# 分片集群的部署 分片集群的部署有两种方式，一是把参数写在命令行中，而是把参数写在配置文件中，这两种方式是等价的，下面的链接是命令行参数与配置文件参数对应表： 下面讲述的是如何在单机上部署我们的测试分片集群（采用配置文件方式）： ## 1.配置说明 | 组件名 | IP | 端口 | 备注 | | ------------ | --------- | ----- | ------------- | | configserver | 127.0.0.1 | 27001 | 最先启动，副本集单成员形式 | | shard1 | 127.0.0.1 | 27003 | 第二启动，副本集单成员形式 | | shard2 | 127.0.0.1 | 27005 | 第三启动，副本集单成员形式 | | mongos | 127.0.0.1 | 27007 | 最后启动 | ## 2.创建对应的文件夹： ```python mkdir -p /home/hongxin/mongodb/conf mkdir -p /home/hongxin/mongodb/mongos/log mkdir -p /home/hongxin/mongodb/config/data mkdir -p /home/hongxin/mongodb/config/log mkdir -p /home/hongxin/mongodb/shard1/data mkdir -p /home/hongxin/mongodb/shard1/log mkdir -p /home/hongxin/mongodb/shard2/data mkdir -p /home/hongxin/mongodb/shard2/log ``` ## 3.新建config server 副本集 注：本次为了测试，该副本集只包含一个成员，实际生产环境应至少有三个成员） ### 配置文件说明 从MongoDB2.6开始，配置文件开始采用YAML的格式（YAML支持空格，不支持Tab） ```python # 2.6版本之前的写法（键值对） configsvr = true replSet=configs # 2.6版本以后的写法 sharding: clusterRole: configsvr replication: replSetName: ``` 配置参数说明： ### 新建config.conf文件 ```python >vim /home/hongxin/mongodb/conf/config.conf # 填入以下内容 systemLog: destination: file path: /home/hongxin/mongodb/config/log/configsrv.log logAppend: true storage: dbPath: /home/hongxin/mongodb/config/data journal: enabled: true processManagement: pidFilePath: /home/hongxin/mongodb/config/log/configsrv.pid fork: true net: bindIp: 127.0.0.1 port: 27001 sharding: clusterRole: configsvr # 代表此进程是config server replication: replSetName: configs # 副本集名称 ``` ### 运行并初始化 ```python # 创建进程 [hongxin@localhost conf]$ mongod -f /home/hongxin/mongodb/conf/config.conf 2018-12-02T15:29:15.608+0800 I CONTROL [main] Automatically disabling TLS 1.0, to force-enable TLS 1.0 specify --sslDisabledProtocols 'none' about to fork child process, waiting until server is ready for connections. forked process: 9587 child process started successfully, parent exiting # 连接 [hongxin@localhost conf]$ mongo --host 127.0.0.1 --port 27001 # 初始化，默认创建一个成员 >rs.initiate() # 如果要创建多个成员，参见下面 rs.initiate( { _id: "", configsvr: true, members: [ { _id : 0, host : "cfg1.example.net:27019" }, { _id : 1, host : "cfg2.example.net:27019" }, { _id : 2, host : "cfg3.example.net:27019" } ] } ) ``` ## 4.新建shard副本集 shard的创建方式与config server大同小异 ### 新建shard1.conf文件 ```python systemLog: destination: file path: /home/hongxin/mongodb/shard1/log/shard1.log logAppend: true storage: dbPath: /home/hongxin/mongodb/shard1/data journal: enabled: true processManagement: pidFilePath: /home/hongxin/mongodb/shard1/log/shard1.pid fork: true net: bindIp: 127.0.0.1 port: 27003 sharding: clusterRole: shardsvr replication: replSetName: shard1 ``` ### 创建连接及初始化 ```python [hongxin@localhost conf]$ mongod -f /home/hongxin/mongodb/conf/shard1.conf [hongxin@localhost conf]$ mongo --host 127.0.0.1 --port 27003 >rs.initiate() ``` shard2 的创建与上面类似 注： shard 副本集的名字与 config server的名字不能一样！ ## 5.新建mongos ### 新建 mongos.conf 注意 mongos.conf 没有存储选项 ```python systemLog: destination: file path: /home/hongxin/mongodb/mongos/log/mongos.log logAppend: true processManagement: pidFilePath: /home/hongxin/mongodb/mongos/log/mongos.pid fork: true net: bindIp: 127.0.0.1 port: 27007 sharding: configDB: configs/127.0.0.1:27001 ``` ### 创建连接及初始化 ```python [hongxin@localhost conf]$ mongos -f /home/hongxin/mongodb/conf/mongos.conf [hongxin@localhost conf]$ mongo --host 127.0.0.1 --port 27007 mongos> sh.addShard("shard1/127.0.0.1:27003, 127.0.0.1:27005") ``` **至此，单机分片集群已搭建完毕** ## 测试 ```python mongos> use testdb switched to db testdb mongos> db.mycoll.insertOne({id: 1, "test": "test_shard"}) { "acknowledged" : true, "insertedId" : ObjectId("5c039517071e1ba7e339fe24") } # 首先要让数据库启动分片 mongos> sh.enableSharding("testdb") { "ok" : 1, "operationTime" : Timestamp(1543738671, 3), "$clusterTime" : { .... } mongos> db.mycoll.find({}) { "_id" : ObjectId("5c039517071e1ba7e339fe24"), "id" : 1, "test" : "test_shard" } # 因为集合不为空，所以在执行shardCollection前要先创建索引，如果集合为空，会自动创建索引 mongos> db.mycoll.createIndex({id: 1}) { "raw" : { "shard1/127.0.0.1:27003" : { "createdCollectionAutomatically" : false, "numIndexesBefore" : 1, "numIndexesAfter" : 2, "ok" : 1 } }, ... } # 查看索引 mongos> db.mycoll.getIndexes() [ { "v" : 2, "key" : { "_id" : 1 }, "name" : "_id_", "ns" : "testdb.mycoll" }, { "v" : 2, "key" : { "id" : 1 }, "name" : "id_1", "ns" : "testdb.mycoll" } ] # 使用 _id 作为分片键 mongos> sh.shardCollection("testdb.mycoll", {_id:1}) { "collectionsharded" : "testdb.mycoll", "collectionUUID" : UUID("cffb9588-52df-4df4-bb1a-9e461602cabd"), "ok" : 1, ... } # 因为先使用 _id 作为分片键，再使用 id 会报错 mongos> sh.shardCollection("testdb.mycoll", {id:1}) { "ok" : 0, "errmsg" : "sharding already enabled for collection testdb.mycoll with options { _id: \"testdb.mycoll\", lastmodEpoch: ObjectId('5c039722733b2ba5a2bafebd'), lastmod: new Date(4294967296), dropped: false, key: { _id: 1.0 }, unique: false, uuid: UUID(\"cffb9588-52df-4df4-bb1a-9e461602cabd\") }", "code" : 23, "codeName" : "AlreadyInitialized", ... } # 插入100000条数据 mongos> for (var i = 1; i <= 100000; i++) db.mycoll.save({id:i,"test1":"test_shard"}); WriteResult({ "nInserted" : 1 }) # 查看结果（不知道是不是因为分片键为_id的缘故，数据并没有分散到shard2中） mongos> db.mycoll.stats() ``` ## 后期运维 mongodb的启动顺序是，先启动配置服务器，在启动分片，最后启动mongos. mongod -f /usr/local/mongodb/conf/config.conf\ mongod -f /usr/local/mongodb/conf/shard1.conf\ mongod -f /usr/local/mongodb/conf/shard2.conf\ mongod -f /usr/local/mongodb/conf/shard3.conf\ mongod -f /usr/local/mongodb/conf/mongos.conf 关闭时，直接killall杀掉所有进程\ killall mongod\ killall mongos 参考： \ --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://xiehongxin.gitbook.io/mongodb/sharding/shard_cluster_deploy.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.