搭建mongodb分片集群

 

 

 

 

 

 

 

 注意：mongos、config、shard三个角色的实例的keyfile内容保证完全一致：

 

如果搭建副本集时，出错，那么删掉

 

 

 

 

 config副本集配置文件内容：使用mongod启动：

[work@xxx etc]$ cat mongodb.conf
systemLog:
  destination: file
  path: /home/work/mongodb/mongo_40000/log/mongodb.log
  logAppend: true

cloud:
   monitoring:
      free:
         state: off

#net Options
net:
  maxIncomingConnections: 10240
  port: 40000
  bindIp: 10.10.10.40,localhost

#security Options
security:
  authorization: 'enabled'
  keyFile: /home/work/mongodb/mongo_40000/etc/igoodful_config
  clusterAuthMode: "keyFile"

#storage Options
storage:
  engine: "wiredTiger"
  directoryPerDB: true
  dbPath: /home/work/mongodb/mongo_40000/data
  journal:
    enabled: true
    commitIntervalMs: 100
  wiredTiger:
    engineConfig:
      directoryForIndexes: true
      cacheSizeGB: 4
      journalCompressor: "snappy"
    collectionConfig:
      blockCompressor: "snappy"
    indexConfig:
      prefixCompression: true
    #wiredTigerCollectionConfigString: lsm
    #wiredTigerIndexConfigString: lsm

#replication Options
replication:
  oplogSizeMB: 81920 #80GB
  replSetName: igoodful_config

#operationProfiling Options
operationProfiling:
  slowOpThresholdMs: 100
  mode: "slowOp"

processManagement:
  fork: true
  pidFilePath: /home/work/mongodb/mongo_40000/tmp/mongo_40000.pid

sharding:
  clusterRole: configsvr

 

 

 

 

mongos配置文件：使用mongos启动：

[work@hostname etc]$ cat mongodb.conf
systemLog:
  destination: file
  path: /home/work/mongodb_dba/mongo_30000/log/mongodb_dba.log
  logAppend: true

#net Options
net:
  maxIncomingConnections: 10240
  port: 30000
  bindIp: 10.10.10.30,localhost
  serviceExecutor : adaptive

#security Options
security:
  keyFile: /home/work/mongodb_dba/mongo_30000/etc/test.mongos
  clusterAuthMode: "keyFile"

sharding:
  configDB: "igoodful_config/10.10.10.40:40000,10.10.10.41:40000,10.10.10.42:40000"

processManagement:
  fork: true
  pidFilePath: /home/work/mongodb_dba/mongo_30000/tmp/mongo_30000.pid

 

 

分片副本集配置：使用mongod启动：

启动命令：/home/work/mongodb/4.0.14/bin/mongod --config /home/work/mongodb/mongo_28001/etc/mongodb.conf


[work@xxx etc]$ cat mongodb.conf 
systemLog:
  destination: file
  path: "/home/work/mongodb/mongo_28001/log/mongodb.log"
  logAppend: true

cloud:
   monitoring:
      free:
         state: off

#net Options
net:
  maxIncomingConnections: 10240
  port: 28001
  bindIp: 10.10.10.50,localhost
  serviceExecutor : adaptive

#security Options
security:
  authorization: 'enabled'
  keyFile: /home/work/mongodb/mongo_28001/etc/igoodful_shard
  clusterAuthMode: "keyFile"

#storage Options
storage:
  engine: "wiredTiger"
  directoryPerDB: true
  dbPath: /home/work/mongodb/mongo_28001/data
  journal:
    enabled: true
    commitIntervalMs: 100
  wiredTiger:
    engineConfig:
      directoryForIndexes: true
      cacheSizeGB: 25
      journalCompressor: "snappy"
    collectionConfig:
      blockCompressor: "snappy"
    indexConfig:
      prefixCompression: true
    #wiredTigerCollectionConfigString: lsm
    #wiredTigerIndexConfigString: lsm

#replication Options
replication:
  oplogSizeMB: 81920 #80GB
  replSetName: igoodful_shard

#operationProfiling Options
operationProfiling:
  slowOpThresholdMs: 100
  mode: "slowOp"

processManagement:
  fork: true
  pidFilePath: /home/work/mongodb/mongo_28001/tmp/mongo_28001.pid

sharding:
  clusterRole: shardsvr

 

 

 

 

 

mongos启动：（注意：不是用mongod来启动）

 

注意一：mongos启动

[work@hostname mongodb_dba]$ /home/work/mongodb_dba/4.0/bin/mongos --config /home/work/mongodb_dba/mongo_30000/etc/mongodb.conf 
2020-10-27T14:49:34.383+0800 W SHARDING [main] Running a sharded cluster with fewer than 3 config servers should only be done for testing purposes and is not recommended for production.
about to fork child process, waiting until server is ready for connections.
forked process: 30389
child process started successfully, parent exiting

 

注意二：配置的config需要有相同的keyfile认证，即config开启认证

 

[work@tj1-ai-stag-db-i1-01 mongodb_dba]$ /home/work/mongodb_dba/4.0/bin/mongos --config /home/work/mongodb_dba/mongo_30000/etc/mongodb.conf 
2020-10-27T16:40:09.882+0800 W SHARDING [main] Running a sharded cluster with fewer than 3 config servers should only be done for testing purposes and is not recommended for production.
about to fork child process, waiting until server is ready for connections.
forked process: 300974

 

 登录mongos：

 

/home/work/mongodb_dba/4.0/bin/mongo  --host 10.38.167.100 --port 30000  -u mongo_dba  -p 123456 --authenticationDatabase admin


mongos> use admin
switched to db admin
mongos> sh.addShard("test_shard1/10.10.10.10:28001");
{
"shardAdded" : "test_shard1",
"ok" : 1,
"operationTime" : Timestamp(1603788736, 3),
"$clusterTime" : {
"clusterTime" : Timestamp(1603788736, 3),
"signature" : {
"hash" : BinData(0,"32np1vC8xMXEKRFRgE5MGjZvKqM="),
"keyId" : NumberLong("6888216464256401439")
}
}
}
mongos>

mongos> sh.status()
--- Sharding Status --- 
sharding version: {
"_id" : 1,
"minCompatibleVersion" : 5,
"currentVersion" : 6,
"clusterId" : ObjectId("5f97dc61cd2244b6a298aaf4")
}
shards:
{ "_id" : "igoodful_shard", "host" : "igoodful_shard/10.10.10.50:28001,10.10.10.51:28001,10.10.10.52:28001", "state" : 1 }
active mongoses:
"4.0.17-10" : 1
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 }
config.system.sessions
shard key: { "_id" : 1 }
unique: false
balancing: true
chunks:
test_shard1 1
{ "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : test_shard1 Timestamp(1, 0)

mongos>

 

 

向分片集群添加分片副本集：登录mongos操作：

#######################################################
sh.addShard("igoodful_shard/10.10.10.50:28001,10.10.10.51:28001,10.10.10.52");

sh.addShard("分片集群的副本集名称/该副本集的ip列表")

 

 

 

 对数据库启用分片：

# 对数据库启动分片：

sh.enableSharding("<database>")

 

 对集合启用分片：

# 对集合分片，该集合所在数据库必须先启用分片

db.adminCommand( { shardCollection: "records.people", key: { zipcode: 1 } } )

{

shardCollection: "<database>.<collection>",

key: { <field1>: <1|"hashed">, ... },

unique: <boolean>,

numInitialChunks: <integer>,

presplitHashedZones: <boolean>,

collation: { locale: "simple" }

}

 

 

 

# 查看balance是否已经开启


 sh.getBalancerState()

# 关闭均衡器：sh.stopBalancer()

# 开启均衡器：sh.startBalancer()

 

 

查看当前分片集群中的所有分片副本集：

db.adminCommand( { listShards: 1 } )

sh.status() 

db.printShardingStatus() 

 

在 admin 数据库中,运行 removeShard 命令.运行之后会开始将这个分片的数据块”转移”到其他分片的过程,比如,对一个命名为 mongodb0 的分片运行：db.runCommand( { removeShard: "mongodb0" } )

use admin

db.runCommand( { removeShard: "mongodb0" } )

# 这个操作是立刻返回的，返回数据如下：

{
    "msg" : "draining started successfully",
    "state" : "started",
    "shard" : "mongodb0",
    "ok" : 1
}

 

检查迁移状态：

检查迁移的状态,再次在 admin 数据库运行 removeShard 命令,比如,对一个命名为 mongodb0 的分片,运行:

use admin
db.runCommand( { removeShard: "mongodb0" } )
这条命令返回类似如下的输出:

{
     "msg" : "draining ongoing",
    "state" : "ongoing",
    "remaining" : {
        "chunks" : 42,
        "dbs" : 1
    },
    "ok" : 1
}
在输出结果中, remaining 文档显示的是MongoDB必须迁移到其他分片的数据块中剩余的数据块数量与”primary”在这个分片的数据库数量.

在 remaining 字段变为0之前,持续运行 removeShard 命令检查状态.这个命令需要在 admin 数据库上运行,在其他库可以使用 sh._adminCommand 命令操作.

 

 

迁移没有分片的数据：

如果这个分片是一个或多个数据库的 primary shard ,上面会存储没有分片的数据,如果不是,则跳过 完成迁移 任务.

在集群中,没有分片的数据库只会将数据存放在一个分片上,这个分片就是这个数据库的主分片.(不同的数据库可以有不同的主分片.)

 

 查看数据库对应的主分片：

sh.status()
db.printShardingStatus()
在返回的文档中, databases 字段列出了所有数据库和它的主分片.比如,以下的 databases 字段显示了 products 数据库使用 mongodb0 作为主分片.

{  "_id" : "products",  "partitioned" : true,  "primary" : "mongodb0" }
将数据库迁移到另一个分片,需要使用 movePrimary 命令.使用如下命令将所有的剩余的未分片的数据从 mongodb0 迁移到 mongodb1 上.

 

 迁移主分片：db.runCommand( { movePrimary: "products", to: "mongodb1" })

db.runCommand( { movePrimary: "products", to: "mongodb1" })
This command does not return until MongoDB completes moving all data, which may take a long time. The response from this command will resemble the following:

{ "primary" : "mongodb1", "ok" : 1 }
警告

这个命令直到全部数据迁移完成才会返回,可能会花费较长时间.最后返回的结果类似这样:

 

完成迁移：

为了清除所有的元信息,并结束删除分片的过程,再次执行 removeShard , 比如,对 mongodb0 这个分片,执行:

use admin
db.runCommand( { removeShard: "mongodb0" } )
在完成时会显示出成功的信息:

{
    "msg" : "removeshard completed successfully",
    "state" : "completed",
    "shard" : "mongodb0",
    "ok" : 1
}
一旦 state 的值变为 “completed”,就可以安全地停止 mongodb0` 分片上的monod进程.

 

 

 

 

 

############################