal` [#5111](https://github.com/pingcap/tiflow/pull/5111) - - send one heartbeat for successive skipped GTID when enable relay log [#5092](https://github.com/pingcap/tiflow/pull/5092) - - fix bug of relay log may turn off after master reboot in v5.3.1 [#4877](https://github.com/pingcap/tiflow/pull/4877) - - Syncer will use working directory of DM-worker rather than /tmp to write internal files, and clean it after task is stopped [#4804](https://github.com/pingcap/tiflow/pull/4804) - - Fix stability problem in workerpool, which is used by Unified Sorter. [#4682](https://github.com/pingcap/tiflow/pull/4682) - - Fix a bug that sequence should not be replicated even if force-replication is true. -Fix data loss when upstream transaction conflicts during cdc reconnection. [#4666](https://github.com/pingcap/tiflow/pull/4666) - - `None`. [#4650](https://github.com/pingcap/tiflow/pull/4650) - - Please add a release note. -`None`. [#4647](https://github.com/pingcap/tiflow/pull/4647) - - `None`. [#4345](https://github.com/pingcap/tiflow/pull/4345) - - -## Bug Fixes - -+ PingCAP/TiDB - - - fix the problem that dumpling can't dump with --compress and s3 output directory. [#35356](https://github.com/pingcap/tidb/pull/35356) - - Fix the data inconsistency caused by invalid usage of lazy existence check and untouch key optimization. [#30912](https://github.com/pingcap/tidb/pull/30912) - - Fix the bug that sql got cancel if including json column joins char column. [#30777](https://github.com/pingcap/tidb/pull/30777) - - -+ PingCAP/TiFlash - - - Fix potential wrong result after a lot of insert and delete operations [#4967](https://github.com/pingcap/tiflash/pull/4967) - - Fix the metadata corruption issue when `Prepare Merge` is triggered after a new election is finished but the isolated peer is not informed [#4495](https://github.com/pingcap/tiflash/pull/4495) - - Fix the issue that a query containing `JOIN` could be hung if an error was encountered [#4269](https://github.com/pingcap/tiflash/pull/4269) - - -## Compatibility Changes - -+ PingCAP/TiDB - - - Fix an issue that REPLACE statement changing other rows when the auto ID is out of range [#32324](https://github.com/pingcap/tidb/pull/32324) - - -## Improvements - -+ TiKV/TiKV - - - Delay in raft client to reduce syscalls and improve CPU efficiency [#11761](https://github.com/tikv/tikv/pull/11761) - - +--- +title: TiDB 5.3.2 Release Notes +--- + +# TiDB 5.3.2 Release Notes + +Release Date: June xx, 2022 + +TiDB version: 5.3.2 + +## __unsorted + ++ PingCAP/TiDB + + - release-note [#34722](https://github.com/pingcap/tidb/issues/34722) + - release-note [#33892](https://github.com/pingcap/tidb/issues/33892) + - release-note [#34906](https://github.com/pingcap/tidb/issues/34906) + - release-note [#35340](https://github.com/pingcap/tidb/issues/35340) + - release-note [#29283](https://github.com/pingcap/tidb/issues/29283) + - Fixed an issue where extra datums may break binlog. [#33608](https://github.com/pingcap/tidb/issues/33608) + - release-note [#34447](https://github.com/pingcap/tidb/issues/34447) + - release-note [#34320](https://github.com/pingcap/tidb/issues/34320) + - release-note [#34417](https://github.com/pingcap/tidb/issues/34417) + - release-note [#33509](https://github.com/pingcap/tidb/issues/33509) + - release-note [#34350](https://github.com/pingcap/tidb/issues/34350) + - release-note [#27937](https://github.com/pingcap/tidb/issues/27937) + - lightning: split and scatter regions in batches [#33618](https://github.com/pingcap/tidb/issues/33618) + - release-note [#34213](https://github.com/pingcap/tidb/issues/34213) + - Fix the issue that the table attributes don't support index and won't be updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) + - Fix a bug of duplicate primary key when insert record into table after incremental restoration. [#33596](https://github.com/pingcap/tidb/issues/33596) + - release-note [#33893](https://github.com/pingcap/tidb/issues/33893) + - release-note [#33335](https://github.com/pingcap/tidb/issues/33335) + - release-note [#33908](https://github.com/pingcap/tidb/issues/33908) + - Fix the issue that the schedulers won't be resumed after BR/Lightning exits abnormally. [#33546](https://github.com/pingcap/tidb/issues/33546) + - Fix the issue that privilege-related operations may fail for upgraded clusters. [#33588](https://github.com/pingcap/tidb/issues/33588) + - Fix the issue that NewCollationEnable config not checked during restoration. [#33422](https://github.com/pingcap/tidb/issues/33422) + - Fix a bug that BR incremental restore return error by mistake caused by ddl job with empty query. [#33322](https://github.com/pingcap/tidb/issues/33322) + - Fix the issue that BR not retry enough when region not consistency during restoration. [#33419](https://github.com/pingcap/tidb/issues/33419) + - release-note [#33310](https://github.com/pingcap/tidb/issues/33310) + - Fix a bug that caused BR get stuck when restore meets some unrecoverable error. [#33200](https://github.com/pingcap/tidb/issues/33200) + - executor: fix wrong result of delete multiple tables using left join [#31321](https://github.com/pingcap/tidb/issues/31321) + - planner: Fix the issue that TiDB may dispatch duplicated tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) + - lightning: fix checksum encountered “GC life time is shorter than transaction duration” error [#32733](https://github.com/pingcap/tidb/issues/32733) + - Fix BR failure on backup rawkv. [#32607](https://github.com/pingcap/tidb/issues/32607) + - Fix some connections and goroutines leak caused by not closed HTTP response body [#30571](https://github.com/pingcap/tidb/issues/30571) + + ++ TiKV/TiKV + + - Fix bug which causes frequent pd client reconnection [#12345](https://github.com/tikv/tikv/issues/12345) + - Report bad health status if raftstore stops working. [#12398](https://github.com/tikv/tikv/issues/12398) + - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12739](https://github.com/tikv/tikv/issues/12739) + - Fix tikv crash when conv empty string [#12673](https://github.com/tikv/tikv/issues/12673) + - Fix possible duplicate commit record in async-commit pessimistic transactions. [#12615](https://github.com/tikv/tikv/issues/12615) + - Fix a bug that sometimes generates a message with zero store id when doing follower read [#12478](https://github.com/tikv/tikv/issues/12478) + - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#12457](https://github.com/pingcap/tidb/issues/30087) + - Report bad health status if raftstore stops working. [#12398](https://github.com/tikv/tikv/issues/12398) + - fix race between split check and destroy [#12368](https://github.com/tikv/tikv/issues/12368) + - Fixes that successfully committed optimistic transactions may report false WriteConflict on network errors. [#34066](https://github.com/pingcap/tidb/issues/34066) + - fix tikv panic and peer unexpected destroy due to fake merge target [#12232](https://github.com/tikv/tikv/issues/12232) + - Fix stale message cause panic [#12023](https://github.com/tikv/tikv/issues/12023) + - Solve the problem of raft msg memory metrics overflow, which will cause intermittent packet loss and oom. [#12160](https://github.com/tikv/tikv/issues/12160) + - Fix crash when profiling in Ubuntu 18.04. [#9765](https://github.com/tikv/tikv/issues/9765) + - Fix logic of error string match in `bad-ssts`. [#12329](https://github.com/tikv/tikv/issues/12329) + - Pass leader transferee to cdc observer to reduce TiCDC latency spike. [#12111](https://github.com/tikv/tikv/issues/12111) + - Fix potential linearizability violation in replica reads. [#12109](https://github.com/tikv/tikv/issues/12109) + - fix panic when target peer is replaced with an destroyed uninitialized peer during merge [#12048](https://github.com/tikv/tikv/issues/12048) + - Fixes the bug that TiKV keep running over 2 years may panic. [#11940](https://github.com/tikv/tikv/issues/11940) + - None. [#10540](https://github.com/tikv/tikv/issues/10540) + - None. [#11374](https://github.com/tikv/tikv/issues/11374) + + ++ PingCAP/TiFlash + + - Fix potential query error when select on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) + - Fix bug that TiFlash query will meet keepalive timeout error randomly. [#4192](https://github.com/pingcap/tiflash/issues/4192) + - Avoid leaving data on tiflash node which doesn't corresponding to any region range [#4414](https://github.com/pingcap/tiflash/issues/4414) + - Fix a bug that MPP tasks may leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) + - Fix the problem that empty segments cannot be merged after gc [#4511](https://github.com/pingcap/tiflash/issues/4511) + - Fix wrong result of cast(float as decimal) when overflow happens [#3998](https://github.com/pingcap/tiflash/issues/3998) + - Fix the potential crash issue that occurs when TLS is enabled [#23144](https://github.com/grpc/grpc/issues/23144) + - fix the problem that expired data was not recycled timely due to slow gc speed [#4146](https://github.com/pingcap/tiflash/issues/4146) + - Fix the bug that canceled MPP query may cause tasks hang forever when local tunnel is enabled. [#4229](https://github.com/pingcap/tiflash/issues/4229) + - Fix cast datetime to decimal wrong result bug [#4151](https://github.com/pingcap/tiflash/issues/4151) + - Fix the bug that invalid storage dir configurations lead to unexpected behavior [#4093](https://github.com/pingcap/tics/issues/4093) + - Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) + - Fix potential query error after add column under heavy read workload [#3967](https://github.com/pingcap/tics/issues/3967) + - Fix str_to_date() function incorrectly handles leading zeros when parsing Microseconds [#3557](https://github.com/pingcap/tics/issues/3557) + - Fix the bug that castStringAsReal has different behaivor between tiflash and tikv/tidb. [#3475](https://github.com/pingcap/tics/issues/3475) + - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3713](https://github.com/pingcap/tiflash/issues/3713) + - Fix random `EstablishMPPConnection` fail after TiFlash server restart. [#3615](https://github.com/pingcap/tiflash/issues/3615) + - Fix the bug that results of `where ` is wrong because it will be converted to int type. [#3447](https://github.com/pingcap/tics/issues/3447) + + ++ PD + + - server: disable swagger server [#4932](https://github.com/tikv/pd/issues/4932) + - Fix the issue that the hot region may cause panic due to no leader [#5005](https://github.com/tikv/pd/issues/5005) + - None. [#4946](https://github.com/tikv/pd/issues/4946) + - Fix the issue that scheduling cannot immediately start after PD leader transfers [#4769](https://github.com/tikv/pd/issues/4769) + - Fix the issue that the removed tombstone store shows again after transferring the PD leader [#4941](https://github.com/tikv/pd/issues/4941) + - Fix the corner case that may cause TSO fallback. [#4884](https://github.com/tikv/pd/issues/4884) + - None. [#4805](https://github.com/tikv/pd/issues/4805) + - Fix the issue that the label distribution has residual labels [#4825](https://github.com/tikv/pd/issues/4825) + + ++ Tools + + + PingCAP/TiCDC + + - Fix TiCDC incorrectly display stale metrics data on dashboard. [#4774](https://github.com/pingcap/tiflow/issues/4774) + - Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#4287](https://github.com/pingcap/tiflow/issues/4287) + - Fix a bug that after auto resume, DM will use more disk space. [#3734](https://github.com/pingcap/tiflow/issues/3734) + - Fix a bug in redo log manager that flush log executed before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Fix a bug that resolved ts moves too fast when part of tables are not maintained redo writer. [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Add uuid suffix to redo log file name to prevent name conflict, which may cause data loss. [#5486](https://github.com/pingcap/tiflow/issues/5486) + - `Fix a bug that mysql sink may save a wrong checkpointTs`. [#5107](https://github.com/pingcap/tiflow/issues/5107) + - Please add a release note. Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#5344](https://github.com/pingcap/tiflow/issues/5344) + - Fix DM can't replicate uppercase tables when the task has case-sensitive: false [#5255](https://github.com/pingcap/tiflow/issues/5255) + - Fix an issue where TiCDC clusters may panic after an upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) + - save table checkpoint after a DDL is filtered [#5272](https://github.com/pingcap/tiflow/issues/5272) + - `None`. [#4464](https://github.com/pingcap/tiflow/issues/4464) + - fix the issue that ticdc failed to start when connects to multiple pd endpoints with tls-enabled and the 1st endpoint is not available [#4777](https://github.com/pingcap/tiflow/issues/4777) + - fix tracker panic when pk of downstream table orders behind [#5159](https://github.com/pingcap/tiflow/issues/5159) + - `None` [#2792](https://github.com/pingcap/tiflow/issues/2792) + - fix tracker panic when pk of downstream table orders behind [#5159](https://github.com/pingcap/tiflow/issues/5159) + - `Fix a bug that openapi may be stuck when pd is abnormal` [#4778](https://github.com/pingcap/tiflow/issues/4778) + - send one heartbeat for successive skipped GTID when enable relay log [#5063](https://github.com/pingcap/tiflow/issues/5063) + - fix bug of relay log may turn off after master reboot in v5.3.1 [#4803](https://github.com/pingcap/tiflow/issues/4803) + - Syncer will use working directory of DM-worker rather than /tmp to write internal files, and clean it after task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) + - Fix stability problem in workerpool, which is used by Unified Sorter. [#4447](https://github.com/pingcap/tiflow/issues/4447) + - Fix a bug that sequence should not be replicated even if force-replication is true. Fix data loss when upstream transaction conflicts during cdc reconnection. [#4552](https://github.com/pingcap/tiflow/issues/4552) + - `None`. [#4554](https://github.com/pingcap/tiflow/issues/4554) + - Please add a release note. `None`. [#4565](https://github.com/pingcap/tiflow/issues/4565) + - `None`. [#4135](https://github.com/pingcap/tiflow/issues/4135) + + + +## Compatibility Changes + ++ TiDB + + - Fix an issue that REPLACE statement changing other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) + + +## Improvements + ++ TiDB + + - + ++ TiKV + + - Delay in raft client to reduce syscalls and improve CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) + ++ PD + + - + ++ TiDB Dashboard + + - + ++ TiFlash + + - + ++ Tools + + + Backup & Restore (BR) + + - + + + TiCDC + + - + + + Dumpling + + - + + + TiDB Binlog + + - + + + TiDB Data Migration (DM) + + - + + + TiDB Lightning + + - + + +## Bug Fixes + ++ TiDB + + - fix the problem that dumpling can't dump with --compress and s3 output directory. [#30534](https://github.com/pingcap/tidb/issues/30534) + - Fix the data inconsistency caused by invalid usage of lazy existence check and untouch key optimization. [#30410](https://github.com/pingcap/tidb/issues/30410) + - Fix the bug that sql got cancel if including json column joins char column. [#29401](https://github.com/pingcap/tidb/issues/29401) + ++ TiKV + + - + ++ PD + + - + ++ TiDB Dashboard + + - + ++ TiFlash + + - Fix potential wrong result after a lot of insert and delete operations [#4956](https://github.com/pingcap/tiflash/issues/4956) + - Fix the metadata corruption issue when `Prepare Merge` is triggered after a new election is finished but the isolated peer is not informed [#4437](https://github.com/pingcap/tiflash/issues/4437) + - Fix the issue that a query containing `JOIN` could be hung if an error was encountered [#4195](https://github.com/pingcap/tiflash/issues/4195) + ++ Tools + + + Backup & Restore (BR) + + - + + + TiCDC + + - + + + Dumpling + + - + + + TiDB Binlog + + - + + + TiDB Data Migration (DM) + + - + + + TiDB Lightning + + - From e16e180b082428a126fef0dc74c51aac2a58113a Mon Sep 17 00:00:00 2001 From: Ran Date: Mon, 20 Jun 2022 17:56:36 +0800 Subject: [PATCH 04/38] mark duplicated notes Signed-off-by: Ran --- releases/release-5.3.2.md | 146 +++++++++++++++++++------------------- 1 file changed, 73 insertions(+), 73 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 1cb8233be17c..8d25242be0db 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -18,118 +18,118 @@ TiDB version: 5.3.2 - release-note [#35340](https://github.com/pingcap/tidb/issues/35340) - release-note [#29283](https://github.com/pingcap/tidb/issues/29283) - Fixed an issue where extra datums may break binlog. [#33608](https://github.com/pingcap/tidb/issues/33608) - - release-note [#34447](https://github.com/pingcap/tidb/issues/34447) + (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) - release-note [#34320](https://github.com/pingcap/tidb/issues/34320) - release-note [#34417](https://github.com/pingcap/tidb/issues/34417) - - release-note [#33509](https://github.com/pingcap/tidb/issues/33509) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) - release-note [#34350](https://github.com/pingcap/tidb/issues/34350) - - release-note [#27937](https://github.com/pingcap/tidb/issues/27937) - - lightning: split and scatter regions in batches [#33618](https://github.com/pingcap/tidb/issues/33618) - - release-note [#34213](https://github.com/pingcap/tidb/issues/34213) - - Fix the issue that the table attributes don't support index and won't be updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) - - Fix a bug of duplicate primary key when insert record into table after incremental restoration. [#33596](https://github.com/pingcap/tidb/issues/33596) - - release-note [#33893](https://github.com/pingcap/tidb/issues/33893) + (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Lightning)- Fix the issue of Local-backend import failure caused by out-of-bounds data in the `auto_increment` column [#29737](https://github.com/pingcap/tidb/issues/27937) + (dup: release-6.1.0.md > Improvements> Tools> TiDB Lightning)- Optimize Scatter Region to batch mode to improve the stability of the Scatter Region process [#33618](https://github.com/pingcap/tidb/issues/33618) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the issue that the precheck does not check local disk resources and cluster availability [#34213](https://github.com/pingcap/tidb/issues/34213) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix duplicate primary keys when inserting a record into a table after incremental restoration [#33596](https://github.com/pingcap/tidb/issues/33596) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the TiDB server might run out of memory when the `INFORMATION_SCHEMA.CLUSTER_SLOW_QUERY` table is queried. This issue can be triggered when you check slow queries on the Grafana dashboard [#33893](https://github.com/pingcap/tidb/issues/33893) - release-note [#33335](https://github.com/pingcap/tidb/issues/33335) - - release-note [#33908](https://github.com/pingcap/tidb/issues/33908) - - Fix the issue that the schedulers won't be resumed after BR/Lightning exits abnormally. [#33546](https://github.com/pingcap/tidb/issues/33546) - - Fix the issue that privilege-related operations may fail for upgraded clusters. [#33588](https://github.com/pingcap/tidb/issues/33588) + (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that some DDL statements might be stuck for a period after the PD node of a cluster is replaced [#33908](https://github.com/pingcap/tidb/issues/33908) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that schedulers do not resume after BR or TiDB Lightning exits abnormally [#33546](https://github.com/pingcap/tidb/issues/33546) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that granting the `all` privilege might fail in clusters that are upgraded from v4.0 [#33588](https://github.com/pingcap/tidb/issues/33588) - Fix the issue that NewCollationEnable config not checked during restoration. [#33422](https://github.com/pingcap/tidb/issues/33422) - - Fix a bug that BR incremental restore return error by mistake caused by ddl job with empty query. [#33322](https://github.com/pingcap/tidb/issues/33322) - - Fix the issue that BR not retry enough when region not consistency during restoration. [#33419](https://github.com/pingcap/tidb/issues/33419) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix a bug that BR incremental restore returns errors mistakenly due to DDL jobs with empty query [#33322](https://github.com/pingcap/tidb/issues/33322) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) - release-note [#33310](https://github.com/pingcap/tidb/issues/33310) - - Fix a bug that caused BR get stuck when restore meets some unrecoverable error. [#33200](https://github.com/pingcap/tidb/issues/33200) - - executor: fix wrong result of delete multiple tables using left join [#31321](https://github.com/pingcap/tidb/issues/31321) - - planner: Fix the issue that TiDB may dispatch duplicated tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) - - lightning: fix checksum encountered “GC life time is shorter than transaction duration” error [#32733](https://github.com/pingcap/tidb/issues/32733) - - Fix BR failure on backup rawkv. [#32607](https://github.com/pingcap/tidb/issues/32607) - - Fix some connections and goroutines leak caused by not closed HTTP response body [#30571](https://github.com/pingcap/tidb/issues/30571) + (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Backup & Restore (BR))- Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the checksum error "GC life time is shorter than transaction duration" [#32733](https://github.com/pingcap/tidb/issues/32733) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) + (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) + TiKV/TiKV - Fix bug which causes frequent pd client reconnection [#12345](https://github.com/tikv/tikv/issues/12345) - - Report bad health status if raftstore stops working. [#12398](https://github.com/tikv/tikv/issues/12398) + (dup: release-6.1.0.md > Improvements> TiKV)- Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12739](https://github.com/tikv/tikv/issues/12739) - Fix tikv crash when conv empty string [#12673](https://github.com/tikv/tikv/issues/12673) - Fix possible duplicate commit record in async-commit pessimistic transactions. [#12615](https://github.com/tikv/tikv/issues/12615) - - Fix a bug that sometimes generates a message with zero store id when doing follower read [#12478](https://github.com/tikv/tikv/issues/12478) + (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#12457](https://github.com/pingcap/tidb/issues/30087) - - Report bad health status if raftstore stops working. [#12398](https://github.com/tikv/tikv/issues/12398) - - fix race between split check and destroy [#12368](https://github.com/tikv/tikv/issues/12368) - - Fixes that successfully committed optimistic transactions may report false WriteConflict on network errors. [#34066](https://github.com/pingcap/tidb/issues/34066) - - fix tikv panic and peer unexpected destroy due to fake merge target [#12232](https://github.com/tikv/tikv/issues/12232) - - Fix stale message cause panic [#12023](https://github.com/tikv/tikv/issues/12023) - - Solve the problem of raft msg memory metrics overflow, which will cause intermittent packet loss and oom. [#12160](https://github.com/tikv/tikv/issues/12160) - - Fix crash when profiling in Ubuntu 18.04. [#9765](https://github.com/tikv/tikv/issues/9765) - - Fix logic of error string match in `bad-ssts`. [#12329](https://github.com/tikv/tikv/issues/12329) - - Pass leader transferee to cdc observer to reduce TiCDC latency spike. [#12111](https://github.com/tikv/tikv/issues/12111) - - Fix potential linearizability violation in replica reads. [#12109](https://github.com/tikv/tikv/issues/12109) - - fix panic when target peer is replaced with an destroyed uninitialized peer during merge [#12048](https://github.com/tikv/tikv/issues/12048) - - Fixes the bug that TiKV keep running over 2 years may panic. [#11940](https://github.com/tikv/tikv/issues/11940) - - None. [#10540](https://github.com/tikv/tikv/issues/10540) - - None. [#11374](https://github.com/tikv/tikv/issues/11374) + (dup: release-6.1.0.md > Improvements> TiKV)- Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) + (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that TiKV panics and destroys peers unexpectedly because the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that stale messages cause TiKV to panic [#12023](https://github.com/tikv/tikv/issues/12023) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the potential panic issue that occurs when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) + (dup: release-5.2.4.md > Bug fixes> TiKV)- Fix the issue that tikv-ctl returns an incorrect result due to its wrong string match [#12329](https://github.com/tikv/tikv/issues/12329) + (dup: release-5.2.4.md > Improvements> TiKV)- Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that replica reads might violate the linearizability [#12109](https://github.com/tikv/tikv/issues/12109) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- 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](https://github.com/tikv/tikv/issues/12048) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that TiKV might panic if it has been running for 2 years or more [#11940](https://github.com/tikv/tikv/issues/11940) + (dup: release-5.3.0.md > Improvements> TiKV> Enhance disk space protection to improve storage stability)- Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) + (dup: release-5.0.6.md > Improvements> TiKV)- Add more metrics for the garbage collection module of Raft logs to locate performance problems in the module [#11374](https://github.com/tikv/tikv/issues/11374) + PingCAP/TiFlash - - Fix potential query error when select on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) - - Fix bug that TiFlash query will meet keepalive timeout error randomly. [#4192](https://github.com/pingcap/tiflash/issues/4192) - - Avoid leaving data on tiflash node which doesn't corresponding to any region range [#4414](https://github.com/pingcap/tiflash/issues/4414) - - Fix a bug that MPP tasks may leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) - - Fix the problem that empty segments cannot be merged after gc [#4511](https://github.com/pingcap/tiflash/issues/4511) - - Fix wrong result of cast(float as decimal) when overflow happens [#3998](https://github.com/pingcap/tiflash/issues/3998) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that MPP tasks might leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that empty segments cannot be merged after GC [#4511](https://github.com/pingcap/tiflash/issues/4511) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - Fix the potential crash issue that occurs when TLS is enabled [#23144](https://github.com/grpc/grpc/issues/23144) - - fix the problem that expired data was not recycled timely due to slow gc speed [#4146](https://github.com/pingcap/tiflash/issues/4146) - - Fix the bug that canceled MPP query may cause tasks hang forever when local tunnel is enabled. [#4229](https://github.com/pingcap/tiflash/issues/4229) - - Fix cast datetime to decimal wrong result bug [#4151](https://github.com/pingcap/tiflash/issues/4151) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that expired data is recycled slowly [#4146](https://github.com/pingcap/tiflash/issues/4146) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) - Fix the bug that invalid storage dir configurations lead to unexpected behavior [#4093](https://github.com/pingcap/tics/issues/4093) - - Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) - Fix potential query error after add column under heavy read workload [#3967](https://github.com/pingcap/tics/issues/3967) - Fix str_to_date() function incorrectly handles leading zeros when parsing Microseconds [#3557](https://github.com/pingcap/tics/issues/3557) - Fix the bug that castStringAsReal has different behaivor between tiflash and tikv/tidb. [#3475](https://github.com/pingcap/tics/issues/3475) - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3713](https://github.com/pingcap/tiflash/issues/3713) - - Fix random `EstablishMPPConnection` fail after TiFlash server restart. [#3615](https://github.com/pingcap/tiflash/issues/3615) + (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that TiFlash might return the `EstablishMPPConnection` error after it is restarted [#3615](https://github.com/pingcap/tiflash/issues/3615) - Fix the bug that results of `where ` is wrong because it will be converted to int type. [#3447](https://github.com/pingcap/tics/issues/3447) + PD - - server: disable swagger server [#4932](https://github.com/tikv/pd/issues/4932) + (dup: release-6.1.0.md > Improvements> PD)- Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) - Fix the issue that the hot region may cause panic due to no leader [#5005](https://github.com/tikv/pd/issues/5005) - None. [#4946](https://github.com/tikv/pd/issues/4946) - - Fix the issue that scheduling cannot immediately start after PD leader transfers [#4769](https://github.com/tikv/pd/issues/4769) - - Fix the issue that the removed tombstone store shows again after transferring the PD leader [#4941](https://github.com/tikv/pd/issues/4941) - - Fix the corner case that may cause TSO fallback. [#4884](https://github.com/tikv/pd/issues/4884) - - None. [#4805](https://github.com/tikv/pd/issues/4805) - - Fix the issue that the label distribution has residual labels [#4825](https://github.com/tikv/pd/issues/4825) + (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that scheduling cannot start immediately after the PD leader transfer [4769](https://github.com/tikv/pd/issues/4769) + (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) + (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) + (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) + (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) + Tools + PingCAP/TiCDC - - Fix TiCDC incorrectly display stale metrics data on dashboard. [#4774](https://github.com/pingcap/tiflow/issues/4774) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix incorrect metrics caused by owner changes [#4774](https://github.com/pingcap/tiflow/issues/4774) - Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#4287](https://github.com/pingcap/tiflow/issues/4287) - - Fix a bug that after auto resume, DM will use more disk space. [#3734](https://github.com/pingcap/tiflow/issues/3734) + (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix the issue that DM occupies more disk space after the task automatically resumes [#3734](https://github.com/pingcap/tiflow/issues/3734) [#5344](https://github.com/pingcap/tiflow/issues/5344) - Fix a bug in redo log manager that flush log executed before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) - Fix a bug that resolved ts moves too fast when part of tables are not maintained redo writer. [#5486](https://github.com/pingcap/tiflow/issues/5486) - Add uuid suffix to redo log file name to prevent name conflict, which may cause data loss. [#5486](https://github.com/pingcap/tiflow/issues/5486) - `Fix a bug that mysql sink may save a wrong checkpointTs`. [#5107](https://github.com/pingcap/tiflow/issues/5107) - Please add a release note. Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#5344](https://github.com/pingcap/tiflow/issues/5344) - - Fix DM can't replicate uppercase tables when the task has case-sensitive: false [#5255](https://github.com/pingcap/tiflow/issues/5255) + (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix an issue that the uppercase table cannot be replicated when `case-sensitive: true` is not set [#5255](https://github.com/pingcap/tiflow/issues/5255) - Fix an issue where TiCDC clusters may panic after an upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) - - save table checkpoint after a DDL is filtered [#5272](https://github.com/pingcap/tiflow/issues/5272) - - `None`. [#4464](https://github.com/pingcap/tiflow/issues/4464) - - fix the issue that ticdc failed to start when connects to multiple pd endpoints with tls-enabled and the 1st endpoint is not available [#4777](https://github.com/pingcap/tiflow/issues/4777) - - fix tracker panic when pk of downstream table orders behind [#5159](https://github.com/pingcap/tiflow/issues/5159) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that in some cases manually executing the filtered DDL in the downstream might cause task resumption failure [#5272](https://github.com/pingcap/tiflow/issues/5272) + (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix the issue that changefeed gets stuck when tables are repeatedly scheduled in the same node [#4464](https://github.com/pingcap/tiflow/issues/4464) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix the issue that TiCDC fails to start when the first PD set in `--pd` is not available after TLS is enabled [#4777](https://github.com/pingcap/tiflow/issues/4777) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) - `None` [#2792](https://github.com/pingcap/tiflow/issues/2792) - - fix tracker panic when pk of downstream table orders behind [#5159](https://github.com/pingcap/tiflow/issues/5159) - - `Fix a bug that openapi may be stuck when pd is abnormal` [#4778](https://github.com/pingcap/tiflow/issues/4778) - - send one heartbeat for successive skipped GTID when enable relay log [#5063](https://github.com/pingcap/tiflow/issues/5063) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) + (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix a bug that querying status through open API may be blocked when the PD node is abnormal [#4778](https://github.com/pingcap/tiflow/issues/4778) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that CPU usage may increase and a large amount of log is printed when GTID is enabled or when the task is automatically resumed [#5063](https://github.com/pingcap/tiflow/issues/5063) - fix bug of relay log may turn off after master reboot in v5.3.1 [#4803](https://github.com/pingcap/tiflow/issues/4803) - - Syncer will use working directory of DM-worker rather than /tmp to write internal files, and clean it after task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) - - Fix stability problem in workerpool, which is used by Unified Sorter. [#4447](https://github.com/pingcap/tiflow/issues/4447) - - Fix a bug that sequence should not be replicated even if force-replication is true. Fix data loss when upstream transaction conflicts during cdc reconnection. [#4552](https://github.com/pingcap/tiflow/issues/4552) + (dup: release-5.4.1.md > Improvements> Tools> TiDB Data Migration (DM))- Support Syncer using the working directory of the DM-worker rather than `/tmp` to write internal files, and cleaning the directory after the task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a stability problem in workerpool used by Unified Sorter [#4447](https://github.com/pingcap/tiflow/issues/4447) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a bug that sequence is incorrectly replicated in some cases [#4563](https://github.com/pingcap/tiflow/issues/4552) - `None`. [#4554](https://github.com/pingcap/tiflow/issues/4554) - Please add a release note. `None`. [#4565](https://github.com/pingcap/tiflow/issues/4565) - `None`. [#4135](https://github.com/pingcap/tiflow/issues/4135) @@ -140,7 +140,7 @@ TiDB version: 5.3.2 + TiDB - - Fix an issue that REPLACE statement changing other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) + (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the `REPLACE` statement incorrectly changes other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) ## Improvements @@ -151,7 +151,7 @@ TiDB version: 5.3.2 + TiKV - - Delay in raft client to reduce syscalls and improve CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) + (dup: release-5.4.0.md > Improvements> TiKV)- Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) + PD @@ -196,9 +196,9 @@ TiDB version: 5.3.2 + TiDB - - fix the problem that dumpling can't dump with --compress and s3 output directory. [#30534](https://github.com/pingcap/tidb/issues/30534) - - Fix the data inconsistency caused by invalid usage of lazy existence check and untouch key optimization. [#30410](https://github.com/pingcap/tidb/issues/30410) - - Fix the bug that sql got cancel if including json column joins char column. [#29401](https://github.com/pingcap/tidb/issues/29401) + (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Dumpling)- Fix the issue that Amazon S3 cannot correctly calculate the size of compressed data [#30534](https://github.com/pingcap/tidb/issues/30534) + (dup: release-5.4.0.md > Bug fixes> TiDB)- Fix the issue of potential data index inconsistency in optimistic transaction mode [#30410](https://github.com/pingcap/tidb/issues/30410) + (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that a SQL operation is canceled when its JSON type column joins its `CHAR` type column [#29401](https://github.com/pingcap/tidb/issues/29401) + TiKV @@ -214,9 +214,9 @@ TiDB version: 5.3.2 + TiFlash - - Fix potential wrong result after a lot of insert and delete operations [#4956](https://github.com/pingcap/tiflash/issues/4956) - - Fix the metadata corruption issue when `Prepare Merge` is triggered after a new election is finished but the isolated peer is not informed [#4437](https://github.com/pingcap/tiflash/issues/4437) - - Fix the issue that a query containing `JOIN` could be hung if an error was encountered [#4195](https://github.com/pingcap/tiflash/issues/4195) + (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix possible metadata corruption caused by Region merge on a lagging Region peer [#4437](https://github.com/pingcap/tiflash/issues/4437) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that a query containing `JOIN` might be hung if an error occurs [#4195](https://github.com/pingcap/tiflash/issues/4195) + Tools From 81aece75072cd427f824e34dd14e6961e118f196 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 10:20:21 +0800 Subject: [PATCH 05/38] update tidb notes Signed-off-by: Ran --- releases/release-5.3.2.md | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 8d25242be0db..05ba9ff0da86 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -12,38 +12,26 @@ TiDB version: 5.3.2 + PingCAP/TiDB - - release-note [#34722](https://github.com/pingcap/tidb/issues/34722) - release-note [#33892](https://github.com/pingcap/tidb/issues/33892) - release-note [#34906](https://github.com/pingcap/tidb/issues/34906) - release-note [#35340](https://github.com/pingcap/tidb/issues/35340) - release-note [#29283](https://github.com/pingcap/tidb/issues/29283) - - Fixed an issue where extra datums may break binlog. [#33608](https://github.com/pingcap/tidb/issues/33608) - (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) - release-note [#34320](https://github.com/pingcap/tidb/issues/34320) - release-note [#34417](https://github.com/pingcap/tidb/issues/34417) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) - release-note [#34350](https://github.com/pingcap/tidb/issues/34350) - (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Lightning)- Fix the issue of Local-backend import failure caused by out-of-bounds data in the `auto_increment` column [#29737](https://github.com/pingcap/tidb/issues/27937) + (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Lightning)- Fix the issue of Local-backend import failure caused by out-of-bounds data in the `auto_increment` column [#27937](https://github.com/pingcap/tidb/issues/27937) (dup: release-6.1.0.md > Improvements> Tools> TiDB Lightning)- Optimize Scatter Region to batch mode to improve the stability of the Scatter Region process [#33618](https://github.com/pingcap/tidb/issues/33618) (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the issue that the precheck does not check local disk resources and cluster availability [#34213](https://github.com/pingcap/tidb/issues/34213) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix duplicate primary keys when inserting a record into a table after incremental restoration [#33596](https://github.com/pingcap/tidb/issues/33596) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the TiDB server might run out of memory when the `INFORMATION_SCHEMA.CLUSTER_SLOW_QUERY` table is queried. This issue can be triggered when you check slow queries on the Grafana dashboard [#33893](https://github.com/pingcap/tidb/issues/33893) - release-note [#33335](https://github.com/pingcap/tidb/issues/33335) - (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that some DDL statements might be stuck for a period after the PD node of a cluster is replaced [#33908](https://github.com/pingcap/tidb/issues/33908) (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that schedulers do not resume after BR or TiDB Lightning exits abnormally [#33546](https://github.com/pingcap/tidb/issues/33546) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that granting the `all` privilege might fail in clusters that are upgraded from v4.0 [#33588](https://github.com/pingcap/tidb/issues/33588) - Fix the issue that NewCollationEnable config not checked during restoration. [#33422](https://github.com/pingcap/tidb/issues/33422) (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix a bug that BR incremental restore returns errors mistakenly due to DDL jobs with empty query [#33322](https://github.com/pingcap/tidb/issues/33322) (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) - release-note [#33310](https://github.com/pingcap/tidb/issues/33310) (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Backup & Restore (BR))- Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the checksum error "GC life time is shorter than transaction duration" [#32733](https://github.com/pingcap/tidb/issues/32733) (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) - (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) - + TiKV/TiKV @@ -142,7 +130,6 @@ TiDB version: 5.3.2 (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the `REPLACE` statement incorrectly changes other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) - ## Improvements + TiDB @@ -199,6 +186,17 @@ TiDB version: 5.3.2 (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Dumpling)- Fix the issue that Amazon S3 cannot correctly calculate the size of compressed data [#30534](https://github.com/pingcap/tidb/issues/30534) (dup: release-5.4.0.md > Bug fixes> TiDB)- Fix the issue of potential data index inconsistency in optimistic transaction mode [#30410](https://github.com/pingcap/tidb/issues/30410) (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that a SQL operation is canceled when its JSON type column joins its `CHAR` type column [#29401](https://github.com/pingcap/tidb/issues/29401) + - In the event that a network connectivity event occurred, TiDB did not always free the resources of the disconnected session correctly. This has been fixed to ensure that open transactions are rolled back, as well as other associated resources. [#34722](https://github.com/pingcap/tidb/issues/34722) + - Fixed an issue where extra datums may break binlog. [#33608](https://github.com/pingcap/tidb/issues/33608) + - Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) + - Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) + - Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) + - Fix the issue that the TiDB server might run out of memory when the `INFORMATION_SCHEMA.CLUSTER_SLOW_QUERY` table is queried. This issue can be triggered when you check slow queries on the Grafana dashboard [#33893](https://github.com/pingcap/tidb/issues/33893) + - Fix the issue that some DDL statements might be stuck for a period after the PD node of a cluster is replaced [#33908](https://github.com/pingcap/tidb/issues/33908) + - Fix the issue that granting the `all` privilege might fail in clusters that are upgraded from v4.0 [#33588](https://github.com/pingcap/tidb/issues/33588) + - Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) + - Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) + - Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) + TiKV From 4bace2a484b32e4c1cdd16c995ce0708f96bd96b Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 11:50:42 +0800 Subject: [PATCH 06/38] add tikv notes Signed-off-by: Ran --- releases/release-5.3.2.md | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 05ba9ff0da86..503992b24199 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -139,6 +139,10 @@ TiDB version: 5.3.2 + TiKV (dup: release-5.4.0.md > Improvements> TiKV)- Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) + - Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) + - Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) + - Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) + - Add more metrics for the garbage collection module of Raft logs [#11374](https://github.com/tikv/tikv/issues/11374) + PD @@ -200,7 +204,21 @@ TiDB version: 5.3.2 + TiKV - - + - Fix the bug that causes frequent pd client reconnection [#12345](https://github.com/tikv/tikv/issues/12345) + - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12739](https://github.com/tikv/tikv/issues/12739) + - Fix the bug that converting empty string causes TiKV crash [#12673](https://github.com/tikv/tikv/issues/12673) + - Fix possible duplicate commit records in async-commit pessimistic transactions [#12615](https://github.com/tikv/tikv/issues/12615) + - Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) + - Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) + - Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) + - Fix the issue that TiKV panics and destroys peers unexpectedly when the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) + - Fix the bug that stale messages cause TiKV panic [#12023](https://github.com/tikv/tikv/issues/12023) + - Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) + - Fix a potential panic issue when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) + - Fix the issue that tikv-ctl returns an incorrect result due to incorrect string match pattern [#12329](https://github.com/tikv/tikv/issues/12329) + - Fix a bug that replica reads might violate the linearizability [#12109](https://github.com/tikv/tikv/issues/12109) + - Fix the TiKV panic issue that the target peer is replaced with the peer that is destroyed without being initialized when merging a Region [#12048](https://github.com/tikv/tikv/issues/12048) + - Fix a bug that TiKV might panic if it has been running for 2 years or more [#11940](https://github.com/tikv/tikv/issues/11940) + PD From b2df1a563cd7a50369476e99586ba4b395fd3002 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 11:59:31 +0800 Subject: [PATCH 07/38] remove duplicated tikv notes in unsorted Signed-off-by: Ran --- releases/release-5.3.2.md | 33 ++++++--------------------------- 1 file changed, 6 insertions(+), 27 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 503992b24199..c522faf2d1cd 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -35,28 +35,7 @@ TiDB version: 5.3.2 + TiKV/TiKV - - Fix bug which causes frequent pd client reconnection [#12345](https://github.com/tikv/tikv/issues/12345) - (dup: release-6.1.0.md > Improvements> TiKV)- Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) - - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12739](https://github.com/tikv/tikv/issues/12739) - - Fix tikv crash when conv empty string [#12673](https://github.com/tikv/tikv/issues/12673) - - Fix possible duplicate commit record in async-commit pessimistic transactions. [#12615](https://github.com/tikv/tikv/issues/12615) - (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) - - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#12457](https://github.com/pingcap/tidb/issues/30087) - (dup: release-6.1.0.md > Improvements> TiKV)- Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) - (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that TiKV panics and destroys peers unexpectedly because the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that stale messages cause TiKV to panic [#12023](https://github.com/tikv/tikv/issues/12023) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the potential panic issue that occurs when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) - (dup: release-5.2.4.md > Bug fixes> TiKV)- Fix the issue that tikv-ctl returns an incorrect result due to its wrong string match [#12329](https://github.com/tikv/tikv/issues/12329) - (dup: release-5.2.4.md > Improvements> TiKV)- Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that replica reads might violate the linearizability [#12109](https://github.com/tikv/tikv/issues/12109) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- 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](https://github.com/tikv/tikv/issues/12048) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that TiKV might panic if it has been running for 2 years or more [#11940](https://github.com/tikv/tikv/issues/11940) - (dup: release-5.3.0.md > Improvements> TiKV> Enhance disk space protection to improve storage stability)- Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) - (dup: release-5.0.6.md > Improvements> TiKV)- Add more metrics for the garbage collection module of Raft logs to locate performance problems in the module [#11374](https://github.com/tikv/tikv/issues/11374) - + - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#30087](https://github.com/pingcap/tidb/issues/30087) + PingCAP/TiFlash @@ -139,7 +118,7 @@ TiDB version: 5.3.2 + TiKV (dup: release-5.4.0.md > Improvements> TiKV)- Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) - - Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) + (dup: release-6.1.0.md > Improvements> TiKV) - Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) - Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) - Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) - Add more metrics for the garbage collection module of Raft logs [#11374](https://github.com/tikv/tikv/issues/11374) @@ -208,10 +187,10 @@ TiDB version: 5.3.2 - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12739](https://github.com/tikv/tikv/issues/12739) - Fix the bug that converting empty string causes TiKV crash [#12673](https://github.com/tikv/tikv/issues/12673) - Fix possible duplicate commit records in async-commit pessimistic transactions [#12615](https://github.com/tikv/tikv/issues/12615) - - Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) - - Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) - - Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) - - Fix the issue that TiKV panics and destroys peers unexpectedly when the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) + (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) + (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that TiKV panics and destroys peers unexpectedly when the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) - Fix the bug that stale messages cause TiKV panic [#12023](https://github.com/tikv/tikv/issues/12023) - Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) - Fix a potential panic issue when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) From cb74358b3c7ec3ff0cddc873e66e8917de9de526 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 12:07:01 +0800 Subject: [PATCH 08/38] add tiflash notes --- releases/release-5.3.2.md | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index c522faf2d1cd..ce640a3ed0c1 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -44,20 +44,13 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that MPP tasks might leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that empty segments cannot be merged after GC [#4511](https://github.com/pingcap/tiflash/issues/4511) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - - Fix the potential crash issue that occurs when TLS is enabled [#23144](https://github.com/grpc/grpc/issues/23144) + - Fix the potential crash issue that occurs when TLS is enabled [#4196](https://github.com/pingcap/tiflash/issues/4196) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that expired data is recycled slowly [#4146](https://github.com/pingcap/tiflash/issues/4146) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) - - Fix the bug that invalid storage dir configurations lead to unexpected behavior [#4093](https://github.com/pingcap/tics/issues/4093) + - Fix the bug that invalid storage dir configurations lead to unexpected behavior [#4093](https://github.com/pingcap/tiflash/issues/4093) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) - - Fix potential query error after add column under heavy read workload [#3967](https://github.com/pingcap/tics/issues/3967) - - Fix str_to_date() function incorrectly handles leading zeros when parsing Microseconds [#3557](https://github.com/pingcap/tics/issues/3557) - - Fix the bug that castStringAsReal has different behaivor between tiflash and tikv/tidb. [#3475](https://github.com/pingcap/tics/issues/3475) - - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3713](https://github.com/pingcap/tiflash/issues/3713) + - Fix potential query error after add column under heavy read workload [#3967](https://github.com/pingcap/tiflash/issues/3967) + - Fix str_to_date() function incorrectly handles leading zeros when parsing Microseconds [#3557](https://github.com/pingcap/tiflash/issues/3557) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that TiFlash might return the `EstablishMPPConnection` error after it is restarted [#3615](https://github.com/pingcap/tiflash/issues/3615) - - Fix the bug that results of `where ` is wrong because it will be converted to int type. [#3447](https://github.com/pingcap/tics/issues/3447) - + PD @@ -211,7 +204,18 @@ TiDB version: 5.3.2 (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix possible metadata corruption caused by Region merge on a lagging Region peer [#4437](https://github.com/pingcap/tiflash/issues/4437) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that a query containing `JOIN` might be hung if an error occurs [#4195](https://github.com/pingcap/tiflash/issues/4195) + - Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [4229](https://github.com/pingcap/tiflash/issues/4229) + - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [3713](https://github.com/pingcap/tiflash/issues/3713) + - Fix bug that MPP query may failed due to grpc keepalive timeout [4662](https://github.com/pingcap/tiflash/issues/4662) + - Fix bug that MPP query may hang forever if there are many retry in exchange receiver [3473](https://github.com/pingcap/tiflash/pull/3473) + - Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [4151](https://github.com/pingcap/tiflash/issues/4151) + - Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [3998](https://github.com/pingcap/tiflash/issues/3998) + - Fix potential `index out of bounds` error if calling `json_length` with empty string [2705](https://github.com/pingcap/tiflash/issues/2705) + - Fix decimal compare will return wrong result in some corner cae [4942](https://github.com/pingcap/tiflash/pull/4942) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix bug that MPP query may hang forever if query failed in join build stage [4195](https://github.com/pingcap/tiflash/issues/4195) + - Fix the bug that results of `where ` is wrong because it will be converted to int type. [3447](https://github.com/pingcap/tiflash/issues/3447) + - Fix the bug that castStringAsReal has different behaivor between tiflash and tikv/tidb. [3475](https://github.com/pingcap/tiflash/issues/3475) + - Fix the bug that when cast string to datetime, the `microsecond` field may get wrong result [3556](https://github.com/pingcap/tiflash/issues/3556) + Tools From a4dcc3c68cc141907d23ec2f78139ec87c0b5f51 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 12:22:40 +0800 Subject: [PATCH 09/38] add dm/ticdc notes Signed-off-by: Ran --- releases/release-5.3.2.md | 96 +++++++++++++++++++-------------------- 1 file changed, 46 insertions(+), 50 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index ce640a3ed0c1..197ff4949233 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -44,12 +44,12 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that MPP tasks might leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that empty segments cannot be merged after GC [#4511](https://github.com/pingcap/tiflash/issues/4511) - - Fix the potential crash issue that occurs when TLS is enabled [#4196](https://github.com/pingcap/tiflash/issues/4196) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the panic issue that occurs when TLS is enabled [#4196](https://github.com/pingcap/tiflash/issues/4196) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that expired data is recycled slowly [#4146](https://github.com/pingcap/tiflash/issues/4146) - - Fix the bug that invalid storage dir configurations lead to unexpected behavior [#4093](https://github.com/pingcap/tiflash/issues/4093) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) - - Fix potential query error after add column under heavy read workload [#3967](https://github.com/pingcap/tiflash/issues/3967) - - Fix str_to_date() function incorrectly handles leading zeros when parsing Microseconds [#3557](https://github.com/pingcap/tiflash/issues/3557) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the potential query error after adding columns under heavy read workload [#3967](https://github.com/pingcap/tiflash/issues/3967) + (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the bug that the `STR_TO_DATE()` function incorrectly handles leading zeros when parsing microseconds [#3557](https://github.com/pingcap/tiflash/issues/3557) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that TiFlash might return the `EstablishMPPConnection` error after it is restarted [#3615](https://github.com/pingcap/tiflash/issues/3615) + PD @@ -68,28 +68,9 @@ TiDB version: 5.3.2 + PingCAP/TiCDC - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix incorrect metrics caused by owner changes [#4774](https://github.com/pingcap/tiflow/issues/4774) - Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#4287](https://github.com/pingcap/tiflow/issues/4287) - (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix the issue that DM occupies more disk space after the task automatically resumes [#3734](https://github.com/pingcap/tiflow/issues/3734) [#5344](https://github.com/pingcap/tiflow/issues/5344) - - Fix a bug in redo log manager that flush log executed before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) - - Fix a bug that resolved ts moves too fast when part of tables are not maintained redo writer. [#5486](https://github.com/pingcap/tiflow/issues/5486) - - Add uuid suffix to redo log file name to prevent name conflict, which may cause data loss. [#5486](https://github.com/pingcap/tiflow/issues/5486) - - `Fix a bug that mysql sink may save a wrong checkpointTs`. [#5107](https://github.com/pingcap/tiflow/issues/5107) - Please add a release note. Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#5344](https://github.com/pingcap/tiflow/issues/5344) - (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix an issue that the uppercase table cannot be replicated when `case-sensitive: true` is not set [#5255](https://github.com/pingcap/tiflow/issues/5255) - - Fix an issue where TiCDC clusters may panic after an upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that in some cases manually executing the filtered DDL in the downstream might cause task resumption failure [#5272](https://github.com/pingcap/tiflow/issues/5272) - (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix the issue that changefeed gets stuck when tables are repeatedly scheduled in the same node [#4464](https://github.com/pingcap/tiflow/issues/4464) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix the issue that TiCDC fails to start when the first PD set in `--pd` is not available after TLS is enabled [#4777](https://github.com/pingcap/tiflow/issues/4777) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) - `None` [#2792](https://github.com/pingcap/tiflow/issues/2792) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) - (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix a bug that querying status through open API may be blocked when the PD node is abnormal [#4778](https://github.com/pingcap/tiflow/issues/4778) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that CPU usage may increase and a large amount of log is printed when GTID is enabled or when the task is automatically resumed [#5063](https://github.com/pingcap/tiflow/issues/5063) - - fix bug of relay log may turn off after master reboot in v5.3.1 [#4803](https://github.com/pingcap/tiflow/issues/4803) - (dup: release-5.4.1.md > Improvements> Tools> TiDB Data Migration (DM))- Support Syncer using the working directory of the DM-worker rather than `/tmp` to write internal files, and cleaning the directory after the task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a stability problem in workerpool used by Unified Sorter [#4447](https://github.com/pingcap/tiflow/issues/4447) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a bug that sequence is incorrectly replicated in some cases [#4563](https://github.com/pingcap/tiflow/issues/4552) - `None`. [#4554](https://github.com/pingcap/tiflow/issues/4554) - Please add a release note. `None`. [#4565](https://github.com/pingcap/tiflow/issues/4565) - `None`. [#4135](https://github.com/pingcap/tiflow/issues/4135) @@ -112,9 +93,9 @@ TiDB version: 5.3.2 (dup: release-5.4.0.md > Improvements> TiKV)- Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) (dup: release-6.1.0.md > Improvements> TiKV) - Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) - - Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) - - Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) - - Add more metrics for the garbage collection module of Raft logs [#11374](https://github.com/tikv/tikv/issues/11374) + (dup: release-5.2.4.md > Improvements> TiKV)- Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) + (dup: release-5.3.0.md > Improvements> TiKV> Enhance disk space protection to improve storage stability)- Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) + (dup: release-5.0.6.md > Improvements> TiKV)- Add more metrics for the garbage collection module of Raft logs to locate performance problems in the module [#11374](https://github.com/tikv/tikv/issues/11374) + PD @@ -148,7 +129,7 @@ TiDB version: 5.3.2 + TiDB Data Migration (DM) - - + (dup: release-5.4.1.md > Improvements> Tools> TiDB Data Migration (DM))- Support Syncer using the working directory of the DM-worker rather than `/tmp` to write internal files, and cleaning the directory after the task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) + TiDB Lightning @@ -164,15 +145,15 @@ TiDB version: 5.3.2 (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that a SQL operation is canceled when its JSON type column joins its `CHAR` type column [#29401](https://github.com/pingcap/tidb/issues/29401) - In the event that a network connectivity event occurred, TiDB did not always free the resources of the disconnected session correctly. This has been fixed to ensure that open transactions are rolled back, as well as other associated resources. [#34722](https://github.com/pingcap/tidb/issues/34722) - Fixed an issue where extra datums may break binlog. [#33608](https://github.com/pingcap/tidb/issues/33608) - - Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) - - Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) - - Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) - - Fix the issue that the TiDB server might run out of memory when the `INFORMATION_SCHEMA.CLUSTER_SLOW_QUERY` table is queried. This issue can be triggered when you check slow queries on the Grafana dashboard [#33893](https://github.com/pingcap/tidb/issues/33893) - - Fix the issue that some DDL statements might be stuck for a period after the PD node of a cluster is replaced [#33908](https://github.com/pingcap/tidb/issues/33908) - - Fix the issue that granting the `all` privilege might fail in clusters that are upgraded from v4.0 [#33588](https://github.com/pingcap/tidb/issues/33588) - - Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) - - Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) - - Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) + (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the TiDB server might run out of memory when the `INFORMATION_SCHEMA.CLUSTER_SLOW_QUERY` table is queried. This issue can be triggered when you check slow queries on the Grafana dashboard [#33893](https://github.com/pingcap/tidb/issues/33893) + (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that some DDL statements might be stuck for a period after the PD node of a cluster is replaced [#33908](https://github.com/pingcap/tidb/issues/33908) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that granting the `all` privilege might fail in clusters that are upgraded from v4.0 [#33588](https://github.com/pingcap/tidb/issues/33588) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) + (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) + (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) + TiKV @@ -184,13 +165,13 @@ TiDB version: 5.3.2 (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that TiKV panics and destroys peers unexpectedly when the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) - - Fix the bug that stale messages cause TiKV panic [#12023](https://github.com/tikv/tikv/issues/12023) - - Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) - - Fix a potential panic issue when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) - - Fix the issue that tikv-ctl returns an incorrect result due to incorrect string match pattern [#12329](https://github.com/tikv/tikv/issues/12329) - - Fix a bug that replica reads might violate the linearizability [#12109](https://github.com/tikv/tikv/issues/12109) - - Fix the TiKV panic issue that the target peer is replaced with the peer that is destroyed without being initialized when merging a Region [#12048](https://github.com/tikv/tikv/issues/12048) - - Fix a bug that TiKV might panic if it has been running for 2 years or more [#11940](https://github.com/tikv/tikv/issues/11940) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that stale messages cause TiKV to panic [#12023](https://github.com/tikv/tikv/issues/12023) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the potential panic issue that occurs when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) + (dup: release-5.2.4.md > Bug fixes> TiKV)- Fix the issue that tikv-ctl returns an incorrect result due to its wrong string match [#12329](https://github.com/tikv/tikv/issues/12329) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that replica reads might violate the linearizability [#12109](https://github.com/tikv/tikv/issues/12109) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- 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](https://github.com/tikv/tikv/issues/12048) + (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that TiKV might panic if it has been running for 2 years or more [#11940](https://github.com/tikv/tikv/issues/11940) + PD @@ -204,17 +185,17 @@ TiDB version: 5.3.2 (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix possible metadata corruption caused by Region merge on a lagging Region peer [#4437](https://github.com/pingcap/tiflash/issues/4437) - - Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [4229](https://github.com/pingcap/tiflash/issues/4229) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [3713](https://github.com/pingcap/tiflash/issues/3713) - Fix bug that MPP query may failed due to grpc keepalive timeout [4662](https://github.com/pingcap/tiflash/issues/4662) - Fix bug that MPP query may hang forever if there are many retry in exchange receiver [3473](https://github.com/pingcap/tiflash/pull/3473) - - Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [4151](https://github.com/pingcap/tiflash/issues/4151) - - Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [3998](https://github.com/pingcap/tiflash/issues/3998) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - Fix potential `index out of bounds` error if calling `json_length` with empty string [2705](https://github.com/pingcap/tiflash/issues/2705) - Fix decimal compare will return wrong result in some corner cae [4942](https://github.com/pingcap/tiflash/pull/4942) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix bug that MPP query may hang forever if query failed in join build stage [4195](https://github.com/pingcap/tiflash/issues/4195) - - Fix the bug that results of `where ` is wrong because it will be converted to int type. [3447](https://github.com/pingcap/tiflash/issues/3447) - - Fix the bug that castStringAsReal has different behaivor between tiflash and tikv/tidb. [3475](https://github.com/pingcap/tiflash/issues/3475) + (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix possible wrong results when a query contains the `where ` clause [#3447](https://github.com/pingcap/tiflash/issues/3447) + (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that the `CastStringAsReal` behavior is inconsistent in TiFlash and in TiDB or TiKV [#3475](https://github.com/pingcap/tiflash/issues/3475) - Fix the bug that when cast string to datetime, the `microsecond` field may get wrong result [3556](https://github.com/pingcap/tiflash/issues/3556) + Tools @@ -225,7 +206,17 @@ TiDB version: 5.3.2 + TiCDC - - + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix incorrect metrics caused by owner changes [#4774](https://github.com/pingcap/tiflow/issues/4774) + - Fix a bug in redo log manager that flush log executed before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Fix a bug that resolved ts moves too fast when part of tables are not maintained redo writer. [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Add uuid suffix to redo log file name to prevent name conflict, which may cause data loss. [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Fix a bug that mysql sink may save a wrong checkpointTs. [#5107](https://github.com/pingcap/tiflow/issues/5107) + - Fix an issue where TiCDC clusters may panic after an upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) + (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix the issue that changefeed gets stuck when tables are repeatedly scheduled in the same node [#4464](https://github.com/pingcap/tiflow/issues/4464) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix the issue that TiCDC fails to start when the first PD set in `--pd` is not available after TLS is enabled [#4777](https://github.com/pingcap/tiflow/issues/4777) + (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix a bug that querying status through open API may be blocked when the PD node is abnormal [#4778](https://github.com/pingcap/tiflow/issues/4778) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a stability problem in workerpool used by Unified Sorter [#4447](https://github.com/pingcap/tiflow/issues/4447) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a bug that sequence is incorrectly replicated in some cases [#4563](https://github.com/pingcap/tiflow/issues/4552) + Dumpling @@ -237,7 +228,12 @@ TiDB version: 5.3.2 + TiDB Data Migration (DM) - - + (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix the issue that DM occupies more disk space after the task automatically resumes [#3734](https://github.com/pingcap/tiflow/issues/3734) [#5344](https://github.com/pingcap/tiflow/issues/5344) + (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix an issue that the uppercase table cannot be replicated when `case-sensitive: true` is not set [#5255](https://github.com/pingcap/tiflow/issues/5255) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that in some cases manually executing the filtered DDL in the downstream might cause task resumption failure [#5272](https://github.com/pingcap/tiflow/issues/5272) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that CPU usage may increase and a large amount of log is printed when GTID is enabled or when the task is automatically resumed [#5063](https://github.com/pingcap/tiflow/issues/5063) + - fix bug of relay log may turn off after master reboot in v5.3.1 [#4803](https://github.com/pingcap/tiflow/issues/4803) + TiDB Lightning From 14bd6a3a074d6b436e7c3b66a306bbc10a7425a4 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 15:40:45 +0800 Subject: [PATCH 10/38] remove ticdc prs with no release notes --- releases/release-5.3.2.md | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 197ff4949233..7345c8f4cdbf 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -63,20 +63,6 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) - -+ Tools - - + PingCAP/TiCDC - - - Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#4287](https://github.com/pingcap/tiflow/issues/4287) - - Please add a release note. Please refer to [Release Notes Language Style Guide](https://pingcap.github.io/tidb-dev-guide/contribute-to-tidb/release-notes-style-guide.html) to write a quality release note. If you don't think this PR needs a release note then fill it with `None`. [#5344](https://github.com/pingcap/tiflow/issues/5344) - - `None` [#2792](https://github.com/pingcap/tiflow/issues/2792) - - `None`. [#4554](https://github.com/pingcap/tiflow/issues/4554) - - Please add a release note. `None`. [#4565](https://github.com/pingcap/tiflow/issues/4565) - - `None`. [#4135](https://github.com/pingcap/tiflow/issues/4135) - - - ## Compatibility Changes + TiDB From 7565dac8c193a2ae428a0eb4f2abdd9fa20f0486 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 16:56:38 +0800 Subject: [PATCH 11/38] clean up Signed-off-by: Ran --- releases/release-5.3.2.md | 107 +++++++++++--------------------------- 1 file changed, 29 insertions(+), 78 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 7345c8f4cdbf..09b34ff8a7f2 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -15,53 +15,17 @@ TiDB version: 5.3.2 - release-note [#33892](https://github.com/pingcap/tidb/issues/33892) - release-note [#34906](https://github.com/pingcap/tidb/issues/34906) - release-note [#35340](https://github.com/pingcap/tidb/issues/35340) - - release-note [#29283](https://github.com/pingcap/tidb/issues/29283) - release-note [#34320](https://github.com/pingcap/tidb/issues/34320) - release-note [#34417](https://github.com/pingcap/tidb/issues/34417) - release-note [#34350](https://github.com/pingcap/tidb/issues/34350) - (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Lightning)- Fix the issue of Local-backend import failure caused by out-of-bounds data in the `auto_increment` column [#27937](https://github.com/pingcap/tidb/issues/27937) - (dup: release-6.1.0.md > Improvements> Tools> TiDB Lightning)- Optimize Scatter Region to batch mode to improve the stability of the Scatter Region process [#33618](https://github.com/pingcap/tidb/issues/33618) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the issue that the precheck does not check local disk resources and cluster availability [#34213](https://github.com/pingcap/tidb/issues/34213) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix duplicate primary keys when inserting a record into a table after incremental restoration [#33596](https://github.com/pingcap/tidb/issues/33596) - - release-note [#33335](https://github.com/pingcap/tidb/issues/33335) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that schedulers do not resume after BR or TiDB Lightning exits abnormally [#33546](https://github.com/pingcap/tidb/issues/33546) - Fix the issue that NewCollationEnable config not checked during restoration. [#33422](https://github.com/pingcap/tidb/issues/33422) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix a bug that BR incremental restore returns errors mistakenly due to DDL jobs with empty query [#33322](https://github.com/pingcap/tidb/issues/33322) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) - - release-note [#33310](https://github.com/pingcap/tidb/issues/33310) - (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Backup & Restore (BR))- Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the checksum error "GC life time is shorter than transaction duration" [#32733](https://github.com/pingcap/tidb/issues/32733) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) - -+ TiKV/TiKV - - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#30087](https://github.com/pingcap/tidb/issues/30087) -+ PingCAP/TiFlash - - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that MPP tasks might leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that empty segments cannot be merged after GC [#4511](https://github.com/pingcap/tiflash/issues/4511) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the panic issue that occurs when TLS is enabled [#4196](https://github.com/pingcap/tiflash/issues/4196) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that expired data is recycled slowly [#4146](https://github.com/pingcap/tiflash/issues/4146) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the potential query error after adding columns under heavy read workload [#3967](https://github.com/pingcap/tiflash/issues/3967) - (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the bug that the `STR_TO_DATE()` function incorrectly handles leading zeros when parsing microseconds [#3557](https://github.com/pingcap/tiflash/issues/3557) - (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that TiFlash might return the `EstablishMPPConnection` error after it is restarted [#3615](https://github.com/pingcap/tiflash/issues/3615) - + PD - (dup: release-6.1.0.md > Improvements> PD)- Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) + - Fix the issue that the hot region may cause panic due to no leader [#5005](https://github.com/tikv/pd/issues/5005) - None. [#4946](https://github.com/tikv/pd/issues/4946) - (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that scheduling cannot start immediately after the PD leader transfer [4769](https://github.com/tikv/pd/issues/4769) - (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) - (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) - (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) - (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) ## Compatibility Changes @@ -85,41 +49,17 @@ TiDB version: 5.3.2 + PD - - - -+ TiDB Dashboard - - - - -+ TiFlash - - - + (dup: release-6.1.0.md > Improvements> PD)- Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) + Tools - + Backup & Restore (BR) - - - - - + TiCDC - - - - - + Dumpling - - - - - + TiDB Binlog - - - - + TiDB Data Migration (DM) (dup: release-5.4.1.md > Improvements> Tools> TiDB Data Migration (DM))- Support Syncer using the working directory of the DM-worker rather than `/tmp` to write internal files, and cleaning the directory after the task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) + TiDB Lightning - - + (dup: release-6.1.0.md > Improvements> Tools> TiDB Lightning)- Optimize Scatter Region to batch mode to improve the stability of the Scatter Region process [#33618](https://github.com/pingcap/tidb/issues/33618) ## Bug Fixes @@ -161,11 +101,11 @@ TiDB version: 5.3.2 + PD - - - -+ TiDB Dashboard - - - + (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that scheduling cannot start immediately after the PD leader transfer [4769](https://github.com/tikv/pd/issues/4769) + (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) + (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) + (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) + (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) + TiFlash @@ -183,12 +123,29 @@ TiDB version: 5.3.2 (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix possible wrong results when a query contains the `where ` clause [#3447](https://github.com/pingcap/tiflash/issues/3447) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that the `CastStringAsReal` behavior is inconsistent in TiFlash and in TiDB or TiKV [#3475](https://github.com/pingcap/tiflash/issues/3475) - Fix the bug that when cast string to datetime, the `microsecond` field may get wrong result [3556](https://github.com/pingcap/tiflash/issues/3556) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that MPP tasks might leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that empty segments cannot be merged after GC [#4511](https://github.com/pingcap/tiflash/issues/4511) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the panic issue that occurs when TLS is enabled [#4196](https://github.com/pingcap/tiflash/issues/4196) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that expired data is recycled slowly [#4146](https://github.com/pingcap/tiflash/issues/4146) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the potential query error after adding columns under heavy read workload [#3967](https://github.com/pingcap/tiflash/issues/3967) + (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the bug that the `STR_TO_DATE()` function incorrectly handles leading zeros when parsing microseconds [#3557](https://github.com/pingcap/tiflash/issues/3557) + (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that TiFlash might return the `EstablishMPPConnection` error after it is restarted [#3615](https://github.com/pingcap/tiflash/issues/3615) + Tools + Backup & Restore (BR) - - + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix duplicate primary keys when inserting a record into a table after incremental restoration [#33596](https://github.com/pingcap/tidb/issues/33596) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that schedulers do not resume after BR or TiDB Lightning exits abnormally [#33546](https://github.com/pingcap/tidb/issues/33546) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix a bug that BR incremental restore returns errors mistakenly due to DDL jobs with empty query [#33322](https://github.com/pingcap/tidb/issues/33322) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) + (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Backup & Restore (BR))- Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) + (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) + TiCDC @@ -204,14 +161,6 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a stability problem in workerpool used by Unified Sorter [#4447](https://github.com/pingcap/tiflow/issues/4447) (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a bug that sequence is incorrectly replicated in some cases [#4563](https://github.com/pingcap/tiflow/issues/4552) - + Dumpling - - - - - + TiDB Binlog - - - - + TiDB Data Migration (DM) (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix the issue that DM occupies more disk space after the task automatically resumes [#3734](https://github.com/pingcap/tiflow/issues/3734) [#5344](https://github.com/pingcap/tiflow/issues/5344) @@ -223,4 +172,6 @@ TiDB version: 5.3.2 + TiDB Lightning - - + (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Lightning)- Fix the issue of Local-backend import failure caused by out-of-bounds data in the `auto_increment` column [#27937](https://github.com/pingcap/tidb/issues/27937) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the issue that the precheck does not check local disk resources and cluster availability [#34213](https://github.com/pingcap/tidb/issues/34213) + (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the checksum error "GC life time is shorter than transaction duration" [#32733](https://github.com/pingcap/tidb/issues/32733) From 28b54ab811ae553517f920ad1f795ecff8e5ee57 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 19:46:25 +0800 Subject: [PATCH 12/38] add tidb note Signed-off-by: Ran --- releases/release-5.3.2.md | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 09b34ff8a7f2..69b4548afd77 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -12,11 +12,6 @@ TiDB version: 5.3.2 + PingCAP/TiDB - - release-note [#33892](https://github.com/pingcap/tidb/issues/33892) - - release-note [#34906](https://github.com/pingcap/tidb/issues/34906) - - release-note [#35340](https://github.com/pingcap/tidb/issues/35340) - - release-note [#34320](https://github.com/pingcap/tidb/issues/34320) - - release-note [#34417](https://github.com/pingcap/tidb/issues/34417) - release-note [#34350](https://github.com/pingcap/tidb/issues/34350) - Fix the issue that NewCollationEnable config not checked during restoration. [#33422](https://github.com/pingcap/tidb/issues/33422) - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#30087](https://github.com/pingcap/tidb/issues/30087) @@ -80,6 +75,7 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) + - Fix the problem that tidb-server may encounter crash because of `fatal error: concurrent map read and map write` [#35340](https://github.com/pingcap/tidb/issues/35340) + TiKV From f630c402f4b05561ce461c0225d6adfd0df6328a Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 19:51:58 +0800 Subject: [PATCH 13/38] remove duplicated notes Signed-off-by: Ran --- releases/release-5.3.2.md | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 69b4548afd77..747d33b1ba30 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -105,8 +105,10 @@ TiDB version: 5.3.2 + TiFlash - (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix possible metadata corruption caused by Region merge on a lagging Region peer [#4437](https://github.com/pingcap/tiflash/issues/4437) + - Fix some configuration issues [#4093](https://github.com/pingcap/tiflash/issues/4093), [#4091](https://github.com/pingcap/tiflash/issues/4091) + - Fix `TiFlash_schema_error` caused by `NOT NULL` columns [#4596](https://github.com/pingcap/tiflash/issues/4596) + - Fix repeated crash caused by`commit state jump backward` errors [#2576](https://github.com/pingcap/tiflash/issues/2576) + - Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [3713](https://github.com/pingcap/tiflash/issues/3713) - Fix bug that MPP query may failed due to grpc keepalive timeout [4662](https://github.com/pingcap/tiflash/issues/4662) From 97e3824c300ff51348242f2d00a58110a668bafc Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 20:37:37 +0800 Subject: [PATCH 14/38] add pd/br notes Signed-off-by: Ran --- releases/release-5.3.2.md | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 747d33b1ba30..295df6508f99 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -12,27 +12,20 @@ TiDB version: 5.3.2 + PingCAP/TiDB - - release-note [#34350](https://github.com/pingcap/tidb/issues/34350) - Fix the issue that NewCollationEnable config not checked during restoration. [#33422](https://github.com/pingcap/tidb/issues/33422) - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#30087](https://github.com/pingcap/tidb/issues/30087) -+ PD - - - - Fix the issue that the hot region may cause panic due to no leader [#5005](https://github.com/tikv/pd/issues/5005) - - None. [#4946](https://github.com/tikv/pd/issues/4946) - ## Compatibility Changes + TiDB (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the `REPLACE` statement incorrectly changes other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) -## Improvements ++ PD -+ TiDB + - Disable compile swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) - - +## Improvements + TiKV @@ -56,7 +49,6 @@ TiDB version: 5.3.2 (dup: release-6.1.0.md > Improvements> Tools> TiDB Lightning)- Optimize Scatter Region to batch mode to improve the stability of the Scatter Region process [#33618](https://github.com/pingcap/tidb/issues/33618) - ## Bug Fixes + TiDB @@ -144,6 +136,7 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Backup & Restore (BR))- Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) + - Fix the issue that backup cannot handle s3 internal error. [#34350](https://github.com/pingcap/tidb/issues/34350) + TiCDC From 4d7a561f38ed92d5185112ef072ce43c6e085997 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 20:38:01 +0800 Subject: [PATCH 15/38] Update releases/release-5.3.2.md Co-authored-by: ShuNing --- releases/release-5.3.2.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 295df6508f99..cfe88dbd0d7d 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -94,6 +94,8 @@ TiDB version: 5.3.2 (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) + - Fix `SchedulerMaxWaitingOperator` set to 1, no scheduling issue [#4946](https://github.com/tikv/pd/issues/4946) + - Fix PD Panic problem caused by Hot Region without Leader [#5005](https://github.com/tikv/pd/issues/5005) + TiFlash From 19fc9329c819c7f712720920f92674023163d866 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 20:52:30 +0800 Subject: [PATCH 16/38] add missing notes from chinese --- releases/release-5.3.2.md | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 295df6508f99..26c4a26abd9e 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -89,10 +89,12 @@ TiDB version: 5.3.2 + PD - (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that scheduling cannot start immediately after the PD leader transfer [4769](https://github.com/tikv/pd/issues/4769) + - 修复 PD 由于 Hot Region 没有 Leader 导致的 Panic 的问题 [#5005](https://github.com/tikv/pd/issues/5005) + (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that scheduling cannot start immediately after the PD leader transfer [#4769](https://github.com/tikv/pd/issues/4769) (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) + - 修复 `SchedulerMaxWaitingOperator ` 设置为 1 时，不产生调度的问题 [#4946](https://github.com/tikv/pd/issues/4946) (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) + TiFlash @@ -102,17 +104,17 @@ TiDB version: 5.3.2 - Fix repeated crash caused by`commit state jump backward` errors [#2576](https://github.com/pingcap/tiflash/issues/2576) - Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [3713](https://github.com/pingcap/tiflash/issues/3713) - - Fix bug that MPP query may failed due to grpc keepalive timeout [4662](https://github.com/pingcap/tiflash/issues/4662) - - Fix bug that MPP query may hang forever if there are many retry in exchange receiver [3473](https://github.com/pingcap/tiflash/pull/3473) + - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3713](https://github.com/pingcap/tiflash/issues/3713) + - Fix bug that MPP query may failed due to grpc keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) + - Fix bug that MPP query may hang forever if there are many retry in exchange receiver [#3473](https://github.com/pingcap/tiflash/pull/3473) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - - Fix potential `index out of bounds` error if calling `json_length` with empty string [2705](https://github.com/pingcap/tiflash/issues/2705) - - Fix decimal compare will return wrong result in some corner cae [4942](https://github.com/pingcap/tiflash/pull/4942) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix bug that MPP query may hang forever if query failed in join build stage [4195](https://github.com/pingcap/tiflash/issues/4195) + - Fix potential `index out of bounds` error if calling `json_length` with empty string [#2705](https://github.com/pingcap/tiflash/issues/2705) + - Fix decimal compare will return wrong result in some corner cae [#4942](https://github.com/pingcap/tiflash/pull/4942) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix bug that MPP query may hang forever if query failed in join build stage [#4195](https://github.com/pingcap/tiflash/issues/4195) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix possible wrong results when a query contains the `where ` clause [#3447](https://github.com/pingcap/tiflash/issues/3447) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that the `CastStringAsReal` behavior is inconsistent in TiFlash and in TiDB or TiKV [#3475](https://github.com/pingcap/tiflash/issues/3475) - - Fix the bug that when cast string to datetime, the `microsecond` field may get wrong result [3556](https://github.com/pingcap/tiflash/issues/3556) + - Fix the bug that when cast string to datetime, the `microsecond` field may get wrong result [#3556](https://github.com/pingcap/tiflash/issues/3556) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) From fc186ce489de438042cdb892283ee4728e957d77 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 20:56:09 +0800 Subject: [PATCH 17/38] mark dup notes --- releases/release-5.3.2.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 26c4a26abd9e..b7f404cb4aff 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -23,7 +23,7 @@ TiDB version: 5.3.2 + PD - - Disable compile swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) + (dup: release-6.1.0.md > Improvements> PD)- Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) ## Improvements @@ -99,10 +99,10 @@ TiDB version: 5.3.2 + TiFlash - - Fix some configuration issues [#4093](https://github.com/pingcap/tiflash/issues/4093), [#4091](https://github.com/pingcap/tiflash/issues/4091) + (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) - Fix `TiFlash_schema_error` caused by `NOT NULL` columns [#4596](https://github.com/pingcap/tiflash/issues/4596) - Fix repeated crash caused by`commit state jump backward` errors [#2576](https://github.com/pingcap/tiflash/issues/2576) - - Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) + (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3713](https://github.com/pingcap/tiflash/issues/3713) - Fix bug that MPP query may failed due to grpc keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) From a117fd70d76fad14a295f70e9c2e1458227470e4 Mon Sep 17 00:00:00 2001 From: Ran Date: Thu, 23 Jun 2022 21:14:35 +0800 Subject: [PATCH 18/38] remove unsorted notes Signed-off-by: Ran --- releases/release-5.3.2.md | 7 ------- 1 file changed, 7 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 247bc266970f..4afcea4b4f5e 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -8,13 +8,6 @@ Release Date: June xx, 2022 TiDB version: 5.3.2 -## __unsorted - -+ PingCAP/TiDB - - - Fix the issue that NewCollationEnable config not checked during restoration. [#33422](https://github.com/pingcap/tidb/issues/33422) - - Add a new hidden config `s3_multi_part_size` to make backup can control the part size when upload big sst file to s3. [#30087](https://github.com/pingcap/tidb/issues/30087) - ## Compatibility Changes + TiDB From 8b2a13b023e080efb0ae8c9612115072c1e5005b Mon Sep 17 00:00:00 2001 From: TomShawn <41534398+TomShawn@users.noreply.github.com> Date: Mon, 27 Jun 2022 11:31:53 +0800 Subject: [PATCH 19/38] Apply suggestions from code review --- releases/release-5.3.2.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 4afcea4b4f5e..0a06f2d769e1 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -49,8 +49,8 @@ TiDB version: 5.3.2 (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Dumpling)- Fix the issue that Amazon S3 cannot correctly calculate the size of compressed data [#30534](https://github.com/pingcap/tidb/issues/30534) (dup: release-5.4.0.md > Bug fixes> TiDB)- Fix the issue of potential data index inconsistency in optimistic transaction mode [#30410](https://github.com/pingcap/tidb/issues/30410) (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that a SQL operation is canceled when its JSON type column joins its `CHAR` type column [#29401](https://github.com/pingcap/tidb/issues/29401) - - In the event that a network connectivity event occurred, TiDB did not always free the resources of the disconnected session correctly. This has been fixed to ensure that open transactions are rolled back, as well as other associated resources. [#34722](https://github.com/pingcap/tidb/issues/34722) - - Fixed an issue where extra datums may break binlog. [#33608](https://github.com/pingcap/tidb/issues/33608) + - Previously, when a network connectivity issue occurred, TiDB did not always correctly free the resources held by the disconnected session. This issue has been fixed so that open transactions can be rolled back and other associated resources can be released. [#34722](https://github.com/pingcap/tidb/issues/34722) + - Fix the issue of the `data and columnID count not match` error that occurs when inserting duplicated values with TiDB Binlog enabled [#33608](https://github.com/pingcap/tidb/issues/33608) (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) From d08abc2b6f14d35680ce0f1dabeaa92f6f4847ad Mon Sep 17 00:00:00 2001 From: Aolin Date: Mon, 27 Jun 2022 13:10:51 +0800 Subject: [PATCH 20/38] Apply suggestions from code review --- releases/release-5.3.2.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 0a06f2d769e1..f50a9efbe053 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -64,9 +64,9 @@ TiDB version: 5.3.2 + TiKV - - Fix the bug that causes frequent pd client reconnection [#12345](https://github.com/tikv/tikv/issues/12345) - - Fix a wrong check in datetime when the datetime has a fraction and 'Z' [#12739](https://github.com/tikv/tikv/issues/12739) - - Fix the bug that converting empty string causes TiKV crash [#12673](https://github.com/tikv/tikv/issues/12673) + - Fix the bug that causes frequent PD client reconnection [#12345](https://github.com/tikv/tikv/issues/12345) + - Fix a wrong check in `DATETIME` when the datetime has a fraction and `Z` [#12739](https://github.com/tikv/tikv/issues/12739) + - Fix the bug that converting empty string causes TiKV panic [#12673](https://github.com/tikv/tikv/issues/12673) - Fix possible duplicate commit records in async-commit pessimistic transactions [#12615](https://github.com/tikv/tikv/issues/12615) (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) From c02f83a578776f713c22c5c25787bc05775612c3 Mon Sep 17 00:00:00 2001 From: Aolin Date: Mon, 27 Jun 2022 13:35:08 +0800 Subject: [PATCH 21/38] Apply suggestions from code review --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index f50a9efbe053..2cd84ffb3f4d 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -23,7 +23,7 @@ TiDB version: 5.3.2 + TiKV (dup: release-5.4.0.md > Improvements> TiKV)- Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) - (dup: release-6.1.0.md > Improvements> TiKV) - Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) + (dup: release-6.1.0.md > Improvements> TiKV)- Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) (dup: release-5.2.4.md > Improvements> TiKV)- Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) (dup: release-5.3.0.md > Improvements> TiKV> Enhance disk space protection to improve storage stability)- Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) (dup: release-5.0.6.md > Improvements> TiKV)- Add more metrics for the garbage collection module of Raft logs to locate performance problems in the module [#11374](https://github.com/tikv/tikv/issues/11374) From 97cf6ce8ba899af05d207c8a11456bfddf05d264 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Tue, 28 Jun 2022 09:45:23 +0800 Subject: [PATCH 22/38] Apply suggestions from code review --- releases/release-5.3.2.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 2cd84ffb3f4d..458271b164d8 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -82,12 +82,12 @@ TiDB version: 5.3.2 + PD - - 修复 PD 由于 Hot Region 没有 Leader 导致的 Panic 的问题 [#5005](https://github.com/tikv/pd/issues/5005) + - Fix the PD panic that occurs when a hot region has no leader [#5005](https://github.com/tikv/pd/issues/5005) (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that scheduling cannot start immediately after the PD leader transfer [#4769](https://github.com/tikv/pd/issues/4769) (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) - - 修复 `SchedulerMaxWaitingOperator ` 设置为 1 时，不产生调度的问题 [#4946](https://github.com/tikv/pd/issues/4946) + - Fix the issue that schedulers do not work when `SchedulerMaxWaitingOperator ` is set to `1` [#4946](https://github.com/tikv/pd/issues/4946) (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) - Fix `SchedulerMaxWaitingOperator` set to 1, no scheduling issue [#4946](https://github.com/tikv/pd/issues/4946) - Fix PD Panic problem caused by Hot Region without Leader [#5005](https://github.com/tikv/pd/issues/5005) From 98df936c277452d67a08b88d5336962592b080b7 Mon Sep 17 00:00:00 2001 From: shichun-0415 <89768198+shichun-0415@users.noreply.github.com> Date: Tue, 28 Jun 2022 11:19:46 +0800 Subject: [PATCH 23/38] Apply suggestions from code review --- releases/release-5.3.2.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 458271b164d8..65ff7716a56c 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -95,21 +95,21 @@ TiDB version: 5.3.2 + TiFlash (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) - - Fix `TiFlash_schema_error` caused by `NOT NULL` columns [#4596](https://github.com/pingcap/tiflash/issues/4596) + - Fix `TiFlash_schema_error` reported when adding `NOT NULL` columns [#4596](https://github.com/pingcap/tiflash/issues/4596) - Fix repeated crash caused by`commit state jump backward` errors [#2576](https://github.com/pingcap/tiflash/issues/2576) (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - - Avoid false alert of `DB::Exception: Encode type of coprocessor response is not CHBlock` [#3713](https://github.com/pingcap/tiflash/issues/3713) - - Fix bug that MPP query may failed due to grpc keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) + - Fix false report of inconsistent TiFlash version when TiFlash uses remote read [#3713](https://github.com/pingcap/tiflash/issues/3713) + - Fix a bug that MPP query fails due to random gRPC keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) - Fix bug that MPP query may hang forever if there are many retry in exchange receiver [#3473](https://github.com/pingcap/tiflash/pull/3473) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - - Fix potential `index out of bounds` error if calling `json_length` with empty string [#2705](https://github.com/pingcap/tiflash/issues/2705) - - Fix decimal compare will return wrong result in some corner cae [#4942](https://github.com/pingcap/tiflash/pull/4942) + - Fix the potential `index out of bounds` error if calling `json_length` with empty string [#2705](https://github.com/pingcap/tiflash/issues/2705) + - Fix wrong decimal comparison result in corner cases [#4942](https://github.com/pingcap/tiflash/pull/4942) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix bug that MPP query may hang forever if query failed in join build stage [#4195](https://github.com/pingcap/tiflash/issues/4195) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix possible wrong results when a query contains the `where ` clause [#3447](https://github.com/pingcap/tiflash/issues/3447) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that the `CastStringAsReal` behavior is inconsistent in TiFlash and in TiDB or TiKV [#3475](https://github.com/pingcap/tiflash/issues/3475) - - Fix the bug that when cast string to datetime, the `microsecond` field may get wrong result [#3556](https://github.com/pingcap/tiflash/issues/3556) + - Fix incorrect `microsecond` when casting string to datetime [#3556](https://github.com/pingcap/tiflash/issues/3556) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) @@ -133,7 +133,7 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Backup & Restore (BR))- Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) - - Fix the issue that backup cannot handle s3 internal error. [#34350](https://github.com/pingcap/tidb/issues/34350) + - Fix the issue that BR cannot handle S3 internal errors [#34350](https://github.com/pingcap/tidb/issues/34350) + TiCDC From 37dd7065cbecbc7b9c98d5dd800b52675bc45404 Mon Sep 17 00:00:00 2001 From: shichun-0415 <89768198+shichun-0415@users.noreply.github.com> Date: Tue, 28 Jun 2022 11:26:57 +0800 Subject: [PATCH 24/38] Update releases/release-5.3.2.md --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 65ff7716a56c..72923027bbff 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -101,7 +101,7 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - Fix false report of inconsistent TiFlash version when TiFlash uses remote read [#3713](https://github.com/pingcap/tiflash/issues/3713) - Fix a bug that MPP query fails due to random gRPC keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) - - Fix bug that MPP query may hang forever if there are many retry in exchange receiver [#3473](https://github.com/pingcap/tiflash/pull/3473) + - Fix a bug that MPP query might hang forever if there are retries in the exchange receiver [#3473](https://github.com/pingcap/tiflash/pull/3473) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - Fix the potential `index out of bounds` error if calling `json_length` with empty string [#2705](https://github.com/pingcap/tiflash/issues/2705) From ff4e3a9660675662f1ca949945f557def769f5dc Mon Sep 17 00:00:00 2001 From: Aolin Date: Tue, 28 Jun 2022 13:22:50 +0800 Subject: [PATCH 25/38] Apply suggestions from code review --- releases/release-5.3.2.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 72923027bbff..65958e91187b 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -64,10 +64,10 @@ TiDB version: 5.3.2 + TiKV - - Fix the bug that causes frequent PD client reconnection [#12345](https://github.com/tikv/tikv/issues/12345) - - Fix a wrong check in `DATETIME` when the datetime has a fraction and `Z` [#12739](https://github.com/tikv/tikv/issues/12739) - - Fix the bug that converting empty string causes TiKV panic [#12673](https://github.com/tikv/tikv/issues/12673) - - Fix possible duplicate commit records in async-commit pessimistic transactions [#12615](https://github.com/tikv/tikv/issues/12615) + - Fix the bug that the PD client reconnects frequently [#12345](https://github.com/tikv/tikv/issues/12345) + - Fix a wrong check for `DATETIME` values with fractional seconds and `Z` [#12739](https://github.com/tikv/tikv/issues/12739) + - Fix the issue that TiKV panics when converting empty string [#12673](https://github.com/tikv/tikv/issues/12673) + - Fix possible duplicate commit records in pessimistic transactions when async commit is enabled [#12615](https://github.com/tikv/tikv/issues/12615) (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) From 2c45be2d28b641ccabf6828cc72a2b84e28f1e30 Mon Sep 17 00:00:00 2001 From: TomShawn <41534398+TomShawn@users.noreply.github.com> Date: Tue, 28 Jun 2022 14:11:07 +0800 Subject: [PATCH 26/38] Update releases/release-5.3.2.md --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 65958e91187b..7c5dd6db154f 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -60,7 +60,7 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) - - Fix the problem that tidb-server may encounter crash because of `fatal error: concurrent map read and map write` [#35340](https://github.com/pingcap/tidb/issues/35340) + - Fix the panic issue caused by the `fatal error: concurrent map read and map write` error [#35340](https://github.com/pingcap/tidb/issues/35340) + TiKV From cb6d2c50168c36d7b73484fe3650ab900bbeb50b Mon Sep 17 00:00:00 2001 From: Aolin Date: Tue, 28 Jun 2022 14:17:17 +0800 Subject: [PATCH 27/38] Apply suggestions from code review Co-authored-by: TomShawn <41534398+TomShawn@users.noreply.github.com> --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 7c5dd6db154f..e80bca06ffae 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -64,7 +64,7 @@ TiDB version: 5.3.2 + TiKV - - Fix the bug that the PD client reconnects frequently [#12345](https://github.com/tikv/tikv/issues/12345) + - Fix the issue of frequent PD client reconnection that occurs when the PD client meets an error [#12345](https://github.com/tikv/tikv/issues/12345) - Fix a wrong check for `DATETIME` values with fractional seconds and `Z` [#12739](https://github.com/tikv/tikv/issues/12739) - Fix the issue that TiKV panics when converting empty string [#12673](https://github.com/tikv/tikv/issues/12673) - Fix possible duplicate commit records in pessimistic transactions when async commit is enabled [#12615](https://github.com/tikv/tikv/issues/12615) From 59daeed62c25bbde43db46a48569198b08ab8617 Mon Sep 17 00:00:00 2001 From: Aolin Date: Tue, 28 Jun 2022 14:26:57 +0800 Subject: [PATCH 28/38] Apply suggestions from code review Co-authored-by: TomShawn <41534398+TomShawn@users.noreply.github.com> --- releases/release-5.3.2.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index e80bca06ffae..777bfc49f336 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -65,9 +65,9 @@ TiDB version: 5.3.2 + TiKV - Fix the issue of frequent PD client reconnection that occurs when the PD client meets an error [#12345](https://github.com/tikv/tikv/issues/12345) - - Fix a wrong check for `DATETIME` values with fractional seconds and `Z` [#12739](https://github.com/tikv/tikv/issues/12739) - - Fix the issue that TiKV panics when converting empty string [#12673](https://github.com/tikv/tikv/issues/12673) - - Fix possible duplicate commit records in pessimistic transactions when async commit is enabled [#12615](https://github.com/tikv/tikv/issues/12615) + - Fix the issue of time parsing error that occurs when the `DATETIME` values contain a fraction and `Z` [#12739](https://github.com/tikv/tikv/issues/12739) + - Fix the issue that TiKV panics when performing type conversion for an empty string [#12673](https://github.com/tikv/tikv/issues/12673) + - Fix the possible duplicate commit records in pessimistic transactions when async commit is enabled [#12615](https://github.com/tikv/tikv/issues/12615) (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) From 45387c745bb8cb31d8c51073918c45cee64e0e68 Mon Sep 17 00:00:00 2001 From: Ran Date: Tue, 28 Jun 2022 16:07:57 +0800 Subject: [PATCH 29/38] Update releases/release-5.3.2.md --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 777bfc49f336..0ed0dd32b80b 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -156,7 +156,7 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that in some cases manually executing the filtered DDL in the downstream might cause task resumption failure [#5272](https://github.com/pingcap/tiflow/issues/5272) (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that CPU usage may increase and a large amount of log is printed when GTID is enabled or when the task is automatically resumed [#5063](https://github.com/pingcap/tiflow/issues/5063) - - fix bug of relay log may turn off after master reboot in v5.3.1 [#4803](https://github.com/pingcap/tiflow/issues/4803) + - Fix the issue that the relay log may be disabled after the DM-master reboots [#4803](https://github.com/pingcap/tiflow/issues/4803) + TiDB Lightning From d9e45eadce23e4d3c0d6a2464da9d61107bce68a Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Tue, 28 Jun 2022 20:04:45 +0800 Subject: [PATCH 30/38] Update releases/release-5.3.2.md Co-authored-by: shichun-0415 <89768198+shichun-0415@users.noreply.github.com> --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 0ed0dd32b80b..e47959842754 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -87,7 +87,7 @@ TiDB version: 5.3.2 (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) - - Fix the issue that schedulers do not work when `SchedulerMaxWaitingOperator ` is set to `1` [#4946](https://github.com/tikv/pd/issues/4946) + - Fix the issue that schedulers do not work when `SchedulerMaxWaitingOperator` is set to `1` [#4946](https://github.com/tikv/pd/issues/4946) (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) - Fix `SchedulerMaxWaitingOperator` set to 1, no scheduling issue [#4946](https://github.com/tikv/pd/issues/4946) - Fix PD Panic problem caused by Hot Region without Leader [#5005](https://github.com/tikv/pd/issues/5005) From b6da4245cacc0cc1add7077d93db80ef9332a8be Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Tue, 28 Jun 2022 20:09:18 +0800 Subject: [PATCH 31/38] Update releases/release-5.3.2.md --- releases/release-5.3.2.md | 2 -- 1 file changed, 2 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index e47959842754..406370d5ed79 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -89,8 +89,6 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) - Fix the issue that schedulers do not work when `SchedulerMaxWaitingOperator` is set to `1` [#4946](https://github.com/tikv/pd/issues/4946) (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) - - Fix `SchedulerMaxWaitingOperator` set to 1, no scheduling issue [#4946](https://github.com/tikv/pd/issues/4946) - - Fix PD Panic problem caused by Hot Region without Leader [#5005](https://github.com/tikv/pd/issues/5005) + TiFlash From c5ac5e954e28235673868be4977923202432a322 Mon Sep 17 00:00:00 2001 From: xixirangrang Date: Tue, 28 Jun 2022 21:34:32 +0800 Subject: [PATCH 32/38] Update releases/release-5.3.2.md --- releases/release-5.3.2.md | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 406370d5ed79..022cf233e1a0 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -136,11 +136,11 @@ TiDB version: 5.3.2 + TiCDC (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix incorrect metrics caused by owner changes [#4774](https://github.com/pingcap/tiflow/issues/4774) - - Fix a bug in redo log manager that flush log executed before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) - - Fix a bug that resolved ts moves too fast when part of tables are not maintained redo writer. [#5486](https://github.com/pingcap/tiflow/issues/5486) - - Add uuid suffix to redo log file name to prevent name conflict, which may cause data loss. [#5486](https://github.com/pingcap/tiflow/issues/5486) - - Fix a bug that mysql sink may save a wrong checkpointTs. [#5107](https://github.com/pingcap/tiflow/issues/5107) - - Fix an issue where TiCDC clusters may panic after an upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) + - Fix the bug that the redo log manager flushes logs before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Fix the bug that the resolved ts moves too fast when some tables are not maintained by the redo writer [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Fix the issue that file name conflicts may cause data loss by adding the UUID suffix to the redo log file name [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Fix the bug that MySQL Sink may save a wrong checkpointTs [#5107](https://github.com/pingcap/tiflow/issues/5107) + - Fix the issue that TiCDC clusters may panic after upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix the issue that changefeed gets stuck when tables are repeatedly scheduled in the same node [#4464](https://github.com/pingcap/tiflow/issues/4464) (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix the issue that TiCDC fails to start when the first PD set in `--pd` is not available after TLS is enabled [#4777](https://github.com/pingcap/tiflow/issues/4777) (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix a bug that querying status through open API may be blocked when the PD node is abnormal [#4778](https://github.com/pingcap/tiflow/issues/4778) From cabcca44f7af5343364e8a7df8cab4ff7e440244 Mon Sep 17 00:00:00 2001 From: xixirangrang Date: Tue, 28 Jun 2022 21:46:43 +0800 Subject: [PATCH 33/38] Update releases/release-5.3.2.md --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 022cf233e1a0..c8f57215cdb5 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -138,7 +138,7 @@ TiDB version: 5.3.2 (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix incorrect metrics caused by owner changes [#4774](https://github.com/pingcap/tiflow/issues/4774) - Fix the bug that the redo log manager flushes logs before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) - Fix the bug that the resolved ts moves too fast when some tables are not maintained by the redo writer [#5486](https://github.com/pingcap/tiflow/issues/5486) - - Fix the issue that file name conflicts may cause data loss by adding the UUID suffix to the redo log file name [#5486](https://github.com/pingcap/tiflow/issues/5486) + - Add the UUID suffix to the redo log file name to fix the issue that file name conflicts may cause data loss [#5486](https://github.com/pingcap/tiflow/issues/5486) - Fix the bug that MySQL Sink may save a wrong checkpointTs [#5107](https://github.com/pingcap/tiflow/issues/5107) - Fix the issue that TiCDC clusters may panic after upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix the issue that changefeed gets stuck when tables are repeatedly scheduled in the same node [#4464](https://github.com/pingcap/tiflow/issues/4464) From 58cb88bb70defccbaf8bbe87c9294687877cfc48 Mon Sep 17 00:00:00 2001 From: shichun-0415 <89768198+shichun-0415@users.noreply.github.com> Date: Tue, 28 Jun 2022 23:40:47 +0800 Subject: [PATCH 34/38] Apply suggestions from code review Co-authored-by: Grace Cai --- releases/release-5.3.2.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index c8f57215cdb5..192ec3659291 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -93,21 +93,21 @@ TiDB version: 5.3.2 + TiFlash (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) - - Fix `TiFlash_schema_error` reported when adding `NOT NULL` columns [#4596](https://github.com/pingcap/tiflash/issues/4596) - - Fix repeated crash caused by`commit state jump backward` errors [#2576](https://github.com/pingcap/tiflash/issues/2576) + - Fix `TiFlash_schema_error` reported when `NOT NULL` columns are added [#4596](https://github.com/pingcap/tiflash/issues/4596) + - Fix repeated crashes caused by the `commit state jump backward` errors [#2576](https://github.com/pingcap/tiflash/issues/2576) (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - - Fix false report of inconsistent TiFlash version when TiFlash uses remote read [#3713](https://github.com/pingcap/tiflash/issues/3713) - - Fix a bug that MPP query fails due to random gRPC keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) - - Fix a bug that MPP query might hang forever if there are retries in the exchange receiver [#3473](https://github.com/pingcap/tiflash/pull/3473) + - Fix false reports of inconsistent TiFlash versions when TiFlash uses remote read [#3713](https://github.com/pingcap/tiflash/issues/3713) + - Fix a bug that an MPP query might fail due to random gRPC keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) + - Fix a bug that an MPP query might hang forever if there are retries in the exchange receiver [#3444](https://github.com/pingcap/tiflash/issues/3444) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - Fix the potential `index out of bounds` error if calling `json_length` with empty string [#2705](https://github.com/pingcap/tiflash/issues/2705) - - Fix wrong decimal comparison result in corner cases [#4942](https://github.com/pingcap/tiflash/pull/4942) + - Fix wrong decimal comparison results in corner cases [#4512](https://github.com/pingcap/tiflash/issues/4512) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix bug that MPP query may hang forever if query failed in join build stage [#4195](https://github.com/pingcap/tiflash/issues/4195) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix possible wrong results when a query contains the `where ` clause [#3447](https://github.com/pingcap/tiflash/issues/3447) (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that the `CastStringAsReal` behavior is inconsistent in TiFlash and in TiDB or TiKV [#3475](https://github.com/pingcap/tiflash/issues/3475) - - Fix incorrect `microsecond` when casting string to datetime [#3556](https://github.com/pingcap/tiflash/issues/3556) + - Fix incorrect `microsecond` when casting string to datetime [#3556](https://github.com/pingcap/tiflash/issues/3556) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) From 4e7bc4c345aeba120cf325c77ecc38654f9663ec Mon Sep 17 00:00:00 2001 From: Ran Date: Wed, 29 Jun 2022 09:04:18 +0800 Subject: [PATCH 35/38] Update releases/release-5.3.2.md --- releases/release-5.3.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 192ec3659291..69f732fe886c 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -4,7 +4,7 @@ title: TiDB 5.3.2 Release Notes # TiDB 5.3.2 Release Notes -Release Date: June xx, 2022 +Release Date: June 29, 2022 TiDB version: 5.3.2 From 46927fc26df9a54fe092167a7113e5930887d0e4 Mon Sep 17 00:00:00 2001 From: Ran Date: Wed, 29 Jun 2022 09:47:40 +0800 Subject: [PATCH 36/38] replace duplicated notes Signed-off-by: Ran --- releases/release-5.3.2.md | 159 ++++++++++++++++++-------------------- 1 file changed, 76 insertions(+), 83 deletions(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 69f732fe886c..a2b14b2519f8 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -10,57 +10,50 @@ TiDB version: 5.3.2 ## Compatibility Changes -+ TiDB - - (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the `REPLACE` statement incorrectly changes other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) - + PD - (dup: release-6.1.0.md > Improvements> PD)- Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) + - Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) ## Improvements + TiKV - (dup: release-5.4.0.md > Improvements> TiKV)- Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) - (dup: release-6.1.0.md > Improvements> TiKV)- Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) - (dup: release-5.2.4.md > Improvements> TiKV)- Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) + - Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) + - Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) + - Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) (dup: release-5.3.0.md > Improvements> TiKV> Enhance disk space protection to improve storage stability)- Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) - (dup: release-5.0.6.md > Improvements> TiKV)- Add more metrics for the garbage collection module of Raft logs to locate performance problems in the module [#11374](https://github.com/tikv/tikv/issues/11374) - -+ PD - - (dup: release-6.1.0.md > Improvements> PD)- Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) + - Add more metrics for the garbage collection module of Raft logs to locate performance problems in the module [#11374](https://github.com/tikv/tikv/issues/11374) + Tools + TiDB Data Migration (DM) - (dup: release-5.4.1.md > Improvements> Tools> TiDB Data Migration (DM))- Support Syncer using the working directory of the DM-worker rather than `/tmp` to write internal files, and cleaning the directory after the task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) + - Support Syncer using the working directory of the DM-worker rather than `/tmp` to write internal files, and cleaning the directory after the task is stopped [#4107](https://github.com/pingcap/tiflow/issues/4107) + TiDB Lightning - (dup: release-6.1.0.md > Improvements> Tools> TiDB Lightning)- Optimize Scatter Region to batch mode to improve the stability of the Scatter Region process [#33618](https://github.com/pingcap/tidb/issues/33618) + - Optimize Scatter Region to batch mode to improve the stability of the Scatter Region process [#33618](https://github.com/pingcap/tidb/issues/33618) ## Bug Fixes + TiDB - (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Dumpling)- Fix the issue that Amazon S3 cannot correctly calculate the size of compressed data [#30534](https://github.com/pingcap/tidb/issues/30534) - (dup: release-5.4.0.md > Bug fixes> TiDB)- Fix the issue of potential data index inconsistency in optimistic transaction mode [#30410](https://github.com/pingcap/tidb/issues/30410) - (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that a SQL operation is canceled when its JSON type column joins its `CHAR` type column [#29401](https://github.com/pingcap/tidb/issues/29401) + - Fix the issue that Amazon S3 cannot correctly calculate the size of compressed data [#30534](https://github.com/pingcap/tidb/issues/30534) + - Fix the issue of potential data index inconsistency in optimistic transaction mode [#30410](https://github.com/pingcap/tidb/issues/30410) + - Fix the issue that a SQL operation is canceled when its JSON type column joins its `CHAR` type column [#29401](https://github.com/pingcap/tidb/issues/29401) - Previously, when a network connectivity issue occurred, TiDB did not always correctly free the resources held by the disconnected session. This issue has been fixed so that open transactions can be rolled back and other associated resources can be released. [#34722](https://github.com/pingcap/tidb/issues/34722) - Fix the issue of the `data and columnID count not match` error that occurs when inserting duplicated values with TiDB Binlog enabled [#33608](https://github.com/pingcap/tidb/issues/33608) - (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that the TiDB server might run out of memory when the `INFORMATION_SCHEMA.CLUSTER_SLOW_QUERY` table is queried. This issue can be triggered when you check slow queries on the Grafana dashboard [#33893](https://github.com/pingcap/tidb/issues/33893) - (dup: release-6.1.0.md > Bug fixes> TiDB)- Fix the issue that some DDL statements might be stuck for a period after the PD node of a cluster is replaced [#33908](https://github.com/pingcap/tidb/issues/33908) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix the issue that granting the `all` privilege might fail in clusters that are upgraded from v4.0 [#33588](https://github.com/pingcap/tidb/issues/33588) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) - (dup: release-5.4.1.md > Bug Fixes> TiDB)- Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) - (dup: release-5.2.4.md > Bug fixes> TiDB)- Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) + - Fix the issue that query result might be wrong when Plan Cache is started in the RC isolation level [#34447](https://github.com/pingcap/tidb/issues/34447) + - Fix the session panic that occurs when executing the prepared statement after table schema change with the MySQL binary protocol [#33509](https://github.com/pingcap/tidb/issues/33509) + - Fix the issue that the table attributes are not indexed when a new partition is added and the issue that the table range information is not updated when the partition changes [#33929](https://github.com/pingcap/tidb/issues/33929) + - Fix the issue that the TiDB server might run out of memory when the `INFORMATION_SCHEMA.CLUSTER_SLOW_QUERY` table is queried. This issue can be triggered when you check slow queries on the Grafana dashboard [#33893](https://github.com/pingcap/tidb/issues/33893) + - Fix the issue that some DDL statements might be stuck for a period after the PD node of a cluster is replaced [#33908](https://github.com/pingcap/tidb/issues/33908) + - Fix the issue that granting the `all` privilege might fail in clusters that are upgraded from v4.0 [#33588](https://github.com/pingcap/tidb/issues/33588) + - Fix wrong results of deleting data of multiple tables using `left join` [#31321](https://github.com/pingcap/tidb/issues/31321) + - Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) + - Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) - Fix the panic issue caused by the `fatal error: concurrent map read and map write` error [#35340](https://github.com/pingcap/tidb/issues/35340) + - Fix the issue that the `REPLACE` statement incorrectly changes other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) + TiKV @@ -68,96 +61,96 @@ TiDB version: 5.3.2 - Fix the issue of time parsing error that occurs when the `DATETIME` values contain a fraction and `Z` [#12739](https://github.com/tikv/tikv/issues/12739) - Fix the issue that TiKV panics when performing type conversion for an empty string [#12673](https://github.com/tikv/tikv/issues/12673) - Fix the possible duplicate commit records in pessimistic transactions when async commit is enabled [#12615](https://github.com/tikv/tikv/issues/12615) - (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) - (dup: release-6.1.0.md > Bug fixes> TiKV)- Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue that TiKV panics and destroys peers unexpectedly when the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that stale messages cause TiKV to panic [#12023](https://github.com/tikv/tikv/issues/12023) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix the potential panic issue that occurs when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) - (dup: release-5.2.4.md > Bug fixes> TiKV)- Fix the issue that tikv-ctl returns an incorrect result due to its wrong string match [#12329](https://github.com/tikv/tikv/issues/12329) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that replica reads might violate the linearizability [#12109](https://github.com/tikv/tikv/issues/12109) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- 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](https://github.com/tikv/tikv/issues/12048) - (dup: release-5.4.1.md > Bug Fixes> TiKV)- Fix a bug that TiKV might panic if it has been running for 2 years or more [#11940](https://github.com/tikv/tikv/issues/11940) + - Fix the bug that TiKV reports the `invalid store ID 0` error when using Follower Read [#12478](https://github.com/tikv/tikv/issues/12478) + - Fix the issue of TiKV panic caused by the race between destroying peers and batch splitting Regions [#12368](https://github.com/tikv/tikv/issues/12368) + - Fix the issue that successfully committed optimistic transactions may report the `Write Conflict` error when the network is poor [#34066](https://github.com/pingcap/tidb/issues/34066) + - Fix the issue that TiKV panics and destroys peers unexpectedly when the target Region to be merged is invalid [#12232](https://github.com/tikv/tikv/issues/12232) + - Fix a bug that stale messages cause TiKV to panic [#12023](https://github.com/tikv/tikv/issues/12023) + - Fix the issue of intermittent packet loss and out of memory (OOM) caused by the overflow of memory metrics [#12160](https://github.com/tikv/tikv/issues/12160) + - Fix the potential panic issue that occurs when TiKV performs profiling on Ubuntu 18.04 [#9765](https://github.com/tikv/tikv/issues/9765) + - Fix the issue that tikv-ctl returns an incorrect result due to its wrong string match [#12329](https://github.com/tikv/tikv/issues/12329) + - Fix a bug that replica reads might violate the linearizability [#12109](https://github.com/tikv/tikv/issues/12109) + - 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](https://github.com/tikv/tikv/issues/12048) + - Fix a bug that TiKV might panic if it has been running for 2 years or more [#11940](https://github.com/tikv/tikv/issues/11940) + PD - Fix the PD panic that occurs when a hot region has no leader [#5005](https://github.com/tikv/pd/issues/5005) - (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that scheduling cannot start immediately after the PD leader transfer [#4769](https://github.com/tikv/pd/issues/4769) - (dup: release-6.1.0.md > Bug fixes> PD)- Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) - (dup: release-6.1.0.md > Bug fixes> PD)- Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) - (dup: release-5.4.1.md > Bug Fixes> PD)- 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](https://github.com/tikv/pd/issues/4805) + - Fix the issue that scheduling cannot start immediately after the PD leader transfer [#4769](https://github.com/tikv/pd/issues/4769) + - Fix the issue that a removed tombstone store appears again after the PD leader transfer ​​[#4941](https://github.com/tikv/pd/issues/4941) + - Fix a bug of TSO fallback in some corner cases [#4884](https://github.com/tikv/pd/issues/4884) + - 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](https://github.com/tikv/pd/issues/4805) - Fix the issue that schedulers do not work when `SchedulerMaxWaitingOperator` is set to `1` [#4946](https://github.com/tikv/pd/issues/4946) - (dup: release-5.4.1.md > Bug Fixes> PD)- Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) + - Fix the issue that the label distribution has residual labels in the metrics [#4825](https://github.com/tikv/pd/issues/4825) + TiFlash - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) + - Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) - Fix `TiFlash_schema_error` reported when `NOT NULL` columns are added [#4596](https://github.com/pingcap/tiflash/issues/4596) - Fix repeated crashes caused by the `commit state jump backward` errors [#2576](https://github.com/pingcap/tiflash/issues/2576) - (dup: release-6.1.0.md > Bug fixes> TiFlash)- Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) + - Fix potential data inconsistency after a lot of INSERT and DELETE operations [#4956](https://github.com/pingcap/tiflash/issues/4956) + - Fix a bug that canceled MPP queries might cause tasks to hang forever when the local tunnel is enabled [#4229](https://github.com/pingcap/tiflash/issues/4229) - Fix false reports of inconsistent TiFlash versions when TiFlash uses remote read [#3713](https://github.com/pingcap/tiflash/issues/3713) - Fix a bug that an MPP query might fail due to random gRPC keepalive timeout [#4662](https://github.com/pingcap/tiflash/issues/4662) - Fix a bug that an MPP query might hang forever if there are retries in the exchange receiver [#3444](https://github.com/pingcap/tiflash/issues/3444) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) + - Fix the wrong result that occurs when casting `DATETIME` to `DECIMAL` [#4151](https://github.com/pingcap/tiflash/issues/4151) + - Fix the overflow that occurs when casting `FLOAT` to `DECIMAL` [#3998](https://github.com/pingcap/tiflash/issues/3998) - Fix the potential `index out of bounds` error if calling `json_length` with empty string [#2705](https://github.com/pingcap/tiflash/issues/2705) - Fix wrong decimal comparison results in corner cases [#4512](https://github.com/pingcap/tiflash/issues/4512) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix bug that MPP query may hang forever if query failed in join build stage [#4195](https://github.com/pingcap/tiflash/issues/4195) - (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix possible wrong results when a query contains the `where ` clause [#3447](https://github.com/pingcap/tiflash/issues/3447) - (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that the `CastStringAsReal` behavior is inconsistent in TiFlash and in TiDB or TiKV [#3475](https://github.com/pingcap/tiflash/issues/3475) + - Fix bug that MPP query may hang forever if query failed in join build stage [#4195](https://github.com/pingcap/tiflash/issues/4195) + - Fix possible wrong results when a query contains the `where ` clause [#3447](https://github.com/pingcap/tiflash/issues/3447) + - Fix the issue that the `CastStringAsReal` behavior is inconsistent in TiFlash and in TiDB or TiKV [#3475](https://github.com/pingcap/tiflash/issues/3475) - Fix incorrect `microsecond` when casting string to datetime [#3556](https://github.com/pingcap/tiflash/issues/3556) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that MPP tasks might leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix a bug that empty segments cannot be merged after GC [#4511](https://github.com/pingcap/tiflash/issues/4511) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the panic issue that occurs when TLS is enabled [#4196](https://github.com/pingcap/tiflash/issues/4196) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the issue that expired data is recycled slowly [#4146](https://github.com/pingcap/tiflash/issues/4146) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) - (dup: release-5.4.1.md > Bug Fixes> TiFlash)- Fix the potential query error after adding columns under heavy read workload [#3967](https://github.com/pingcap/tiflash/issues/3967) - (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the bug that the `STR_TO_DATE()` function incorrectly handles leading zeros when parsing microseconds [#3557](https://github.com/pingcap/tiflash/issues/3557) - (dup: release-5.2.4.md > Bug fixes> TiFlash)- Fix the issue that TiFlash might return the `EstablishMPPConnection` error after it is restarted [#3615](https://github.com/pingcap/tiflash/issues/3615) + - Fix potential errors when querying on a table with many delete operations [#4747](https://github.com/pingcap/tiflash/issues/4747) + - Fix a bug that TiFlash reports many "Keepalive watchdog fired" errors randomly [#4192](https://github.com/pingcap/tiflash/issues/4192) + - Fix a bug that data not matching any region range remains on a TiFlash node [#4414](https://github.com/pingcap/tiflash/issues/4414) + - Fix a bug that MPP tasks might leak threads forever [#4238](https://github.com/pingcap/tiflash/issues/4238) + - Fix a bug that empty segments cannot be merged after GC [#4511](https://github.com/pingcap/tiflash/issues/4511) + - Fix the panic issue that occurs when TLS is enabled [#4196](https://github.com/pingcap/tiflash/issues/4196) + - Fix the issue that expired data is recycled slowly [#4146](https://github.com/pingcap/tiflash/issues/4146) + - Fix the bug that invalid storage directory configurations lead to unexpected behaviors [#4093](https://github.com/pingcap/tiflash/issues/4093) + - Fix the bug that some exceptions are not handled properly [#4101](https://github.com/pingcap/tiflash/issues/4101) + - Fix the potential query error after adding columns under heavy read workload [#3967](https://github.com/pingcap/tiflash/issues/3967) + - Fix the bug that the `STR_TO_DATE()` function incorrectly handles leading zeros when parsing microseconds [#3557](https://github.com/pingcap/tiflash/issues/3557) + - Fix the issue that TiFlash might return the `EstablishMPPConnection` error after it is restarted [#3615](https://github.com/pingcap/tiflash/issues/3615) + Tools + Backup & Restore (BR) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix duplicate primary keys when inserting a record into a table after incremental restoration [#33596](https://github.com/pingcap/tidb/issues/33596) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that schedulers do not resume after BR or TiDB Lightning exits abnormally [#33546](https://github.com/pingcap/tidb/issues/33546) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix a bug that BR incremental restore returns errors mistakenly due to DDL jobs with empty query [#33322](https://github.com/pingcap/tidb/issues/33322) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) - (dup: release-6.0.0-dmr.md > Bug fixes> Tools> Backup & Restore (BR))- Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) - (dup: release-5.4.1.md > Bug Fixes> Tools> Backup & Restore (BR))- Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) + - Fix duplicate primary keys when inserting a record into a table after incremental restoration [#33596](https://github.com/pingcap/tidb/issues/33596) + - Fix the issue that schedulers do not resume after BR or TiDB Lightning exits abnormally [#33546](https://github.com/pingcap/tidb/issues/33546) + - Fix a bug that BR incremental restore returns errors mistakenly due to DDL jobs with empty query [#33322](https://github.com/pingcap/tidb/issues/33322) + - Fix the issue that BR does not retry enough times when Regions are not consistent during restoration [#33419](https://github.com/pingcap/tidb/issues/33419) + - Fix a bug that BR gets stuck when the restore operation meets some unrecoverable errors [#33200](https://github.com/pingcap/tidb/issues/33200) + - Fix the issue that BR fails to back up RawKV [#32607](https://github.com/pingcap/tidb/issues/32607) - Fix the issue that BR cannot handle S3 internal errors [#34350](https://github.com/pingcap/tidb/issues/34350) + TiCDC - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix incorrect metrics caused by owner changes [#4774](https://github.com/pingcap/tiflow/issues/4774) + - Fix incorrect metrics caused by owner changes [#4774](https://github.com/pingcap/tiflow/issues/4774) - Fix the bug that the redo log manager flushes logs before writing logs [#5486](https://github.com/pingcap/tiflow/issues/5486) - Fix the bug that the resolved ts moves too fast when some tables are not maintained by the redo writer [#5486](https://github.com/pingcap/tiflow/issues/5486) - Add the UUID suffix to the redo log file name to fix the issue that file name conflicts may cause data loss [#5486](https://github.com/pingcap/tiflow/issues/5486) - Fix the bug that MySQL Sink may save a wrong checkpointTs [#5107](https://github.com/pingcap/tiflow/issues/5107) - Fix the issue that TiCDC clusters may panic after upgrade [#5266](https://github.com/pingcap/tiflow/issues/5266) - (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix the issue that changefeed gets stuck when tables are repeatedly scheduled in the same node [#4464](https://github.com/pingcap/tiflow/issues/4464) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix the issue that TiCDC fails to start when the first PD set in `--pd` is not available after TLS is enabled [#4777](https://github.com/pingcap/tiflow/issues/4777) - (dup: release-5.2.4.md > Bug fixes> Tools> TiCDC)- Fix a bug that querying status through open API may be blocked when the PD node is abnormal [#4778](https://github.com/pingcap/tiflow/issues/4778) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a stability problem in workerpool used by Unified Sorter [#4447](https://github.com/pingcap/tiflow/issues/4447) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiCDC)- Fix a bug that sequence is incorrectly replicated in some cases [#4563](https://github.com/pingcap/tiflow/issues/4552) + - Fix the issue that changefeed gets stuck when tables are repeatedly scheduled in the same node [#4464](https://github.com/pingcap/tiflow/issues/4464) + - Fix the issue that TiCDC fails to start when the first PD set in `--pd` is not available after TLS is enabled [#4777](https://github.com/pingcap/tiflow/issues/4777) + - Fix a bug that querying status through open API may be blocked when the PD node is abnormal [#4778](https://github.com/pingcap/tiflow/issues/4778) + - Fix a stability problem in workerpool used by Unified Sorter [#4447](https://github.com/pingcap/tiflow/issues/4447) + - Fix a bug that sequence is incorrectly replicated in some cases [#4563](https://github.com/pingcap/tiflow/issues/4552) + TiDB Data Migration (DM) - (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix the issue that DM occupies more disk space after the task automatically resumes [#3734](https://github.com/pingcap/tiflow/issues/3734) [#5344](https://github.com/pingcap/tiflow/issues/5344) - (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Data Migration (DM))- Fix an issue that the uppercase table cannot be replicated when `case-sensitive: true` is not set [#5255](https://github.com/pingcap/tiflow/issues/5255) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that in some cases manually executing the filtered DDL in the downstream might cause task resumption failure [#5272](https://github.com/pingcap/tiflow/issues/5272) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Data Migration (DM))- Fix the issue that CPU usage may increase and a large amount of log is printed when GTID is enabled or when the task is automatically resumed [#5063](https://github.com/pingcap/tiflow/issues/5063) + - Fix the issue that DM occupies more disk space after the task automatically resumes [#3734](https://github.com/pingcap/tiflow/issues/3734) [#5344](https://github.com/pingcap/tiflow/issues/5344) + - Fix an issue that the uppercase table cannot be replicated when `case-sensitive: true` is not set [#5255](https://github.com/pingcap/tiflow/issues/5255) + - Fix the issue that in some cases manually executing the filtered DDL in the downstream might cause task resumption failure [#5272](https://github.com/pingcap/tiflow/issues/5272) + - Fix the DM worker panic issue that occurs when the primary key is not first in the index returned by the `SHOW CREATE TABLE` statement [#5159](https://github.com/pingcap/tiflow/issues/5159) + - Fix the issue that CPU usage may increase and a large amount of log is printed when GTID is enabled or when the task is automatically resumed [#5063](https://github.com/pingcap/tiflow/issues/5063) - Fix the issue that the relay log may be disabled after the DM-master reboots [#4803](https://github.com/pingcap/tiflow/issues/4803) + TiDB Lightning - (dup: release-6.1.0.md > Bug fixes> Tools> TiDB Lightning)- Fix the issue of Local-backend import failure caused by out-of-bounds data in the `auto_increment` column [#27937](https://github.com/pingcap/tidb/issues/27937) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the issue that the precheck does not check local disk resources and cluster availability [#34213](https://github.com/pingcap/tidb/issues/34213) - (dup: release-5.4.1.md > Bug Fixes> Tools> TiDB Lightning)- Fix the checksum error "GC life time is shorter than transaction duration" [#32733](https://github.com/pingcap/tidb/issues/32733) + - Fix the issue of Local-backend import failure caused by out-of-bounds data in the `auto_increment` column [#27937](https://github.com/pingcap/tidb/issues/27937) + - Fix the issue that the precheck does not check local disk resources and cluster availability [#34213](https://github.com/pingcap/tidb/issues/34213) + - Fix the checksum error "GC life time is shorter than transaction duration" [#32733](https://github.com/pingcap/tidb/issues/32733) From 16dfe0bbcc63f83f909a7b63055c9809facc2e60 Mon Sep 17 00:00:00 2001 From: Ran Date: Wed, 29 Jun 2022 10:05:31 +0800 Subject: [PATCH 37/38] remove unnecessary tikv note --- releases/release-5.3.2.md | 1 - 1 file changed, 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index a2b14b2519f8..96c1f6aad6cd 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -21,7 +21,6 @@ TiDB version: 5.3.2 - Reduce the system call by the Raft client and increase CPU efficiency [#11309](https://github.com/tikv/tikv/issues/11309) - Improve the health check to detect unavailable Raftstore, so that the TiKV client can update Region Cache in time [#12398](https://github.com/tikv/tikv/issues/12398) - Transfer the leadership to CDC observer to reduce latency jitter [#12111](https://github.com/tikv/tikv/issues/12111) - (dup: release-5.3.0.md > Improvements> TiKV> Enhance disk space protection to improve storage stability)- Reduce the write latency by separating I/O operations from Raftstore thread pool (disabled by default). For more information about tuning, see [Tune TiKV Thread Pool Performance](/tune-tikv-thread-performance.md) [#10540](https://github.com/tikv/tikv/issues/10540) - Add more metrics for the garbage collection module of Raft logs to locate performance problems in the module [#11374](https://github.com/tikv/tikv/issues/11374) + Tools From 7d1cad753f8234eb2f397197979468b0a3c82d3e Mon Sep 17 00:00:00 2001 From: Ran Date: Wed, 29 Jun 2022 10:13:38 +0800 Subject: [PATCH 38/38] add compatibility change notes Signed-off-by: Ran --- releases/release-5.3.2.md | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/releases/release-5.3.2.md b/releases/release-5.3.2.md index 96c1f6aad6cd..290b2759d310 100644 --- a/releases/release-5.3.2.md +++ b/releases/release-5.3.2.md @@ -10,6 +10,10 @@ TiDB version: 5.3.2 ## Compatibility Changes ++ TiDB + + - Fix the issue that the `REPLACE` statement incorrectly changes other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) + + PD - Disable compiling swagger server by default [#4932](https://github.com/tikv/pd/issues/4932) @@ -52,7 +56,6 @@ TiDB version: 5.3.2 - Fix a bug that TiDB may dispatch duplicate tasks to TiFlash [#32814](https://github.com/pingcap/tidb/issues/32814) - Fix the issue that the background HTTP service of TiDB might not exit successfully and makes the cluster in an abnormal state [#30571](https://github.com/pingcap/tidb/issues/30571) - Fix the panic issue caused by the `fatal error: concurrent map read and map write` error [#35340](https://github.com/pingcap/tidb/issues/35340) - - Fix the issue that the `REPLACE` statement incorrectly changes other rows when the auto ID is out of range [#29483](https://github.com/pingcap/tidb/issues/29483) + TiKV