mongodb 复制集
mongodb的复制集，大致就是在主从复制的基础上，加入了高可用控制，类似高可用mysql集群
在主节点出现问题后，从节点会重新找出来一个主节点，系统自动完成，mysql如果也这样就完美了，mysql弄这个好复杂
1.
在linux 下面做3个账号连接，然后依次开启3个mongodb的进程（非守护进行，守护进程只能开启一个）
2.
/usr/local/mongodb2.6/bin/mongod --dbpath=/mongodb/data1 -logpath=/mongodb/log1/log.log -port=10001 --replSet terry /usr/local/mongodb2.6/bin/mongod --dbpath=/mongodb/data2 -logpath=/mongodb/log2/log.log -port=10002 --replSet terry /usr/local/mongodb2.6/bin/mongod --dbpath=/mongodb/data3 -logpath=/mongodb/log3/log.log -port=10003 --replSet terry
查看是否运行成功：
[root@grande ~]# ps -ef | grep mongo root 2345 1 0 Nov27 ? 01:34:52 /usr/local/mongodb2.6/bin/mongod --dbpath=/usr/local/mongodb2.6/data -logpath=/usr/local/mongodb2.6/log -fork -port=27017 root 25111 24779 0 09:13 pts/1 00:00:12 /usr/local/mongodb2.6/bin/mongod --dbpath=/mongodb/data2 -logpath=/mongodb/log2/log.log -port=10002 --replSet terry root 25157 24679 0 09:13 pts/0 00:00:12 /usr/local/mongodb2.6/bin/mongod --dbpath=/mongodb/data1 -logpath=/mongodb/log1/log.log -port=10001 --replSet terry root 25203 24809 0 09:14 pts/2 00:00:11 /usr/local/mongodb2.6/bin/mongod --dbpath=/mongodb/data3 -logpath=/mongodb/log3/log.log -port=10003 --replSet terry root 25341 24854 0 09:17 pts/4 00:00:00 mongo 10.10.10.252:10001/admin root 25895 25288 0 09:38 pts/5 00:00:00 mongo 10.10.10.252:10002 root 26089 26069 0 09:44 pts/6 00:00:00 grep mongo
3. 通过上面的运行，我们还需要初始化：
mongo 10.10.10.252:10001/admin
db.runCommand({"replSetInitiate":{
"_id":"terry",
"members":[
{
"_id":1,
"host":"10.10.10.252:10001"
},
{
"_id":2,
"host":"10.10.10.252:10002"
}
]
}
}
)
如果成功： 显示
{
"info" : "Config now saved locally. Should come online in about a minute.",
"ok" : 1
}
显示复制集状态：
rs.status()
> rs.status()
{
"set" : "terry",
"date" : ISODate("2015-12-08T01:17:08Z"),
"myState" : 1,
"members" : [
{
"_id" : 1,
"name" : "10.10.10.252:10001",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 192,
"optime" : Timestamp(1449537265, 1),
"optimeDate" : ISODate("2015-12-08T01:14:25Z"),
"electionTime" : Timestamp(1449537274, 1),
"electionDate" : ISODate("2015-12-08T01:14:34Z"),
"self" : true
},
{
"_id" : 2,
"name" : "10.10.10.252:10002",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 162,
"optime" : Timestamp(1449537265, 1),
"optimeDate" : ISODate("2015-12-08T01:14:25Z"),
"lastHeartbeat" : ISODate("2015-12-08T01:17:06Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T01:17:06Z"),
"pingMs" : 0,
"syncingTo" : "10.10.10.252:10001"
}
],
"ok" : 1
}
把10003添加到副本集： mongo 10.10.10.252:10001/admin （仲裁服务器）
rs.addArb("10.10.10.252:10003")
terry:PRIMARY> rs.addArb("10.10.10.252:10003")
{ "ok" : 1 }
terry:PRIMARY> rs.status
function () { return db._adminCommand("replSetGetStatus"); }
terry:PRIMARY> rs.status()
{
"set" : "terry",
"date" : ISODate("2015-12-08T01:18:29Z"),
"myState" : 1,
"members" : [
{
"_id" : 1,
"name" : "10.10.10.252:10001",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 273,
"optime" : Timestamp(1449537501, 1),
"optimeDate" : ISODate("2015-12-08T01:18:21Z"),
"electionTime" : Timestamp(1449537274, 1),
"electionDate" : ISODate("2015-12-08T01:14:34Z"),
"self" : true
},
{
"_id" : 2,
"name" : "10.10.10.252:10002",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 243,
"optime" : Timestamp(1449537501, 1),
"optimeDate" : ISODate("2015-12-08T01:18:21Z"),
"lastHeartbeat" : ISODate("2015-12-08T01:18:28Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T01:18:28Z"),
"pingMs" : 0,
"syncingTo" : "10.10.10.252:10001"
},
{
"_id" : 3,
"name" : "10.10.10.252:10003",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 8,
"lastHeartbeat" : ISODate("2015-12-08T01:18:27Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T01:18:28Z"),
"pingMs" : 0
}
],
"ok" : 1
}
查看到3个进程的mongodb的复制集完成，如果添加其他的，还可以陆续添加
4. 执行查询会报错：
但此时在从库中查询报如下错误：
这是正常的现象，对于SECONDARY节点默认是不可读的。因为SECONDARY是不允许读和写的，在写多读少的应用中，使用Replica Sets来实现读写分离。通过在连接时指定或者在主库指定slaveOk，由SECONDARY来分担读的压力，PRIMARY只承担些操作。
解决方法：
第一种：在从节点设置slaveOk（）：
rs.slaveOk();
第二种：在主节点设置：
db.getMongo().setSlaveOk();
这样就可以查询了。
5. 增加新节点：
rs.add("10.10.10.252:10004");
6.删除新增节点：
terry:PRIMARY> rs.remove("10.10.10.252:10004");
2015-12-08T10:30:15.336+0800 DBClientCursor::init call() failed
2015-12-08T10:30:15.339+0800 Error: error doing query: failed at src/mongo/shell/query.js:81
2015-12-08T10:30:15.341+0800 trying reconnect to 10.10.10.252:10001 (10.10.10.252) failed
2015-12-08T10:30:15.341+0800 reconnect 10.10.10.252:10001 (10.10.10.252) ok
terry:PRIMARY> rs.status();
{
"set" : "terry",
"date" : ISODate("2015-12-08T02:30:37Z"),
"myState" : 1,
"members" : [
{
"_id" : 1,
"name" : "10.10.10.252:10001",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 4601,
"optime" : Timestamp(1449541815, 1),
"optimeDate" : ISODate("2015-12-08T02:30:15Z"),
"electionTime" : Timestamp(1449537274, 1),
"electionDate" : ISODate("2015-12-08T01:14:34Z"),
"self" : true
},
{
"_id" : 2,
"name" : "10.10.10.252:10002",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 22,
"optime" : Timestamp(1449541815, 1),
"optimeDate" : ISODate("2015-12-08T02:30:15Z"),
"lastHeartbeat" : ISODate("2015-12-08T02:30:37Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T02:30:36Z"),
"pingMs" : 0,
"lastHeartbeatMessage" : "syncing to: 10.10.10.252:10001",
"syncingTo" : "10.10.10.252:10001"
},
{
"_id" : 3,
"name" : "10.10.10.252:10003",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 22,
"lastHeartbeat" : ISODate("2015-12-08T02:30:37Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T02:30:36Z"),
"pingMs" : 0
}
],
"ok" : 1
}
7.测试切换
主节点停掉：
terry:PRIMARY> rs.status();
{
"set" : "terry",
"date" : ISODate("2015-12-08T03:03:14Z"),
"myState" : 1,
"members" : [
{
"_id" : 1,
"name" : "10.10.10.252:10001",
"health" : 0,
"state" : 8,
"stateStr" : "(not reachable/healthy)",
"uptime" : 0,
"optime" : Timestamp(1449541815, 1),
"optimeDate" : ISODate("2015-12-08T02:30:15Z"),
"lastHeartbeat" : ISODate("2015-12-08T03:03:13Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T02:32:45Z"),
"pingMs" : 0
},
{
"_id" : 2,
"name" : "10.10.10.252:10002",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 6565,
"optime" : Timestamp(1449543736, 1),
"optimeDate" : ISODate("2015-12-08T03:02:16Z"),
"electionTime" : Timestamp(1449541974, 1),
"electionDate" : ISODate("2015-12-08T02:32:54Z"),
"self" : true
},
{
"_id" : 3,
"name" : "10.10.10.252:10003",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 1979,
"lastHeartbeat" : ISODate("2015-12-08T03:03:12Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T03:03:13Z"),
"pingMs" : 0
},
{
"_id" : 4,
"name" : "10.10.10.252:10004",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 58,
"optime" : Timestamp(1449543736, 1),
"optimeDate" : ISODate("2015-12-08T03:02:16Z"),
"lastHeartbeat" : ISODate("2015-12-08T03:03:12Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T03:03:12Z"),
"pingMs" : 0,
"lastHeartbeatMessage" : "syncing to: 10.10.10.252:10002",
"syncingTo" : "10.10.10.252:10002"
}
],
"ok" : 1
}
可以看到 id为1的状态为"stateStr" : "(not reachable/healthy)",
id为2的变为PRIMARY
在yii2下链接是没有问题的
关于副本集的yii2的配置参看：
http://blog.csdn.net/terry_water/article/details/50215887
重新启动id为1的mongo：
terry:PRIMARY> rs.status();
{
"set" : "terry",
"date" : ISODate("2015-12-08T03:06:09Z"),
"myState" : 1,
"members" : [
{
"_id" : 1,
"name" : "10.10.10.252:10001",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 4,
"optime" : Timestamp(1449541815, 1),
"optimeDate" : ISODate("2015-12-08T02:30:15Z"),
"lastHeartbeat" : ISODate("2015-12-08T03:06:07Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T03:06:08Z"),
"pingMs" : 52
},
{
"_id" : 2,
"name" : "10.10.10.252:10002",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 6740,
"optime" : Timestamp(1449543736, 1),
"optimeDate" : ISODate("2015-12-08T03:02:16Z"),
"electionTime" : Timestamp(1449541974, 1),
"electionDate" : ISODate("2015-12-08T02:32:54Z"),
"self" : true
},
{
"_id" : 3,
"name" : "10.10.10.252:10003",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 2154,
"lastHeartbeat" : ISODate("2015-12-08T03:06:08Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T03:06:09Z"),
"pingMs" : 0
},
{
"_id" : 4,
"name" : "10.10.10.252:10004",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 233,
"optime" : Timestamp(1449543736, 1),
"optimeDate" : ISODate("2015-12-08T03:02:16Z"),
"lastHeartbeat" : ISODate("2015-12-08T03:06:08Z"),
"lastHeartbeatRecv" : ISODate("2015-12-08T03:06:08Z"),
"pingMs" : 0,
"syncingTo" : "10.10.10.252:10002"
}
],
"ok" : 1
}
可以看到启动成功了
副本集是强大！！！！配置如此简单。