mongodb的集群结构如上图
网上有个mongo3.0的集群例子:
http://www.zzvips.com/uploads/allimg/kheocx2o22o
router提供入口,mongo客户端通过router连入集群(本例只配置一个route集群)
Config Servers辅助记录数据分片(一个集群)
Shard为数据分片集群(本例中配置两个,用于验证分片),
本例中,为每个集群(shard config)三个mongo实例
config与shard为同一个类型的进程mongod
route则为mongos进程
下载mongo二进制压缩包
https://fastdl.mongodb.org/linux/mongodb-linux-x86_64-rhel70-5.0.6.tgz
解压可以看到
用openssl生成一个key,用于mongo集群内部数据通迅
openssl rand -base64 123 > keyfile
mongod配置为件(config与shard通用)
mongo_node.conf:
storage: engine: wiredTiger directoryPerDB: true journal: enabled: true systemLog: destination: file logAppend: true operationProfiling: slowOpThresholdMs: 10000 replication: oplogSizeMB: 10240 processManagement: fork: true security: authorization: "disabled"
mongos的配置文件(即图中的route)
mongos.conf:
systemLog: destination: file logAppend: true processManagement: fork: true
启动config集群(3个mongod进程)
WORK_DIR=/home/???/go/mongodb/mongo_test KEYFILE=$WORK_DIR/key/keyfile cat $KEYFILE CONFFILE=$WORK_DIR/conf/mongo_node.conf cat $CONFFILE MONGOD=mongod echo $MONGOD $MONGOD --port 26001 --bind_ip_all --configsvr --replSet configReplSet --keyFile $KEYFILE --dbpath $WORK_DIR/config_cluster/conf_n1/data --pidfilepath $WORK_DIR/config_cluster/conf_n1/db.pid --logpath $WORK_DIR/config_cluster/conf_n1/db.log --config $CONFFILE $MONGOD --port 26002 --bind_ip_all --configsvr --replSet configReplSet --keyFile $KEYFILE --dbpath $WORK_DIR/config_cluster/conf_n2/data --pidfilepath $WORK_DIR/config_cluster/conf_n2/db.pid --logpath $WORK_DIR/config_cluster/conf_n2/db.log --config $CONFFILE $MONGOD --port 26003 --bind_ip_all --configsvr --replSet configReplSet --keyFile $KEYFILE --dbpath $WORK_DIR/config_cluster/conf_n3/data --pidfilepath $WORK_DIR/config_cluster/conf_n3/db.pid --logpath $WORK_DIR/config_cluster/conf_n3/db.log --config $CONFFILE
启动成功后
用命令mongo --port 26001 --host 127.0.0.1
如下图,进入mongo的shell
在shell中输入如下js代码 设置config集群
cfg={ _id:"configReplSet", configsvr: true, members:[ {_id:0, host:'127.0.0.1:26001'}, {_id:1, host:'127.0.0.1:26002'}, {_id:2, host:'127.0.0.1:26003'} ]}; rs.initiate(cfg);
三个config mongo进程会自动选出一个primary,过一会再进入shell就会发现 shell提示变成primary
顺便给config添加一个admin用户,(一个集群只要在primary进程添加一次,会自动同步给secondary)
use admin db.createUser({ user:'admin',pwd:'123456', roles:[ {role:'clusterAdmin',db:'admin'}, {role:'userAdminAnyDatabase',db:'admin'}, {role:'dbAdminAnyDatabase',db:'admin'}, {role:'readWriteAnyDatabase',db:'admin'} ]})
同样之后shard也做同样的添加用户操作,便于后继观察数据
启动shard
WORK_DIR=/home/???/go/mongodb/mongo_test KEYFILE=$WORK_DIR/key/keyfile cat $KEYFILE CONFFILE=$WORK_DIR/conf/mongo_node.conf cat $CONFFILE MONGOD=mongod echo $MONGOD echo "start shard1 replicaset" $MONGOD --port 27001 --bind_ip_all --shardsvr --replSet shard1 --keyFile $KEYFILE --dbpath $WORK_DIR/shard1/sh1_n1/data --pidfilepath $WORK_DIR/shard1/sh1_n1/db.pid --logpath $WORK_DIR/shard1/sh1_n1/db.log --config $CONFFILE $MONGOD --port 27002 --bind_ip_all --shardsvr --replSet shard1 --keyFile $KEYFILE --dbpath $WORK_DIR/shard1/sh1_n2/data --pidfilepath $WORK_DIR/shard1/sh1_n2/db.pid --logpath $WORK_DIR/shard1/sh1_n2/db.log --config $CONFFILE $MONGOD --port 27003 --bind_ip_all --shardsvr --replSet shard1 --keyFile $KEYFILE --dbpath $WORK_DIR/shard1/sh1_n3/data --pidfilepath $WORK_DIR/shard1/sh1_n3/db.pid --logpath $WORK_DIR/shard1/sh1_n3/db.log --config $CONFFILE
用mongo --port 27001 --host 127.0.0.1进入mongo shell
cfg={ _id:"shard1", members:[ {_id:0, host:'127.0.0.1:27001'}, {_id:1, host:'127.0.0.1:27002'}, {_id:2, host:'127.0.0.1:27003'} ]}; rs.initiate(cfg);
同样用之前的添加用户的js
并用同样的方法启动shard2集群,用于实验数据分片
对应的目录与分片名改成shard2
启动route
WORK_DIR=/home/???/go/mongodb/mongo_test KEYFILE=$WORK_DIR/key/keyfile cat $KEYFILE CONFFILE=$WORK_DIR/conf/mongos.conf cat $CONFFILE MONGOS=mongos echo $MONGOS echo "start mongos route instances" $MONGOS --port=25001 --bind_ip_all --configdb configReplSet/127.0.0.1:26001,127.0.0.1:26002,127.0.0.1:26003 --keyFile $KEYFILE --pidfilepath $WORK_DIR/route/r_n1/db.pid --logpath $WORK_DIR/route/r_n1/db.log --config $CONFFILE $MONGOS --port 25002 --bind_ip_all --configdb configReplSet/127.0.0.1:26001,127.0.0.1:26002,127.0.0.1:26003 --keyFile $KEYFILE --pidfilepath $WORK_DIR/route/r_n2/db.pid --logpath $WORK_DIR/route/r_n2/db.log --config $CONFFILE $MONGOS --port 25003 --bind_ip_all --configdb configReplSet/127.0.0.1:26001,127.0.0.1:26002,127.0.0.1:26003 --keyFile $KEYFILE --pidfilepath $WORK_DIR/route/r_n3/db.pid --logpath $WORK_DIR/route/r_n3/db.log --config $CONFFILE
路由添加分片
用mongo --port 25001 --host 127.0.0.1 -u admin -p 123456进入shell
或者这样也可 mongo mongodb://admin:123456@127.0.0.1:25001
在mongo shell分别执行以下两行js
sh.addShard("shard1/127.0.0.1:27001") sh.addShard("shard2/127.0.0.1:27011")
创建一个mongo database与collection
并设置分片
use test sh.enableSharding("test") db.createCollection("test_shard") sh.shardCollection("test.test_shard", {_id:"hashed"}, false, { numInitialChunks: 4} )
在mongo shell用以下js添加数据,可以修改循环次数避免测试时间过长
var cnt = 0; for(var i=0; i<1000; i++){ var dl = []; for(var j=0; j<100; j++){ dl.push({ "bookId" : "BBK-" + i + "-" + j, "type" : "Revision", "version" : "IricSoneVB0001", "title" : "Jackson's Life", "subCount" : 10, "location" : "China CN Shenzhen Futian District", "author" : { "name" : 50, "email" : "RichardFoo@yahoo.com", "gender" : "female" }, "createTime" : new Date() }); } cnt += dl.length; db.test_shard.insertMany(dl); print("insert ", cnt); }
在windows下安装mongodb,利用自带的compass客户端观察两个shard集群
会发现数据分流到两个集群了
也可以直接连route观察数据
补充:
把js存到文件里给shell执行会比较方便
执行js命令如下:
mongo mongodb://admin:123456@127.0.0.1:25001 ./test.js
示例js代码:
print('=========WECOME=========='); conn = new Mongo("mongodb://admin:123456@192.168.2.129:25001"); db = conn.getDB("testjs") sh.enableSharding("testjs") db.createCollection("testjs_col") sh.shardCollection("testjs.testjs_col", {_id:"hashed"}, false, { numInitialChunks: 4} ) var dl = []; for(var j=0; j<10; j++){ dl.push({ "bookId" : "BBK-" + 0 + "-" + j, "type" : "Revision", "version" : "IricSoneVB0001", "title" : "Jackson's Life", "subCount" : 10, "location" : "China CN Shenzhen Futian District", "author" : { "name" : 50, "email" : "RichardFoo@yahoo.com", "gender" : "female" }, "createTime" : new Date() }); } db.testjs_col.insertMany(dl); cursor = db.testjs_col.find(); printjson(cursor.toArray());