MySQL InnoDB Cluster环境搭建和简单测试

InnoDB Cluster初印象

  记得MySQL Group Replicatioin 刚开始的时候，MySQL界很是轰动，等待了多年，终于有了官方的这个高可用解决方案。你要说还有一些方案补充，比如MySQL Cluster,MySQL Proxy,这些的使用率个人感觉还是不高，也就是经受的考验还不够，原因有很多，就不赘述了。

   不久，我和一个MySQL DBA有了下面的一个基本对话。

   我： MySQL GR GA之后，里面的自动切换功能确实很赞，能够做到读写分离，原本MHA的方案现在MGR也可以做了。

   MySQL DBA:如果数据库发生了故障，这个自动切换的过程，其实对于应用不是透明的，因为读写节点相当于漂移到了另外一台服务器上，除非再做个中间件。

   我：单纯MGR目前还做不了这个，它目前只是保证数据库层面的这种切换和高可用。

   MySQL DBA:所以说MGR的企业级应用还是需要一些辅助，这样才算是一个完整的解决方案。

    不久，MySQL InnoDB Cluster推出，我觉得这个方案想比原来的MGR更进一步，说实话，我很看好这个方案，尽管目前愿意真正去用的用户确实不多。

    如果你看一下官方的这个架构图，就会发现，MGR本身就是Innodb Cluster的一部分，还有两个组件，MySQL Shell,MySQL Router,这三板斧就是InnoDB Cluster的一个核心组件，而正如我之前所说，可以看到MySQL的一个格局和定位，他正在很努力去解决以前诟病的问题。

安装前先保证Python满足要求

    要安装InnoDB Cluster,环境的一个基本要求就是Python,我看了下，很多默认的系统版本是2.6，而它的最低要求是2.7及以上，所以还是需要提前准备下这个部分。

    如果你的系统是Python 2.6版本的，可以考虑升级到2.7，参考如下的方法。

    下载安装包，部署

 wget http://python.org/ftp/python/2.7/Python-2.7.tar.bz2     --no-check-certificate
./configure   
make all   
make install   
make clean   
make distclean   
查看Python的版本
# /usr/local/bin/python2.7 -V
Python 2.7
做基本的环境设置，替换旧的Python
mv /usr/bin/python /usr/bin/python2.6
ln -s /usr/local/bin/python2.7 /usr/bin/python

sandbox安装部署InnoDB Cluster

搭建InnoDB Cluster显而易见需要多台服务器，而如果在一台服务器上练习测试，也是全然没有问题，如果想更快更方便的测试模拟，还可以使用sandbox来做，首先你得有sandbox，接着InnoDB Cluster的三大组件是MGR,MySQL Shell,MySQL Router,所以你可以从官网直接下载下来。

然后我们开启安装之旅。

使用MySQL Shell的命令mysqlsh开始部署，创建一个端口为3310的实例

mysql-js> dba.deploySandboxInstance(3310)
A new MySQL sandbox instance will be created on this host in
/root/mysql-sandboxes/3310

输入密码之后，一个3310端口的MySQL服务就启动了。
Please enter a MySQL root password for the new instance:
Deploying new MySQL instance...
Instance localhost:3310 successfully deployed and started.
Use shell.connect('root@localhost:3310'); to connect to the instance.
接着创建另外两个节点 3320,3330
dba.deploySandboxInstance(3320)
dba.deploySandboxInstance(3330)
我们切换到3310的MySQL实例，准备开始创建Cluster
mysql-js>  \connect root@localhost:3310
Creating a Session to 'root@localhost:3310'
Enter password:
Closing old connection...
Classic Session successfully established. No default schema selected.

定义一个Cluster变量，节点1就开启了Cluster创建之旅，可以从下面的信息看出，至少需要3个节点

mysql-js>  var cluster = dba.createCluster('testCluster')
A new InnoDB cluster will be created on instance 'root@localhost:3310'.
Creating InnoDB cluster 'testCluster' on 'root@localhost:3310'...
Adding Seed Instance...
Cluster successfully created. Use Cluster.addInstance() to add MySQL instances.
At least 3 instances are needed for the cluster to be able to withstand up to  one server failure.
接着把另外两个节点加入进来，先加入端口为3320的节点
mysql-js> cluster.addInstance('root@localhost:3320')
A new instance will be added to the InnoDB cluster. Depending on the amount of
data on the cluster this might take from a few seconds to several hours.
Please provide the password for 'root@localhost:3320':
Adding instance to the cluster ...
加入端口为3330的节点，日志和节点2相似。
mysql-js> cluster.addInstance('root@localhost:3330')
这个时候Cluster就创建好了。

这个时候，我们再配置一下MySQL Router，创建个软链接，保证能够正常调用。

# ln -s /home/innodb_cluster/mysql-router-2.1.3-linux-glibc2.12-x86-64bit/bin/mysqlrouter   /usr/bin/mysqlroute
# which mysqlroute
/usr/bin/mysqlroute
配置MySQL Router的启动节点为端口3310的实例
# mysqlrouter --bootstrap root@localhost:3310 --user=mysql

这个时候还是要输入密码，成功之后，这个绑定就打通了。

Please enter MySQL password for root:
Bootstrapping system MySQL Router instance...
MySQL Router  has now been configured for the InnoDB cluster 'testCluster'.
The following connection information can be used to connect to the cluster.
Classic MySQL protocol connections to cluster 'testCluster':
- Read/Write Connections: localhost:6446
- Read/Only Connections: localhost:6447
X protocol connections to cluster 'testCluster':
- Read/Write Connections: localhost:64460
- Read/Only Connections: localhost:64470
可以从上面的日志看出来，分配的读写端口是6446，只读端口是6447，还有x协议连接的端口为64460,64470
启动MySQL Router
# mysqlrouter &
[1] 2913

如果对MySQL Router还有些疑问，可以看看安装目录下，会生成下面的配置文件，我们就看里面的.conf文件，里面的一部分内容如下：
[routing:testCluster_default_rw]
bind_address=0.0.0.0
bind_port=6446
destinations=metadata-cache://testCluster/default?role=PRIMARY
mode=read-write
protocol=classic

验证测试

我们尝试使用6446来连接登录，这个时候就通过MySQL Shell开启了连接入口，MySQL Router做了转接，连接到了里面的读写节点3310

# mysqlsh --uri root@localhost:6446
Creating a Session to 'root@localhost:6446'
Enter password:
Classic Session successfully established. No default schema selected.
Welcome to MySQL Shell 1.0.9
切换到sql模式，查看端口就知道是哪个节点了。
mysql-js> \sql
Switching to SQL mode... Commands end with ;
mysql-sql> select @@port;
+--------+
| @@port |
+--------+
|   3310 |
+--------+
1 row in set (0.00 sec)
如果切换为脚本模式查看实例的状态，可以使用里面定义的API来做，输出都是JSON串。
mysql-js> dba.configureLocalInstance('root@127.0.0.1:3310')
Please provide the password for 'root@127.0.0.1:3310':
Detected as sandbox instance.
Validating MySQL configuration file at: /root/mysql-sandboxes/3310/my.cnf
Validating instance...
The instance '127.0.0.1:3310' is valid for Cluster usage
You can now use it in an InnoDB Cluster.
{
    "status": "ok"
}
如果查看Cluster的信息，可以看到下面的读写节点，只读节点的状态信息
mysql-js> dba.getCluster()
<Cluster:testCluster>

var cluster = dba.getCluster()

得到Cluster的信息

mysql-js> cluster.status()
{
    "clusterName": "testCluster",
    "defaultReplicaSet": {
        "name": "default",
        "primary": "localhost:3310",
        "status": "OK",
        "statusText": "Cluster is ONLINE and can tolerate up to ONE failure.",
        "topology": {
            "localhost:3310": {
                "address": "localhost:3310",
                "mode": "R/W",
                "readReplicas": {},
                "role": "HA",
                "status": "ONLINE"
            },
            "localhost:3320": {
                "address": "localhost:3320",
                "mode": "R/O",
                "readReplicas": {},
                "role": "HA",
                "status": "ONLINE"
            },
            "localhost:3330": {
                "address": "localhost:3330",
                "mode": "R/O",
                "readReplicas": {},
                "role": "HA",
                "status": "ONLINE"
            }
        }
    }
}
也可以使用describe得到一些基本的信息
mysql-js> cluster.describe();
{
    "clusterName": "testCluster",
    "defaultReplicaSet": {
        "instances": [
            {
                "host": "localhost:3310",
                "label": "localhost:3310",
                "role": "HA"
            },
            {
                "host": "localhost:3320",
                "label": "localhost:3320",
                "role": "HA"
            },
            {
                "host": "localhost:3330",
                "label": "localhost:3330",
                "role": "HA"
            }
        ],
        "name": "default"
    }
}

切换测试

当然光看不练还是假把式，我们切换一下，看看好使不？

模拟一个节点出现问题，可以使用killSandboxInstance方法。
mysql-js> dba.killSandboxInstance(3310)
The MySQL sandbox instance on this host in
/root/mysql-sandboxes/3310 will be killed
Killing MySQL instance...
Instance localhost:3310 successfully killed.

节点被清理了，没有任何进程存在。
# ps -ef|grep mysql|grep 3310
#
我们还是使用6446的端口来统一连接，这个时候就切换到了端口3320的MySQL服务
# mysqlsh --uri root@localhost:6446
Creating a Session to 'root@localhost:6446'
Enter password:
Classic Session successfully established. No default schema selected.
Welcome to MySQL Shell 1.0.9
mysql-js> \sql
Switching to SQL mode... Commands end with ;
mysql-sql> select @@port;
+--------+
| @@port |
+--------+
|   3320 |
+--------+
1 row in set (0.00 sec)

所以切换的部分没有问题，我们再次把“迷失”的节点启动起来。
# mysqlsh --uri root@localhost:6446
mysql-js> dba.startSandboxInstance(3310)
The MySQL sandbox instance on this host in
/root/mysql-sandboxes/3310 will be started
Starting MySQL instance...
Instance localhost:3310 successfully started.
这个时候再次查看Cluster的状态，3320就是主了，3310就是只读节点了。
mysql-js> dba.getCluster()
<Cluster:testCluster>

把节点2纳入到Cluster中

mysql-js> cluster.rejoinInstance('root@localhost:3310')
Rejoining the instance to the InnoDB cluster. Depending on the original
problem that made the instance unavailable, the rejoin operation might not be
successful and further manual steps will be needed to fix the underlying
problem.
Please monitor the output of the rejoin operation and take necessary action if
the instance cannot rejoin.
Please provide the password for 'root@localhost:3310':
Rejoining instance to the cluster ...
The instance 'root@localhost:3310' was successfully rejoined on the cluster.
The instance 'localhost:3310' was successfully added to the MySQL Cluster.
mysql-js>

可以想象如果是一个生产系统，这么多的日志，这个过程真是让人纠结。
最后来一个切换后的Cluster状态
mysql-js> cluster.status()
{
    "clusterName": "testCluster",
    "defaultReplicaSet": {
        "name": "default",
        "primary": "localhost:3320",
        "status": "OK",
        "statusText": "Cluster is ONLINE and can tolerate up to ONE failure.",
        "topology": {
            "localhost:3310": {
                "address": "localhost:3310",
                "mode": "R/O",
                "readReplicas": {},
                "role": "HA",
                "status": "ONLINE"
            },
            "localhost:3320": {
                "address": "localhost:3320",
                "mode": "R/W",
                "readReplicas": {},
                "role": "HA",
                "status": "ONLINE"
            },
            "localhost:3330": {
                "address": "localhost:3330",
                "mode": "R/O",
                "readReplicas": {},
                "role": "HA",
                "status": "ONLINE"
            }
        }
    }
}