MongoDB Replica Sets（复制集）部署
复制集和副本集有所区别，复制集是有自动故障恢复功能的主从集群。在复制集中，当主节点出现故障而宕机，集群会立刻通过选举，选出新的主节点，保证整个集群继续可用。
复制集具有如下优点：
(1) 故障自动恢复
(2) 高可用性
(3) 数据安全
复制集正常情况下:
当主节点出现故障，在从节点选举出新的主节点
复制集的部署
环境说明：
系统：Windows 10专业版
集群规模：三个节点，一个主节点两个从节点，一般集群节点个数为奇数个。
注意：
由于资源有限，只在一台主机上开启3个mongod进程模拟3个节点。在实际生产中，最好一个主机上运行一个节点，以保证集群的性能。
为了便于测试直接将日志输出到终端。在实际生产中，将日志输出到日志文件，便于分析。
约定：
| 集群rs1 |
IP |
Port |
Dbpath |
Logpath |
| 节点1 |
127.0.0.1 |
27017 |
C:\data\db |
终端 |
| 节点2 |
127.0.0.1 |
27018 |
D:\data\db |
终端 |
| 节点3 |
127.0.0.1 |
27019 |
E:\data\db |
终端 |
在启动集群之前，保证数据目录存在，且目录为空。
启动mongod进程：
mongod --dbpath c:\data\db --bind_ip127.0.0.1 --port 27017 --replSet rs1
mongod --dbpath D:\data\db --bind_ip127.0.0.1 --port 27018 --replSet rs1
mongod --dbpath E:\data\db --bind_ip127.0.0.1 --port 27019 --replSet rs1
--replSet指定复制集名，三个节点属于同一个复制集，指定为同一个名称。
登陆任一节点，配置复制集，
mongo 127.0.0.1:27017
编写配置对象：
> config={
... _id:'rs1',
... members:[
... {_id:1,host:'127.0.0.1:27017'},
... {_id:2,host:'127.0.0.1:27018'},
... {_id:3,host:'127.0.0.1:27019'}
... ]
... }
{
"_id" : "rs1",
"members" : [
{
"_id" : 1,
"host" :"127.0.0.1:27017"
},
{
"_id" : 2,
"host" :"127.0.0.1:27018"
},
{
"_id" : 3,
"host" :"127.0.0.1:27019"
}
]
}
加载配置：
> rs.initiate(config)
{ "ok" : 1 }
rs1:OTHER>
rs1:PRIMARY> rs.isMaster()
{
"hosts" : [
"127.0.0.1:27017",
"127.0.0.1:27018",
"127.0.0.1:27019"
],
"setName" : "rs1",
"setVersion" : 1,
"ismaster" : true,
"secondary" : false,
"primary" : "127.0.0.1:27017",
"me" : "127.0.0.1:27017",
"electionId" : ObjectId("7fffffff0000000000000001"),
"lastWrite" : {
"opTime" : {
"ts" :Timestamp(1498831736, 1),
"t" :NumberLong(1)
},
"lastWriteDate" :ISODate("2017-06-30T14:08:56Z")
},
"maxBsonObjectSize" : 16777216,
"maxMessageSizeBytes" : 48000000,
"maxWriteBatchSize" : 1000,
"localTime" : ISODate("2017-06-30T14:09:05.756Z"),
"maxWireVersion" : 5,
"minWireVersion" : 0,
"readOnly" : false,
"ok" : 1
}
rs1:PRIMARY>
此节点是一个主节点。
连接另一个节点：
>mongo 127.0.0.1:27018
MongoDB shell version v3.4.5
connecting to: 127.0.0.1:27018
MongoDB server version: 3.4.5
Server has startup warnings:
2017-06-30T21:48:39.401+0800 I CONTROL [initandlisten]
2017-06-30T21:48:39.404+0800 I CONTROL [initandlisten] ** WARNING: Access control isnot enabled for the database.
2017-06-30T21:48:39.407+0800 I CONTROL [initandlisten] ** Read and write access to data and configurationis unrestricted.
2017-06-30T21:48:39.410+0800 I CONTROL [initandlisten]
rs1:SECONDARY>
可以看到这是一个从节点。
在主节点创建数据库和集合并插入数据：
rs1:PRIMARY> show dbs
admin 0.000GB
local 0.000GB
rs1:PRIMARY> use test
switched to db test
rs1:PRIMARY>db.user.insert({name:'gyb'})
WriteResult({ "nInserted" : 1 })
rs1:PRIMARY>db.user.insert({name:'bobo'})
WriteResult({ "nInserted" : 1 })
在从节点查看数据：
rs1:SECONDARY> rs.slaveOk()
rs1:SECONDARY> show dbs
admin 0.000GB
local 0.000GB
test 0.000GB
rs1:SECONDARY> use test
switched to db test
rs1:SECONDARY> db.user.find()
{ "_id" :ObjectId("59566ffe6a3a31e1ea4abec7"), "name" :"gyb" }
{ "_id" :ObjectId("5956700e6a3a31e1ea4abec8"), "name" :"bobo" }
rs1:SECONDARY>
可以看到数据和主节点同步。
注意：从节点只能读数据，无法写入数据。在写入数据时会有如下错误：
rs1:SECONDARY>db.user.insert({name:'1123'})
WriteResult({ "writeError" : {"code" : 10107, "errmsg" : "not master" } })
rs1:SECONDARY>
故障恢复验证：
现在通过手动终止主节点，模拟宕机
在主节点Ctrl+C终止进程，
登陆127.0.0.1：27019
>mongo 127.0.0.1:27019
MongoDB shell version v3.4.5
connecting to: 127.0.0.1:27019
MongoDB server version: 3.4.5
Server has startup warnings:
2017-06-30T21:51:02.535+0800 I CONTROL [initandlisten]
2017-06-30T21:51:02.538+0800 I CONTROL [initandlisten] ** WARNING: Access control isnot enabled for the database.
2017-06-30T21:51:02.542+0800 I CONTROL [initandlisten] ** Read and write access to data andconfiguration is unrestricted.
2017-06-30T21:51:02.548+0800 I CONTROL [initandlisten]
rs1:PRIMARY>
rs1:PRIMARY>
发现此节点被升级为主节点。
查看数据文件
rs1:PRIMARY> show dbs
admin 0.000GB
local 0.000GB
test 0.000GB
rs1:PRIMARY> use test
switched to db test
rs1:PRIMARY> db.user.find()
{ "_id" :ObjectId("59566ffe6a3a31e1ea4abec7"), "name" :"gyb" }
{ "_id" :ObjectId("5956700e6a3a31e1ea4abec8"), "name" :"bobo" }
rs1:PRIMARY>
可以看到数据还在，这就体现了复制集的数据安全性以及高可用性。
查看集群状态：
rs1:PRIMARY> rs.status()
{
"set" : "rs1",
"date" : ISODate("2017-06-30T15:48:41.835Z"),
"myState" : 1,
"term" : NumberLong(2),
"heartbeatIntervalMillis" : NumberLong(2000),
"optimes" : {
"lastCommittedOpTime": {
"ts" :Timestamp(1498837720, 1),
"t" : NumberLong(2)
},
"appliedOpTime" : {
"ts" :Timestamp(1498837720, 1),
"t" :NumberLong(2)
},
"durableOpTime" : {
"ts" : Timestamp(1498837720,1),
"t" :NumberLong(2)
}
},
"members" : [
{
"_id" : 1,
"name" :"127.0.0.1:27017",
"health" : 0,
"state" : 8,
"stateStr" :"(not reachable/healthy)",
"uptime" : 0,
"optime" : {
"ts": Timestamp(0, 0),
"t" :NumberLong(-1)
},
"optimeDurable" : {
"ts": Timestamp(0, 0),
"t" :NumberLong(-1)
},
"optimeDate": ISODate("1970-01-01T00:00:00Z"),
"optimeDurableDate" :ISODate("1970-01-01T00:00:00Z"),
"lastHeartbeat" :ISODate("2017-06-30T15:48:37.579Z"),
"lastHeartbeatRecv" :ISODate("2017-06-30T15:43:41.461Z"),
"pingMs" :NumberLong(0),
"lastHeartbeatMessage" : "����Ŀ�����������ܾ��������ӡ�",
"configVersion" : -1
},
{
"_id" : 2,
"name" :"127.0.0.1:27018",
"health" : 1,
"state" : 2,
"stateStr" :"SECONDARY",
"uptime" :216,
"optime" : {
"ts": Timestamp(1498837720, 1),
"t" :NumberLong(2)
},
"optimeDurable": {
"ts": Timestamp(1498837720, 1),
"t" :NumberLong(2)
},
"optimeDate": ISODate("2017-06-30T15:48:40Z"),
"optimeDurableDate": ISODate("2017-06-30T15:48:40Z"),
"lastHeartbeat" :ISODate("2017-06-30T15:48:41.094Z"),
"lastHeartbeatRecv" :ISODate("2017-06-30T15:48:40.046Z"),
"pingMs" :NumberLong(0),
"syncingTo" :"127.0.0.1:27019",
"configVersion" : 1
},
{
"_id" : 3,
"name" :"127.0.0.1:27019",
"health" : 1,
"state" : 1,
"stateStr" :"PRIMARY",
"uptime" :7061,
"optime" : {
"ts": Timestamp(1498837720, 1),
"t" : NumberLong(2)
},
"optimeDate": ISODate("2017-06-30T15:48:40Z"),
"electionTime" : Timestamp(1498837508, 1),
"electionDate" : ISODate("2017-06-30T15:45:08Z"),
"configVersion" : 1,
"self" : true
}
],
"ok" : 1
}
rs1:PRIMARY>
可以看到127.0.0.1：27017成员节点已处于宕机状态。
现在重新开启127.0.0.1：27017mongod 服务：
mongod --dbpath C:\data\db --bind_ip127.0.0.1 --port 27017 --replSet rs1
再次查看集群状态：
rs1:PRIMARY> rs.status()
{
"set" : "rs1",
"date" : ISODate("2017-06-30T15:52:27.392Z"),
"myState" : 1,
"term" : NumberLong(2),
"heartbeatIntervalMillis" : NumberLong(2000),
"optimes" : {
"lastCommittedOpTime": {
"ts" :Timestamp(1498837940, 1),
"t" :NumberLong(2)
},
"appliedOpTime" : {
"ts" :Timestamp(1498837940, 1),
"t" :NumberLong(2)
},
"durableOpTime" : {
"ts" :Timestamp(1498837940, 1),
"t" :NumberLong(2)
}
},
"members" : [
{
"_id" : 1,
"name" :"127.0.0.1:27017",
"health" : 1,
"state" : 2,
"stateStr" :"SECONDARY",
"uptime" : 11,
"optime" : {
"ts": Timestamp(1498837940, 1),
"t" :NumberLong(2)
},
"optimeDurable" : {
"ts": Timestamp(1498837940, 1),
"t" :NumberLong(2)
},
"optimeDate": ISODate("2017-06-30T15:52:20Z"),
"optimeDurableDate": ISODate("2017-06-30T15:52:20Z"),
"lastHeartbeat" :ISODate("2017-06-30T15:52:25.910Z"),
"lastHeartbeatRecv" :ISODate("2017-06-30T15:52:25.964Z"),
"pingMs" :NumberLong(1),
"syncingTo" :"127.0.0.1:27018",
"configVersion" : 1
},
{
"_id" : 2,
"name" :"127.0.0.1:27018",
"health" : 1,
"state" : 2,
"stateStr" :"SECONDARY",
"uptime" :441,
"optime" : {
"ts": Timestamp(1498837940, 1),
"t" :NumberLong(2)
},
"optimeDurable" : {
"ts": Timestamp(1498837940, 1),
"t" :NumberLong(2)
},
"optimeDate": ISODate("2017-06-30T15:52:20Z"),
"optimeDurableDate" :ISODate("2017-06-30T15:52:20Z"),
"lastHeartbeat" :ISODate("2017-06-30T15:52:27.254Z"),
"lastHeartbeatRecv": ISODate("2017-06-30T15:52:26.202Z"),
"pingMs" :NumberLong(0),
"syncingTo" :"127.0.0.1:27019",
"configVersion" : 1
},
{
"_id" : 3,
"name" :"127.0.0.1:27019",
"health" : 1,
"state" : 1,
"stateStr" :"PRIMARY",
"uptime" : 7287,
"optime" : {
"ts": Timestamp(1498837940, 1),
"t" :NumberLong(2)
},
"optimeDate": ISODate("2017-06-30T15:52:20Z"),
"electionTime" : Timestamp(1498837508, 1),
"electionDate" : ISODate("2017-06-30T15:45:08Z"),
"configVersion" : 1,
"self" : true
}
],
"ok" : 1
}
rs1:PRIMARY>
可以看到id为1的成员已经恢复，成为了一个从节点。这就体现了复制集的故障恢复以及容错能力。
另外，可以在复制集中动态的增加删除节点。这里就不再演示。
使用rs.add(‘IP：port’).