MongoDB集群部署之Replica Set（副本集）、Sharding（分片）【转】

 

一、集群模式介绍

            Mongodb是时下流行的NoSql数据库，它的存储方式是文档式存储，并不是Key-Value形式。MongoDB 集群部署有三种模式，分别为主从复制（Master-Slaver）、副本集（Replica Set）和分片（Sharding）模式。

• Master-Slaver 是一种主从副本的模式，目前已经不推荐使用。
• Replica Set 模式取代了 Master-Slaver 模式，是一种互为主从的关系。Replica Set 将数据复制多份保存，不同服务器保存同一份数据，在出现故障时自动切换，实现故障转移，在实际生产中非常实用。
• Sharding 模式适合处理大量数据，它将数据分开存储，不同服务器保存不同的数据，所有服务器数据的总和即为整个数据集。

Sharding 模式追求的是高性能，而且是三种集群中最复杂的。在实际生产环境中，通常将 Replica Set 和 Sharding 两种技术结合使用。

二、部署Replica Set集群

              Replica Set集群当中包含了多份数据，保证主节点挂掉了，备节点能继续提供数据服务，提供的前提就是数据需要和主节点一致。

• Mongodb(M)表示主节点，Mongodb(S)表示备节点，Mongodb(A)表示仲裁节点。主备节点存储数据，仲裁节点不存储数据。客户端同时连接主节点与备节点，不连接仲裁节点。
• 默认设置主节点提供所有增删查改服务，备节点不提供任何服务。但是可以通过设置使备节点提供查询服务，这样就可以减少主节点的压力，当客户端进行数据查询时，请求自动转到备节点上。这个设置叫做Read Preference Modes。
• 仲裁节点是一种特殊的节点，它本身并不存储数据，主要的作用是决定哪一个备节点在主节点挂掉之后提升为主节点，所以客户端不需要连接此节点。这里虽然只有一个备节点，但是仍然需要一个仲裁节点来提升备节点级别。

1、下载软件包

官网地址：https://www.mongodb.com/try/download/community

 ①创建所需目录

mkdir -p /app/mongodb4.2/{logs,conf,data}
mkdir -p /app/mongodb4.2/data/{master,slaver,arbiter}

②解压软件包

tar xf mongodb-linux-x86_64-rhel70-4.2.13.tgz -C /app/mongodb4.2/
mv /app/mongodb4.2/mongodb-linux-x86_64-rhel70-4.2.13/* /app/mongodb4.2/
rm -fr /app/mongodb4.2/mongodb-linux-x86_64-rhel70-4.2.13

③创建MongoDB配置文件、日志文件

touch /app/mongodb4.2/logs/mongodb.log
touch /app/mongodb4.2/conf/mongodb.conf

④配置mongodb.conf

port=27018
#以后台方式运行
fork=true
#数据库日志存放目录
logpath=/app/mongodb4.2/logs/mongodb.log
#以追加的方式记录日志
logappend=true
#replica set的名字
replSet=MongoDB-rs
#为每一个数据库按照数据库名建立文件夹存放
directoryperdb=true
#mongodb操作日志文件的最大大小。单位为Mb，默认为硬盘剩余空间的5%
oplogSize=10000
#数据库数据存放目录
dbpath=/app/mongodb4.2/data/master   #分发到节点，此处需要修改
#开启用户认证true
auth=false
#过滤一些无用的日志信息，若需要调试使用请设置为false
quiet=true
#auth=true
##mongodb所绑定的ip地址
bind_ip = 0.0.0.0

⑤分发软件到节点

scp -rp mongodb4.2 192.168.56.175:/app/
scp -rp mongodb4.2 192.168.56.182:/app/

2、启动MongoDB服务

①启动服务(三个节点都执行)

[root@k8s-node1 app]# /app/mongodb4.2/bin/mongod -f /app/mongodb4.2/conf/mongodb.conf
about to fork child process, waiting until server is ready for connections.
forked process: 3317
child process started successfully, parent exiting

②配置主、备、仲裁节点

连接任意一台节点

[root@k8s-node1 app]# /app/mongodb4.2/bin/mongo 192.168.56.175:27018
MongoDB shell version v4.2.13
connecting to: mongodb://192.168.56.175:27018/test?compressors=disabled&gssapiServiceName=mongodb
Implicit session: session { "id" : UUID("01ac5aa9-4ae0-446e-bda7-084d12c48683") }
MongoDB server version: 4.2.13
Welcome to the MongoDB shell.

执行节点配置命令

> use admin
switched to db admin
> cfg={ _id:"MongoDB-rs", members:[ {_id:0,host:'192.168.56.175:27018',priority:2}, {_id:1,host:'192.168.56.176:27018',priority:1}, {_id:2,host:'192.168.56.182:27018',arbiterOnly:true}] };
{
    "_id" : "MongoDB-rs",
    "members" : [
        {
            "_id" : 0,
            "host" : "192.168.56.175:27018",
            "priority" : 2
        },
        {
            "_id" : 1,
            "host" : "192.168.56.176:27018",
            "priority" : 1
        },
        {
            "_id" : 2,
            "host" : "192.168.56.182:27018",
            "arbiterOnly" : true
        }
    ]
}
> rs.initiate(cfg) #使配置生效
{ "ok" : 1 }

cfg是可以任意的名字,相当于定义一个变量。
_id表示replica set的名字
members里包含的是所有节点的地址以及优先级。优先级最高的即成为主节点，即这里的192.168.56.175:27018。
特别注意第三个节点的配置arbiterOnly:true，这个千万不能少了，不然主备模式就不能生效。

③查看集群状态

MongoDB-rs:SECONDARY> rs.status()
{
    "set" : "MongoDB-rs",
    "date" : ISODate("2021-03-31T03:04:11.467Z"),
    "myState" : 1,
    "term" : NumberLong(1),
    "syncingTo" : "",
    "syncSourceHost" : "",
    "syncSourceId" : -1,
    "heartbeatIntervalMillis" : NumberLong(2000),
    "majorityVoteCount" : 2,
    "writeMajorityCount" : 2,
    "optimes" : {
        "lastCommittedOpTime" : {
            "ts" : Timestamp(1617159844, 1),
            "t" : NumberLong(1)
        },
        "lastCommittedWallTime" : ISODate("2021-03-31T03:04:04.141Z"),
        "readConcernMajorityOpTime" : {
            "ts" : Timestamp(1617159844, 1),
            "t" : NumberLong(1)
        },
        "readConcernMajorityWallTime" : ISODate("2021-03-31T03:04:04.141Z"),
        "appliedOpTime" : {
            "ts" : Timestamp(1617159844, 1),
            "t" : NumberLong(1)
        },
        "durableOpTime" : {
            "ts" : Timestamp(1617159844, 1),
            "t" : NumberLong(1)
        },
        "lastAppliedWallTime" : ISODate("2021-03-31T03:04:04.141Z"),
        "lastDurableWallTime" : ISODate("2021-03-31T03:04:04.141Z")
    },
    "lastStableRecoveryTimestamp" : Timestamp(1617159814, 1),
    "lastStableCheckpointTimestamp" : Timestamp(1617159814, 1),
    "electionCandidateMetrics" : {
        "lastElectionReason" : "electionTimeout",
        "lastElectionDate" : ISODate("2021-03-31T02:59:34.104Z"),
        "electionTerm" : NumberLong(1),
        "lastCommittedOpTimeAtElection" : {
            "ts" : Timestamp(0, 0),
            "t" : NumberLong(-1)
        },
        "lastSeenOpTimeAtElection" : {
            "ts" : Timestamp(1617159562, 1),
            "t" : NumberLong(-1)
        },
        "numVotesNeeded" : 2,
        "priorityAtElection" : 2,
        "electionTimeoutMillis" : NumberLong(10000),
        "numCatchUpOps" : NumberLong(0),
        "newTermStartDate" : ISODate("2021-03-31T02:59:34.134Z"),
        "wMajorityWriteAvailabilityDate" : ISODate("2021-03-31T02:59:35.580Z")
    },
    "members" : [
        {
            "_id" : 0,
            "name" : "192.168.56.175:27018",
            "health" : 1,
            "state" : 1,
            "stateStr" : "PRIMARY",
            "uptime" : 820,
            "optime" : {
                "ts" : Timestamp(1617159844, 1),
                "t" : NumberLong(1)
            },
            "optimeDate" : ISODate("2021-03-31T03:04:04Z"),
            "syncingTo" : "",
            "syncSourceHost" : "",
            "syncSourceId" : -1,
            "infoMessage" : "",
            "electionTime" : Timestamp(1617159574, 1),
            "electionDate" : ISODate("2021-03-31T02:59:34Z"),
            "configVersion" : 1,
            "self" : true,
            "lastHeartbeatMessage" : ""
        },
        {
            "_id" : 1,
            "name" : "192.168.56.176:27018",
            "health" : 1,
            "state" : 2,
            "stateStr" : "SECONDARY",
            "uptime" : 288,
            "optime" : {
                "ts" : Timestamp(1617159844, 1),
                "t" : NumberLong(1)
            },
            "optimeDurable" : {
                "ts" : Timestamp(1617159844, 1),
                "t" : NumberLong(1)
            },
            "optimeDate" : ISODate("2021-03-31T03:04:04Z"),
            "optimeDurableDate" : ISODate("2021-03-31T03:04:04Z"),
            "lastHeartbeat" : ISODate("2021-03-31T03:04:10.275Z"),
            "lastHeartbeatRecv" : ISODate("2021-03-31T03:04:09.717Z"),
            "pingMs" : NumberLong(1),
            "lastHeartbeatMessage" : "",
            "syncingTo" : "192.168.56.175:27018",
            "syncSourceHost" : "192.168.56.175:27018",
            "syncSourceId" : 0,
            "infoMessage" : "",
            "configVersion" : 1
        },
        {
            "_id" : 2,
            "name" : "192.168.56.182:27018",
            "health" : 1,
            "state" : 7,
            "stateStr" : "ARBITER",
            "uptime" : 288,
            "lastHeartbeat" : ISODate("2021-03-31T03:04:10.275Z"),
            "lastHeartbeatRecv" : ISODate("2021-03-31T03:04:09.560Z"),
            "pingMs" : NumberLong(1),
            "lastHeartbeatMessage" : "",
            "syncingTo" : "",
            "syncSourceHost" : "",
            "syncSourceId" : -1,
            "infoMessage" : "",
            "configVersion" : 1
        }
    ],
    "ok" : 1,
    "$clusterTime" : {
        "clusterTime" : Timestamp(1617159844, 1),
        "signature" : {
            "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
            "keyId" : NumberLong(0)
        }
    },
    "operationTime" : Timestamp(1617159844, 1)
}

至此Replica Set集群模式部署完毕

三、增加安全认证机制KeyFile

1、生成Key

mkdir -p /app/mongodb4.2/key
openssl rand -base64 745 > /app/mongodb4.2/key/mongodb-keyfile
chmod 600 /app/mongodb4.2/key/mongodb-keyfile

2、分发Key文件到节点

scp /app/mongodb4.2/key/mongodb-keyfile 192.168.56.176:/app/mongodb4.2/key/
scp /app/mongodb4.2/key/mongodb-keyfile 192.168.56.182:/app/mongodb4.2/key/

3、修改配置文件

vim mongodb.conf
keyFile=/app/mongodb4.2/key/mongodb-keyfile

4、启动服务

 /app/mongodb4.2/bin/mongod -f /app/mongodb4.2/conf/mongodb.conf

5、关闭服务

/app/mongodb4.2/bin/mongod -f /app/mongodb4.2/conf/mongodb.conf --shutdown

6、查看集群状态

MongoDB-rs:PRIMARY> rs.status()
{
    "set" : "MongoDB-rs",
    "date" : ISODate("2021-03-31T08:27:22.429Z"),
    "myState" : 1,
    "term" : NumberLong(4),
    "syncingTo" : "",
    "syncSourceHost" : "",
    "syncSourceId" : -1,
    "heartbeatIntervalMillis" : NumberLong(2000),
    "majorityVoteCount" : 2,
    "writeMajorityCount" : 2,
    "optimes" : {
        "lastCommittedOpTime" : {
            "ts" : Timestamp(1617179232, 1),
            "t" : NumberLong(4)
        },
        "lastCommittedWallTime" : ISODate("2021-03-31T08:27:12.669Z"),
        "readConcernMajorityOpTime" : {
            "ts" : Timestamp(1617179232, 1),
            "t" : NumberLong(4)
        },
        "readConcernMajorityWallTime" : ISODate("2021-03-31T08:27:12.669Z"),
        "appliedOpTime" : {
            "ts" : Timestamp(1617179232, 1),
            "t" : NumberLong(4)
        },
        "durableOpTime" : {
            "ts" : Timestamp(1617179232, 1),
            "t" : NumberLong(4)
        },
        "lastAppliedWallTime" : ISODate("2021-03-31T08:27:12.669Z"),
        "lastDurableWallTime" : ISODate("2021-03-31T08:27:12.669Z")
    },
    "lastStableRecoveryTimestamp" : Timestamp(1617179202, 1),
    "lastStableCheckpointTimestamp" : Timestamp(1617179202, 1),
    "electionCandidateMetrics" : {
        "lastElectionReason" : "priorityTakeover",
        "lastElectionDate" : ISODate("2021-03-31T08:26:12.634Z"),
        "electionTerm" : NumberLong(4),
        "lastCommittedOpTimeAtElection" : {
            "ts" : Timestamp(1617179171, 1),
            "t" : NumberLong(3)
        },
        "lastSeenOpTimeAtElection" : {
            "ts" : Timestamp(1617179171, 1),
            "t" : NumberLong(3)
        },
        "numVotesNeeded" : 2,
        "priorityAtElection" : 2,
        "electionTimeoutMillis" : NumberLong(10000),
        "priorPrimaryMemberId" : 1,
        "numCatchUpOps" : NumberLong(0),
        "newTermStartDate" : ISODate("2021-03-31T08:26:12.667Z"),
        "wMajorityWriteAvailabilityDate" : ISODate("2021-03-31T08:26:14.696Z")
    },
    "electionParticipantMetrics" : {
        "votedForCandidate" : true,
        "electionTerm" : NumberLong(3),
        "lastVoteDate" : ISODate("2021-03-31T08:26:01.824Z"),
        "electionCandidateMemberId" : 1,
        "voteReason" : "",
        "lastAppliedOpTimeAtElection" : {
            "ts" : Timestamp(1617177014, 1),
            "t" : NumberLong(1)
        },
        "maxAppliedOpTimeInSet" : {
            "ts" : Timestamp(1617177014, 1),
            "t" : NumberLong(1)
        },
        "priorityAtElection" : 2
    },
    "members" : [
        {
            "_id" : 0,
            "name" : "192.168.56.175:27018",
            "health" : 1,
            "state" : 1,
            "stateStr" : "PRIMARY",
            "uptime" : 97,
            "optime" : {
                "ts" : Timestamp(1617179232, 1),
                "t" : NumberLong(4)
            },
            "optimeDate" : ISODate("2021-03-31T08:27:12Z"),
            "syncingTo" : "",
            "syncSourceHost" : "",
            "syncSourceId" : -1,
            "infoMessage" : "",
            "electionTime" : Timestamp(1617179172, 1),
            "electionDate" : ISODate("2021-03-31T08:26:12Z"),
            "configVersion" : 1,
            "self" : true,
            "lastHeartbeatMessage" : ""
        },
        {
            "_id" : 1,
            "name" : "192.168.56.176:27018",
            "health" : 1,
            "state" : 2,
            "stateStr" : "SECONDARY",
            "uptime" : 91,
            "optime" : {
                "ts" : Timestamp(1617179232, 1),
                "t" : NumberLong(4)
            },
            "optimeDurable" : {
                "ts" : Timestamp(1617179232, 1),
                "t" : NumberLong(4)
            },
            "optimeDate" : ISODate("2021-03-31T08:27:12Z"),
            "optimeDurableDate" : ISODate("2021-03-31T08:27:12Z"),
            "lastHeartbeat" : ISODate("2021-03-31T08:27:20.698Z"),
            "lastHeartbeatRecv" : ISODate("2021-03-31T08:27:20.867Z"),
            "pingMs" : NumberLong(0),
            "lastHeartbeatMessage" : "",
            "syncingTo" : "192.168.56.175:27018",
            "syncSourceHost" : "192.168.56.175:27018",
            "syncSourceId" : 0,
            "infoMessage" : "",
            "configVersion" : 1
        },
        {
            "_id" : 2,
            "name" : "192.168.56.182:27018",
            "health" : 1,
            "state" : 7,
            "stateStr" : "ARBITER",
            "uptime" : 87,
            "lastHeartbeat" : ISODate("2021-03-31T08:27:20.698Z"),
            "lastHeartbeatRecv" : ISODate("2021-03-31T08:27:20.797Z"),
            "pingMs" : NumberLong(0),
            "lastHeartbeatMessage" : "",
            "syncingTo" : "",
            "syncSourceHost" : "",
            "syncSourceId" : -1,
            "infoMessage" : "",
            "configVersion" : 1
        }
    ],
    "ok" : 1,
    "$clusterTime" : {
        "clusterTime" : Timestamp(1617179232, 1),
        "signature" : {
            "hash" : BinData(0,"uXViIfbqr0npQMxp/pJ4jAO/GfY="),
            "keyId" : NumberLong("6945647482743291907")
        }
    },
    "operationTime" : Timestamp(1617179232, 1)
}

         MongoDB启用认证后，不仅客户端连接需要认证，服务内部不同节点之间，也是需要相互认证的，如果仅仅是启用认证方式，不设定服务内部之间的认证参数，则服务内部各节点之间是无法成功通信的，就造成了当前登录节点一直是恢复中的状态，而其他节点应该是 not reachable的状态，所以会显示为 not reachable/healthy

mongodb提供了两种方式：一种是 keyfile的方式，一种是证书方式，详细官方介绍 https://docs.mongodb.com/manual/core/security-internal-authentication/

四、分片集群部署

1、分片集群架构图

 2、分片技术介绍

          分片（sharding）是MongoDB用来将大型集合分割到不同服务器（或者说一个集群）上所采用的方法。MongoDB的最大特点在于它几乎能自动完成所有事情，只要告诉MongoDB要分配数据，它就能自动维护数据在不同服务器之间的均衡。MongoDB分片集群，英文名称为： Sharded Cluster旨在通过横向扩展，来提高数据吞吐性能、增大数据存储量。分片集群由三个组件：“mongos”, “config server”, “shard” 组成。

• mongos ：mongos充当查询路由器，在客户端应用程序和分片群集之间提供接口。
• config server：配置服务器存储集群的元数据和配置设置。从MongoDB 3.4开始，配置服务器必须部署为副本集（CSRS）。
• shard：真正的数据存储位置，以chunk为单位存数据，每个shard都可以部署为副本集。

 

服务器分配

 

3、config server服务部署

①创建所需目录

mkdir -p /app/mongodb-cluster/{mongos,config,shard1,shard2,shard3}/{data,log,conf}

├── config
│   ├── conf
│   ├── data
│   └── log
├── mongos
│   ├── conf
│   ├── data
│   └── log
├── shard1
│   ├── conf
│   ├── data
│   └── log
├── shard2
│   ├── conf
│   ├── data
│   └── log
└── shard3
    ├── conf
    ├── data
    └── log

②分片集群各部分组件搭建顺序

“config server” -> 2. “shard” -> 3. “mongos”
#启动服务顺序一样
"config server"由三台主机组成，每台主机上运行一个mongod进程，三台主机的mongod组成一个副本集。

③创建MongoDB配置文件、日志文件

touch /app/mongodb-cluster/{mongos,config,shard1,shard2,shard3}/conf/mongodb.conf
touch /app/mongodb-cluster/{mongos,config,shard1,shard2,shard3}/log/mongodb.log

④config server配置文件

port=27017
#以后台方式运行
fork=true
#数据库日志存放目录
logpath=/app/mongodb-cluster/config/log/mongodb.log
#数据库存放目录
dbpath=/app/mongodb-cluster/config/data
#以追加的方式记录日志
logappend=true
#replica set的名字
replSet=rs-config
#必须配置
configsvr = true
#为每一个数据库按照数据库名建立文件夹存放
directoryperdb=true
#mongodb操作日志文件的最大大小。单位为Mb，默认为硬盘剩余空间的5%
oplogSize=10000
#数据库数据存放目录
#开启用户认证true
auth=false
#过滤一些无用的日志信息，若需要调试使用请设置为false
quiet=true
#auth=true
##mongodb所绑定的ip地址
bind_ip = 0.0.0.0

三台主机的mongodb.conf配置相同,分发上面配置文件到节点

scp -rp /app/mongodb-cluster 192.168.56.176:/app/
scp -rp /app/mongodb-cluster 192.168.56.182:/app/

⑤启动config server(三个节点)

/app/mongodb-cluster/bin/mongod -f /app/mongodb-cluster/config/conf/mongodb.conf

⑥初始化config server副本集

任意节点登录执行

/app/mongodb-cluster/bin/mongo 192.168.56.175:27017

> rs.initiate(
... {
... "_id" : "rs-config",
... "members" : [
... {"_id" : 0, "host" : "192.168.56.175:27017"},
... {"_id" : 1, "host" : "192.168.56.176:27017"},
... {"_id" : 2, "host" : "192.168.56.182:27017"}
... ]
... }
... )

#以下为输出
{
"ok" : 1,
"$gleStats" : {
"lastOpTime" : Timestamp(1617785674, 1),
"electionId" : ObjectId("000000000000000000000000")
},
"lastCommittedOpTime" : Timestamp(0, 0)
}

出现 ok 1 则成功

 4、shard分片服务部署

①创建shard1配置文件

port=27018
#以后台方式运行
fork=true
#数据库日志存放目录
logpath=/app/mongodb-cluster/shard1/log/mongodb.log
#数据库存放目录
dbpath=/app/mongodb-cluster/shard1/data
#以追加的方式记录日志
logappend=true
#replica set的名字
replSet=rs-shard1
#必须配置
shardsvr=true
#为每一个数据库按照数据库名建立文件夹存放
directoryperdb=true
#mongodb操作日志文件的最大大小。单位为Mb，默认为硬盘剩余空间的5%
oplogSize=10000
#数据库数据存放目录
#开启用户认证true
auth=false
#过滤一些无用的日志信息，若需要调试使用请设置为false
quiet=true
#auth=true
##mongodb所绑定的ip地址
bind_ip = 0.0.0.0

②分发配置文件到节点

scp mongodb.conf 192.168.56.176:/app/mongodb-cluster/shard1/conf/
scp mongodb.conf 192.168.56.182:/app/mongodb-cluster/shard1/conf/

③启动三台服务器的shard1 server

/app/mongodb-cluster/bin/mongod -f /app/mongodb-cluster/shard1/conf/mongodb.conf

任选分片内的一个节点，执行初始化命令

/app/mongodb-cluster/bin/mongo 192.168.56.175:27018

> rs.initiate(
...   {
...      "_id" : "rs-shard1", 
...      "members" : [
...        {"_id" : 0, "host" : "192.168.56.175:27018"},
...        {"_id" : 1, "host" : "192.168.56.176:27018"},
...        {"_id" : 2, "host" : "192.168.56.182:27018"}
...      ]
...   }
... )
#输出结果
{ "ok" : 1 }

④创建shard2配置文件

port=27019
#以后台方式运行
fork=true
#数据库日志存放目录
logpath=/app/mongodb-cluster/shard2/log/mongodb.log
#数据库存放目录
dbpath=/app/mongodb-cluster/shard2/data
#以追加的方式记录日志
logappend=true
#replica set的名字
replSet=rs-shard2
#必须配置
shardsvr=true
#为每一个数据库按照数据库名建立文件夹存放
directoryperdb=true
#mongodb操作日志文件的最大大小。单位为Mb，默认为硬盘剩余空间的5%
oplogSize=10000
#数据库数据存放目录
#开启用户认证true
auth=false
#过滤一些无用的日志信息，若需要调试使用请设置为false
quiet=true
#auth=true
##mongodb所绑定的ip地址
bind_ip = 0.0.0.0

分发配置文件到节点

scp mongodb.conf 192.168.56.176:/app/mongodb-cluster/shard2/conf/
scp mongodb.conf 192.168.56.182:/app/mongodb-cluster/shard2/conf/

启动三台服务器的shard2 server

/app/mongodb-cluster/bin/mongod -f /app/mongodb-cluster/shard2/conf/mongodb.conf

任选分片内的一个节点，执行初始化命令

/app/mongodb-cluster/bin/mongo 192.168.56.175:27019

> rs.initiate(
...   {
...      "_id" : "rs-shard2", 
...      "members" : [
...        {"_id" : 0, "host" : "192.168.56.175:27019"},
...        {"_id" : 1, "host" : "192.168.56.176:27019"},
...        {"_id" : 2, "host" : "192.168.56.182:27019"}
...      ]
...   }
... )
#输出结果
{ "ok" : 1 }

5、mongos服务部署

①创建mongos配置文件

port=27020
#以后台方式运行
fork=true
#数据库日志存放目录
logpath=/app/mongodb-cluster/mongos/log/mongodb.log
#以追加的方式记录日志
logappend=true
#Mongos路由配置
configdb =rs-config/192.168.29.175:27017,192.168.29.176:27017,192.168.29.182:27017
#过滤一些无用的日志信息，若需要调试使用请设置为false
quiet=true
##mongodb所绑定的ip地址
bind_ip = 0.0.0.0
# max connections
maxConns=200

mongos可以为1个，也可以为多个。配置了三个节点，每个mongos配置相对独立，配置文件相同

②分发配置文件

scp mongodb.conf 192.168.56.176:/app/mongodb-cluster/mongos/conf/
scp mongodb.conf 192.168.56.182:/app/mongodb-cluster/mongos/conf/

③启动mongos

/app/mongodb-cluster/bin/mongos -f /app/mongodb-cluster/mongos/conf/mongodb.conf

④添加分片

连接任意一个mongos，添加分片信息

/app/mongodb-cluster/bin/mongo 192.168.56.175:27020

mongos> use admin
switched to db admin
mongos> sh.addShard("rs-shard1/192.168.56.175:27018,192.168.56.176:27018,192.168.56.182:27018")
{
    "shardAdded" : "rs-shard1",
    "ok" : 1,
    "operationTime" : Timestamp(1618540303, 8),
    "$clusterTime" : {
        "clusterTime" : Timestamp(1618540303, 8),
        "signature" : {
            "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
            "keyId" : NumberLong(0)
        }
    }
}
mongos> sh.addShard("rs-shard2/192.168.56.175:27019,192.168.56.175:27019,192.168.56.182:27019")
{
    "shardAdded" : "rs-shard2",
    "ok" : 1,
    "operationTime" : Timestamp(1618540329, 2),
    "$clusterTime" : {
        "clusterTime" : Timestamp(1618540329, 2),
        "signature" : {
            "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
            "keyId" : NumberLong(0)
        }
    }
}

⑤查看集群分片状态

mongos> sh.status()
--- Sharding Status --- 
  sharding version: {
      "_id" : 1,
      "minCompatibleVersion" : 5,
      "currentVersion" : 6,
      "clusterId" : ObjectId("606d735537ca11c12cc9925c")
  }
  shards:
        {  "_id" : "rs-shard1",  "host" : "rs-shard1/192.168.56.175:27018,192.168.56.176:27018,192.168.56.182:27018",  "state" : 1 }
        {  "_id" : "rs-shard2",  "host" : "rs-shard2/192.168.56.175:27019,192.168.56.176:27019,192.168.56.182:27019",  "state" : 1 }
  active mongoses:
        "4.2.13" : 3
  autosplit:
        Currently enabled: yes
  balancer:
        Currently enabled:  yes
        Currently running:  no
        Failed balancer rounds in last 5 attempts:  0
        Migration Results for the last 24 hours: 
                No recent migrations
  databases:
        {  "_id" : "config",  "primary" : "config",  "partitioned" : true }

至此，一个基本的"sharded cluster"，就已经搭建完成

6、使用分片

①开启数据库分片

mongos> use admin
switched to db admin
mongos> sh.enableSharding("test") #以test库为例
{
    "ok" : 1,
    "operationTime" : Timestamp(1618540808, 6),
    "$clusterTime" : {
        "clusterTime" : Timestamp(1618540808, 7),
        "signature" : {
            "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
            "keyId" : NumberLong(0)
        }
    }
}

②开启集合分片

mongos> sh.shardCollection("test.coll", {"name" : "hashed"}) #以集合test.coll为例
{
    "collectionsharded" : "test.coll",
    "collectionUUID" : UUID("9263ed96-7b4f-49ca-8351-9f8f990375d6"),
    "ok" : 1,
    "operationTime" : Timestamp(1618540982, 54),
    "$clusterTime" : {
        "clusterTime" : Timestamp(1618540982, 54),
        "signature" : {
            "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
            "keyId" : NumberLong(0)
        }
    }
}

说明：
(1) 第一个参数为集合的完整namespace名称
(2) 第二个参数为片键，指定根据哪个字段进行分片

③插入数据验证数据分片

use test

for (var i = 1; i <= 100000; i++){
  db.coll.insert({"id" : i, "name" : "name" + i});
}

④查看状态验证

mongos> sh.status()
--- Sharding Status --- 
  sharding version: {
      "_id" : 1,
      "minCompatibleVersion" : 5,
      "currentVersion" : 6,
      "clusterId" : ObjectId("606d735537ca11c12cc9925c")
  }
  shards:
        {  "_id" : "rs-shard1",  "host" : "rs-shard1/192.168.56.175:27018,192.168.56.176:27018,192.168.56.182:27018",  "state" : 1 }
        {  "_id" : "rs-shard2",  "host" : "rs-shard2/192.168.56.175:27019,192.168.56.176:27019,192.168.56.182:27019",  "state" : 1 }
  active mongoses:
        "4.2.13" : 3
  autosplit:
        Currently enabled: yes
  balancer:
        Currently enabled:  yes
        Currently running:  no
        Failed balancer rounds in last 5 attempts:  0
        Migration Results for the last 24 hours: 
                512 : Success
  databases:
        {  "_id" : "config",  "primary" : "config",  "partitioned" : true }
                config.system.sessions
                        shard key: { "_id" : 1 }
                        unique: false
                        balancing: true
                        chunks:
                                rs-shard1    512
                                rs-shard2    512
                        too many chunks to print, use verbose if you want to force print
        {  "_id" : "test",  "primary" : "rs-shard2",  "partitioned" : true,  "version" : {  "uuid" : UUID("44ca8ee0-262b-475a-9a98-f94fe37488c4"),  "lastMod" : 1 } }
                test.coll
                        shard key: { "name" : "hashed" }
                        unique: false
                        balancing: true
                        chunks:
                                rs-shard1    2
                                rs-shard2    2
                        { "name" : { "$minKey" : 1 } } -->> { "name" : NumberLong("-4611686018427387902") } on : rs-shard1 Timestamp(1, 0) 
                        { "name" : NumberLong("-4611686018427387902") } -->> { "name" : NumberLong(0) } on : rs-shard1 Timestamp(1, 1) 
                        { "name" : NumberLong(0) } -->> { "name" : NumberLong("4611686018427387902") } on : rs-shard2 Timestamp(1, 2) 
                        { "name" : NumberLong("4611686018427387902") } -->> { "name" : { "$maxKey" : 1 } } on : rs-shard2 Timestamp(1, 3) 

说明：分片1和分片2各有两个chunks分片，数据已均匀分布在两个分片上。

五、添加集群认证

1、生成并分发密钥文件keyfile

mkdir /app/mongodb-cluster/key -p
openssl rand -base64 745 > /app/mongodb-cluster/key/mongodb-keyfile
chmod 600 /app/mongodb-cluster/key/mongodb-keyfile 

scp -rp /app/mongodb-cluster/key 192.168.56.176:/app/mongodb-cluster/
scp -rp /app/mongodb-cluster/key 192.168.56.182:/app/mongodb-cluster/

2、添加超级管理员用户

连接任意一个mongos，创建超级管理员用户

db.createUser(
    {
        user: "root",
        pwd: "root",
        roles: [{role : "root", db : "admin"}]
    }
)

输出信息：

mongos> use admin
switched to db admin
mongos> db.createUser(
...     {
...         user: "root",
...         pwd: "root",
...         roles: [{role : "root", db : "admin"}]
...     }
... )
Successfully added user: {
    "user" : "root",
    "roles" : [
        {
            "role" : "root",
            "db" : "admin"
        }
    ]
}

在"mongos"上添加用户，用户信息实际保存在"config server"上，"mongos"本身不存储任何数据，包括用户信息。然而，"mongos"上创建的用户，是不会自动添加到"shard"分片服务器上的。为了以后方便维护shard分片服务器，分别登录到每个分片服务器的"primary"节点，添加管理员用户。

3、为分片集群创建账号

①shard1配置

use admin

db.createUser(
  {
    user: "admin",
    pwd: "admin",
    roles: [ { role: "userAdminAnyDatabase", db: "admin" } ]
  }
)

输出信息：

/app/mongodb-cluster/bin/mongo 192.168.56.175:27018

rs-shard1:PRIMARY> use admin
switched to db admin
rs-shard1:PRIMARY> db.createUser(
...   {
...     user: "admin",
...     pwd: "admin",
...     roles: [ { role: "userAdminAnyDatabase", db: "admin" } ]
...   }
... )
Successfully added user: {
    "user" : "admin",
    "roles" : [
        {
            "role" : "userAdminAnyDatabase",
            "db" : "admin"
        }
    ]
}

②shard2配置

/app/mongodb-cluster/bin/mongo 192.168.56.175:27019

rs-shard2:PRIMARY> use admin
switched to db admin
rs-shard2:PRIMARY> db.createUser(
...   {
...     user: "admin",
...     pwd: "admin",
...     roles: [ { role: "userAdminAnyDatabase", db: "admin" } ]
...   }
... )
Successfully added user: {
    "user" : "admin",
    "roles" : [
        {
            "role" : "userAdminAnyDatabase",
            "db" : "admin"
        }
    ]
}

4、开启安全认证

①修改mongos,config,shard1,shard2配置文件添加认证参数

auth = true
keyFile = /app/mongodb-cluster/key/mongodb-keyfile

vim config/conf/mongodb.conf 
vim mongos/conf/mongodb.conf #mongos不加参数auth=true
vim shard1/conf/mongodb.conf 
vim shard2/conf/mongodb.conf 

②停止所有mongodb程序

pkill mongod && pkill mongos
#杀不掉就多杀几次

③启动服务

 启动顺序：1、config server -> shard1、shard2 -> mongos

/app/mongodb-cluster/bin/mongod -f /app/mongodb-cluster/config/conf/mongodb.conf
/app/mongodb-cluster/bin/mongod -f /app/mongodb-cluster/shard1/conf/mongodb.conf
/app/mongodb-cluster/bin/mongod -f /app/mongodb-cluster/shard2/conf/mongodb.conf
/app/mongodb-cluster/bin/mongos -f /app/mongodb-cluster/mongos/conf/mongodb.conf

④验证用户访问

任意连接一个mongos

[root@k8s-node1 ~]# /app/mongodb-cluster/bin/mongo --port 27020 -u "root" -p "root" --authenticationDatabase "admin"
MongoDB shell version v4.2.13
connecting to: mongodb://127.0.0.1:27020/?authSource=admin&compressors=disabled&gssapiServiceName=mongodb
Implicit session: session { "id" : UUID("699a4543-3d62-421f-bbb3-32f1f250a9d4") }
MongoDB server version: 4.2.13
Server has startup warnings: 
2021-04-16T11:45:38.158+0800 I  CONTROL  [main] ** WARNING: You are running this process as the root user, which is not recommended.
2021-04-16T11:45:38.158+0800 I  CONTROL  [main] 
mongos> show dbs;
admin   0.000GB
config  0.003GB
test    0.005GB

 转自

linux运维、架构之路-MongoDB集群部署 - 闫新江 - 博客园
https://www.cnblogs.com/yanxinjiang/p/14600568.html