MongoDB搭建高可用集群的完整步骤（3个分片+3个副本）

share1 10.0.0.7:30011:/data/share_rs/share_rs1/share1/data/share2 10.0.0.7:40011:/data/share_rs/share_rs1/share2/data/share3 10.0.0.7:50011:/data/share_rs/share_rs1/share3/data/

share1 10.0.0.7:30012:/data/share_rs/share_rs2/share1/data/share2 10.0.0.7:40012:/data/share_rs/share_rs2/share2/data/share3 10.0.0.7:50012:/data/share_rs/share_rs2/share3/data/

share1 10.0.0.7:30013:/data/share_rs/share_rs3/share1/data/share2 10.0.0.7:40013:/data/share_rs/share_rs3/share2/data/share3 10.0.0.7:50013:/data/share_rs/share_rs3/share3/data/

config1 10.0.0.7:30002:/data/share_rs/config/config1/data/config2 10.0.0.7:30002:/data/share_rs/config/config2/data/config3 10.0.0.7:30002:/data/share_rs/config/config3/data/

mongos1 10.0.0.7:30001:/data/share_rs/mongos/mongos1/data/mongos2 10.0.0.7:30001:/data/share_rs/mongos/mongos2/data/mongos3 10.0.0.7:30001:/data/share_rs/mongos/mongos3/data/

mkdir -p /data/share_rs/{share_rs1,share_rs2,share_rs3}/{share1,share2,share3}/{data,log}mkdir -p /data/share_rs/mongos/{mongos1,mongos2,mongos3}/{data,log}mkdir -p /data/share_rs/config/{config1,config2,config3}/{data,log}

[mongo@mongo config1]$ cat mongo.confdbpath=/data/share_rs/config/config1/data/logpath=/data/share_rs/config/config1/log/mongo.loglogappend=trueport=30002fork=truerest=truehttpinterface=trueconfigsvr=true[mongo@mongo mongs1]$ cat mongo.conf logpath=/data/share_rs/mongos/mongos1/log/mongo.loglogappend=trueport=30001fork=trueconfigdb=10.0.0.7:30002,10.0.0.7:40002,10.0.0.7:50002chunkSize=1

mongod -f /data/share_rs/config/config1/mongo.confmongod -f /data/share_rs/config/config2/mongo.confmongod -f /data/share_rs/config/config3/mongo.confmongos -f /data/share_rs/mongos/mongos1/mongo.confmongos -f /data/share_rs/mongos/mongos2/mongo.confmongos -f /data/share_rs/mongos/mongos3/mongo.conf

第一个副本集的一个分片[mongo@mongo share_rs1]$ cat share1/mongo.confdbpath=/data/share_rs/share_rs1/share1/datalogpath=/data/share_rs/share_rs1/share1/log/mongo.loglogappend=trueport=30011fork=truerest=truehttpinterface=truereplSet=rs1shardsvr=true第二个副本集的一个分片[mongo@mongo share_rs2]$ cat share1/mongo.confdbpath=/data/share_rs/share_rs2/share1/datalogpath=/data/share_rs/share_rs2/share1/log/mongo.loglogappend=trueport=30012fork=truerest=truehttpinterface=truereplSet=rs2shardsvr=true第三个副本集的一个分片[mongo@mongo share_rs1]$ cat share1/mongo.confdbpath=/data/share_rs/share_rs3/share1/datalogpath=/data/share_rs/share_rs3/share1/log/mongo.loglogappend=trueport=30013fork=truerest=truehttpinterface=truereplSet=rs3shardsvr=true

mongod -f /data/share_rs/share_rs1/share1/mongo.confmongod -f /data/share_rs/share_rs1/share2/mongo.confmongod -f /data/share_rs/share_rs1/share3/mongo.confmongod -f /data/share_rs/share_rs2/share1/mongo.confmongod -f /data/share_rs/share_rs2/share2/mongo.confmongod -f /data/share_rs/share_rs2/share3/mongo.confmongod -f /data/share_rs/share_rs3/share1/mongo.confmongod -f /data/share_rs/share_rs3/share2/mongo.confmongod -f /data/share_rs/share_rs3/share3/mongo.conf[mongo@mongo share_rs]$ ps -ef | grep mongo | grep share | grep -v grepmongo  2480  1 0 12:50 ?  00:00:03 mongod -f /data/share_rs/share_rs1/share1/mongo.confmongo  2506  1 0 12:50 ?  00:00:03 mongod -f /data/share_rs/share_rs1/share2/mongo.confmongo  2532  1 0 12:50 ?  00:00:02 mongod -f /data/share_rs/share_rs1/share3/mongo.confmongo  2558  1 0 12:50 ?  00:00:03 mongod -f /data/share_rs/share_rs2/share1/mongo.confmongo  2584  1 0 12:50 ?  00:00:03 mongod -f /data/share_rs/share_rs2/share2/mongo.confmongo  2610  1 0 12:50 ?  00:00:02 mongod -f /data/share_rs/share_rs2/share3/mongo.confmongo  2636  1 0 12:50 ?  00:00:01 mongod -f /data/share_rs/share_rs3/share1/mongo.confmongo  2662  1 0 12:50 ?  00:00:01 mongod -f /data/share_rs/share_rs3/share2/mongo.confmongo  2688  1 0 12:50 ?  00:00:01 mongod -f /data/share_rs/share_rs3/share3/mongo.confmongo  3469  1 0 13:17 ?  00:00:00 mongod -f /data/share_rs/config/config1/mongo.confmongo  3485  1 0 13:17 ?  00:00:00 mongod -f /data/share_rs/config/config2/mongo.confmongo  3513  1 0 13:17 ?  00:00:00 mongod -f /data/share_rs/config/config3/mongo.confmongo  3535  1 0 13:18 ?  00:00:00 mongos -f /data/share_rs/mongos/mongos1/mongo.confmongo  3629  1 0 13:22 ?  00:00:00 mongos -f /data/share_rs/mongos/mongos2/mongo.confmongo  3678  1 0 13:22 ?  00:00:00 mongos -f /data/share_rs/mongos/mongos3/mongo.conf

1.登录第一个副本的一个分片，为其设置副本集mongo 127.0.0.1:30011/adminconfig = { _id:"rs1", members:[      {_id:0,host:"10.0.0.7:30011"},      {_id:1,host:"10.0.0.7:40011"},      {_id:2,host:"10.0.0.7:50011",arbiterOnly:true}    ]   }-- >; 注意：这里id rs1 需要与副本集中的名称一样即replSet的值rs.initiate(config){ "ok" : 1 } -- >; 提示这个说明初始化成功 2.登录第二个副本的一个分片，为其设置副本集mongo 127.0.0.1:30012/adminconfig = { _id:"rs2", members:[      {_id:0,host:"10.0.0.7:30012"},      {_id:1,host:"10.0.0.7:40012"},      {_id:2,host:"10.0.0.7:50012",arbiterOnly:true}    ]   }rs.initiate(config){ "ok" : 1 } -- >; 提示这个说明初始化成功3.登录第三个副本的一个分片，为其设置副本集mongo 127.0.0.1:30013/adminconfig = { _id:"rs3", members:[      {_id:0,host:"10.0.0.7:30013"},      {_id:1,host:"10.0.0.7:40013"},      {_id:2,host:"10.0.0.7:50013",arbiterOnly:true}    ]   }rs.initiate(config){ "ok" : 1 } -- >; 提示这个说明初始化成功

连接到第一个mongos上mongo 10.0.0.7:30001/admindb.runCommand({addshard:"rs1/10.0.0.7:30011,10.0.0.7:40011,10.0.0.7:50011",allowLocal:true}); db.runCommand({addshard:"rs2/10.0.0.7:30012,10.0.0.7:40012,10.0.0.7:50012"});db.runCommand({addshard:"rs3/10.0.0.7:30013,10.0.0.7:40013,10.0.0.7:50013"}); -- >; 将第一个分片的所有副本全部加入-- >; 如里shard是单台服务器，用 db.runCommand( { addshard : "[: ]" } )这样的命令加入-- >; 如果shard是副本集，用db.runCommand( { addshard : "replicaSetName/[:port][,serverhostname2[:port],…]" });这样的格式表示.mongos>; sh.status()--- Sharding Status --- sharding version: {  "_id" : 1,  "minCompatibleVersion" : 5,  "currentVersion" : 6,  "clusterId" : ObjectId("57f33f4d35d9c494714adfa7")} shards:  { "_id" : "rs1", "host" : "rs1/10.0.0.7:30011,10.0.0.7:40011" }  { "_id" : "rs2", "host" : "rs2/10.0.0.7:30012,10.0.0.7:40012" }  { "_id" : "rs3", "host" : "rs3/10.0.0.7:30013,10.0.0.7:40013" } active mongoses:  "3.2.7" : 3 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:

db.runCommand({enablesharding:"testcol"});-- >; 指定testdb分片生效db.runCommand({shardcollection: "testcol.testdoc",key : {id: 1} } )-- >; 指定数据库里需要分片的集合和片键-->; 插入测试数据for (var i = 1; i <;= 100000; i++){ db.testdoc.save({id:i,"name":"harvey"})}; -- >; 查看该集合的状态db.testcol.stats();