MySQL Cluster 开发环境简明部署(中文)

主要内容

1. 获取; 2. 安装; 3. 配置; 4. 运行; 5. 测试; 6. 停止; 7. 总结

1. 获取

这个不必多言吧，到 MySQL 网站上面下载就是了。MySQL Cluster 的英文部署测试手册很简明，这里和其内容基本一样。喜欢读英文版的朋友可以直接下载，略过本文。

2. 安装

先解压，然后创建一个 symbolic link：

michael@linux:~$ tar xvf mysql-cluster-gpl-7.2.4-linux2.6-x86_64.tar  michael@linux:~$ ln -s mysql-cluster-gpl-7.2.4-linux2.6-x86_64 mysqlc

如果你愿意，可以将~/mysqlc/bin加入到你的 path 里，方便使用。

3. 配置

作为在开发环境上第一次部署，还是以测试为主要目的。一个完整的 MySQL Cluster 由 MySQL Server，Data Nodes，Management Node 三部分组成。首先我们为它们创建一些必须的目录：

michael@linux:~$ mkdir mysql-cluster  michael@linux:~$ cd mysql-cluster  michael@linux:~$ mkdir conf ndb_data mysqld_data

然后在conf目录下创建如下两个文件，分别是config.ini和my.cnf，内容如下：

config.ini

用于 MySQL Server 的配置，端口号port根据你自己的情况设定。

[mysqld]  ndbcluster  datadir=/home/user1/my_cluster/mysqld_data  basedir=/home/user1/mysqlc  port=5050

my.cnf

该文件用于配置各结点的 NodeId 和 Data Nodes 与 Management Node 的数据目录。

[ndb_mgmd]  hostname=localhost  datadir=/home/user1/my_cluster/ndb_data  NodeId=1    [ndbd default]  noofreplicas=2  datadir=/home/user1/my_cluster/ndb_data    [ndbd]  hostname=localhost  NodeId=3    [ndbd]  hostname=localhost  NodeId=4    [mysqld]  NodeId=50

这时你的目录结构应该如下：

~  +-- /mysql-cluster-gpl-7.2.4-linux2.6-x86_64  +-- /mysqlc -> mysql-cluster-gpl-7.2.4-linux2.6-x86_64  +-- /mysql-cluster      +-- /conf      +-- /ndb_data      +-- /mysqld_data

4. 运行

MySQL Cluster 的启动顺序是有要求的，如下：

Management Node
Data Nodes
MySQL Server

命令如下：

michael@linux:~$ cd ../mysql-cluster  michael@linux:~/mysql-cluster$ $HOME/mysqlc/bin/ndb_mgmd -f conf/config.ini --initial --configdir=$HOME/mysql-cluster/conf/  michael@linux:~/mysql-cluster$ $HOME/mysqlc/bin/ndbd -c localhost:1186  michael@linux:~/mysql-cluster$ $HOME/mysqlc/bin/ndbd -c localhost:1186

检查已经启动的结点的状态，命令为：

poecahnt@linux:~$ $HOME/mysqlc/bin/ndb_mgm -e show

输出如下：

Connected to Management Server at: localhost:1186  Cluster Configuration  ---------------------  [ndbd(NDB)]     2 node(s)  id=3    @127.0.0.1  (mysql-5.5.19 ndb-7.2.4, Nodegroup: 0, Master)  id=4    @127.0.0.1  (mysql-5.5.19 ndb-7.2.4, Nodegroup: 0)    [ndb_mgmd(MGM)] 1 node(s)  id=1    @127.0.0.1  (mysql-5.5.19 ndb-7.2.4)    [mysqld(API)]   1 node(s)  id=50   @127.0.0.1  (mysql-5.5.19 ndb-7.2.4)

表示已经可以启动 MySQL Server 了。最后启动 MySQL Server，命令为：

michael@linux:~/mysql-cluseter$ $HOME/mysqlc/bin/mysqld --defaults-file=conf/my.cnf &

输出信息如下：

120223 15:29:02 InnoDB: The InnoDB memory heap is disabled  120223 15:29:02 InnoDB: Mutexes and rw_locks use GCC atomic builtins  120223 15:29:02 InnoDB: Compressed tables use zlib 1.2.3  120223 15:29:02 InnoDB: Using Linux native AIO  120223 15:29:02 InnoDB: Initializing buffer pool, size = 128.0M  120223 15:29:02 InnoDB: Completed initialization of buffer pool  120223 15:29:02 InnoDB: highest supported file format is Barracuda.  120223 15:29:02  InnoDB: Waiting for the background threads to start  120223 15:29:03 InnoDB: 1.1.8 started; log sequence number 1595675  120223 15:29:04 [Note] NDB: NodeID is 50, management server 'localhost:1186'  120223 15:29:04 [Note] NDB[0]: NodeID: 50, all storage nodes connected  120223 15:29:04 [Warning] NDB: server id set to zero - changes logged to bin log with   server id zero will be logged with another server id by slave mysqlds  120223 15:29:04 [Note] Starting Cluster Binlog Thread  120223 15:29:04 [Note] Event Scheduler: Loaded 0 events  120223 15:29:04 [Note] $HOME/mysqlc/bin/mysqld: ready for connections.  Version: '5.5.19-ndb-7.2.4-gpl'  socket: '/tmp/mysql.sock'  port: 5050  MySQL Cluster Community Server (GPL)  120223 15:29:05 [Note] NDB: Creating mysql.ndb_schema  120223 15:29:08 [Note] NDB Binlog: CREATE TABLE Event: REPL$mysql/ndb_schema  120223 15:29:09 [Note] NDB Binlog: logging ./mysql/ndb_schema (UPDATED,USE_WRITE)  120223 15:29:09 [Note] NDB: Creating mysql.ndb_apply_status  120223 15:29:09 [Note] NDB Binlog: CREATE TABLE Event: REPL$mysql/ndb_apply_status  120223 15:29:09 [Note] NDB Binlog: logging ./mysql/ndb_apply_status (UPDATED,USE_WRITE)  120223 15:29:09 [Note] NDB: missing frm for mysql.ndb_index_stat_sample, discovering...  120223 15:29:09 [Note] NDB: missing frm for mysql.ndb_index_stat_head, discovering...  2012-02-23 15:29:10 [NdbApi] INFO     -- Flushing incomplete GCI:s < 579/14  2012-02-23 15:29:10 [NdbApi] INFO     -- Flushing incomplete GCI:s < 579/14  120223 15:29:10 [Note] NDB Binlog: starting log at epoch 579/14  120223 15:29:10 [Note] NDB Binlog: ndb tables writable

5. 测试

连接 MySQL Server 进行测试，确认可以用ndb存储引擎来创建数据库中的表，如下：

michael@linux:~$ $HOME/mysqlc/bin/mysql -h 127.0.0.1 -P 5050  mysql> create database clusterdb;  mysql> use clusterdb;  mysql> insert into simples values (1),(2),(3),(4);  mysql> select * from simples;            +----+          | id |          +----+          |  3 |          |  1 |          |  2 |          |  4 |          +----+

6. 停止

MySQL Cluster 必须手动停止，Data Nodes 可以用 ndb_mgm 来停止：

michael@linux:~$ $HOME/mysqlc/bin/mysqladmin -h 127.0.0.1 -P 5050 shutdown

如果提示：

/home/michael/mysqlc/bin/mysqladmin: shutdown failed; error: 'Access denied; you need (at least one of) the SHUTDOWN privilege(s) for this operation'

则在shutdown命令前加上sudo。

michael@linux:~$ $HOME/mysqlc/bin/ndb_mgm -e shutdown

正常停止的信息类似如下：

120223 16:44:11 [Note] /home/michael/mysqlc/bin/mysqld: Normal shutdown    michael@linux:~/mysql-cluster$ 120223 16:44:11 [Note] Event Scheduler: Purging the queue. 0 events  120223 16:44:13 [Warning] /home/michael/mysqlc/bin/mysqld: Forcing close of thread 2  user: 'michael'    120223 16:44:13 [Note] Stopping Cluster Utility thread  120223 16:44:13 [Note] Stopping Cluster Index Stats thread  120223 16:44:13 [Note] Stopping Cluster Binlog  120223 16:44:13 [Note] Stopping Cluster Index Statistics thread  120223 16:44:14  InnoDB: Starting shutdown...  120223 16:44:15  InnoDB: Shutdown completed; log sequence number 1595675  120223 16:44:15 [Note] /home/michael/mysqlc/bin/mysqld: Shutdown complete