From c8795986279429c62d1984439adbcd0812384f50 Mon Sep 17 00:00:00 2001 From: qiancai <“qqzczy@126.com”> Date: Mon, 15 Mar 2021 15:02:15 +0800 Subject: [PATCH 01/35] correct-the-title-level Updated the title level of "Concentrated reads or writes" from the 2nd to the 3rd. --- dashboard/dashboard-key-visualizer.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dashboard/dashboard-key-visualizer.md b/dashboard/dashboard-key-visualizer.md index 9d4044dc4af17..db7c1984187c3 100644 --- a/dashboard/dashboard-key-visualizer.md +++ b/dashboard/dashboard-key-visualizer.md @@ -156,7 +156,7 @@ In the heatmap above, bright and dark colors are a fine-grained mix. This indica In the heatmap above, there is an alternating brightness and darkness along the X-axis (time) but the brightness is relatively even along the Y-axis (Region). This indicates that the reads and writes change periodically, which might occur in scenarios of periodically scheduled tasks. For example, the big data platform periodically extracts data from TiDB every day. In this kind of scenarios, pay attention to whether the resources are sufficient during peak usage. -## Concentrated reads or writes +### Concentrated reads or writes ![Concentrated](/media/dashboard/dashboard-keyviz-continue.png) From 656259a2523782759293b1fa80ac9ae6a9d2fe08 Mon Sep 17 00:00:00 2001 From: qiancai <“qqzczy@126.com”> Date: Mon, 15 Mar 2021 15:04:58 +0800 Subject: [PATCH 02/35] Revert "correct-the-title-level" This reverts commit c8795986279429c62d1984439adbcd0812384f50. --- dashboard/dashboard-key-visualizer.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dashboard/dashboard-key-visualizer.md b/dashboard/dashboard-key-visualizer.md index db7c1984187c3..9d4044dc4af17 100644 --- a/dashboard/dashboard-key-visualizer.md +++ b/dashboard/dashboard-key-visualizer.md @@ -156,7 +156,7 @@ In the heatmap above, bright and dark colors are a fine-grained mix. This indica In the heatmap above, there is an alternating brightness and darkness along the X-axis (time) but the brightness is relatively even along the Y-axis (Region). This indicates that the reads and writes change periodically, which might occur in scenarios of periodically scheduled tasks. For example, the big data platform periodically extracts data from TiDB every day. In this kind of scenarios, pay attention to whether the resources are sufficient during peak usage. -### Concentrated reads or writes +## Concentrated reads or writes ![Concentrated](/media/dashboard/dashboard-keyviz-continue.png) From b0db8e6602eb08cd1106fa053d4d012209db5296 Mon Sep 17 00:00:00 2001 From: qiancai Date: Mon, 19 Apr 2021 12:21:01 +0800 Subject: [PATCH 03/35] TiUP: Fix broken links and errors --- tiup/tiup-component-cluster-start.md | 6 +++--- tiup/tiup-component-cluster-stop.md | 10 +++++----- tiup/tiup-component-dm-display.md | 6 +++--- tiup/tiup-component-dm-upgrade.md | 2 +- 4 files changed, 12 insertions(+), 12 deletions(-) diff --git a/tiup/tiup-component-cluster-start.md b/tiup/tiup-component-cluster-start.md index de41844f773f9..60c2299438af6 100644 --- a/tiup/tiup-component-cluster-start.md +++ b/tiup/tiup-component-cluster-start.md @@ -12,7 +12,7 @@ The `tiup cluster start` command is used to start all services or some services tiup cluster start [flags] ``` -`` is the name of the cluster to operate on. If you forget the cluster name, check it using the [cluster list](/tiup/tiup-component-cluster-list.md) command. +`` is the name of the cluster to operate on. If you forget the cluster name, you can check it using the [`tiup cluster list`](/tiup/tiup-component-cluster-list.md) command. ## Options @@ -24,7 +24,7 @@ tiup cluster start [flags] > **Note:** > -> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. +> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are started. ### -R, --role @@ -34,7 +34,7 @@ tiup cluster start [flags] > **Note:** > -> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. +> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are started. ### -h, --help diff --git a/tiup/tiup-component-cluster-stop.md b/tiup/tiup-component-cluster-stop.md index a7bbe4510114a..c208b82feddde 100644 --- a/tiup/tiup-component-cluster-stop.md +++ b/tiup/tiup-component-cluster-stop.md @@ -16,29 +16,29 @@ The `tiup cluster stop` command is used to stop all services or some services of tiup cluster stop [flags] ``` -`` is the name of the cluster to operate on. If you forget the cluster name, check it with the [cluster list](/tiup/tiup-component-cluster-list.md) command. +`` is the name of the cluster to operate on. If you forget the cluster name, you can check it using the [`tiup cluster list`](/tiup/tiup-component-cluster-list.md) command. ## Options ### -N, --node -- Specifies the nodes to be stoped. The value of this option is a comma-separated list of node IDs. You can get the node IDs from the first column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. +- Specifies the nodes to be stopped. The value of this option is a comma-separated list of node IDs. You can get the node IDs from the first column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. - Data type: `STRING` - Default: []. If this option is not specified in the command, the command stops all the nodes. > **Note:** > -> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. +> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are stopped. ### -R, --role -- Specifies the roles of nodes to be stoped. The value of this option is a comma-separated list of the roles of the nodes. You can get the roles of the nodes from the second column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. +- Specifies the roles of nodes to be stopped. The value of this option is a comma-separated list of the roles of the nodes. You can get the roles of the nodes from the second column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. - Data type: `STRING` - Default: []. If this option is not specified in the command, the command stops all the roles. > **Note:** > -> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. +> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are stopped. ### -h, --help diff --git a/tiup/tiup-component-dm-display.md b/tiup/tiup-component-dm-display.md index 1945ae76a8476..dbc6e403a2595 100644 --- a/tiup/tiup-component-dm-display.md +++ b/tiup/tiup-component-dm-display.md @@ -12,7 +12,7 @@ If you want to check the operational status of each component in a DM cluster, i tiup dm display [flags] ``` -`` is the name of the cluster to be operated. If you forget the cluster name, you can use the `[tiup dm list](/tiup/tiup-component-dm-list.md)` command to check it. +`` is the name of the cluster to be operated. If you forget the cluster name, you can check it using the [`tiup dm list`](/tiup/tiup-component-dm-list.md) command. ## Options @@ -24,7 +24,7 @@ tiup dm display [flags] > **Note:** > -> If `-R, --role` is also specified, only the services in the intersection of the specified nodes and roles is queried. +> If `-R, --role` is also specified, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are queried. ### -R, --role @@ -34,7 +34,7 @@ tiup dm display [flags] > **Note:** > -> If `-N, --node` is also specified, only the services in the intersection of the specified nodes and roles is queried. +> If `-N, --node` is also specified, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are queried. ### -h, --help diff --git a/tiup/tiup-component-dm-upgrade.md b/tiup/tiup-component-dm-upgrade.md index 31bcb8aa2e90b..3540d619eb9eb 100644 --- a/tiup/tiup-component-dm-upgrade.md +++ b/tiup/tiup-component-dm-upgrade.md @@ -12,7 +12,7 @@ The `tiup dm upgrade` command is used to upgrade a specified cluster to a specif tiup dm upgrade [flags] ``` -- `` is the name of the cluster to be operated on. If you forget the cluster name, you can use the `[tiup dm list](/tiup/tiup-component-dm-list.md)` command to check it. +- `` is the name of the cluster to be operated on. If you forget the cluster name, you can check it using the [`tiup dm list`](/tiup/tiup-component-dm-list.md) command. - `` is the target version to be upgraded to. Currently, only upgrading to a later version is allowed, and upgrading to an earlier version is not allowed, which means the downgrade is not allowed. Upgrading to a nightly version is not allowed either. ## Options From 1e4666ccfeeeab05d203a78d3aa7b3d854e07d2d Mon Sep 17 00:00:00 2001 From: qiancai Date: Mon, 19 Apr 2021 12:22:22 +0800 Subject: [PATCH 04/35] Revert "TiUP: Fix broken links and errors" This reverts commit b0db8e6602eb08cd1106fa053d4d012209db5296. --- tiup/tiup-component-cluster-start.md | 6 +++--- tiup/tiup-component-cluster-stop.md | 10 +++++----- tiup/tiup-component-dm-display.md | 6 +++--- tiup/tiup-component-dm-upgrade.md | 2 +- 4 files changed, 12 insertions(+), 12 deletions(-) diff --git a/tiup/tiup-component-cluster-start.md b/tiup/tiup-component-cluster-start.md index 60c2299438af6..de41844f773f9 100644 --- a/tiup/tiup-component-cluster-start.md +++ b/tiup/tiup-component-cluster-start.md @@ -12,7 +12,7 @@ The `tiup cluster start` command is used to start all services or some services tiup cluster start [flags] ``` -`` is the name of the cluster to operate on. If you forget the cluster name, you can check it using the [`tiup cluster list`](/tiup/tiup-component-cluster-list.md) command. +`` is the name of the cluster to operate on. If you forget the cluster name, check it using the [cluster list](/tiup/tiup-component-cluster-list.md) command. ## Options @@ -24,7 +24,7 @@ tiup cluster start [flags] > **Note:** > -> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are started. +> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. ### -R, --role @@ -34,7 +34,7 @@ tiup cluster start [flags] > **Note:** > -> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are started. +> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. ### -h, --help diff --git a/tiup/tiup-component-cluster-stop.md b/tiup/tiup-component-cluster-stop.md index c208b82feddde..a7bbe4510114a 100644 --- a/tiup/tiup-component-cluster-stop.md +++ b/tiup/tiup-component-cluster-stop.md @@ -16,29 +16,29 @@ The `tiup cluster stop` command is used to stop all services or some services of tiup cluster stop [flags] ``` -`` is the name of the cluster to operate on. If you forget the cluster name, you can check it using the [`tiup cluster list`](/tiup/tiup-component-cluster-list.md) command. +`` is the name of the cluster to operate on. If you forget the cluster name, check it with the [cluster list](/tiup/tiup-component-cluster-list.md) command. ## Options ### -N, --node -- Specifies the nodes to be stopped. The value of this option is a comma-separated list of node IDs. You can get the node IDs from the first column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. +- Specifies the nodes to be stoped. The value of this option is a comma-separated list of node IDs. You can get the node IDs from the first column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. - Data type: `STRING` - Default: []. If this option is not specified in the command, the command stops all the nodes. > **Note:** > -> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are stopped. +> If the `-R, --role` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. ### -R, --role -- Specifies the roles of nodes to be stopped. The value of this option is a comma-separated list of the roles of the nodes. You can get the roles of the nodes from the second column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. +- Specifies the roles of nodes to be stoped. The value of this option is a comma-separated list of the roles of the nodes. You can get the roles of the nodes from the second column of the [cluster status table](/tiup/tiup-component-cluster-display.md) returned by the `tiup cluster display` command. - Data type: `STRING` - Default: []. If this option is not specified in the command, the command stops all the roles. > **Note:** > -> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are stopped. +> If the `-N, --node` option is specified at the same time, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are restarted. ### -h, --help diff --git a/tiup/tiup-component-dm-display.md b/tiup/tiup-component-dm-display.md index dbc6e403a2595..1945ae76a8476 100644 --- a/tiup/tiup-component-dm-display.md +++ b/tiup/tiup-component-dm-display.md @@ -12,7 +12,7 @@ If you want to check the operational status of each component in a DM cluster, i tiup dm display [flags] ``` -`` is the name of the cluster to be operated. If you forget the cluster name, you can check it using the [`tiup dm list`](/tiup/tiup-component-dm-list.md) command. +`` is the name of the cluster to be operated. If you forget the cluster name, you can use the `[tiup dm list](/tiup/tiup-component-dm-list.md)` command to check it. ## Options @@ -24,7 +24,7 @@ tiup dm display [flags] > **Note:** > -> If `-R, --role` is also specified, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are queried. +> If `-R, --role` is also specified, only the services in the intersection of the specified nodes and roles is queried. ### -R, --role @@ -34,7 +34,7 @@ tiup dm display [flags] > **Note:** > -> If `-N, --node` is also specified, only the service nodes that match both the specifications of `-N, --node` and `-R, --role` are queried. +> If `-N, --node` is also specified, only the services in the intersection of the specified nodes and roles is queried. ### -h, --help diff --git a/tiup/tiup-component-dm-upgrade.md b/tiup/tiup-component-dm-upgrade.md index 3540d619eb9eb..31bcb8aa2e90b 100644 --- a/tiup/tiup-component-dm-upgrade.md +++ b/tiup/tiup-component-dm-upgrade.md @@ -12,7 +12,7 @@ The `tiup dm upgrade` command is used to upgrade a specified cluster to a specif tiup dm upgrade [flags] ``` -- `` is the name of the cluster to be operated on. If you forget the cluster name, you can check it using the [`tiup dm list`](/tiup/tiup-component-dm-list.md) command. +- `` is the name of the cluster to be operated on. If you forget the cluster name, you can use the `[tiup dm list](/tiup/tiup-component-dm-list.md)` command to check it. - `` is the target version to be upgraded to. Currently, only upgrading to a later version is allowed, and upgrading to an earlier version is not allowed, which means the downgrade is not allowed. Upgrading to a nightly version is not allowed either. ## Options From 0449095575e8b99f6f3d9acd6892f62eda717823 Mon Sep 17 00:00:00 2001 From: ti-srebot <66930949+ti-srebot@users.noreply.github.com> Date: Wed, 20 Oct 2021 14:24:49 +0800 Subject: [PATCH 05/35] releases: add TiDB 5.2.2-14-24-47 release notes * releases: add TiDB 5.2.2-14-24-47 release notes * Update releases/release-5.2.2-14-24-47.md --- releases/release-5.2.2-14-24-47.md | 117 +++++++++++++++++++++++++++++ 1 file changed, 117 insertions(+) create mode 100644 releases/release-5.2.2-14-24-47.md diff --git a/releases/release-5.2.2-14-24-47.md b/releases/release-5.2.2-14-24-47.md new file mode 100644 index 0000000000000..d31945de47819 --- /dev/null +++ b/releases/release-5.2.2-14-24-47.md @@ -0,0 +1,117 @@ +--- +title: TiDB 5.2.2 Release Notes +category: Releases +--- + + + +# TiDB 5.2.2 Release Notes + +Release Date: October 20, 2021 + +TiDB version: 5.2.2 + +## __unsorted + ++ PingCAP/TiDB + + - Fix wrong result of hour function in vectorized expression [#28874](https://github.com/pingcap/tidb/pull/28874) + - Fixed a bug where MySQL 5.1 and older clients had issues authenticating [#28734](https://github.com/pingcap/tidb/pull/28734) + - fix auto analyze may get triggered out of specified time. [#28725](https://github.com/pingcap/tidb/pull/28725) + - Fix the bug that setting any session variable will make `tidb_snapshot` unwork. [#28683](https://github.com/pingcap/tidb/pull/28683) + - Fixed a bug that caused BR get stuck when many missing-peer regions in cluster. [#28680](https://github.com/pingcap/tidb/pull/28680) + - Fix unexpected error like `tidb_cast to Int32 is not supported` when unsupported cast is pushed down to TiFlash [#28654](https://github.com/pingcap/tidb/pull/28654) + - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#28439](https://github.com/pingcap/tidb/pull/28439) + - Fix bug that mpp node availability detect does not work in some corner cases [#28289](https://github.com/pingcap/tidb/pull/28289) + - sessionctx: fix data-race bug when alloc task id [#28283](https://github.com/pingcap/tidb/pull/28283) + - Fix index out of bound bug when empty dual table is remove for mpp query [#28280](https://github.com/pingcap/tidb/pull/28280) + - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) + - Fix `can not found column in Schema column` error for mpp queries [#28149](https://github.com/pingcap/tidb/pull/28149) + - Fix a bug that TiDB may crash when TiFlash is shutting down. [#28140](https://github.com/pingcap/tidb/pull/28140) + - None. [#27924](https://github.com/pingcap/tidb/pull/27924) + - Support for 3DES based TLS ciphersuites was removed [#27859](https://github.com/pingcap/tidb/pull/27859) + - Do not check multi-ingest when store is offline. [#27826](https://github.com/pingcap/tidb/pull/27826) + - Please add a release note, or a 'None' if it is not needed. [#27727](https://github.com/pingcap/tidb/pull/27727) + - Fix the issue that pre-check cost too much time when import too many files for tables. [#27623](https://github.com/pingcap/tidb/pull/27623) + - fix expression rewrite makes between expr infers wrong collation. [#27550](https://github.com/pingcap/tidb/pull/27550) + - make `group_concat` function consider the collation [#27530](https://github.com/pingcap/tidb/pull/27530) + - expression: fix extract bug when argument is a negative duration [#27366](https://github.com/pingcap/tidb/pull/27366) + - fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#27100](https://github.com/pingcap/tidb/pull/27100) + + ++ TiKV/TiKV + + - Fix frequent CDC incremental scan retry due to `Congest` error. [#11092](https://github.com/tikv/tikv/pull/11092) + - Simplify the algorithm of L0 flow control [#11081](https://github.com/tikv/tikv/pull/11081) + - fix channel full could break the raft connection [#11073](https://github.com/tikv/tikv/pull/11073) + - Fix the issue #9714. Now RaftClient will check the size of RaftMessage.extra_ctx and RaftMessage.message.context size as part of its message size estimate. [#11066](https://github.com/tikv/tikv/pull/11066) + - rename config resource-metering.agent_address to resource-metering.receiver_address [#11023](https://github.com/tikv/tikv/pull/11023) + - resolved_ts: fix coroutine leaking [#11020](https://github.com/tikv/tikv/pull/11020) + - Hide untouched storage commands' metrics in grafana dashboard [#11003](https://github.com/tikv/tikv/pull/11003) + - Fix panic in coprocessor when response size exceeds 4GiB [#10993](https://github.com/tikv/tikv/pull/10993) + - - improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) + - RaftStore Snapshot GC fix: fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10874](https://github.com/tikv/tikv/pull/10874) + - - TiKV coprocessor slow log will only consider time spent on processing the request. +- Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) + - Bug fix: fix an unexpected panic when exceeds deadline on processing copr requests. [#10857](https://github.com/tikv/tikv/pull/10857) + - None. [#10809](https://github.com/tikv/tikv/pull/10809) + + ++ PingCAP/TiFlash + + - Please add a release note, or a 'None' if it is not needed. [#2820](https://github.com/pingcap/tics/pull/2820) + + ++ PD + + - speed scheduler exit [#4199](https://github.com/tikv/pd/pull/4199) + - Fix the data race problem of hot region config [#4170](https://github.com/tikv/pd/pull/4170) + - allow empty region to be scheduled and use a sperate tolerance config in scatter range scheduler [#4118](https://github.com/tikv/pd/pull/4118) + - None. [#4028](https://github.com/tikv/pd/pull/4028) + + ++ Tools + + + PingCAP/TiCDC + + - change Kafka sink default `MaxMessageBytes` to 1MB. [#3104](https://github.com/pingcap/ticdc/pull/3104) + - fix the bug that fallback resolvedTs event will block the progress of resolve lock when occur region merging [#3102](https://github.com/pingcap/ticdc/pull/3102) + - Close gPRC stream and re-create it when meeting `ErrPrewriteNotMatch` to avoid duplicated request error [#3093](https://github.com/pingcap/ticdc/pull/3093) + - Fix a bug that will lead cdc owner to consume cpu hugely. [#3077](https://github.com/pingcap/ticdc/pull/3077) + - fix kafka sink can not send message due to constraint by `max-message-size` option. [#3049](https://github.com/pingcap/ticdc/pull/3049) + - Nond [#3045](https://github.com/pingcap/ticdc/pull/3045) + - Add metrics to observe incremental scan remaining time [#3035](https://github.com/pingcap/ticdc/pull/3035) + - Fix possible deadlocking when Kafka producer reports an error. [#3018](https://github.com/pingcap/ticdc/pull/3018) + - Please add a release note. +If you don't think this PR needs a release note then fill it with `None`. [#2944](https://github.com/pingcap/ticdc/pull/2944) + - Fix dml is not replicated after adding partition in partition table without valid index [#2866](https://github.com/pingcap/ticdc/pull/2866) + - `None` [#2861](https://github.com/pingcap/ticdc/pull/2861) + - Extend creating service gc safepoint ttl to 1 hr to support creating changefeeds that needs long initialization time. [#2854](https://github.com/pingcap/ticdc/pull/2854) + - Fix json encoding could panic when processing a string type value in some cases. [#2784](https://github.com/pingcap/ticdc/pull/2784) + + +## Bug Fixes + ++ PingCAP/TiDB + + - planner: fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28837](https://github.com/pingcap/tidb/pull/28837) + - 'None' [#28820](https://github.com/pingcap/tidb/pull/28820) + - planner: fix the issue that planner may cache invalid plans for joins in some cases [#28447](https://github.com/pingcap/tidb/pull/28447) + - Fix wrong index hash join when hash col is enum. [#28082](https://github.com/pingcap/tidb/pull/28082) + - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) + - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) + - Fix the wrong result for date_add and date_sub in some cases. [#27454](https://github.com/pingcap/tidb/pull/27454) + + ++ PingCAP/TiFlash + + - Fix the issue that TiFlash fails to start up under platform without library `nsl` [#3207](https://github.com/pingcap/tics/pull/3207) + + ++ PD + + - Fix the issue that PD may not elect leader as soon as leader step down [#4220](https://github.com/tikv/pd/pull/4220) + - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) + - Fix the bug that PD would not fix down-peer in time. [#4084](https://github.com/tikv/pd/pull/4084) + + From 1a93c7f0e1c224a7763ef61393cc3dbe88f8c16b Mon Sep 17 00:00:00 2001 From: qiancai Date: Wed, 20 Oct 2021 15:11:44 +0800 Subject: [PATCH 06/35] add_5.2.2_to_toc --- TOC.md | 1 + ...ase-5.2.2-14-24-47.md => release-5.2.2.md} | 77 +++++++++---------- releases/release-notes.md | 1 + 3 files changed, 38 insertions(+), 41 deletions(-) rename releases/{release-5.2.2-14-24-47.md => release-5.2.2.md} (91%) diff --git a/TOC.md b/TOC.md index 9308397b8707e..76ceb55911da3 100644 --- a/TOC.md +++ b/TOC.md @@ -558,6 +558,7 @@ + Release Notes + [All Releases](/releases/release-notes.md) + v5.2 + + [5.2.2](/releases/release-5.2.2.md) + [5.2.1](/releases/release-5.2.1.md) + [5.2.0](/releases/release-5.2.0.md) + v5.1 diff --git a/releases/release-5.2.2-14-24-47.md b/releases/release-5.2.2.md similarity index 91% rename from releases/release-5.2.2-14-24-47.md rename to releases/release-5.2.2.md index d31945de47819..64ba1ab535665 100644 --- a/releases/release-5.2.2-14-24-47.md +++ b/releases/release-5.2.2.md @@ -1,19 +1,46 @@ --- title: TiDB 5.2.2 Release Notes -category: Releases --- # TiDB 5.2.2 Release Notes -Release Date: October 20, 2021 +Release Date: October 27, 2021 TiDB version: 5.2.2 +## Compatibility change(s) + +## Feature enhancement(s) + +## Improvements + +## Bug Fixes + ++ TiDB + + - planner: fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28837](https://github.com/pingcap/tidb/pull/28837) + - 'None' [#28820](https://github.com/pingcap/tidb/pull/28820) + - planner: fix the issue that planner may cache invalid plans for joins in some cases [#28447](https://github.com/pingcap/tidb/pull/28447) + - Fix wrong index hash join when hash col is enum. [#28082](https://github.com/pingcap/tidb/pull/28082) + - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) + - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) + - Fix the wrong result for date_add and date_sub in some cases. [#27454](https://github.com/pingcap/tidb/pull/27454) + ++ TiFlash + + - Fix the issue that TiFlash fails to start up under platform without library `nsl` [#3207](https://github.com/pingcap/tics/pull/3207) + ++ PD + + - Fix the issue that PD may not elect leader as soon as leader step down [#4220](https://github.com/tikv/pd/pull/4220) + - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) + - Fix the bug that PD would not fix down-peer in time. [#4084](https://github.com/tikv/pd/pull/4084) + ## __unsorted -+ PingCAP/TiDB ++ TiDB - Fix wrong result of hour function in vectorized expression [#28874](https://github.com/pingcap/tidb/pull/28874) - Fixed a bug where MySQL 5.1 and older clients had issues authenticating [#28734](https://github.com/pingcap/tidb/pull/28734) @@ -38,8 +65,7 @@ TiDB version: 5.2.2 - expression: fix extract bug when argument is a negative duration [#27366](https://github.com/pingcap/tidb/pull/27366) - fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#27100](https://github.com/pingcap/tidb/pull/27100) - -+ TiKV/TiKV ++ TiKV - Fix frequent CDC incremental scan retry due to `Congest` error. [#11092](https://github.com/tikv/tikv/pull/11092) - Simplify the algorithm of L0 flow control [#11081](https://github.com/tikv/tikv/pull/11081) @@ -49,19 +75,17 @@ TiDB version: 5.2.2 - resolved_ts: fix coroutine leaking [#11020](https://github.com/tikv/tikv/pull/11020) - Hide untouched storage commands' metrics in grafana dashboard [#11003](https://github.com/tikv/tikv/pull/11003) - Fix panic in coprocessor when response size exceeds 4GiB [#10993](https://github.com/tikv/tikv/pull/10993) - - - improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) + - improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) - RaftStore Snapshot GC fix: fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10874](https://github.com/tikv/tikv/pull/10874) - - - TiKV coprocessor slow log will only consider time spent on processing the request. -- Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) + - TiKV coprocessor slow log will only consider time spent on processing the request. + - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) - Bug fix: fix an unexpected panic when exceeds deadline on processing copr requests. [#10857](https://github.com/tikv/tikv/pull/10857) - None. [#10809](https://github.com/tikv/tikv/pull/10809) - -+ PingCAP/TiFlash ++ TiFlash - Please add a release note, or a 'None' if it is not needed. [#2820](https://github.com/pingcap/tics/pull/2820) - + PD - speed scheduler exit [#4199](https://github.com/tikv/pd/pull/4199) @@ -69,7 +93,6 @@ TiDB version: 5.2.2 - allow empty region to be scheduled and use a sperate tolerance config in scatter range scheduler [#4118](https://github.com/tikv/pd/pull/4118) - None. [#4028](https://github.com/tikv/pd/pull/4028) - + Tools + PingCAP/TiCDC @@ -82,36 +105,8 @@ TiDB version: 5.2.2 - Nond [#3045](https://github.com/pingcap/ticdc/pull/3045) - Add metrics to observe incremental scan remaining time [#3035](https://github.com/pingcap/ticdc/pull/3035) - Fix possible deadlocking when Kafka producer reports an error. [#3018](https://github.com/pingcap/ticdc/pull/3018) - - Please add a release note. -If you don't think this PR needs a release note then fill it with `None`. [#2944](https://github.com/pingcap/ticdc/pull/2944) + - Please add a release note. If you don't think this PR needs a release note then fill it with `None`. [#2944](https://github.com/pingcap/ticdc/pull/2944) - Fix dml is not replicated after adding partition in partition table without valid index [#2866](https://github.com/pingcap/ticdc/pull/2866) - `None` [#2861](https://github.com/pingcap/ticdc/pull/2861) - Extend creating service gc safepoint ttl to 1 hr to support creating changefeeds that needs long initialization time. [#2854](https://github.com/pingcap/ticdc/pull/2854) - Fix json encoding could panic when processing a string type value in some cases. [#2784](https://github.com/pingcap/ticdc/pull/2784) - - -## Bug Fixes - -+ PingCAP/TiDB - - - planner: fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28837](https://github.com/pingcap/tidb/pull/28837) - - 'None' [#28820](https://github.com/pingcap/tidb/pull/28820) - - planner: fix the issue that planner may cache invalid plans for joins in some cases [#28447](https://github.com/pingcap/tidb/pull/28447) - - Fix wrong index hash join when hash col is enum. [#28082](https://github.com/pingcap/tidb/pull/28082) - - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) - - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) - - Fix the wrong result for date_add and date_sub in some cases. [#27454](https://github.com/pingcap/tidb/pull/27454) - - -+ PingCAP/TiFlash - - - Fix the issue that TiFlash fails to start up under platform without library `nsl` [#3207](https://github.com/pingcap/tics/pull/3207) - - -+ PD - - - Fix the issue that PD may not elect leader as soon as leader step down [#4220](https://github.com/tikv/pd/pull/4220) - - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) - - Fix the bug that PD would not fix down-peer in time. [#4084](https://github.com/tikv/pd/pull/4084) - - diff --git a/releases/release-notes.md b/releases/release-notes.md index b374e82c5989b..f3aac27bc69cd 100644 --- a/releases/release-notes.md +++ b/releases/release-notes.md @@ -7,6 +7,7 @@ aliases: ['/docs/dev/releases/release-notes/','/docs/dev/releases/rn/'] ## 5.2 +- [5.2.2](/releases/release-5.2.2.md) - [5.2.1](/releases/release-5.2.1.md) - [5.2.0](/releases/release-5.2.0.md) From 483af9e9b313b4484cc84fa6c56e795e67c15d8c Mon Sep 17 00:00:00 2001 From: qiancai Date: Wed, 20 Oct 2021 15:12:52 +0800 Subject: [PATCH 07/35] Update release-5.2.2.md --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 64ba1ab535665..103ac745639ac 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -95,7 +95,7 @@ TiDB version: 5.2.2 + Tools - + PingCAP/TiCDC + + TiCDC - change Kafka sink default `MaxMessageBytes` to 1MB. [#3104](https://github.com/pingcap/ticdc/pull/3104) - fix the bug that fallback resolvedTs event will block the progress of resolve lock when occur region merging [#3102](https://github.com/pingcap/ticdc/pull/3102) From 113bf56ad2fad95b249415e7d48a9c34abdcfc91 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Mon, 25 Oct 2021 11:44:54 +0800 Subject: [PATCH 08/35] Update releases/release-5.2.2.md Co-authored-by: Ryan Leung --- releases/release-5.2.2.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 103ac745639ac..2f1748a598874 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -34,6 +34,14 @@ TiDB version: 5.2.2 + PD + - Fix the issue that PD would not fix down peers in time [#4084](https://github.com/tikv/pd/pull/4084) + - Fix the data race issue when setting hot region configuration [#4170](https://github.com/tikv/pd/pull/4170) + - Fix the issue that key manager cost too much CPU [#4153](https://github.com/tikv/pd/pull/4153) + - Fix the issue that PD may remove the wrong pending peer if the peer count exceeds the limit [#4075](https://github.com/tikv/pd/pull/4075) + - Fix the issue that scatter range scheduler cannot schedule the empty region [#4118](https://github.com/tikv/pd/pull/4118) + - Fix the issue that PD may not elect a new leader as soon as the leader steps down [#4220](https://github.com/tikv/pd/pull/4220) + + - Fix the issue that PD may not elect leader as soon as leader step down [#4220](https://github.com/tikv/pd/pull/4220) - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) - Fix the bug that PD would not fix down-peer in time. [#4084](https://github.com/tikv/pd/pull/4084) From efaa42f34e9c32d5a6aa5d09aa6738a550a6f73c Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Mon, 25 Oct 2021 11:46:18 +0800 Subject: [PATCH 09/35] Update releases/release-5.2.2.md Co-authored-by: Ryan Leung --- releases/release-5.2.2.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 2f1748a598874..069c83768798a 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -16,6 +16,13 @@ TiDB version: 5.2.2 ## Improvements ++ PD + + - Add more write query kind in QPS dimension for hot region scheduler [#4028](https://github.com/tikv/pd/pull/4028) + - Dynamically adjust the retry limit to improve the performance of the balance region scheduler [#4046](https://github.com/tikv/pd/pull/4046) + - Support scheduling regions with unhealthy peers for the evict leader scheduler [#4132](https://github.com/tikv/pd/pull/4132) + - Speed up scheduler exit process when receiving the signal [#4199](https://github.com/tikv/pd/pull/4199) + ## Bug Fixes + TiDB From 83c29fe70608ff09afc0b2c09d0052f0a8914381 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Mon, 25 Oct 2021 11:46:54 +0800 Subject: [PATCH 10/35] Update releases/release-5.2.2.md --- releases/release-5.2.2.md | 7 ------- 1 file changed, 7 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 069c83768798a..bf03a8b743bdf 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -101,13 +101,6 @@ TiDB version: 5.2.2 - Please add a release note, or a 'None' if it is not needed. [#2820](https://github.com/pingcap/tics/pull/2820) -+ PD - - - speed scheduler exit [#4199](https://github.com/tikv/pd/pull/4199) - - Fix the data race problem of hot region config [#4170](https://github.com/tikv/pd/pull/4170) - - allow empty region to be scheduled and use a sperate tolerance config in scatter range scheduler [#4118](https://github.com/tikv/pd/pull/4118) - - None. [#4028](https://github.com/tikv/pd/pull/4028) - + Tools + TiCDC From 915c25fb9027b0774859f827068bd5b0fb836519 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Tue, 26 Oct 2021 09:05:17 +0800 Subject: [PATCH 11/35] Apply suggestions from code review Co-authored-by: Lynn --- releases/release-5.2.2.md | 25 +++++++++++++++---------- 1 file changed, 15 insertions(+), 10 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index bf03a8b743bdf..2c0e59276b459 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -27,9 +27,12 @@ TiDB version: 5.2.2 + TiDB - - planner: fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28837](https://github.com/pingcap/tidb/pull/28837) - - 'None' [#28820](https://github.com/pingcap/tidb/pull/28820) - - planner: fix the issue that planner may cache invalid plans for joins in some cases [#28447](https://github.com/pingcap/tidb/pull/28447) + - Fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28837](https://github.com/pingcap/tidb/pull/28837) + + - Fix the wrong partition pruning when some conditions are out of range [#28820](https://github.com/pingcap/tidb/pull/28820) + + - Fix the issue that planner may cache invalid plans for joins in some cases [#28447](https://github.com/pingcap/tidb/pull/28447) + - Fix wrong index hash join when hash col is enum. [#28082](https://github.com/pingcap/tidb/pull/28082) - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) @@ -65,20 +68,22 @@ TiDB version: 5.2.2 - Fix unexpected error like `tidb_cast to Int32 is not supported` when unsupported cast is pushed down to TiFlash [#28654](https://github.com/pingcap/tidb/pull/28654) - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#28439](https://github.com/pingcap/tidb/pull/28439) - Fix bug that mpp node availability detect does not work in some corner cases [#28289](https://github.com/pingcap/tidb/pull/28289) - - sessionctx: fix data-race bug when alloc task id [#28283](https://github.com/pingcap/tidb/pull/28283) + - Fix data-race bug when alloc MPP task ID [#28283](https://github.com/pingcap/tidb/pull/28283) + - Fix index out of bound bug when empty dual table is remove for mpp query [#28280](https://github.com/pingcap/tidb/pull/28280) - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) - Fix `can not found column in Schema column` error for mpp queries [#28149](https://github.com/pingcap/tidb/pull/28149) - Fix a bug that TiDB may crash when TiFlash is shutting down. [#28140](https://github.com/pingcap/tidb/pull/28140) - - None. [#27924](https://github.com/pingcap/tidb/pull/27924) + - Show information about SQL statement in the debug log about coprocessor encountering lock [#27924](https://github.com/pingcap/tidb/pull/27924) - Support for 3DES based TLS ciphersuites was removed [#27859](https://github.com/pingcap/tidb/pull/27859) - Do not check multi-ingest when store is offline. [#27826](https://github.com/pingcap/tidb/pull/27826) - - Please add a release note, or a 'None' if it is not needed. [#27727](https://github.com/pingcap/tidb/pull/27727) + - Add backup and restore size for TiDB SQL brie task . [#27727](https://github.com/pingcap/tidb/pull/27727) - Fix the issue that pre-check cost too much time when import too many files for tables. [#27623](https://github.com/pingcap/tidb/pull/27623) - - fix expression rewrite makes between expr infers wrong collation. [#27550](https://github.com/pingcap/tidb/pull/27550) - - make `group_concat` function consider the collation [#27530](https://github.com/pingcap/tidb/pull/27530) - - expression: fix extract bug when argument is a negative duration [#27366](https://github.com/pingcap/tidb/pull/27366) - - fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#27100](https://github.com/pingcap/tidb/pull/27100) + - Fix expression rewrite makes `between` infers wrong collation. [#27550](https://github.com/pingcap/tidb/pull/27550) + + - Make `group_concat` function consider the collation [#27530](https://github.com/pingcap/tidb/pull/27530) + - Fix extract bug when argument is a negative duration [#27366](https://github.com/pingcap/tidb/pull/27366) + - Fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#27100](https://github.com/pingcap/tidb/pull/27100) + TiKV From 97b51cbd288646aafd561f290a4acfc90e26f4e8 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Tue, 26 Oct 2021 09:11:47 +0800 Subject: [PATCH 12/35] Update releases/release-5.2.2.md Co-authored-by: birdstorm --- releases/release-5.2.2.md | 4 ---- 1 file changed, 4 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 2c0e59276b459..d77563d387b10 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -102,10 +102,6 @@ TiDB version: 5.2.2 - Bug fix: fix an unexpected panic when exceeds deadline on processing copr requests. [#10857](https://github.com/tikv/tikv/pull/10857) - None. [#10809](https://github.com/tikv/tikv/pull/10809) -+ TiFlash - - - Please add a release note, or a 'None' if it is not needed. [#2820](https://github.com/pingcap/tics/pull/2820) - + Tools + TiCDC From a6772e7a3c3534bb041ea11afb918db8ad2afe4d Mon Sep 17 00:00:00 2001 From: qiancai Date: Tue, 26 Oct 2021 09:29:08 +0800 Subject: [PATCH 13/35] implement input --- releases/release-5.2.2.md | 89 +++++++++++++++++---------------------- 1 file changed, 39 insertions(+), 50 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index d77563d387b10..ce30b453b174c 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -2,8 +2,6 @@ title: TiDB 5.2.2 Release Notes --- - - # TiDB 5.2.2 Release Notes Release Date: October 27, 2021 @@ -16,6 +14,11 @@ TiDB version: 5.2.2 ## Improvements ++ TiDB + + - Show information about SQL statement in the debug log about coprocessor encountering lock [#27924](https://github.com/pingcap/tidb/pull/27924) + - Add backup and restore size for TiDB SQL brie task . [#27727](https://github.com/pingcap/tidb/pull/27727) + + PD - Add more write query kind in QPS dimension for hot region scheduler [#4028](https://github.com/tikv/pd/pull/4028) @@ -23,43 +26,21 @@ TiDB version: 5.2.2 - Support scheduling regions with unhealthy peers for the evict leader scheduler [#4132](https://github.com/tikv/pd/pull/4132) - Speed up scheduler exit process when receiving the signal [#4199](https://github.com/tikv/pd/pull/4199) ++ Tools + + + TiCDC + ## Bug Fixes + TiDB - Fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28837](https://github.com/pingcap/tidb/pull/28837) - - Fix the wrong partition pruning when some conditions are out of range [#28820](https://github.com/pingcap/tidb/pull/28820) - - Fix the issue that planner may cache invalid plans for joins in some cases [#28447](https://github.com/pingcap/tidb/pull/28447) - - Fix wrong index hash join when hash col is enum. [#28082](https://github.com/pingcap/tidb/pull/28082) - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) - Fix the wrong result for date_add and date_sub in some cases. [#27454](https://github.com/pingcap/tidb/pull/27454) - -+ TiFlash - - - Fix the issue that TiFlash fails to start up under platform without library `nsl` [#3207](https://github.com/pingcap/tics/pull/3207) - -+ PD - - - Fix the issue that PD would not fix down peers in time [#4084](https://github.com/tikv/pd/pull/4084) - - Fix the data race issue when setting hot region configuration [#4170](https://github.com/tikv/pd/pull/4170) - - Fix the issue that key manager cost too much CPU [#4153](https://github.com/tikv/pd/pull/4153) - - Fix the issue that PD may remove the wrong pending peer if the peer count exceeds the limit [#4075](https://github.com/tikv/pd/pull/4075) - - Fix the issue that scatter range scheduler cannot schedule the empty region [#4118](https://github.com/tikv/pd/pull/4118) - - Fix the issue that PD may not elect a new leader as soon as the leader steps down [#4220](https://github.com/tikv/pd/pull/4220) - - - - Fix the issue that PD may not elect leader as soon as leader step down [#4220](https://github.com/tikv/pd/pull/4220) - - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) - - Fix the bug that PD would not fix down-peer in time. [#4084](https://github.com/tikv/pd/pull/4084) - -## __unsorted - -+ TiDB - - Fix wrong result of hour function in vectorized expression [#28874](https://github.com/pingcap/tidb/pull/28874) - Fixed a bug where MySQL 5.1 and older clients had issues authenticating [#28734](https://github.com/pingcap/tidb/pull/28734) - fix auto analyze may get triggered out of specified time. [#28725](https://github.com/pingcap/tidb/pull/28725) @@ -69,21 +50,47 @@ TiDB version: 5.2.2 - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#28439](https://github.com/pingcap/tidb/pull/28439) - Fix bug that mpp node availability detect does not work in some corner cases [#28289](https://github.com/pingcap/tidb/pull/28289) - Fix data-race bug when alloc MPP task ID [#28283](https://github.com/pingcap/tidb/pull/28283) - - Fix index out of bound bug when empty dual table is remove for mpp query [#28280](https://github.com/pingcap/tidb/pull/28280) - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) - Fix `can not found column in Schema column` error for mpp queries [#28149](https://github.com/pingcap/tidb/pull/28149) - Fix a bug that TiDB may crash when TiFlash is shutting down. [#28140](https://github.com/pingcap/tidb/pull/28140) - - Show information about SQL statement in the debug log about coprocessor encountering lock [#27924](https://github.com/pingcap/tidb/pull/27924) - Support for 3DES based TLS ciphersuites was removed [#27859](https://github.com/pingcap/tidb/pull/27859) - Do not check multi-ingest when store is offline. [#27826](https://github.com/pingcap/tidb/pull/27826) - - Add backup and restore size for TiDB SQL brie task . [#27727](https://github.com/pingcap/tidb/pull/27727) - Fix the issue that pre-check cost too much time when import too many files for tables. [#27623](https://github.com/pingcap/tidb/pull/27623) - Fix expression rewrite makes `between` infers wrong collation. [#27550](https://github.com/pingcap/tidb/pull/27550) - - Make `group_concat` function consider the collation [#27530](https://github.com/pingcap/tidb/pull/27530) - Fix extract bug when argument is a negative duration [#27366](https://github.com/pingcap/tidb/pull/27366) - Fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#27100](https://github.com/pingcap/tidb/pull/27100) + - Avoid expressions with side effects in column pruning and aggregation pushdown. [#27370](https://github.com/pingcap/tidb/pull/27370) + - Remove useless GRPC log in production.[#27239](https://github.com/pingcap/tidb/pull/27239) + - Limit the valid decimal length to fix precision-related issues.[#28649](https://github.com/pingcap/tidb/pull/28649) + - Fix the wrong way to check for overflow in `plus` expression.[#27419](https://github.com/pingcap/tidb/pull/27419) + - Fix `data too long` error when dumping stats from the table with new collation data.[#27302](https://github.com/pingcap/tidb/pull/27302) + - Fix retried transactions' statements are not included in TIDB_TRX.[#28670](https://github.com/pingcap/tidb/pull/28670) + ++ TiKV + ++ PD + + - Fix the issue that PD would not fix down peers in time [#4084](https://github.com/tikv/pd/pull/4084) + - Fix the data race issue when setting hot region configuration [#4170](https://github.com/tikv/pd/pull/4170) + - Fix the issue that key manager cost too much CPU [#4153](https://github.com/tikv/pd/pull/4153) + - Fix the issue that PD may remove the wrong pending peer if the peer count exceeds the limit [#4075](https://github.com/tikv/pd/pull/4075) + - Fix the issue that scatter range scheduler cannot schedule the empty region [#4118](https://github.com/tikv/pd/pull/4118) + - Fix the issue that PD may not elect a new leader as soon as the leader steps down [#4220](https://github.com/tikv/pd/pull/4220) + - Fix the issue that PD may not elect leader as soon as leader step down [#4220](https://github.com/tikv/pd/pull/4220) + - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) + - Fix the bug that PD would not fix down-peer in time. [#4084](https://github.com/tikv/pd/pull/4084) + ++ TiFlash + + - Fix the issue that TiFlash fails to start up under platform without library `nsl` [#3207](https://github.com/pingcap/tics/pull/3207) + ++ Tools + + + TiCDC + +## __unsorted + TiKV @@ -101,21 +108,3 @@ TiDB version: 5.2.2 - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) - Bug fix: fix an unexpected panic when exceeds deadline on processing copr requests. [#10857](https://github.com/tikv/tikv/pull/10857) - None. [#10809](https://github.com/tikv/tikv/pull/10809) - -+ Tools - - + TiCDC - - - change Kafka sink default `MaxMessageBytes` to 1MB. [#3104](https://github.com/pingcap/ticdc/pull/3104) - - fix the bug that fallback resolvedTs event will block the progress of resolve lock when occur region merging [#3102](https://github.com/pingcap/ticdc/pull/3102) - - Close gPRC stream and re-create it when meeting `ErrPrewriteNotMatch` to avoid duplicated request error [#3093](https://github.com/pingcap/ticdc/pull/3093) - - Fix a bug that will lead cdc owner to consume cpu hugely. [#3077](https://github.com/pingcap/ticdc/pull/3077) - - fix kafka sink can not send message due to constraint by `max-message-size` option. [#3049](https://github.com/pingcap/ticdc/pull/3049) - - Nond [#3045](https://github.com/pingcap/ticdc/pull/3045) - - Add metrics to observe incremental scan remaining time [#3035](https://github.com/pingcap/ticdc/pull/3035) - - Fix possible deadlocking when Kafka producer reports an error. [#3018](https://github.com/pingcap/ticdc/pull/3018) - - Please add a release note. If you don't think this PR needs a release note then fill it with `None`. [#2944](https://github.com/pingcap/ticdc/pull/2944) - - Fix dml is not replicated after adding partition in partition table without valid index [#2866](https://github.com/pingcap/ticdc/pull/2866) - - `None` [#2861](https://github.com/pingcap/ticdc/pull/2861) - - Extend creating service gc safepoint ttl to 1 hr to support creating changefeeds that needs long initialization time. [#2854](https://github.com/pingcap/ticdc/pull/2854) - - Fix json encoding could panic when processing a string type value in some cases. [#2784](https://github.com/pingcap/ticdc/pull/2784) From 259b66b614d889cbaa52472c72c62307a0ade215 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Thu, 28 Oct 2021 10:28:10 +0800 Subject: [PATCH 14/35] Update releases/release-5.2.2.md Co-authored-by: birdstorm --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index ce30b453b174c..9155568835528 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -84,7 +84,7 @@ TiDB version: 5.2.2 + TiFlash - - Fix the issue that TiFlash fails to start up under platform without library `nsl` [#3207](https://github.com/pingcap/tics/pull/3207) + - Fix the issue that TiFlash fails to start up under platform without library `nsl` + Tools From 260943abfc30e12e2bfdb8d42ef0609a53ba664e Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Thu, 28 Oct 2021 10:30:39 +0800 Subject: [PATCH 15/35] Update releases/release-5.2.2.md Co-authored-by: Ryan Leung --- releases/release-5.2.2.md | 1 + 1 file changed, 1 insertion(+) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 9155568835528..6875e0bd26fd6 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -23,6 +23,7 @@ TiDB version: 5.2.2 - Add more write query kind in QPS dimension for hot region scheduler [#4028](https://github.com/tikv/pd/pull/4028) - Dynamically adjust the retry limit to improve the performance of the balance region scheduler [#4046](https://github.com/tikv/pd/pull/4046) + - Update TiDB Dashboard to v2021.10.08.1 [#4070](https://github.com/tikv/pd/pull/4070) - Support scheduling regions with unhealthy peers for the evict leader scheduler [#4132](https://github.com/tikv/pd/pull/4132) - Speed up scheduler exit process when receiving the signal [#4199](https://github.com/tikv/pd/pull/4199) From 82577a423631f0b77c8fc9bbf5812fb722ef54fe Mon Sep 17 00:00:00 2001 From: qiancai Date: Thu, 28 Oct 2021 10:51:10 +0800 Subject: [PATCH 16/35] Update release-5.2.2.md --- releases/release-5.2.2.md | 222 +++++++++++++++++++------------------- 1 file changed, 111 insertions(+), 111 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 6875e0bd26fd6..98d0e2f31393d 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -1,111 +1,111 @@ ---- -title: TiDB 5.2.2 Release Notes ---- - -# TiDB 5.2.2 Release Notes - -Release Date: October 27, 2021 - -TiDB version: 5.2.2 - -## Compatibility change(s) - -## Feature enhancement(s) - -## Improvements - -+ TiDB - - - Show information about SQL statement in the debug log about coprocessor encountering lock [#27924](https://github.com/pingcap/tidb/pull/27924) - - Add backup and restore size for TiDB SQL brie task . [#27727](https://github.com/pingcap/tidb/pull/27727) - -+ PD - - - Add more write query kind in QPS dimension for hot region scheduler [#4028](https://github.com/tikv/pd/pull/4028) - - Dynamically adjust the retry limit to improve the performance of the balance region scheduler [#4046](https://github.com/tikv/pd/pull/4046) - - Update TiDB Dashboard to v2021.10.08.1 [#4070](https://github.com/tikv/pd/pull/4070) - - Support scheduling regions with unhealthy peers for the evict leader scheduler [#4132](https://github.com/tikv/pd/pull/4132) - - Speed up scheduler exit process when receiving the signal [#4199](https://github.com/tikv/pd/pull/4199) - -+ Tools - - + TiCDC - -## Bug Fixes - -+ TiDB - - - Fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28837](https://github.com/pingcap/tidb/pull/28837) - - Fix the wrong partition pruning when some conditions are out of range [#28820](https://github.com/pingcap/tidb/pull/28820) - - Fix the issue that planner may cache invalid plans for joins in some cases [#28447](https://github.com/pingcap/tidb/pull/28447) - - Fix wrong index hash join when hash col is enum. [#28082](https://github.com/pingcap/tidb/pull/28082) - - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) - - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) - - Fix the wrong result for date_add and date_sub in some cases. [#27454](https://github.com/pingcap/tidb/pull/27454) - - Fix wrong result of hour function in vectorized expression [#28874](https://github.com/pingcap/tidb/pull/28874) - - Fixed a bug where MySQL 5.1 and older clients had issues authenticating [#28734](https://github.com/pingcap/tidb/pull/28734) - - fix auto analyze may get triggered out of specified time. [#28725](https://github.com/pingcap/tidb/pull/28725) - - Fix the bug that setting any session variable will make `tidb_snapshot` unwork. [#28683](https://github.com/pingcap/tidb/pull/28683) - - Fixed a bug that caused BR get stuck when many missing-peer regions in cluster. [#28680](https://github.com/pingcap/tidb/pull/28680) - - Fix unexpected error like `tidb_cast to Int32 is not supported` when unsupported cast is pushed down to TiFlash [#28654](https://github.com/pingcap/tidb/pull/28654) - - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#28439](https://github.com/pingcap/tidb/pull/28439) - - Fix bug that mpp node availability detect does not work in some corner cases [#28289](https://github.com/pingcap/tidb/pull/28289) - - Fix data-race bug when alloc MPP task ID [#28283](https://github.com/pingcap/tidb/pull/28283) - - Fix index out of bound bug when empty dual table is remove for mpp query [#28280](https://github.com/pingcap/tidb/pull/28280) - - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) - - Fix `can not found column in Schema column` error for mpp queries [#28149](https://github.com/pingcap/tidb/pull/28149) - - Fix a bug that TiDB may crash when TiFlash is shutting down. [#28140](https://github.com/pingcap/tidb/pull/28140) - - Support for 3DES based TLS ciphersuites was removed [#27859](https://github.com/pingcap/tidb/pull/27859) - - Do not check multi-ingest when store is offline. [#27826](https://github.com/pingcap/tidb/pull/27826) - - Fix the issue that pre-check cost too much time when import too many files for tables. [#27623](https://github.com/pingcap/tidb/pull/27623) - - Fix expression rewrite makes `between` infers wrong collation. [#27550](https://github.com/pingcap/tidb/pull/27550) - - Make `group_concat` function consider the collation [#27530](https://github.com/pingcap/tidb/pull/27530) - - Fix extract bug when argument is a negative duration [#27366](https://github.com/pingcap/tidb/pull/27366) - - Fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#27100](https://github.com/pingcap/tidb/pull/27100) - - Avoid expressions with side effects in column pruning and aggregation pushdown. [#27370](https://github.com/pingcap/tidb/pull/27370) - - Remove useless GRPC log in production.[#27239](https://github.com/pingcap/tidb/pull/27239) - - Limit the valid decimal length to fix precision-related issues.[#28649](https://github.com/pingcap/tidb/pull/28649) - - Fix the wrong way to check for overflow in `plus` expression.[#27419](https://github.com/pingcap/tidb/pull/27419) - - Fix `data too long` error when dumping stats from the table with new collation data.[#27302](https://github.com/pingcap/tidb/pull/27302) - - Fix retried transactions' statements are not included in TIDB_TRX.[#28670](https://github.com/pingcap/tidb/pull/28670) - -+ TiKV - -+ PD - - - Fix the issue that PD would not fix down peers in time [#4084](https://github.com/tikv/pd/pull/4084) - - Fix the data race issue when setting hot region configuration [#4170](https://github.com/tikv/pd/pull/4170) - - Fix the issue that key manager cost too much CPU [#4153](https://github.com/tikv/pd/pull/4153) - - Fix the issue that PD may remove the wrong pending peer if the peer count exceeds the limit [#4075](https://github.com/tikv/pd/pull/4075) - - Fix the issue that scatter range scheduler cannot schedule the empty region [#4118](https://github.com/tikv/pd/pull/4118) - - Fix the issue that PD may not elect a new leader as soon as the leader steps down [#4220](https://github.com/tikv/pd/pull/4220) - - Fix the issue that PD may not elect leader as soon as leader step down [#4220](https://github.com/tikv/pd/pull/4220) - - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) - - Fix the bug that PD would not fix down-peer in time. [#4084](https://github.com/tikv/pd/pull/4084) - -+ TiFlash - - - Fix the issue that TiFlash fails to start up under platform without library `nsl` - -+ Tools - - + TiCDC - -## __unsorted - -+ TiKV - - - Fix frequent CDC incremental scan retry due to `Congest` error. [#11092](https://github.com/tikv/tikv/pull/11092) - - Simplify the algorithm of L0 flow control [#11081](https://github.com/tikv/tikv/pull/11081) - - fix channel full could break the raft connection [#11073](https://github.com/tikv/tikv/pull/11073) - - Fix the issue #9714. Now RaftClient will check the size of RaftMessage.extra_ctx and RaftMessage.message.context size as part of its message size estimate. [#11066](https://github.com/tikv/tikv/pull/11066) - - rename config resource-metering.agent_address to resource-metering.receiver_address [#11023](https://github.com/tikv/tikv/pull/11023) - - resolved_ts: fix coroutine leaking [#11020](https://github.com/tikv/tikv/pull/11020) - - Hide untouched storage commands' metrics in grafana dashboard [#11003](https://github.com/tikv/tikv/pull/11003) - - Fix panic in coprocessor when response size exceeds 4GiB [#10993](https://github.com/tikv/tikv/pull/10993) - - improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) - - RaftStore Snapshot GC fix: fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10874](https://github.com/tikv/tikv/pull/10874) - - TiKV coprocessor slow log will only consider time spent on processing the request. - - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) - - Bug fix: fix an unexpected panic when exceeds deadline on processing copr requests. [#10857](https://github.com/tikv/tikv/pull/10857) - - None. [#10809](https://github.com/tikv/tikv/pull/10809) +--- +title: TiDB 5.2.2 Release Notes +--- + +# TiDB 5.2.2 Release Notes + +Release Date: October 27, 2021 + +TiDB version: 5.2.2 + +## Compatibility change(s) + +## Feature enhancement(s) + +## Improvements + ++ TiDB + + - Show information about SQL statement in the debug log about coprocessor encountering lock [#27718](https://github.com/pingcap/tidb/issues/27718) + - Add backup and restore size for TiDB SQL brie task . [#27247](https://github.com/pingcap/tidb/issues/27247) + ++ PD + + - Add more write query kind in QPS dimension for hot region scheduler [#4028](https://github.com/tikv/pd/pull/4028) + - Dynamically adjust the retry limit to improve the performance of the balance region scheduler [#4046](https://github.com/tikv/pd/pull/4046) + - Update TiDB Dashboard to v2021.10.08.1 [#4070](https://github.com/tikv/pd/pull/4070) + - Support scheduling regions with unhealthy peers for the evict leader scheduler [#4132](https://github.com/tikv/pd/pull/4132) + - Speed up scheduler exit process when receiving the signal [#4199](https://github.com/tikv/pd/pull/4199) + ++ Tools + + + TiCDC + +## Bug Fixes + ++ TiDB + + - Fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28254](https://github.com/pingcap/tidb/issues/28254) + - Fix the wrong partition pruning when some conditions are out of range [#28233](https://github.com/pingcap/tidb/issues/28233) + - Fix the issue that planner may cache invalid plans for joins in some cases [#28087](https://github.com/pingcap/tidb/issues/28087) + - Fix wrong index hash join when hash col is enum. [#27893](https://github.com/pingcap/tidb/issues/27893) + - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) + - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) + - Fix the wrong result for date_add and date_sub in some cases. [#27232](https://github.com/pingcap/tidb/issues/27232) + - Fix wrong result of hour function in vectorized expression [#28643](https://github.com/pingcap/tidb/issues/28643) + - Fixed a bug where MySQL 5.1 and older clients had issues authenticating [#27855](https://github.com/pingcap/tidb/issues/27855) + - fix auto analyze may get triggered out of specified time. [#28698](https://github.com/pingcap/tidb/issues/28698) + - Fix the bug that setting any session variable will make `tidb_snapshot` unwork. [#28683](https://github.com/pingcap/tidb/pull/28683) + - Fixed a bug that caused BR get stuck when many missing-peer regions in cluster. [#27534](https://github.com/pingcap/tidb/issues/27534) + - Fix unexpected error like `tidb_cast to Int32 is not supported` when unsupported cast is pushed down to TiFlash [#23907](https://github.com/pingcap/tidb/issues/23907) + - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#27964](https://github.com/pingcap/tidb/issues/27964) + - Fix bug that mpp node availability detect does not work in some corner cases [#3118](https://github.com/pingcap/tics/issues/3118) + - Fix data-race bug when alloc MPP task ID [#27952](https://github.com/pingcap/tidb/issues/27952) + - Fix index out of bound bug when empty dual table is remove for mpp query [#28250](https://github.com/pingcap/tidb/issues/28250) + - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) + - Fix `can not found column in Schema column` error for mpp queries [#28149](https://github.com/pingcap/tidb/pull/28149) + - Fix a bug that TiDB may crash when TiFlash is shutting down. [#28096](https://github.com/pingcap/tidb/issues/28096) + - Support for 3DES based TLS ciphersuites was removed [#27859](https://github.com/pingcap/tidb/pull/27859) + - Do not check multi-ingest when store is offline. [#27826](https://github.com/pingcap/tidb/pull/27826) + - Fix the issue that pre-check cost too much time when import too many files for tables. [#27605](https://github.com/pingcap/tidb/issues/27605) + - Fix expression rewrite makes `between` infers wrong collation. [#27146](https://github.com/pingcap/tidb/issues/27146) + - Make `group_concat` function consider the collation [#27429](https://github.com/pingcap/tidb/issues/27429) + - Fix extract bug when argument is a negative duration [#27236](https://github.com/pingcap/tidb/issues/27236) + - Fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#26765](https://github.com/pingcap/tidb/issues/26765) + - Avoid expressions with side effects in column pruning and aggregation pushdown. [#27106](https://github.com/pingcap/tidb/issues/27106) + - Remove useless GRPC log in production.[#27239](https://github.com/pingcap/tidb/pull/27239) + - Limit the valid decimal length to fix precision-related issues.[#3091](https://github.com/pingcap/tics/issues/3091) + - Fix the wrong way to check for overflow in `plus` expression.[#26977](https://github.com/pingcap/tidb/issues/26977) + - Fix `data too long` error when dumping stats from the table with new collation data.[#27024](https://github.com/pingcap/tidb/issues/27024) + - Fix retried transactions' statements are not included in TIDB_TRX.[#28670](https://github.com/pingcap/tidb/pull/28670) + ++ TiKV + ++ PD + + - Fix the issue that PD would not fix down peers in time [#4077](https://github.com/tikv/pd/issues/4077) + - Fix the data race issue when setting hot region configuration [#4170](https://github.com/tikv/pd/pull/4170) + - Fix the issue that key manager cost too much CPU [#4071](https://github.com/tikv/pd/issues/4071) + - Fix the issue that PD may remove the wrong pending peer if the peer count exceeds the limit [#4045](https://github.com/tikv/pd/issues/4045) + - Fix the issue that scatter range scheduler cannot schedule the empty region [#4118](https://github.com/tikv/pd/pull/4118) + - Fix the issue that PD may not elect a new leader as soon as the leader steps down [#3936](https://github.com/tikv/pd/issues/3936) + - Fix the issue that PD may not elect leader as soon as leader step down [#3936](https://github.com/tikv/pd/issues/3936) + - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) + - Fix the bug that PD would not fix down-peer in time. [#4077](https://github.com/tikv/pd/issues/4077) + ++ TiFlash + + - Fix the issue that TiFlash fails to start up under platform without library `nsl` + ++ Tools + + + TiCDC + +## __unsorted + ++ TiKV + + - Fix frequent CDC incremental scan retry due to `Congest` error. [#11082](https://github.com/tikv/tikv/issues/11082) + - Simplify the algorithm of L0 flow control [#10879](https://github.com/tikv/tikv/issues/10879) + - fix channel full could break the raft connection [#11047](https://github.com/tikv/tikv/issues/11047) + - Fix the issue #9714. Now RaftClient will check the size of RaftMessage.extra_ctx and RaftMessage.message.context size as part of its message size estimate. [#9714](https://github.com/tikv/tikv/issues/9714) + - rename config resource-metering.agent_address to resource-metering.receiver_address [#11023](https://github.com/tikv/tikv/pull/11023) + - resolved_ts: fix coroutine leaking [#10965](https://github.com/tikv/tikv/issues/10965) + - Hide untouched storage commands' metrics in grafana dashboard [#11003](https://github.com/tikv/tikv/pull/11003) + - Fix panic in coprocessor when response size exceeds 4GiB [#9012](https://github.com/tikv/tikv/issues/9012) + - improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) + - RaftStore Snapshot GC fix: fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10813](https://github.com/tikv/tikv/issues/10813) + - TiKV coprocessor slow log will only consider time spent on processing the request. + - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) + - Bug fix: fix an unexpected panic when exceeds deadline on processing copr requests. [#10852](https://github.com/tikv/tikv/issues/10852) + - None. [#10507](https://github.com/tikv/tikv/issues/10507) From 76ee3e797dd0388c3addce12ce0b27a19d028b06 Mon Sep 17 00:00:00 2001 From: qiancai Date: Thu, 28 Oct 2021 11:00:03 +0800 Subject: [PATCH 17/35] add_tikv_notes --- releases/release-5.2.2.md | 34 ++++++++++++++++------------------ 1 file changed, 16 insertions(+), 18 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 98d0e2f31393d..888e9d4e992b8 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -19,6 +19,14 @@ TiDB version: 5.2.2 - Show information about SQL statement in the debug log about coprocessor encountering lock [#27718](https://github.com/pingcap/tidb/issues/27718) - Add backup and restore size for TiDB SQL brie task . [#27247](https://github.com/pingcap/tidb/issues/27247) ++ TiKV + + - Simplify the algorithm of L0 flow control [#11081](https://github.com/tikv/tikv/pull/11081) + - Improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) + - TiKV coprocessor slow log will only consider time spent on processing the request. [#10866](https://github.com/tikv/tikv/pull/10866) + - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) + - Add more write query kind. [#10809](https://github.com/tikv/tikv/pull/10809) + + PD - Add more write query kind in QPS dimension for hot region scheduler [#4028](https://github.com/tikv/pd/pull/4028) @@ -71,6 +79,14 @@ TiDB version: 5.2.2 + TiKV + - Fix frequent CDC incremental scan retry due to `Congest` error. [#11092](https://github.com/tikv/tikv/pull/11092) + - Fix channel full could break the raft connection [#11073](https://github.com/tikv/tikv/pull/11073) + - Fix the issue that batch messages can be too large in raft client implementation [#11066](https://github.com/tikv/tikv/pull/11066) + - Fix coroutine leaking in `resolved_ts`. [#11020](https://github.com/tikv/tikv/pull/11020) + - Fix panic in coprocessor when response size exceeds 4GiB [#10993](https://github.com/tikv/tikv/pull/10993) + - Fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10874](https://github.com/tikv/tikv/pull/10874) + - Fix an unexpected panic when exceeds deadline on processing copr requests. [#10857](https://github.com/tikv/tikv/pull/10857) + + PD - Fix the issue that PD would not fix down peers in time [#4077](https://github.com/tikv/pd/issues/4077) @@ -91,21 +107,3 @@ TiDB version: 5.2.2 + TiCDC -## __unsorted - -+ TiKV - - - Fix frequent CDC incremental scan retry due to `Congest` error. [#11082](https://github.com/tikv/tikv/issues/11082) - - Simplify the algorithm of L0 flow control [#10879](https://github.com/tikv/tikv/issues/10879) - - fix channel full could break the raft connection [#11047](https://github.com/tikv/tikv/issues/11047) - - Fix the issue #9714. Now RaftClient will check the size of RaftMessage.extra_ctx and RaftMessage.message.context size as part of its message size estimate. [#9714](https://github.com/tikv/tikv/issues/9714) - - rename config resource-metering.agent_address to resource-metering.receiver_address [#11023](https://github.com/tikv/tikv/pull/11023) - - resolved_ts: fix coroutine leaking [#10965](https://github.com/tikv/tikv/issues/10965) - - Hide untouched storage commands' metrics in grafana dashboard [#11003](https://github.com/tikv/tikv/pull/11003) - - Fix panic in coprocessor when response size exceeds 4GiB [#9012](https://github.com/tikv/tikv/issues/9012) - - improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) - - RaftStore Snapshot GC fix: fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10813](https://github.com/tikv/tikv/issues/10813) - - TiKV coprocessor slow log will only consider time spent on processing the request. - - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) - - Bug fix: fix an unexpected panic when exceeds deadline on processing copr requests. [#10852](https://github.com/tikv/tikv/issues/10852) - - None. [#10507](https://github.com/tikv/tikv/issues/10507) From eca04b4c817f78e10d23dae27d189f3f5c96e53a Mon Sep 17 00:00:00 2001 From: qiancai Date: Thu, 28 Oct 2021 11:02:38 +0800 Subject: [PATCH 18/35] Update release-5.2.2.md --- releases/release-5.2.2.md | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 888e9d4e992b8..14b7e23575f67 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -21,11 +21,11 @@ TiDB version: 5.2.2 + TiKV - - Simplify the algorithm of L0 flow control [#11081](https://github.com/tikv/tikv/pull/11081) + - Simplify the algorithm of L0 flow control [#10879](https://github.com/tikv/tikv/issues/10879) - Improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) - TiKV coprocessor slow log will only consider time spent on processing the request. [#10866](https://github.com/tikv/tikv/pull/10866) - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) - - Add more write query kind. [#10809](https://github.com/tikv/tikv/pull/10809) + - Add more write query kind. [#10507](https://github.com/tikv/tikv/issues/10507) + PD @@ -79,13 +79,13 @@ TiDB version: 5.2.2 + TiKV - - Fix frequent CDC incremental scan retry due to `Congest` error. [#11092](https://github.com/tikv/tikv/pull/11092) - - Fix channel full could break the raft connection [#11073](https://github.com/tikv/tikv/pull/11073) - - Fix the issue that batch messages can be too large in raft client implementation [#11066](https://github.com/tikv/tikv/pull/11066) - - Fix coroutine leaking in `resolved_ts`. [#11020](https://github.com/tikv/tikv/pull/11020) - - Fix panic in coprocessor when response size exceeds 4GiB [#10993](https://github.com/tikv/tikv/pull/10993) - - Fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10874](https://github.com/tikv/tikv/pull/10874) - - Fix an unexpected panic when exceeds deadline on processing copr requests. [#10857](https://github.com/tikv/tikv/pull/10857) + - Fix frequent CDC incremental scan retry due to `Congest` error. [#11082](https://github.com/tikv/tikv/issues/11082) + - Fix channel full could break the raft connection [#11047](https://github.com/tikv/tikv/issues/11047) + - Fix the issue that batch messages can be too large in raft client implementation [#9714](https://github.com/tikv/tikv/issues/9714) + - Fix coroutine leaking in `resolved_ts`. [#10965](https://github.com/tikv/tikv/issues/10965) + - Fix panic in coprocessor when response size exceeds 4GiB [#9012](https://github.com/tikv/tikv/issues/9012) + - Fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10813](https://github.com/tikv/tikv/issues/10813) + - Fix an unexpected panic when exceeds deadline on processing copr requests. [#10852](https://github.com/tikv/tikv/issues/10852) + PD From 97c2bbc61119a187f5473e6b5ff09542f7514a18 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Thu, 28 Oct 2021 19:28:02 +0800 Subject: [PATCH 19/35] Update releases/release-5.2.2.md MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: Daniël van Eeden --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 14b7e23575f67..13eedd328443e 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -17,7 +17,7 @@ TiDB version: 5.2.2 + TiDB - Show information about SQL statement in the debug log about coprocessor encountering lock [#27718](https://github.com/pingcap/tidb/issues/27718) - - Add backup and restore size for TiDB SQL brie task . [#27247](https://github.com/pingcap/tidb/issues/27247) + - Add backup and restore size for TiDB SQL backup/restore task . [#27247](https://github.com/pingcap/tidb/issues/27247) + TiKV From ea8e90e482dcda4091c4a12cc23d3c2e0e382f57 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Thu, 28 Oct 2021 19:33:49 +0800 Subject: [PATCH 20/35] Apply suggestions from code review MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: Daniël van Eeden --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 13eedd328443e..82df322bd7ae5 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -59,7 +59,7 @@ TiDB version: 5.2.2 - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#27964](https://github.com/pingcap/tidb/issues/27964) - Fix bug that mpp node availability detect does not work in some corner cases [#3118](https://github.com/pingcap/tics/issues/3118) - Fix data-race bug when alloc MPP task ID [#27952](https://github.com/pingcap/tidb/issues/27952) - - Fix index out of bound bug when empty dual table is remove for mpp query [#28250](https://github.com/pingcap/tidb/issues/28250) + - Fix index out of bound bug when empty dual table is remove for MPP query [#28250](https://github.com/pingcap/tidb/issues/28250) - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) - Fix `can not found column in Schema column` error for mpp queries [#28149](https://github.com/pingcap/tidb/pull/28149) - Fix a bug that TiDB may crash when TiFlash is shutting down. [#28096](https://github.com/pingcap/tidb/issues/28096) From f02568c083d32cae3b4a26f4a51f78e768ab466f Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Thu, 28 Oct 2021 20:24:32 +0800 Subject: [PATCH 21/35] Update releases/release-5.2.2.md MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-authored-by: Daniël van Eeden --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 82df322bd7ae5..f509ed437d067 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -57,7 +57,7 @@ TiDB version: 5.2.2 - Fixed a bug that caused BR get stuck when many missing-peer regions in cluster. [#27534](https://github.com/pingcap/tidb/issues/27534) - Fix unexpected error like `tidb_cast to Int32 is not supported` when unsupported cast is pushed down to TiFlash [#23907](https://github.com/pingcap/tidb/issues/23907) - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#27964](https://github.com/pingcap/tidb/issues/27964) - - Fix bug that mpp node availability detect does not work in some corner cases [#3118](https://github.com/pingcap/tics/issues/3118) + - Fix bug that MPP node availability detect does not work in some corner cases [#3118](https://github.com/pingcap/tics/issues/3118) - Fix data-race bug when alloc MPP task ID [#27952](https://github.com/pingcap/tidb/issues/27952) - Fix index out of bound bug when empty dual table is remove for MPP query [#28250](https://github.com/pingcap/tidb/issues/28250) - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) From 7f59bbc23d2776aff7bdb9b78948293d391fb18c Mon Sep 17 00:00:00 2001 From: Enwei Date: Thu, 28 Oct 2021 17:20:20 +0200 Subject: [PATCH 22/35] revise TiKV & PD Co-authored-by: Liuxiaozhen12 <82579298+Liuxiaozhen12@users.noreply.github.com> --- releases/release-5.2.2.md | 77 +++++++++++++++++++-------------------- 1 file changed, 37 insertions(+), 40 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index f509ed437d067..c8d1c1c0a6fc0 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -22,18 +22,18 @@ TiDB version: 5.2.2 + TiKV - Simplify the algorithm of L0 flow control [#10879](https://github.com/tikv/tikv/issues/10879) - - Improve raft client error log report [#10983](https://github.com/tikv/tikv/pull/10983) - - TiKV coprocessor slow log will only consider time spent on processing the request. [#10866](https://github.com/tikv/tikv/pull/10866) - - Drop log instead of blocking threads when slogger thread is overloaded and queue is filled up. [#10866](https://github.com/tikv/tikv/pull/10866) - - Add more write query kind. [#10507](https://github.com/tikv/tikv/issues/10507) + - Improve the error log report in raft client module [#10983](https://github.com/tikv/tikv/pull/10983) + - Make the slow log of TiKV coprocessor only consider the time spent on processing requests [#10841](https://github.com/tikv/tikv/issues/10841) + - Drop log instead of blocking threads when the slogger thread is overloaded and the queue is filled up [#10841](https://github.com/tikv/tikv/issues/10841) + - Add more statistics types of write queries (https://github.com/tikv/tikv/issues/10507) + PD - - Add more write query kind in QPS dimension for hot region scheduler [#4028](https://github.com/tikv/pd/pull/4028) - - Dynamically adjust the retry limit to improve the performance of the balance region scheduler [#4046](https://github.com/tikv/pd/pull/4046) + - Add more types of write queries to QPS dimensions in the hotspot scheduler [#3869](https://github.com/tikv/pd/issues/3869) + - Support dynamically adjusting the retry limit of the balance region scheduler to improve the performance of the scheduler [#3744 (https://github.com/tikv/pd/issues/3744) - Update TiDB Dashboard to v2021.10.08.1 [#4070](https://github.com/tikv/pd/pull/4070) - - Support scheduling regions with unhealthy peers for the evict leader scheduler [#4132](https://github.com/tikv/pd/pull/4132) - - Speed up scheduler exit process when receiving the signal [#4199](https://github.com/tikv/pd/pull/4199) + - Support that the evict leader scheduler can schedule regions with unhealthy peers [#4093 (https://github.com/tikv/pd/issues/4093) + - Speed up the exit process of a scheduler after receiving an end-of-process signal [#4146](https://github.com/tikv/pd/issues/4146) + Tools @@ -60,44 +60,41 @@ TiDB version: 5.2.2 - Fix bug that MPP node availability detect does not work in some corner cases [#3118](https://github.com/pingcap/tics/issues/3118) - Fix data-race bug when alloc MPP task ID [#27952](https://github.com/pingcap/tidb/issues/27952) - Fix index out of bound bug when empty dual table is remove for MPP query [#28250](https://github.com/pingcap/tidb/issues/28250) - - Avoid false positive error log about `invalid cop task execution summaries length` when running MPP query. [#28264](https://github.com/pingcap/tidb/pull/28264) - - Fix `can not found column in Schema column` error for mpp queries [#28149](https://github.com/pingcap/tidb/pull/28149) - - Fix a bug that TiDB may crash when TiFlash is shutting down. [#28096](https://github.com/pingcap/tidb/issues/28096) - - Support for 3DES based TLS ciphersuites was removed [#27859](https://github.com/pingcap/tidb/pull/27859) - - Do not check multi-ingest when store is offline. [#27826](https://github.com/pingcap/tidb/pull/27826) - - Fix the issue that pre-check cost too much time when import too many files for tables. [#27605](https://github.com/pingcap/tidb/issues/27605) - - Fix expression rewrite makes `between` infers wrong collation. [#27146](https://github.com/pingcap/tidb/issues/27146) - - Make `group_concat` function consider the collation [#27429](https://github.com/pingcap/tidb/issues/27429) - - Fix extract bug when argument is a negative duration [#27236](https://github.com/pingcap/tidb/issues/27236) - - Fix a bug that creates partition fail if `NO_UNSIGNED_SUBTRACTION` is set. [#26765](https://github.com/pingcap/tidb/issues/26765) - - Avoid expressions with side effects in column pruning and aggregation pushdown. [#27106](https://github.com/pingcap/tidb/issues/27106) - - Remove useless GRPC log in production.[#27239](https://github.com/pingcap/tidb/pull/27239) - - Limit the valid decimal length to fix precision-related issues.[#3091](https://github.com/pingcap/tics/issues/3091) - - Fix the wrong way to check for overflow in `plus` expression.[#26977](https://github.com/pingcap/tidb/issues/26977) - - Fix `data too long` error when dumping stats from the table with new collation data.[#27024](https://github.com/pingcap/tidb/issues/27024) - - Fix retried transactions' statements are not included in TIDB_TRX.[#28670](https://github.com/pingcap/tidb/pull/28670) + - Fix the issue of false positive error log `invalid cop task execution summaries length` for MPP queries [#1791](https://github.com/pingcap/tics/issues/1791) + - Fix the issue of error log `can not found column in Schema column` for MPP queries [#28149](https://github.com/pingcap/tidb/pull/28149) + - Fix the issue that TiDB might crash when TiFlash shuts down [#28096](https://github.com/pingcap/tidb/issues/28096) + - Remove the support for insecure 3DES (Triple Data Encryption Algorithm) based TLS ciphersuites [#27859](https://github.com/pingcap/tidb/pull/27859) + - Fix the issue that Lightning connects to offline TiKV nodes during pre-check and causes import failures [#27826](https://github.com/pingcap/tidb/pull/27826) + - Fix the issue that pre-check cost too much time when importing many files to tables [#27605](https://github.com/pingcap/tidb/issues/27605) + - Fix the issue that rewriting expressions makes `between` infer wrong collation [#27146](https://github.com/pingcap/tidb/issues/27146) + - Fix the issue that `group_concat` function did not consider the collation [#27429](https://github.com/pingcap/tidb/issues/27429) + - Fix the issue that `extract` function gives wrong results when argument is a negative duration [#27236](https://github.com/pingcap/tidb/issues/27236) + - Fix the issue that creating partition fails if `NO_UNSIGNED_SUBTRACTION` is set [#26765](https://github.com/pingcap/tidb/issues/26765) + - Avoid expressions with side effects in column pruning and aggregation pushdown [#27106](https://github.com/pingcap/tidb/issues/27106) + - Remove useless GRPC log in production [#24190](https://github.com/pingcap/tidb/issues/24190) + - Limit the valid decimal length to fix precision-related issues [#3091](https://github.com/pingcap/tics/issues/3091) + - Fix the issue of a wrong way to check for overflow in `plus` expression [#26977](https://github.com/pingcap/tidb/issues/26977) + - Fix the issue of `data too long` error when dumping statistics from the table with new collation data [#27024](https://github.com/pingcap/tidb/issues/27024) + - Fix the issue that the retried transactions' statements are not included in TIDB_TRX [#28670](https://github.com/pingcap/tidb/pull/28670) + TiKV - - Fix frequent CDC incremental scan retry due to `Congest` error. [#11082](https://github.com/tikv/tikv/issues/11082) - - Fix channel full could break the raft connection [#11047](https://github.com/tikv/tikv/issues/11047) - - Fix the issue that batch messages can be too large in raft client implementation [#9714](https://github.com/tikv/tikv/issues/9714) - - Fix coroutine leaking in `resolved_ts`. [#10965](https://github.com/tikv/tikv/issues/10965) - - Fix panic in coprocessor when response size exceeds 4GiB [#9012](https://github.com/tikv/tikv/issues/9012) - - Fix the issue that snapshot GC missed GC snapshot files when there's one snapshot file failed to be GC-ed. [#10813](https://github.com/tikv/tikv/issues/10813) - - Fix an unexpected panic when exceeds deadline on processing copr requests. [#10852](https://github.com/tikv/tikv/issues/10852) + - Fix the issue that CDC add scan retries frequently due to Congest error [#11082](https://github.com/tikv/tikv/issues/11082) + - Fix that the raft connection is broken when the channel is full [#11047](https://github.com/tikv/tikv/issues/11047) + - Fix the issue that batch messages are too large in Raft client implementation [#9714](https://github.com/tikv/tikv/issues/9714) + - Fix the issue that concurrent leaks in `resolved_ts` [#10965](https://github.com/tikv/tikv/issues/10965) + - Fix a panic issue that occurs to coprocessor when response size exceeds 4 GiB [#9012](https://github.com/tikv/tikv/issues/9012) + - Fix the issue that snapshot Garbage Collection (GC) misses GC snapshot files when snapshot files cannot be garbage collected [#10813](https://github.com/tikv/tikv/issues/10813) + - Fix a panic issue that occurs when processing coprocessor requests times out [#10852](https://github.com/tikv/tikv/issues/10852) + PD - - Fix the issue that PD would not fix down peers in time [#4077](https://github.com/tikv/pd/issues/4077) - - Fix the data race issue when setting hot region configuration [#4170](https://github.com/tikv/pd/pull/4170) - - Fix the issue that key manager cost too much CPU [#4071](https://github.com/tikv/pd/issues/4071) - - Fix the issue that PD may remove the wrong pending peer if the peer count exceeds the limit [#4045](https://github.com/tikv/pd/issues/4045) - - Fix the issue that scatter range scheduler cannot schedule the empty region [#4118](https://github.com/tikv/pd/pull/4118) - - Fix the issue that PD may not elect a new leader as soon as the leader steps down [#3936](https://github.com/tikv/pd/issues/3936) - - Fix the issue that PD may not elect leader as soon as leader step down [#3936](https://github.com/tikv/pd/issues/3936) - - `evict-leader-scheduler` supports schedule the regions with unhealthy peers. [#4132](https://github.com/tikv/pd/pull/4132) - - Fix the bug that PD would not fix down-peer in time. [#4077](https://github.com/tikv/pd/issues/4077) + - Fix the issue that PD incorrectly delete the peers with data and in pending status because the number of peers exceeds the number of configured peers [#4045](https://github.com/tikv/pd/issues/4045) + - Fix the issue that PD does not fix down peers in time [#4077](https://github.com/tikv/pd/issues/4077) + - Fix the issue that the scatter range scheduler cannot schedule empty regions [#4118](https://github.com/tikv/pd/pull/4118) + - Fix the issue that the key manager cost too much CPU [#4071](https://github.com/tikv/pd/issues/4071) + - Fix the data race issue that might occur when setting configurations of hot region scheduler [#4170](https://github.com/tikv/pd/pull/4170) + - Fix slow leader election caused by stucked region syncer[#3936](https://github.com/tikv/pd/issues/3936) + TiFlash From b35ee53445932dd885b2112f6f1aa887b29c76af Mon Sep 17 00:00:00 2001 From: Enwei Date: Thu, 28 Oct 2021 18:36:32 +0200 Subject: [PATCH 23/35] revise format of TiKV and PD parts --- releases/release-5.2.2.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index c8d1c1c0a6fc0..292da0e62840a 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -25,14 +25,14 @@ TiDB version: 5.2.2 - Improve the error log report in raft client module [#10983](https://github.com/tikv/tikv/pull/10983) - Make the slow log of TiKV coprocessor only consider the time spent on processing requests [#10841](https://github.com/tikv/tikv/issues/10841) - Drop log instead of blocking threads when the slogger thread is overloaded and the queue is filled up [#10841](https://github.com/tikv/tikv/issues/10841) - - Add more statistics types of write queries (https://github.com/tikv/tikv/issues/10507) + - Add more statistics types of write queries [#10507](https://github.com/tikv/tikv/issues/10507) + PD - Add more types of write queries to QPS dimensions in the hotspot scheduler [#3869](https://github.com/tikv/pd/issues/3869) - - Support dynamically adjusting the retry limit of the balance region scheduler to improve the performance of the scheduler [#3744 (https://github.com/tikv/pd/issues/3744) + - Support dynamically adjusting the retry limit of the balance region scheduler to improve the performance of the scheduler [#3744] (https://github.com/tikv/pd/issues/3744) - Update TiDB Dashboard to v2021.10.08.1 [#4070](https://github.com/tikv/pd/pull/4070) - - Support that the evict leader scheduler can schedule regions with unhealthy peers [#4093 (https://github.com/tikv/pd/issues/4093) + - Support that the evict leader scheduler can schedule regions with unhealthy peers [#4093] (https://github.com/tikv/pd/issues/4093) - Speed up the exit process of a scheduler after receiving an end-of-process signal [#4146](https://github.com/tikv/pd/issues/4146) + Tools @@ -93,7 +93,7 @@ TiDB version: 5.2.2 - Fix the issue that PD does not fix down peers in time [#4077](https://github.com/tikv/pd/issues/4077) - Fix the issue that the scatter range scheduler cannot schedule empty regions [#4118](https://github.com/tikv/pd/pull/4118) - Fix the issue that the key manager cost too much CPU [#4071](https://github.com/tikv/pd/issues/4071) - - Fix the data race issue that might occur when setting configurations of hot region scheduler [#4170](https://github.com/tikv/pd/pull/4170) + - Fix the data race issue that might occur when setting configurations of hot region scheduler [#4159](https://github.com/tikv/pd/issues/4159) - Fix slow leader election caused by stucked region syncer[#3936](https://github.com/tikv/pd/issues/3936) + TiFlash From d8f3c48a2163fbbd085cc560e8c72da99121c543 Mon Sep 17 00:00:00 2001 From: Enwei Date: Thu, 28 Oct 2021 18:39:19 +0200 Subject: [PATCH 24/35] revise format of TiKV and PD parts --- releases/release-5.2.2.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 292da0e62840a..b8137b8bd01cf 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -30,9 +30,9 @@ TiDB version: 5.2.2 + PD - Add more types of write queries to QPS dimensions in the hotspot scheduler [#3869](https://github.com/tikv/pd/issues/3869) - - Support dynamically adjusting the retry limit of the balance region scheduler to improve the performance of the scheduler [#3744] (https://github.com/tikv/pd/issues/3744) + - Support dynamically adjusting the retry limit of the balance region scheduler to improve the performance of the scheduler [#3744](https://github.com/tikv/pd/issues/3744) - Update TiDB Dashboard to v2021.10.08.1 [#4070](https://github.com/tikv/pd/pull/4070) - - Support that the evict leader scheduler can schedule regions with unhealthy peers [#4093] (https://github.com/tikv/pd/issues/4093) + - Support that the evict leader scheduler can schedule regions with unhealthy peers [#4093](https://github.com/tikv/pd/issues/4093) - Speed up the exit process of a scheduler after receiving an end-of-process signal [#4146](https://github.com/tikv/pd/issues/4146) + Tools From 103bcba1cbad97bba957a99361212fc3e67914f8 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Fri, 29 Oct 2021 09:28:06 +0800 Subject: [PATCH 25/35] Apply suggestions from code review --- releases/release-5.2.2.md | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index b8137b8bd01cf..74396e417dae6 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -16,8 +16,8 @@ TiDB version: 5.2.2 + TiDB - - Show information about SQL statement in the debug log about coprocessor encountering lock [#27718](https://github.com/pingcap/tidb/issues/27718) - - Add backup and restore size for TiDB SQL backup/restore task . [#27247](https://github.com/pingcap/tidb/issues/27247) + - Show the affected SQL statements in the debug log when the coprocessor encounters a lock, which is helpful in diagnosing problems [#27718](https://github.com/pingcap/tidb/issues/27718) + - Support showing the size of the backup and restore data when backing up and restoring data in the SQL logical layer [#27247](https://github.com/pingcap/tidb/issues/27247) + TiKV @@ -43,23 +43,23 @@ TiDB version: 5.2.2 + TiDB - - Fix the issue that plan-cache cannot be aware of changes of unsigned flags [#28254](https://github.com/pingcap/tidb/issues/28254) - - Fix the wrong partition pruning when some conditions are out of range [#28233](https://github.com/pingcap/tidb/issues/28233) - - Fix the issue that planner may cache invalid plans for joins in some cases [#28087](https://github.com/pingcap/tidb/issues/28087) - - Fix wrong index hash join when hash col is enum. [#27893](https://github.com/pingcap/tidb/issues/27893) - - Fix a batch client bug that recycle idle connection may block sending requests in some rare cases. [#27688](https://github.com/pingcap/tidb/pull/27688) - - Fixed Lightning panic when it failed to perform checksum on target cluster. [#27686](https://github.com/pingcap/tidb/pull/27686) - - Fix the wrong result for date_add and date_sub in some cases. [#27232](https://github.com/pingcap/tidb/issues/27232) - - Fix wrong result of hour function in vectorized expression [#28643](https://github.com/pingcap/tidb/issues/28643) - - Fixed a bug where MySQL 5.1 and older clients had issues authenticating [#27855](https://github.com/pingcap/tidb/issues/27855) - - fix auto analyze may get triggered out of specified time. [#28698](https://github.com/pingcap/tidb/issues/28698) - - Fix the bug that setting any session variable will make `tidb_snapshot` unwork. [#28683](https://github.com/pingcap/tidb/pull/28683) - - Fixed a bug that caused BR get stuck when many missing-peer regions in cluster. [#27534](https://github.com/pingcap/tidb/issues/27534) - - Fix unexpected error like `tidb_cast to Int32 is not supported` when unsupported cast is pushed down to TiFlash [#23907](https://github.com/pingcap/tidb/issues/23907) - - Fix error message for DECIMAL overflow is just "ERROR 1690 (22003): %s value is out of range in '%s'" [#27964](https://github.com/pingcap/tidb/issues/27964) - - Fix bug that MPP node availability detect does not work in some corner cases [#3118](https://github.com/pingcap/tics/issues/3118) - - Fix data-race bug when alloc MPP task ID [#27952](https://github.com/pingcap/tidb/issues/27952) - - Fix index out of bound bug when empty dual table is remove for MPP query [#28250](https://github.com/pingcap/tidb/issues/28250) + - Fix the issue that plan-cache cannot detect changes of unsigned flags [#28254](https://github.com/pingcap/tidb/issues/28254) + - Fix the wrong partition pruning when the partition function is out of range [#28233](https://github.com/pingcap/tidb/issues/28233) + - Fix the issue that planner may cache invalid plans for `join` in some cases [#28087](https://github.com/pingcap/tidb/issues/28087) + - Fix wrong index hash join when hash column type is enum [#27893](https://github.com/pingcap/tidb/issues/27893) + - Fix a batch client bug that recycling idle connection may block sending requests in some rare cases [#27688](https://github.com/pingcap/tidb/pull/27688) + - Fixed the TiDB Lightning panic issue when it fails to perform checksum on a target cluster [#27686](https://github.com/pingcap/tidb/pull/27686) + - Fix wrong results of the `date_add` and `date_sub` functions in some cases [#27232](https://github.com/pingcap/tidb/issues/27232) + - Fix wrong results of the `hour` function in vectorized expression [#28643](https://github.com/pingcap/tidb/issues/28643) + - Fix the authenticating issue when connecting to MySQL 5.1 or an older client version [#27855](https://github.com/pingcap/tidb/issues/27855) + - Fix the issue that auto analyze might be triggered out of the specified time when a new index is added [#28698](https://github.com/pingcap/tidb/issues/28698) + - Fix a bug that setting any session variable can make `tidb_snapshot` unwork [#28683](https://github.com/pingcap/tidb/pull/28683) + - Fix a bug that BR is not working for clusters with many missing-peer regions [#27534](https://github.com/pingcap/tidb/issues/27534) + - Fix the unexpected error like `tidb_cast to Int32 is not supported` when the unsupported `cast` is pushed down to TiFlash [#23907](https://github.com/pingcap/tidb/issues/23907) + - Fix the issue that `DECIMAL overflow` is missing in the `%s value is out of range in '%s'`error message [#27964](https://github.com/pingcap/tidb/issues/27964) + - Fix a bug that the availability detection of MPP node does not work in some corner cases [#3118](https://github.com/pingcap/tics/issues/3118) + - Fix the`DATA RACE` issue when assigning`MPP task ID` [#27952](https://github.com/pingcap/tidb/issues/27952) + - Fix the `INDEX OUT OF RANGE` error for a MPP query after deleting an empty `dual table`. [#28250](https://github.com/pingcap/tidb/issues/28250) - Fix the issue of false positive error log `invalid cop task execution summaries length` for MPP queries [#1791](https://github.com/pingcap/tics/issues/1791) - Fix the issue of error log `can not found column in Schema column` for MPP queries [#28149](https://github.com/pingcap/tidb/pull/28149) - Fix the issue that TiDB might crash when TiFlash shuts down [#28096](https://github.com/pingcap/tidb/issues/28096) From ab7a29af6637f627a57a63239d53fad82c65a580 Mon Sep 17 00:00:00 2001 From: Liuxiaozhen12 <82579298+Liuxiaozhen12@users.noreply.github.com> Date: Fri, 29 Oct 2021 10:56:30 +0800 Subject: [PATCH 26/35] Apply suggestions from code review Co-authored-by: xixirangrang Co-authored-by: Enwei --- releases/release-5.2.2.md | 31 ++++++++++++++++++++++--------- 1 file changed, 22 insertions(+), 9 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 74396e417dae6..211318dd0a4c9 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -38,7 +38,11 @@ TiDB version: 5.2.2 + Tools + TiCDC - + - Reduce the default value of the Kafka sink configuration item `MaxMessageBytes` from 64 MB to 1 MB to fix the issue that large messages are rejected by the Kafka Broker [#3104](https://github.com/pingcap/ticdc/pull/3104) + - Reduce memory usage in the synchronization link [#2553](https://github.com/pingcap/ticdc/issues/2553)[#3037](https://github.com/pingcap/ticdc/pull/3037) [#2726](https://github.com/pingcap/ticdc/pull/2726) + - Optimize monitoring items and alert rules to improve observability of synchronous links, memory GC, and stock data scanning processes [#2735](https://github.com/pingcap/ticdc/pull/2735) [#1606](https://github.com/pingcap/ticdc/issues/1606) [#3000](https://github.com/pingcap/ticdc/pull/3000 [#2985](https://github.com/pingcap/ticdc/issues/2985) [#2156](https://github.com/pingcap/ticdc/issues/2156) + - When the sync task status is normal, no more historical error messages are displayed to avoid misleading users [#2242](https://github.com/pingcap/ticdc/issues/2242) + ## Bug Fixes + TiDB @@ -61,21 +65,21 @@ TiDB version: 5.2.2 - Fix the`DATA RACE` issue when assigning`MPP task ID` [#27952](https://github.com/pingcap/tidb/issues/27952) - Fix the `INDEX OUT OF RANGE` error for a MPP query after deleting an empty `dual table`. [#28250](https://github.com/pingcap/tidb/issues/28250) - Fix the issue of false positive error log `invalid cop task execution summaries length` for MPP queries [#1791](https://github.com/pingcap/tics/issues/1791) - - Fix the issue of error log `can not found column in Schema column` for MPP queries [#28149](https://github.com/pingcap/tidb/pull/28149) - - Fix the issue that TiDB might crash when TiFlash shuts down [#28096](https://github.com/pingcap/tidb/issues/28096) - - Remove the support for insecure 3DES (Triple Data Encryption Algorithm) based TLS ciphersuites [#27859](https://github.com/pingcap/tidb/pull/27859) - - Fix the issue that Lightning connects to offline TiKV nodes during pre-check and causes import failures [#27826](https://github.com/pingcap/tidb/pull/27826) + - Fix the issue of error log `cannot found column in Schema column` for MPP queries [#28149](https://github.com/pingcap/tidb/pull/28149) + - Fix the issue that TiDB might panic when TiFlash is shuting down [#28096](https://github.com/pingcap/tidb/issues/28096) + - Remove the support for insecure 3DES (Triple Data Encryption Algorithm) based TLS cipher suites [#27859](https://github.com/pingcap/tidb/pull/27859) + - Fix the issue that Lightning connects to offline TiKV nodes during pre-check and causes import failures [#27826](https://github.com/pingcap/tidb/pull/27826) - Fix the issue that pre-check cost too much time when importing many files to tables [#27605](https://github.com/pingcap/tidb/issues/27605) - Fix the issue that rewriting expressions makes `between` infer wrong collation [#27146](https://github.com/pingcap/tidb/issues/27146) - Fix the issue that `group_concat` function did not consider the collation [#27429](https://github.com/pingcap/tidb/issues/27429) - - Fix the issue that `extract` function gives wrong results when argument is a negative duration [#27236](https://github.com/pingcap/tidb/issues/27236) + - Fix the result wrong that occurs when the argument of the `extract` function is a negative duration [#27236](https://github.com/pingcap/tidb/issues/27236) - Fix the issue that creating partition fails if `NO_UNSIGNED_SUBTRACTION` is set [#26765](https://github.com/pingcap/tidb/issues/26765) - Avoid expressions with side effects in column pruning and aggregation pushdown [#27106](https://github.com/pingcap/tidb/issues/27106) - - Remove useless GRPC log in production [#24190](https://github.com/pingcap/tidb/issues/24190) + - Remove useless gRPC logs [#24190](https://github.com/pingcap/tidb/issues/24190) - Limit the valid decimal length to fix precision-related issues [#3091](https://github.com/pingcap/tics/issues/3091) - Fix the issue of a wrong way to check for overflow in `plus` expression [#26977](https://github.com/pingcap/tidb/issues/26977) - - Fix the issue of `data too long` error when dumping statistics from the table with new collation data [#27024](https://github.com/pingcap/tidb/issues/27024) - - Fix the issue that the retried transactions' statements are not included in TIDB_TRX [#28670](https://github.com/pingcap/tidb/pull/28670) + - Fix the issue of `data too long` error when dumping statistics from the table with `new collation` data [#27024](https://github.com/pingcap/tidb/issues/27024) + - Fix the issue that the retried transactions' statements are not included in `TIDB_TRX` [#28670](https://github.com/pingcap/tidb/pull/28670) + TiKV @@ -103,4 +107,13 @@ TiDB version: 5.2.2 + Tools + TiCDC + - Fix the issue that TiCDC sync task may pause when the upstream TiDB instance unexpectedly exits [#3061](https://github.com/pingcap/ticdc/issues/3061) + - Fix when TiKV sends duplicate requests to the same Region, TiCDC process PANIC `tikv reported duplicated request to the same region, which is not expected` [#2386](https://github.com/pingcap/ticdc/issues/2386) + - Fix useless CPU consumption when verifying downstream TiDB/MySQL availability [#3073](https://github.com/pingcap/ticdc/issues/3073) + - Fix the issue that the volume of Kafka messages generated by TiCDC is not constrained by `max-message-size` [#2962](https://github.com/pingcap/ticdc/issues/2962) + - Fix the issue that TiCDC sync task may pause when an error occurs during writing a Kafka message [#2978](https://github.com/pingcap/ticdc/issues/2978) + - Fix the issue that some partitioned tables without valid indexes may be ignored when `force-replicate` is enabled [#2834](https://github.com/pingcap/ticdc/issues/2834) + - Fix the issue that scanning stock data may fail due to TiKV performing GC when scanning stock data takes too long [#2470](https://github.com/pingcap/ticdc/issues/2470) + - Fix a possible PANIC `interface conversion: interface {} is string, not []uint8` when encoding some types of columns into Open Protocol format [#2758](https://github.com/pingcap/ticdc/issues/2758) + - Fix a possible PANIC `interface conversion: interface {} is uint64, not int64` when encoding some types of columns into Avro format [#2648](https://github.com/pingcap/ticdc/issues/2648) From e96754cf27a332aac63aaee21f46c1fdbd82be5a Mon Sep 17 00:00:00 2001 From: xixirangrang Date: Fri, 29 Oct 2021 12:04:51 +0800 Subject: [PATCH 27/35] Update releases/release-5.2.2.md --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 211318dd0a4c9..9ad6bde9d401d 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -109,7 +109,7 @@ TiDB version: 5.2.2 + TiCDC - Fix the issue that TiCDC sync task may pause when the upstream TiDB instance unexpectedly exits [#3061](https://github.com/pingcap/ticdc/issues/3061) - Fix when TiKV sends duplicate requests to the same Region, TiCDC process PANIC `tikv reported duplicated request to the same region, which is not expected` [#2386](https://github.com/pingcap/ticdc/issues/2386) - - Fix useless CPU consumption when verifying downstream TiDB/MySQL availability [#3073](https://github.com/pingcap/ticdc/issues/3073) + - Fix unnecessary CPU consumption when verifying downstream TiDB/MySQL availability [#3073](https://github.com/pingcap/ticdc/issues/3073) - Fix the issue that the volume of Kafka messages generated by TiCDC is not constrained by `max-message-size` [#2962](https://github.com/pingcap/ticdc/issues/2962) - Fix the issue that TiCDC sync task may pause when an error occurs during writing a Kafka message [#2978](https://github.com/pingcap/ticdc/issues/2978) - Fix the issue that some partitioned tables without valid indexes may be ignored when `force-replicate` is enabled [#2834](https://github.com/pingcap/ticdc/issues/2834) From 105d3c14db101c876b81518f582c24cdf5ab61f5 Mon Sep 17 00:00:00 2001 From: Enwei Date: Fri, 29 Oct 2021 06:27:32 +0200 Subject: [PATCH 28/35] Apply suggestions from code review Co-authored-by: Grace Cai --- releases/release-5.2.2.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 9ad6bde9d401d..d6f9b3c39442a 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -22,7 +22,7 @@ TiDB version: 5.2.2 + TiKV - Simplify the algorithm of L0 flow control [#10879](https://github.com/tikv/tikv/issues/10879) - - Improve the error log report in raft client module [#10983](https://github.com/tikv/tikv/pull/10983) + - Improve the error log report in the raft client module [#10983](https://github.com/tikv/tikv/pull/10983) - Make the slow log of TiKV coprocessor only consider the time spent on processing requests [#10841](https://github.com/tikv/tikv/issues/10841) - Drop log instead of blocking threads when the slogger thread is overloaded and the queue is filled up [#10841](https://github.com/tikv/tikv/issues/10841) - Add more statistics types of write queries [#10507](https://github.com/tikv/tikv/issues/10507) @@ -83,13 +83,13 @@ TiDB version: 5.2.2 + TiKV - - Fix the issue that CDC add scan retries frequently due to Congest error [#11082](https://github.com/tikv/tikv/issues/11082) - - Fix that the raft connection is broken when the channel is full [#11047](https://github.com/tikv/tikv/issues/11047) + - Fix the issue that CDC adds scan retries frequently due to the Congest error [#11082](https://github.com/tikv/tikv/issues/11082) + - Fix the issue that the raft connection is broken when the channel is full [#11047](https://github.com/tikv/tikv/issues/11047) - Fix the issue that batch messages are too large in Raft client implementation [#9714](https://github.com/tikv/tikv/issues/9714) - - Fix the issue that concurrent leaks in `resolved_ts` [#10965](https://github.com/tikv/tikv/issues/10965) - - Fix a panic issue that occurs to coprocessor when response size exceeds 4 GiB [#9012](https://github.com/tikv/tikv/issues/9012) + - Fix the issue that some coroutines leak in `resolved_ts` [#10965](https://github.com/tikv/tikv/issues/10965) + - Fix a panic issue that occurs to the coprocessor when the size of response exceeds 4 GiB [#9012](https://github.com/tikv/tikv/issues/9012) - Fix the issue that snapshot Garbage Collection (GC) misses GC snapshot files when snapshot files cannot be garbage collected [#10813](https://github.com/tikv/tikv/issues/10813) - - Fix a panic issue that occurs when processing coprocessor requests times out [#10852](https://github.com/tikv/tikv/issues/10852) + - Fix a panic issue caused by timeout when processing Coprocessor requests [#10852](https://github.com/tikv/tikv/issues/10852) + PD From 418766ec1fb8e93eac8caf277fa48f6406adef4c Mon Sep 17 00:00:00 2001 From: Enwei Date: Fri, 29 Oct 2021 06:35:05 +0200 Subject: [PATCH 29/35] Apply suggestions from code review Co-authored-by: Xintao Co-authored-by: Ryan Leung --- releases/release-5.2.2.md | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index d6f9b3c39442a..c252db8a6bbce 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -23,8 +23,7 @@ TiDB version: 5.2.2 - Simplify the algorithm of L0 flow control [#10879](https://github.com/tikv/tikv/issues/10879) - Improve the error log report in the raft client module [#10983](https://github.com/tikv/tikv/pull/10983) - - Make the slow log of TiKV coprocessor only consider the time spent on processing requests [#10841](https://github.com/tikv/tikv/issues/10841) - - Drop log instead of blocking threads when the slogger thread is overloaded and the queue is filled up [#10841](https://github.com/tikv/tikv/issues/10841) + - Improve logging threads to avoid them becoming a performance bottleneck [#10841](https://github.com/tikv/tikv/issues/10841) - Add more statistics types of write queries [#10507](https://github.com/tikv/tikv/issues/10507) + PD @@ -33,7 +32,7 @@ TiDB version: 5.2.2 - Support dynamically adjusting the retry limit of the balance region scheduler to improve the performance of the scheduler [#3744](https://github.com/tikv/pd/issues/3744) - Update TiDB Dashboard to v2021.10.08.1 [#4070](https://github.com/tikv/pd/pull/4070) - Support that the evict leader scheduler can schedule regions with unhealthy peers [#4093](https://github.com/tikv/pd/issues/4093) - - Speed up the exit process of a scheduler after receiving an end-of-process signal [#4146](https://github.com/tikv/pd/issues/4146) + - Speed up the exit process of schedulers [#4146](https://github.com/tikv/pd/issues/4146) + Tools From 3fc846a6c8265d1d0339f026f956418f15649bda Mon Sep 17 00:00:00 2001 From: Liuxiaozhen12 <82579298+Liuxiaozhen12@users.noreply.github.com> Date: Fri, 29 Oct 2021 14:28:14 +0800 Subject: [PATCH 30/35] Apply suggestions from code review Co-authored-by: shichun-0415 <89768198+shichun-0415@users.noreply.github.com> Co-authored-by: leoppro Co-authored-by: xixirangrang --- releases/release-5.2.2.md | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index c252db8a6bbce..740dab52838f7 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -47,16 +47,16 @@ TiDB version: 5.2.2 + TiDB - Fix the issue that plan-cache cannot detect changes of unsigned flags [#28254](https://github.com/pingcap/tidb/issues/28254) - - Fix the wrong partition pruning when the partition function is out of range [#28233](https://github.com/pingcap/tidb/issues/28233) + - Fix the wrong partition pruning when the partition function is out of range [#28233](https://github.com/pingcap/tidb/issues/28233) - Fix the issue that planner may cache invalid plans for `join` in some cases [#28087](https://github.com/pingcap/tidb/issues/28087) - Fix wrong index hash join when hash column type is enum [#27893](https://github.com/pingcap/tidb/issues/27893) - Fix a batch client bug that recycling idle connection may block sending requests in some rare cases [#27688](https://github.com/pingcap/tidb/pull/27688) - - Fixed the TiDB Lightning panic issue when it fails to perform checksum on a target cluster [#27686](https://github.com/pingcap/tidb/pull/27686) + - Fix the TiDB Lightning panic issue when it fails to perform checksum on a target cluster [#27686](https://github.com/pingcap/tidb/pull/27686) - Fix wrong results of the `date_add` and `date_sub` functions in some cases [#27232](https://github.com/pingcap/tidb/issues/27232) - Fix wrong results of the `hour` function in vectorized expression [#28643](https://github.com/pingcap/tidb/issues/28643) - Fix the authenticating issue when connecting to MySQL 5.1 or an older client version [#27855](https://github.com/pingcap/tidb/issues/27855) - Fix the issue that auto analyze might be triggered out of the specified time when a new index is added [#28698](https://github.com/pingcap/tidb/issues/28698) - - Fix a bug that setting any session variable can make `tidb_snapshot` unwork [#28683](https://github.com/pingcap/tidb/pull/28683) + - Fix a bug that setting any session variable invalidates `tidb_snapshot` [#28683](https://github.com/pingcap/tidb/pull/28683) - Fix a bug that BR is not working for clusters with many missing-peer regions [#27534](https://github.com/pingcap/tidb/issues/27534) - Fix the unexpected error like `tidb_cast to Int32 is not supported` when the unsupported `cast` is pushed down to TiFlash [#23907](https://github.com/pingcap/tidb/issues/23907) - Fix the issue that `DECIMAL overflow` is missing in the `%s value is out of range in '%s'`error message [#27964](https://github.com/pingcap/tidb/issues/27964) @@ -106,7 +106,7 @@ TiDB version: 5.2.2 + Tools + TiCDC - - Fix the issue that TiCDC sync task may pause when the upstream TiDB instance unexpectedly exits [#3061](https://github.com/pingcap/ticdc/issues/3061) + - Fix the issue that TiCDC replication task may terminate when the upstream TiDB instance unexpectedly exits [#3061](https://github.com/pingcap/ticdc/issues/3061) - Fix when TiKV sends duplicate requests to the same Region, TiCDC process PANIC `tikv reported duplicated request to the same region, which is not expected` [#2386](https://github.com/pingcap/ticdc/issues/2386) - Fix unnecessary CPU consumption when verifying downstream TiDB/MySQL availability [#3073](https://github.com/pingcap/ticdc/issues/3073) - Fix the issue that the volume of Kafka messages generated by TiCDC is not constrained by `max-message-size` [#2962](https://github.com/pingcap/ticdc/issues/2962) @@ -116,3 +116,6 @@ TiDB version: 5.2.2 - Fix a possible PANIC `interface conversion: interface {} is string, not []uint8` when encoding some types of columns into Open Protocol format [#2758](https://github.com/pingcap/ticdc/issues/2758) - Fix a possible PANIC `interface conversion: interface {} is uint64, not int64` when encoding some types of columns into Avro format [#2648](https://github.com/pingcap/ticdc/issues/2648) + + TiDB Binlog + + - Fix the issue that when most tables are filtered out, checkpoint can not be updated under some special load [#1075](https://github.com/pingcap/tidb-binlog/pull/1075) From 35c919a8ccf15cb8de4549c6546a687ebc37b74d Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Fri, 29 Oct 2021 15:02:57 +0800 Subject: [PATCH 31/35] Update releases/release-5.2.2.md Co-authored-by: shichun-0415 <89768198+shichun-0415@users.noreply.github.com> --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 740dab52838f7..7322b605fb858 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -101,7 +101,7 @@ TiDB version: 5.2.2 + TiFlash - - Fix the issue that TiFlash fails to start up under platform without library `nsl` + - Fix the issue that TiFlash fails to start up on some platforms due to the absence of library `nsl` + Tools From a0c5cc40e0615ab082be6c5b497ba1a54bb3cd73 Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Fri, 29 Oct 2021 15:08:44 +0800 Subject: [PATCH 32/35] Update releases/release-5.2.2.md Co-authored-by: leoppro --- releases/release-5.2.2.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 7322b605fb858..6b6619de2bfea 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -38,7 +38,7 @@ TiDB version: 5.2.2 + TiCDC - Reduce the default value of the Kafka sink configuration item `MaxMessageBytes` from 64 MB to 1 MB to fix the issue that large messages are rejected by the Kafka Broker [#3104](https://github.com/pingcap/ticdc/pull/3104) - - Reduce memory usage in the synchronization link [#2553](https://github.com/pingcap/ticdc/issues/2553)[#3037](https://github.com/pingcap/ticdc/pull/3037) [#2726](https://github.com/pingcap/ticdc/pull/2726) + - Reduce memory usage in the relpication pipeline [#2553](https://github.com/pingcap/ticdc/issues/2553)[#3037](https://github.com/pingcap/ticdc/pull/3037) [#2726](https://github.com/pingcap/ticdc/pull/2726) - Optimize monitoring items and alert rules to improve observability of synchronous links, memory GC, and stock data scanning processes [#2735](https://github.com/pingcap/ticdc/pull/2735) [#1606](https://github.com/pingcap/ticdc/issues/1606) [#3000](https://github.com/pingcap/ticdc/pull/3000 [#2985](https://github.com/pingcap/ticdc/issues/2985) [#2156](https://github.com/pingcap/ticdc/issues/2156) - When the sync task status is normal, no more historical error messages are displayed to avoid misleading users [#2242](https://github.com/pingcap/ticdc/issues/2242) From cd272adc98a793b38251887fd746d31d42aaf2d9 Mon Sep 17 00:00:00 2001 From: Liuxiaozhen12 <82579298+Liuxiaozhen12@users.noreply.github.com> Date: Fri, 29 Oct 2021 15:43:33 +0800 Subject: [PATCH 33/35] Apply suggestions from code review Co-authored-by: xixirangrang --- releases/release-5.2.2.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 6b6619de2bfea..1451552d04643 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -107,14 +107,14 @@ TiDB version: 5.2.2 + TiCDC - Fix the issue that TiCDC replication task may terminate when the upstream TiDB instance unexpectedly exits [#3061](https://github.com/pingcap/ticdc/issues/3061) - - Fix when TiKV sends duplicate requests to the same Region, TiCDC process PANIC `tikv reported duplicated request to the same region, which is not expected` [#2386](https://github.com/pingcap/ticdc/issues/2386) + - Fix the issue that TiCDC process might panic when TiKV sends duplicate requests to the same Region [#2386](https://github.com/pingcap/ticdc/issues/2386) - Fix unnecessary CPU consumption when verifying downstream TiDB/MySQL availability [#3073](https://github.com/pingcap/ticdc/issues/3073) - Fix the issue that the volume of Kafka messages generated by TiCDC is not constrained by `max-message-size` [#2962](https://github.com/pingcap/ticdc/issues/2962) - Fix the issue that TiCDC sync task may pause when an error occurs during writing a Kafka message [#2978](https://github.com/pingcap/ticdc/issues/2978) - Fix the issue that some partitioned tables without valid indexes may be ignored when `force-replicate` is enabled [#2834](https://github.com/pingcap/ticdc/issues/2834) - Fix the issue that scanning stock data may fail due to TiKV performing GC when scanning stock data takes too long [#2470](https://github.com/pingcap/ticdc/issues/2470) - - Fix a possible PANIC `interface conversion: interface {} is string, not []uint8` when encoding some types of columns into Open Protocol format [#2758](https://github.com/pingcap/ticdc/issues/2758) - - Fix a possible PANIC `interface conversion: interface {} is uint64, not int64` when encoding some types of columns into Avro format [#2648](https://github.com/pingcap/ticdc/issues/2648) + - Fix a possible panic issue when encoding some types of columns into Open Protocol format [#2758](https://github.com/pingcap/ticdc/issues/2758) + - Fix a possible panic issue when encoding some types of columns into Avro format [#2648](https://github.com/pingcap/ticdc/issues/2648) + TiDB Binlog From 6d5e0cfbd3810a1715bffd19446877a0bd27cf9a Mon Sep 17 00:00:00 2001 From: Grace Cai Date: Fri, 29 Oct 2021 15:47:00 +0800 Subject: [PATCH 34/35] Update releases/release-5.2.2.md --- releases/release-5.2.2.md | 4 ---- 1 file changed, 4 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 1451552d04643..0c7d1b0b87516 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -8,10 +8,6 @@ Release Date: October 27, 2021 TiDB version: 5.2.2 -## Compatibility change(s) - -## Feature enhancement(s) - ## Improvements + TiDB From ca3c2708ada446297712c2eae620f270cef7f515 Mon Sep 17 00:00:00 2001 From: qiancai Date: Fri, 29 Oct 2021 16:33:00 +0800 Subject: [PATCH 35/35] Update release-5.2.2.md --- releases/release-5.2.2.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/releases/release-5.2.2.md b/releases/release-5.2.2.md index 0c7d1b0b87516..3fa7b36dc9422 100644 --- a/releases/release-5.2.2.md +++ b/releases/release-5.2.2.md @@ -35,7 +35,7 @@ TiDB version: 5.2.2 + TiCDC - Reduce the default value of the Kafka sink configuration item `MaxMessageBytes` from 64 MB to 1 MB to fix the issue that large messages are rejected by the Kafka Broker [#3104](https://github.com/pingcap/ticdc/pull/3104) - Reduce memory usage in the relpication pipeline [#2553](https://github.com/pingcap/ticdc/issues/2553)[#3037](https://github.com/pingcap/ticdc/pull/3037) [#2726](https://github.com/pingcap/ticdc/pull/2726) - - Optimize monitoring items and alert rules to improve observability of synchronous links, memory GC, and stock data scanning processes [#2735](https://github.com/pingcap/ticdc/pull/2735) [#1606](https://github.com/pingcap/ticdc/issues/1606) [#3000](https://github.com/pingcap/ticdc/pull/3000 [#2985](https://github.com/pingcap/ticdc/issues/2985) [#2156](https://github.com/pingcap/ticdc/issues/2156) + - Optimize monitoring items and alert rules to improve observability of synchronous links, memory GC, and stock data scanning processes [#2735](https://github.com/pingcap/ticdc/pull/2735) [#1606](https://github.com/pingcap/ticdc/issues/1606) [#3000](https://github.com/pingcap/ticdc/pull/3000) [#2985](https://github.com/pingcap/ticdc/issues/2985) [#2156](https://github.com/pingcap/ticdc/issues/2156) - When the sync task status is normal, no more historical error messages are displayed to avoid misleading users [#2242](https://github.com/pingcap/ticdc/issues/2242) ## Bug Fixes @@ -114,4 +114,4 @@ TiDB version: 5.2.2 + TiDB Binlog - - Fix the issue that when most tables are filtered out, checkpoint can not be updated under some special load [#1075](https://github.com/pingcap/tidb-binlog/pull/1075) + - Fix the issue that when most tables are filtered out, checkpoint can not be updated under some special load [#1075](https://github.com/pingcap/tidb-binlog/pull/1075)