[TiDB] 查看表的分区信息
首先
我在以下每个时间点执行了SHOW TABLE REGIONS命令，并确认了区域信息。
-
- データ追加前
-
- データ追加後
- TiKVの異常終了
确认动作环境
我在以下环境中测试了 TiDB v7.0.0 的运行：
[root@tisim ~]# tiup cluster display demo-cluster
tiup is checking updates for component cluster ...
Starting component `cluster`: /root/.tiup/components/cluster/v1.12.1/tiup-cluster display demo-cluster
Cluster type: tidb
Cluster name: demo-cluster
Cluster version: v7.0.0
Deploy user: tidb
SSH type: builtin
Dashboard URL: http://192.168.3.171:2379/dashboard
Grafana URL: http://192.168.3.171:3000
ID Role Host Ports OS/Arch Status Data Dir Deploy Dir
-- ---- ---- ----- ------- ------ -------- ----------
192.168.3.171:3000 grafana 192.168.3.171 3000 linux/x86_64 Up - /tidb-deploy/grafana-3000
192.168.3.171:2379 pd 192.168.3.171 2379/2380 linux/x86_64 Up|L|UI /tidb-data/pd-2379 /tidb-deploy/pd-2379
192.168.3.171:9090 prometheus 192.168.3.171 9090/12020 linux/x86_64 Up /tidb-data/prometheus-9090 /tidb-deploy/prometheus-9090
192.168.3.171:4000 tidb 192.168.3.171 4000/10080 linux/x86_64 Up - /tidb-deploy/tidb-4000
192.168.3.171:9000 tiflash 192.168.3.171 9000/8123/3930/20170/20292/8234 linux/x86_64 Up /tidb-data/tiflash-9000 /tidb-deploy/tiflash-9000
192.168.3.171:20160 tikv 192.168.3.171 20160/20180 linux/x86_64 Up /tidb-data/tikv-20160 /tidb-deploy/tikv-20160
192.168.3.171:20161 tikv 192.168.3.171 20161/20181 linux/x86_64 Up /tidb-data/tikv-20161 /tidb-deploy/tikv-20161
192.168.3.171:20162 tikv 192.168.3.171 20162/20182 linux/x86_64 Up /tidb-data/tikv-20162 /tidb-deploy/tikv-20162
Total nodes: 8
确认区域信息
使用SHOW TABLE REGIONS语句查看trips表的区域信息。
mysql> SHOW TABLE trips REGIONS;
+-----------+---------------+--------------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS | SCHEDULING_CONSTRAINTS | SCHEDULING_STATE |
+-----------+---------------+--------------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| 3005 | t_94_ | t_94_r_182668 | 3007 | 1 | 3006, 3007, 3008 | 0 | 27 | 0 | 39 | 177442 | | |
| 3009 | t_94_r_182668 | t_94_r_595336 | 3011 | 1 | 3010, 3011, 3012 | 0 | 39 | 0 | 37 | 231736 | | |
| 10 | t_94_r_595336 | t_281474976710654_ | 156 | 1 | 11, 156, 204 | 0 | 35241357 | 149414338 | 37 | 231736 | | |
+-----------+---------------+--------------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
3 rows in set (0.00 sec)
可以通过指定条件来筛选出指定的行。将行的region_id设为3009并执行操作。
mysql> SHOW TABLE trips REGIONS WHERE region_id = 3009 \G
*************************** 1. row ***************************
REGION_ID: 3009
START_KEY: t_94_r_182668
END_KEY: t_94_r_595336
LEADER_ID: 3011
LEADER_STORE_ID: 1
PEERS: 3010, 3011, 3012
SCATTERING: 0
WRITTEN_BYTES: 0
READ_BYTES: 0
APPROXIMATE_SIZE(MB): 60
APPROXIMATE_KEYS: 388612
SCHEDULING_CONSTRAINTS:
SCHEDULING_STATE:
1 row in set (0.00 sec)
从具有region_id为3009的区域信息中可以得知以下事项。
-
- start_keyがt_94_r_182668、かつend_keyがt_94_r_595336である事から、こちらのリージョンのキー値の範囲が、182668から595336である事がわかります。また、プレフィックスのt_94は、テーブルidが94であることを示しています。
-
- leader_idが3011で、peersが3010, 3011, 3012である事から、3011のリージョンがリーダ(読み書き可能)で、3010と3012のリージョンが、リーダのレプリカである事を示しています。
-
- leader_store_idが1である事から、3011のリージョンは、id1のTiKVに格納されている事がわかります。
-
- approximate_size(MB)の60は、こちらのリージョンにおよそ60MBのデータが格納されている事を示しています。
- approximate_keysの388612は、こちらのリージョンにおよそ388612個のキー値が格納されている事を示しています。
每个区域都有一个上限大小（默认为96MiB），如果超过上限，将被分为两个区域。
区域的默认冗余度为3。
区域的默认冗余度为3。
从上述状态下插入了数据。我们将确认数据添加后的区域信息。
mysql> SHOW TABLE trips REGIONS;
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS | SCHEDULING_CONSTRAINTS | SCHEDULING_STATE |
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| 3005 | t_94_ | t_94_r_182668 | 3007 | 1 | 3006, 3007, 3008 | 0 | 0 | 0 | 28 | 182223 | | |
| 3013 | t_94_r_182668 | t_94_r_842765 | 3015 | 1 | 3014, 3015, 3016 | 0 | 39 | 0 | 41 | 243353 | | |
| 3017 | t_94_r_842765 | t_94_r_1245906 | 3019 | 1 | 3018, 3019, 3020 | 0 | 0 | 0 | 21 | 123626 | | |
| 3009 | t_94_r_1245906 | 78000000 | 3011 | 1 | 3010, 3011, 3012 | 0 | 161824266 | 0 | 21 | 123626 | | |
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
4 rows in set (0.01 sec)
在添加数据之前，3009区域管理182,668至595,336的范围。但是，在添加数据之后，划分了3009区域，确认新的3013和3017区域现在管理上述范围的键。
当TiKV发生异常终止时，请确认其操作。
我们将扩展一个 TiKV 节点，并在由 4 个 TiKV 节点组成的集群中，测试在强制停止一个 TiKV 节点时的行为。
反正读cmds就好了
根据TiUP的操作步骤，我们进行了TiDB集群的扩容。首先，需要准备TiKV的配置文件。
[root@tisim ~]# cat scale-out.yml
tikv_servers:
- host: 192.168.3.171
port: 20163
status_port: 20183
config:
server.labels: { host: "logic-host-4" }
如果像作者的环境一样，在一个节点上组建一个 TiKV 集群时，请注意避免与现有的 TiKV 端口冲突。
我会在事前进行检查。
[root@tisim ~]# tiup cluster check demo-cluster scale-out.yml --cluster --user root -p
tiup is checking updates for component cluster ...
Starting component `cluster`: /root/.tiup/components/cluster/v1.12.1/tiup-cluster check demo-cluster scale-out.yml --cluster --user root -p
Input SSH password:
+ Detect CPU Arch Name
- Detecting node 192.168.3.171 Arch info ... Done
+ Detect CPU OS Name
- Detecting node 192.168.3.171 OS info ... Done
+ Download necessary tools
- Downloading check tools for linux/amd64 ... Done
+ Collect basic system information
+ Collect basic system information
- Getting system info of 192.168.3.171:22 ... Done
+ Check time zone
- Checking node 192.168.3.171 ... Done
+ Check system requirements
+ Check system requirements
+ Check system requirements
- Checking node 192.168.3.171 ... Done
- Checking node 192.168.3.171 ... Done
+ Cleanup check files
- Cleanup check files on 192.168.3.171:22 ... Done
Node Check Result Message
---- ----- ------ -------
192.168.3.171 thp Pass THP is disabled
192.168.3.171 cpu-cores Pass number of CPU cores / threads: 4
192.168.3.171 memory Pass memory size is 0MB
192.168.3.171 network Pass network speed of enp0s3 is 1000MB
192.168.3.171 selinux Pass SELinux is disabled
192.168.3.171 command Pass numactl: policy: default
192.168.3.171 os-version Pass OS is CentOS Linux 7 (Core) 7.9.2009
192.168.3.171 cpu-governor Warn Unable to determine current CPU frequency governor policy
192.168.3.171 disk Warn mount point / does not have 'noatime' option set
尽管有一个警告，但我将忽略它并执行扩容操作。
[root@tisim ~]# tiup cluster scale-out demo-cluster scale-out.yml -p
tiup is checking updates for component cluster ...
Starting component `cluster`: /root/.tiup/components/cluster/v1.12.1/tiup-cluster scale-out demo-cluster scale-out.yml -p
Input SSH password:
+ Detect CPU Arch Name
- Detecting node 192.168.3.171 Arch info ... Done
+ Detect CPU OS Name
- Detecting node 192.168.3.171 OS info ... Done
Please confirm your topology:
Cluster type: tidb
Cluster name: demo-cluster
Cluster version: v7.0.0
Role Host Ports OS/Arch Directories
---- ---- ----- ------- -----------
tikv 192.168.3.171 20163/20183 linux/x86_64 /tidb-deploy/tikv-20163,/tidb-data/tikv-20163
Attention:
1. If the topology is not what you expected, check your yaml file.
2. Please confirm there is no port/directory conflicts in same host.
Do you want to continue? [y/N]: (default=N)
输入y。
Do you want to continue? [y/N]: (default=N) y
+ [ Serial ] - SSHKeySet: privateKey=/root/.tiup/storage/cluster/clusters/demo-cluster/ssh/id_rsa, publicKey=/root/.tiup/storage/cluster/clusters/demo-cluster/ssh/id_rsa.pub
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ [Parallel] - UserSSH: user=tidb, host=192.168.3.171
+ Download TiDB components
- Download tikv:v7.0.0 (linux/amd64) ... Done
+ Initialize target host environments
+ Deploy TiDB instance
- Deploy instance tikv -> 192.168.3.171:20163 ... Done
+ Copy certificate to remote host
+ Generate scale-out config
- Generate scale-out config tikv -> 192.168.3.171:20163 ... Done
+ Init monitor config
Enabling component tikv
Enabling instance 192.168.3.171:20163
Enable instance 192.168.3.171:20163 success
Enabling component node_exporter
Enabling instance 192.168.3.171
Enable 192.168.3.171 success
Enabling component blackbox_exporter
Enabling instance 192.168.3.171
Enable 192.168.3.171 success
+ [ Serial ] - Save meta
+ [ Serial ] - Start new instances
Starting component tikv
Starting instance 192.168.3.171:20163
Start instance 192.168.3.171:20163 success
Starting component node_exporter
Starting instance 192.168.3.171
Start 192.168.3.171 success
Starting component blackbox_exporter
Starting instance 192.168.3.171
Start 192.168.3.171 success
+ Refresh components conifgs
- Generate config pd -> 192.168.3.171:2379 ... Done
- Generate config tikv -> 192.168.3.171:20160 ... Done
- Generate config tikv -> 192.168.3.171:20161 ... Done
- Generate config tikv -> 192.168.3.171:20162 ... Done
- Generate config tikv -> 192.168.3.171:20163 ... Done
- Generate config tidb -> 192.168.3.171:4000 ... Done
- Generate config tiflash -> 192.168.3.171:9000 ... Done
- Generate config prometheus -> 192.168.3.171:9090 ... Done
- Generate config grafana -> 192.168.3.171:3000 ... Done
+ Reload prometheus and grafana
- Reload prometheus -> 192.168.3.171:9090 ... Done
- Reload grafana -> 192.168.3.171:3000 ... Done
+ [ Serial ] - UpdateTopology: cluster=demo-cluster
Scaled cluster `demo-cluster` out successfully
确认扩展后的拓扑结构。可以确认tikv-20163如预期般已添加。
[root@tisim ~]# tiup cluster display demo-cluster
tiup is checking updates for component cluster ...
Starting component `cluster`: /root/.tiup/components/cluster/v1.12.1/tiup-cluster display demo-cluster
Cluster type: tidb
Cluster name: demo-cluster
Cluster version: v7.0.0
Deploy user: tidb
SSH type: builtin
Dashboard URL: http://192.168.3.171:2379/dashboard
Grafana URL: http://192.168.3.171:3000
ID Role Host Ports OS/Arch Status Data Dir Deploy Dir
-- ---- ---- ----- ------- ------ -------- ----------
192.168.3.171:3000 grafana 192.168.3.171 3000 linux/x86_64 Up - /tidb-deploy/grafana-3000
192.168.3.171:2379 pd 192.168.3.171 2379/2380 linux/x86_64 Up|L|UI /tidb-data/pd-2379 /tidb-deploy/pd-2379
192.168.3.171:9090 prometheus 192.168.3.171 9090/12020 linux/x86_64 Up /tidb-data/prometheus-9090 /tidb-deploy/prometheus-9090
192.168.3.171:4000 tidb 192.168.3.171 4000/10080 linux/x86_64 Up - /tidb-deploy/tidb-4000
192.168.3.171:9000 tiflash 192.168.3.171 9000/8123/3930/20170/20292/8234 linux/x86_64 Up /tidb-data/tiflash-9000 /tidb-deploy/tiflash-9000
192.168.3.171:20160 tikv 192.168.3.171 20160/20180 linux/x86_64 Up /tidb-data/tikv-20160 /tidb-deploy/tikv-20160
192.168.3.171:20161 tikv 192.168.3.171 20161/20181 linux/x86_64 Up /tidb-data/tikv-20161 /tidb-deploy/tikv-20161
192.168.3.171:20162 tikv 192.168.3.171 20162/20182 linux/x86_64 Up /tidb-data/tikv-20162 /tidb-deploy/tikv-20162
192.168.3.171:20163 tikv 192.168.3.171 20163/20183 linux/x86_64 Up /tidb-data/tikv-20163 /tidb-deploy/tikv-20163
Total nodes: 9
强制停止 TiKV.
查看旅行表的地区信息。
mysql> SHOW TABLE trips REGIONS;
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS | SCHEDULING_CONSTRAINTS | SCHEDULING_STATE |
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| 3005 | 72000001 | t_94_r_182668 | 3008 | 2 | 3006, 3008, 3029 | 0 | 502 | 108038 | 32 | 252517 | | |
| 3013 | t_94_r_182668 | t_94_r_842765 | 3015 | 1 | 3014, 3015, 3016 | 0 | 0 | 0 | 99 | 641578 | | |
| 3017 | t_94_r_842765 | t_94_r_1245906 | 3019 | 1 | 3018, 3019, 3023 | 0 | 155 | 0 | 65 | 423135 | | |
| 3009 | t_94_r_1245906 | 78000000 | 3011 | 1 | 3011, 3012, 3025 | 0 | 140 | 0 | 108 | 530975 | | |
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
4 rows in set (0.01 sec)
我想要将 leader_store_id 为1的 TikV 设定为强制停止的目标。为了确定 leader_store_id 为1的 TiKV，我会检查 information_schema.tikv_store_status。
mysql> SELECT store_id, address FROM information_schema.tikv_store_status;
+----------+---------------------+
| store_id | address |
+----------+---------------------+
| 1 | 192.168.3.171:20160 |
| 2 | 192.168.3.171:20162 |
| 3 | 192.168.3.171:20161 |
| 114 | 192.168.3.171:3930 |
| 3021 | 192.168.3.171:20163 |
+----------+---------------------+
5 rows in set (0.00 sec)
根据上述的情况，我们发现 store_id 为 1 的 TiKV 的地址是 192.168.3.171:20160，因此我们将停止该进程。
[root@tisim ~]# ps -ef |grep tikv-20160|grep -v grep
tidb 1121 1 2 14:44 ? 00:04:06 bin/tikv-server --addr 0.0.0.0:20160 --advertise-addr 192.168.3.171:20160 --status-addr 0.0.0.0:20180 --advertise-status-addr 192.168.3.171:20180 --pd 192.168.3.171:2379 --data-dir /tidb-data/tikv-20160 --config conf/tikv.toml --log-file /tidb-deploy/tikv-20160/log/tikv.log
[root@tisim ~]# kill -9 1121
我們將檢查狀態。tikv-20160的狀態已經變為斷開連接。
[root@tisim ~]# tiup cluster display demo-cluster
(省略)
ID Role Host Ports OS/Arch Status Data Dir Deploy Dir
-- ---- ---- ----- ------- ------ -------- ----------
192.168.3.171:20160 tikv 192.168.3.171 20160/20180 linux/x86_64 Disconnected /tidb-data/tikv-20160 /tidb-deploy/tikv-20160
192.168.3.171:20161 tikv 192.168.3.171 20161/20181 linux/x86_64 Up /tidb-data/tikv-20161 /tidb-deploy/tikv-20161
192.168.3.171:20162 tikv 192.168.3.171 20162/20182 linux/x86_64 Up /tidb-data/tikv-20162 /tidb-deploy/tikv-20162
192.168.3.171:20163 tikv 192.168.3.171 20163/20183 linux/x86_64 Up /tidb-data/tikv-20163 /tidb-deploy/tikv-20163
当重新确认区域信息时，可以从leader_id的值中确认leader_store_id已被更改为非1，并且一个作为副本的区域已被提升为领导者。
mysql> SHOW TABLE trips REGIONS;
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS | SCHEDULING_CONSTRAINTS | SCHEDULING_STATE |
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
| 3005 | 72000001 | t_94_r_182668 | 3008 | 2 | 3006, 3008, 3029 | 0 | 635 | 109725 | 32 | 252517 | | |
| 3013 | t_94_r_182668 | t_94_r_842765 | 3014 | 3 | 3014, 3015, 3016 | 0 | 39 | 0 | 102 | 665224 | | |
| 3017 | t_94_r_842765 | t_94_r_1245906 | 3023 | 3021 | 3018, 3019, 3023 | 0 | 39 | 0 | 62 | 403141 | | |
| 3009 | t_94_r_1245906 | 78000000 | 3012 | 2 | 3011, 3012, 3025 | 0 | 0 | 558135 | 79 | 505124 | | |
+-----------+----------------+----------------+-----------+-----------------+------------------+------------+---------------+------------+----------------------+------------------+------------------------+------------------+
4 rows in set (0.01 sec)
顺便提一下，之前终止的tikv-20160已经自动恢复了，以下是/var/log/messages的摘录。
May 13 17:58:41 tisim systemd: tikv-20160.service: main process exited, code=killed, status=9/KILL
May 13 17:58:41 tisim systemd: Unit tikv-20160.service entered failed state.
May 13 17:58:41 tisim systemd: tikv-20160.service failed.
May 13 17:58:56 tisim systemd: tikv-20160.service holdoff time over, scheduling restart.
May 13 17:58:56 tisim systemd: Stopped tikv service.
May 13 17:58:56 tisim systemd: Started tikv service.
最后
希望本文章能帮助您，我们在本文中确认了以下内容。
-
- SHOW TABLE REGIONS文を使用したリージョン情報
-
- tiupを使用したTiKVのスケールアウト
- TiKV異常終了時のリージョン再配置
