MongoDB分片集群部署详解

 一、环境说明

1、我们prod环境MongoDB的集群架构是做的分片集群的部署，但是目前我们没有分片，即所有数据都在一个分片上，后期如果数量大，需要分配，集群随时可以分片，对业务方透明

2、各个角色的部署情况

3、MongoDB版本

mongos> db.version()
4.0.4-62-g7e345a7

二、基础信息准备

0、系统优化

echo "never" >/sys/kernel/mm/transparent_hugepage/enabled
echo "never" >/sys/kernel/mm/transparent_hugepage/defrag

1、下载MongoDB二进制文件

cd /chj/app
wget ops.chehejia.com:9090/pkg/chj_mongodb_4.0.4.tar.gz
tar -zxvf chj_mongodb_4.0.4.tar.gz

2、相关目录建立

#建立base目录
mkdir /chj/data/mongodb/chj_db
#把MongoDB二进制文件移动到base目录下的bin文件夹
mv chj_mongodb_4.0.4/bin /chj/data/mongodb/chj_db/bin
#建立认证文件目录
mkdir /chj/data/mongodb/chj_db/auth
#建立配置文件目录
mkdir /chj/data/mongodb/chj_db/conf
#建立config server的data和日志目录
mkdir /chj/data/mongodb/chj_db/config/data -p
mkdir /chj/data/mongodb/chj_db/config/log
#建立mongos的日志目录
mkdir /chj/data/mongodb/chj_db/mongos/log -p
#建立数据节点data和日志目录 
mkdir /chj/data/mongodb/chj_db/shard1/data -p
mkdir /chj/data/mongodb/chj_db/shard1/log

3、相关配置文件编写

A、mongos的配置文件编写

vim /chj/data/mongodb/chj_db/conf/mongos.conf
systemLog:
 destination: file
 logAppend: true
 path: /chj/data/mongodb/chj_db/mongos/log/mongos.log

processManagement:
 fork: true # fork and run in background
 pidFilePath: /chj/data/mongodb/chj_db/mongos/log/mongos.pid # location of pidfile
 timeZoneInfo: /usr/share/zoneinfo

net:
 port: 27000
 bindIpAll: true
 maxIncomingConnections: 1000
 unixDomainSocket:
  enabled: true
  pathPrefix: /chj/data/mongodb/chj_db/mongos/log
  filePermissions: 0700

security:
 keyFile: /chj/data/mongodb/chj_db/auth/keyfile.key
# authorization: enabled

#replication:

sharding:
 configDB: repl_configsvr/172.21.244.101:27100,172.21.244.102:27100,172.21.244.94:27100

B、config server的配置文件编写

vim /chj/data/mongodb/chj_db/conf/config.conf
systemLog:
 destination: file
 logAppend: true
 path: /chj/data/mongodb/chj_db/config/log/congigsrv.log

storage:
 dbPath: /chj/data/mongodb/chj_db/config/data
 journal:
  enabled: true
 wiredTiger:
  engineConfig:
   directoryForIndexes: true

processManagement:
 fork: true # fork and run in background
 pidFilePath: /chj/data/mongodb/chj_db/config/log/configsrv.pid # location of pidfile
 timeZoneInfo: /usr/share/zoneinfo

net:
 port: 27100
 bindIp: 0.0.0.0 # Enter 0.0.0.0,:: to bind to all IPv4 and IPv6 addresses or, alternatively, use the net.bindIpAll setting.
 #bindIpAll: true
 maxIncomingConnections: 1000
 unixDomainSocket:
  enabled: true
  pathPrefix: /chj/data/mongodb/chj_db/config/data
  filePermissions: 0700

security:
 keyFile: /chj/data/mongodb/chj_db/auth/keyfile.key
 authorization: enabled

replication:
 replSetName: repl_configsvr
sharding:
 clusterRole: configsvr

C、存储节点的配置文件编写

vim /chj/data/mongodb/chj_db/conf/shard1.conf
systemLog:
 destination: file
 logAppend: true
 path: /chj/data/mongodb/chj_db/shard1/log/shard1.log

storage:
 dbPath: /chj/data/mongodb/chj_db/shard1/data
 journal:
  enabled: true
 wiredTiger:
  engineConfig:
   directoryForIndexes: true

processManagement:
 fork: true # fork and run in background
 pidFilePath: /chj/data/mongodb/chj_db/shard1/log/shard1.pid # location of pidfile
 timeZoneInfo: /usr/share/zoneinfo

net:
 port: 27101
 bindIp: 0.0.0.0 # Enter 0.0.0.0,:: to bind to all IPv4 and IPv6 addresses or, alternatively, use the net.bindIpAll setting.
 #bindIpAll: true
 maxIncomingConnections: 1000
 unixDomainSocket:
  enabled: true
  pathPrefix: /chj/data/mongodb/chj_db/shard1/data
  filePermissions: 0700

security:
 keyFile: /chj/data/mongodb/chj_db/auth/keyfile.key
 authorization: enabled

replication:
 replSetName: shard1
sharding:
 clusterRole: shardsvr

4、生产key认证文件

echo "chj123456" >/chj/data/mongodb/chj_db/auth/keyfile.key
#设置文件的权限为400，不然服务无法启动
chmod 400 /chj/data/mongodb/chj_db/auth/keyfile.key

三、集群初始化

1、启动 config server 服务

/chj/data/mongodb/chj_db/bin/mongod -f /chj/data/mongodb/chj_db/conf/config.conf

2、初始化config server集群

#登录其中一个config server节点
/chj/data/mongodb/chj_db/bin/mongo --port 27100
#配置集群
config = { _id:"repl_configsvr",members:[ {_id:0,host:"172.21.244.101:27100"}, {_id:1,host:"172.21.244.102:27100"}, {_id:2,host:"172.21.244.94:27100"}] }
#初始化集群
rs.initiate(config)
PS：
结果输出如下，说明集群初始化成功，可以通过rs.status()命令查看集群状态
{
    "ok" : 1,
    "$gleStats" : {
        "lastOpTime" : Timestamp(1557538260, 1),
        "electionId" : ObjectId("000000000000000000000000")
    },
    "lastCommittedOpTime" : Timestamp(0, 0)
}

3、启动存储节点服务

/chj/data/mongodb/chj_db/bin/mongod -f /chj/data/mongodb/chj_db/conf/shard1.conf

4、初始化存储集群

#登录你希望是主节点的服务器
/chj/data/mongodb/chj_db/bin/mongo --port 27101
#配置集群
config = { _id:"shard1",members:[ {_id:0,host:"172.21.244.101:27101"}, {_id:1,host:"172.21.244.102:27101"},{_id:2,host:"172.21.244.94:27101",arbiterOnly:true}] }
#初始化集群
rs.initiate(config)
PS：
结果输出如下，说明集群初始化成功，可以通过rs.status()命令查看集群状态
{ "ok" : 1 }

5、添加存储集群的管理账号

登录主节点

/chj/data/mongodb/chj_db/bin/mongo --port 27101
use admin
db.createUser(
{
user: "root",
pwd: "123456",
roles: [ { role: "root", db: "admin" } ]
}
)

6、启动mongos 服务

/chj/data/mongodb/chj_db/bin/mongos -f /chj/data/mongodb/chj_db/conf/mongos.conf

7、添加config server的管理账号

登录任意一个mongos节点

/chj/data/mongodb/chj_db/bin/mongo --port 27000
use admin
db.createUser(
{
user: "root",
pwd: "123456",
roles: [ { role: "root", db: "admin" } ]
}
)

8、把存储节点添加到mongos

登录任意一个mongos节点（如果是在上一步的窗口，需要退出重新登录）

/chj/data/mongodb/chj_db/bin/mongo --port 27000
use admin
db.auth('root','123456')
#添加分片
sh.addShard('shard1/172.21.244.101:27101,172.21.244.102:27101,172.21.244.94:27101')
#查看分片状态
sh.status()

四、交付业务方

1、建立应用账号

登录任意一个mongos节点
/chj/data/mongodb/chj_db/bin/mongo --port 27000
use admin
db.auth('root','123456')
#切到业务数据库
use chj_db
#建立读写账号
db.createUser(
{
  user: "chj_db_rw",
  pwd: "123456",
  roles: [
   { role: "readWrite", db: "chj_db" },
   { role: "dbOwner", db: "chj_db" }
  ]
}
)
#建立只读账号(根据业务需求确认是否需要)
db.createUser(
{
user: "chj_db_r",
pwd: "123456",
roles: [ { role: "read", db: "chj_db" } ]
}
)

2、交付开发人员信息

连接地址：172.21.244.101:27000，172.21.244.102:27000，172.21.244.94:27000
库名：chj_db
账号：chj_db_rw
密码：123456

五、数据库启用分片

如果后期业务量大，需要开启分片，配置如下

#指定需要分片的数据库
mongos> sh.enableSharding("chj_db") 
{
    "ok" : 1,
    "operationTime" : Timestamp(1557546835, 3),
    "$clusterTime" : {
        "clusterTime" : Timestamp(1557546835, 3),
        "signature" : {
            "hash" : BinData(0,"bkrrr8Kxrr9j9udrDc/hURHld38="),
            "keyId" : NumberLong("6689575940508352541")
        }
    }
}
#在chj_db数据库和users集合中创建了name和age为升序的片键
mongos> sh.shardCollection("chj_db.users",{name:1,age:1}) 
{
    "collectionsharded" : "chj_db.users",
    "collectionUUID" : UUID("59c0b99f-efff-4132-b489-f6c7e3d98f42"),
    "ok" : 1,
    "operationTime" : Timestamp(1557546861, 12),
    "$clusterTime" : {
        "clusterTime" : Timestamp(1557546861, 12),
        "signature" : {
            "hash" : BinData(0,"UBB1A/YODnmXwG5eAhgNLcKVzug="),
            "keyId" : NumberLong("6689575940508352541")
        }
    }
}
#查看分片情况
mongos> sh.status() 
--- Sharding Status ---
 sharding version: {
    "_id" : 1,
    "minCompatibleVersion" : 5,
    "currentVersion" : 6,
    "clusterId" : ObjectId("5cd625e0da695346d740f749")
 }
 shards:
    { "_id" : "shard1", "host" : "shard1/172.21.244.101:27101,172.21.244.102:27101", "state" : 1 }
 active mongoses:
    "4.0.4-62-g7e345a7" : 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" : "chj_db", "primary" : "shard1", "partitioned" : true, "version" : { "uuid" : UUID("82088bc7-7b98-4033-843d-7058d8d959f6"), "lastMod" : 1 } }
        chj_db.users
            shard key: { "name" : 1, "age" : 1 }
            unique: false
            balancing: true
            chunks:
                shard1 1
            { "name" : { "$minKey" : 1 }, "age" : { "$minKey" : 1 } } -->> { "name" : { "$maxKey" : 1 }, "age" : { "$maxKey" : 1 } } on : shard1 Timestamp(1, 0)
    { "_id" : "config", "primary" : "config", "partitioned" : true }
        config.system.sessions
            shard key: { "_id" : 1 }
            unique: false
            balancing: true
            chunks:
                shard1 1
            { "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : shard1 Timestamp(1, 0)