MongoDB分片集群搭建

• 什么是分片
  分片是一种跨多台机器分布数据的方法。MongoDB 使用分片来支持超大数据集和高吞吐量操作的部署。
• 什么情况下使用分片

1. 存储容量受单机限制，即磁盘资源遭遇瓶颈。
2. 读写能力受单机限制，可能是CPU、内存或者网卡等资源遭遇瓶颈，导致读写能力无法扩展。

• Mongo分片组件
  
  路由节点(Router) 分片集群的访问入口，其本身并不持久化数据。
  配置节点(Config) 配置服务器包含多个节点，并组成一个复制集结构。
  数据节点(Shard) 分片用于存储真正的数据，并提供最终的数据读写访问。
• Mongo分片方式

1. 基于范围
   
2. 基于哈希
   
3. 自定义Zone（区域/范围）

• 集群设计
  数据是根据分片策略来进行切分的，而分片策略则由 分片键(ShardKey)+分片算法(ShardStrategy)组成。
  个人认为分多少片，分片的大小要结合实际情况来看所以这里不做过多讨论。
  可参考这两篇文章
  https://developer.aliyun.com/article/879873
  https://help.aliyun.com/zh/mongodb/use-cases/introduction-to-apsaradb-for-mongodb-sharded-cluster-instances

• 集群搭建
  可参考这篇博客，十分详细 https://www.cnblogs.com/misakivv/p/18171124
  先搭建shard节点，其次是配置节点，最后是路由节点。

0. 每台机器安装同版本的Mongo

wget https://fastdl.mongodb.org/linux/mongodb-linux-x86_64-rhel70-4.4.6.tgz
tar xzvf mongodb-linux-x86_64-rhel70-4.4.6.tgz -C /usr/local
cd /usr/local/
ln -s /usr/local/mongodb-linux-x86_64-rhel70-4.4.6 /usr/local/mongodb
vim /etc/profileexport PATH=/usr/local/mongodb/bin/:$PATHsource /etc/profile

1. shard1主机操作
   准备存放数据和日志的目录

mkdir -p /usr/local/mongodb/sharded_cluster/myshardrs01_27018/log \
/usr/local/mongodb/sharded_cluster/myshardrs01_27018/data/db \
/usr/local/mongodb/sharded_cluster/myshardrs01_27118/log \
/usr/local/mongodb/sharded_cluster/myshardrs01_27118/data/db \
/usr/local/mongodb/sharded_cluster/myshardrs01_27218/log \
/usr/local/mongodb/sharded_cluster/myshardrs01_27218/data/db

创建配置文件
myshardrs01_27018
记得替换对应的bindIp

vim /usr/local/mongodb/sharded_cluster/myshardrs01_27018/mongod.confsystemLog:destination: filepath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27018/log/mongod.log"logAppend: true
storage:dbPath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27018/data/db"journal:enabled: true
processManagement:fork: truepidFilePath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27018/log/mongod.pid"
net:bindIp: localhost,192.168.112.10port: 27018
replication:replSetName: myshardrs01
sharding:clusterRole: shardsvr

myshardrs01_27118
记得替换对应的bindIp

vim /usr/local/mongodb/sharded_cluster/myshardrs01_27118/mongod.confsystemLog:destination: filepath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27118/log/mongod.log"logAppend: true
storage:dbPath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27118/data/db"journal:enabled: true
processManagement:fork: truepidFilePath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27118/log/mongod.pid"
net:bindIp: localhost,192.168.112.10port: 27118
replication:replSetName: myshardrs01
sharding:clusterRole: shardsvr

myshardrs01_27218
记得替换对应的bindIp

vim /usr/local/mongodb/sharded_cluster/myshardrs01_27218/mongod.confsystemLog:destination: filepath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27218/log/mongod.log"logAppend: true
storage:dbPath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27218/data/db"journal:enabled: true
processManagement:fork: truepidFilePath: "/usr/local/mongodb/sharded_cluster/myshardrs01_27218/log/mongod.pid"
net:bindIp: localhost,192.168.112.10port: 27218
replication:replSetName: myshardrs01
sharding:clusterRole: shardsvr

启动第一套副本集：一主一副本一仲裁

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/sharded_cluster/myshardrs01_27018/mongod.conf
/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/sharded_cluster/myshardrs01_27118/mongod.conf
/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/sharded_cluster/myshardrs01_27218/mongod.conf

验证一下进程是否正常
ps -ef|grep mongod
连接主节点

mongo --host ip.ip.ip.ip --port 27018

rs.initiate()
{"info2" : "no configuration specified. Using a default configuration for the set","me" : "192.168.112.10:27018","ok" : 1
}
myshardrs01:SECONDARY>
myshardrs01:PRIMARY>

添加副节点和仲裁节点

rs.add("192.168.112.10:27118")
{"ok" : 1,"$clusterTime" : {"clusterTime" : Timestamp(1714658449, 1),"signature" : {"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),"keyId" : NumberLong(0)}},"operationTime" : Timestamp(1714658449, 1)
}

rs.addArb("192.168.112.10:27218")
{"ok" : 1,"$clusterTime" : {"clusterTime" : Timestamp(1714658509, 1),"signature" : {"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),"keyId" : NumberLong(0)}},"operationTime" : Timestamp(1714658509, 1)
}

查看副本集
rs.conf()
rs.status()

2. shard2主机操作
   类似shard1
3. 配置节点副本集的创建
   准备存放数据和日志的目录

mkdir -p /usr/local/mongodb/sharded_cluster/myconfigrs_27019/log \
/usr/local/mongodb/sharded_cluster/myconfigrs_27019/data/db \
/usr/local/mongodb/sharded_cluster/myconfigrs_27119/log \
/usr/local/mongodb/sharded_cluster/myconfigrs_27119/data/db \
/usr/local/mongodb/sharded_cluster/myconfigrs_27219/log \
/usr/local/mongodb/sharded_cluster/myconfigrs_27219/data/db

创建配置文件
myconfigrs_27019

systemLog:destination: filepath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27019/log/mongod.log"logAppend: true
storage:dbPath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27019/data/db"journal:enabled: true
processManagement:fork: truepidFilePath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27019/log/mongod.pid"
net:bindIp: localhost,192.168.112.30port: 27019
replication:replSetName: myconfigr
sharding:clusterRole: configsvr

myconfigrs_27119

systemLog:destination: filepath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27119/log/mongod.log"logAppend: true
storage:dbPath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27119/data/db"journal:enabled: true
processManagement:fork: truepidFilePath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27119/log/mongod.pid"
net:bindIp: localhost,192.168.112.30port: 27119
replication:replSetName: myconfigr
sharding:clusterRole: configsvr

myconfigrs_27219

systemLog:destination: filepath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27219/log/mongod.log"logAppend: true
storage:dbPath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27219/data/db"journal:enabled: true
processManagement:fork: truepidFilePath: "/usr/local/mongodb/sharded_cluster/myconfigrs_27219/log/mongod.pid"
net:bindIp: localhost,192.168.112.30port: 27219
replication:replSetName: myconfigr
sharding:clusterRole: configsvr

启动副本集

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/sharded_cluster/myconfigrs_27019/mongod.conf
/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/sharded_cluster/myconfigrs_27119/mongod.conf
/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/sharded_cluster/myconfigrs_27219/mongod.conf

ps检查进程是否正常

连接主节点
mongo --host 192.168.112.30 --port 27019
初始化副本集

rs.initiate()
{"info2" : "no configuration specified. Using a default configuration for the set","me" : "192.168.112.30:27019","ok" : 1,"$gleStats" : {"lastOpTime" : Timestamp(1714702204, 1),"electionId" : ObjectId("000000000000000000000000")},"lastCommittedOpTime" : Timestamp(0, 0)
}

添加副本节点

rs.add("192.168.112.30:27119")
{"ok" : 1,"$gleStats" : {"lastOpTime" : {"ts" : Timestamp(1714702333, 2),"t" : NumberLong(1)},"electionId" : ObjectId("7fffffff0000000000000001")},"lastCommittedOpTime" : Timestamp(1714702333, 2),"$clusterTime" : {"clusterTime" : Timestamp(1714702335, 1),"signature" : {"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),"keyId" : NumberLong(0)}},"operationTime" : Timestamp(1714702333, 2)
}

添加副本节点

rs.add("192.168.112.30:27219")
{"ok" : 1,"$gleStats" : {"lastOpTime" : {"ts" : Timestamp(1714702365, 2),"t" : NumberLong(1)},"electionId" : ObjectId("7fffffff0000000000000001")},"lastCommittedOpTime" : Timestamp(1714702367, 1),"$clusterTime" : {"clusterTime" : Timestamp(1714702367, 1),"signature" : {"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),"keyId" : NumberLong(0)}},"operationTime" : Timestamp(1714702365, 2)
}

4. 路由节点的创建和操作

mkdir -p /usr/local/mongodb/sharded_cluster/mymongos_27017/log

创建配置文件

vim /usr/local/mongodb/sharded_cluster/mymongos_27017/mongos.confsystemLog:destination: filepath: "/usr/local/mongodb/sharded_cluster/mymongos_27017/log/mongod.log"logAppend: true
processManagement:fork: truepidFilePath: "/usr/local/mongodb/sharded_cluster/mymongos_27017/log/mongod.pid"
net:bindIp: localhost,192.168.112.40port: 27017
sharding:configDB: myconfigrs/192.168.112.30:27019,192.168.112.30:27119,192.168.112.30:27219

启动
/usr/local/mongodb/bin/mongos -f /usr/local/mongodb/sharded_cluster/mymongos_27017/mongos.conf

客户端登录mongo
mongo --port 27017

此时写入不了数据
通过路由节点操作，现在只是连接了配置节点
还没有连接分片数据节点，因此无法写入业务数据

5. 添加分片副本集

第一套分片加进来

sh.addShard("myshardrs01/192.168.112.10:27018,192.168.112.10:27118,192.168.112.10:27218")
{"shardAdded" : "myshardrs01","ok" : 1,"operationTime" : Timestamp(1714705107, 3),"$clusterTime" : {"clusterTime" : Timestamp(1714705107, 3),"signature" : {"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),"keyId" : NumberLong(0)}}
}

查看分片状态

sh.status()
--- Sharding Status ---sharding version: {"_id" : 1,"minCompatibleVersion" : 5,"currentVersion" : 6,"clusterId" : ObjectId("6634477c7677719ed2388706")}shards:{  "_id" : "myshardrs01",  "host" : "myshardrs01/192.168.112.10:27018,192.168.112.10:27118",  "state" : 1 }most recently active mongoses:"4.4.6" : 1autosplit:Currently enabled: yesbalancer:Currently enabled:  yesCurrently running:  noFailed balancer rounds in last 5 attempts:  0Migration Results for the last 24 hours:No recent migrationsdatabases:{  "_id" : "config",  "primary" : "config",  "partitioned" : true }

第二套分片加进来

sh.addShard("myshardrs02/192.168.112.20:27318,192.168.112.20:27418,192.168.112.20:27518")
{"shardAdded" : "myshardrs02","ok" : 1,"operationTime" : Timestamp(1714705261, 2),"$clusterTime" : {"clusterTime" : Timestamp(1714705261, 2),"signature" : {"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),"keyId" : NumberLong(0)}}
}

查看分片状态

sh.status()
--- Sharding Status ---sharding version: {"_id" : 1,"minCompatibleVersion" : 5,"currentVersion" : 6,"clusterId" : ObjectId("6634477c7677719ed2388706")}shards:{  "_id" : "myshardrs01",  "host" : "myshardrs01/192.168.112.10:27018,192.168.112.10:27118",  "state" : 1 }{  "_id" : "myshardrs02",  "host" : "myshardrs02/192.168.112.20:27318,192.168.112.20:27418",  "state" : 1 }most recently active mongoses:"4.4.6" : 1autosplit:Currently enabled: yesbalancer:Currently enabled:  yesCurrently running:  noFailed balancer rounds in last 5 attempts:  0Migration Results for the last 24 hours:34 : Successdatabases:{  "_id" : "config",  "primary" : "config",  "partitioned" : true }config.system.sessionsshard key: { "_id" : 1 }unique: falsebalancing: truechunks:myshardrs01     990myshardrs02     34too many chunks to print, use verbose if you want to force print

移除分片

use db
db.runCommand({removeShard: "myshardrs02"})

6. 开启分片功能
   sh.enableSharding(“库名”)
   sh.shardCollection(“库名.集合名”,{“key”:1})

7. 插入数据测试

mongos> use test
switched to db test
mongos> for(var i=1;i<=1000;i++){db.comment.insert({_id:i+"",nickname:"Test"+i})}
WriteResult({ "nInserted" : 1 })
mongos> db.comment.count()
1000

可以分别登录到两个分片节点上查看文档数量

8. 再添加一个路由节点
   步骤同上

9. 其他可参考的博客

https://www.cnblogs.com/misakivv/p/18171124

https://juejin.cn/post/7160298673182605343
https://mp.weixin.qq.com/s/5oLjCRbXfWNFwmB6jkykZQ