Mongodb分片结合复制集,在实现不同数据库分片的基础上，实现集合中document的分片（使用片键）

一、分片概述；

二、分片存储原理；

三、分片的片键；

四、案例：mongodb分片结合复制集高效存储；

一、分片概述：

概述：分片（sharding）是指将数据库拆分，将其分散在不同的机器上的过程。分片集群（sharded cluster）是一种水平扩展数据库系统性能的方法，能够将数据集分布式存储在不同的分片（shard）上，每个分片只保存数据集的一部分，MongoDB保证各个分片之间不会有重复的数据，所有分片保存的数据之和就是完整的数据集。分片集群将数据集分布式存储，能够将负载分摊到多个分片上，每个分片只负责读写一部分数据，充分利用了各个shard的系统资源，提高数据库系统的吞吐量。

注：mongodb3.2版本后，分片技术必须结合复制集完成；

应用场景：

1.单台机器的磁盘不够用了，使用分片解决磁盘空间的问题。

2.单个mongod已经不能满足写数据的性能要求。通过分片让写压力分散到各个分片上面，使用分片服务器自身的资源。

3.想把大量数据放到内存里提高性能。和上面一样，通过分片使用分片服务器自身的资源。

二、分片存储原理：

Mongodb分片结合复制集,在实现不同数据库分片的基础上，实现集合中document的分片（使用片键）

存储方式：数据集被拆分成数据块（chunk），每个数据块包含多个doc，数据块分布式存储在分片集群中。

角色：

Config server：MongoDB负责追踪数据块在shard上的分布信息，每个分片存储哪些数据块，叫做分片的元数据，保存在config server上的数据库 config中，一般使用3台config server，所有config server中的config数据库必须完全相同（建议将config server部署在不同的服务器，以保证稳定性）；

Shard server：将数据进行分片，拆分成数据块（chunk），每个trunk块的大小默认为64M，数据块真正存放的单位；

Mongos server：数据库集群请求的入口，所有的请求都通过mongos进行协调，查看分片的元数据，查找chunk存放位置，mongos自己就是一个请求分发中心，在生产环境通常有多mongos作为请求的入口，防止其中一个挂掉所有的mongodb请求都没有办法操作。

总结：应用请求mongos来操作mongodb的增删改查，配置服务器存储数据库元信息，并且和mongos做同步，数据最终存入在shard（分片）上，为了防止数据丢失，同步在副本集中存储了一份，仲裁节点在数据存储到分片的时候决定存储到哪个节点。

三、分片的片键；

概述：片键是文档的一个属性字段或是一个复合索引字段，一旦建立后则不可改变，片键是拆分数据的关键的依据，如若在数据极为庞大的场景下，片键决定了数据在分片的过程中数据的存储位置，直接会影响集群的性能；

注：创建片键时，需要有一个支撑片键运行的索引；

片键分类：

1.递增片键：使用时间戳，日期，自增的主键，ObjectId，_id等，此类片键的写入操作集中在一个分片服务器上，写入不具有分散性，这会导致单台服务器压力较大，但分割比较容易，这台服务器可能会成为性能瓶颈；

语法解析：

mongos> use 库名

mongos> db.集合名.ensureIndex({"键名":1}) ##创建索引

mongos> sh.enableSharding("库名") ##开启库的分片

mongos> sh.shardCollection("库名.集合名",{"键名":1}) ##开启集合的分片并指定片键

2.哈希片键：也称之为散列索引，使用一个哈希索引字段作为片键，优点是使数据在各节点分布比较均匀，数据写入可随机分发到每个分片服务器上，把写入的压力分散到了各个服务器上。但是读也是随机的，可能会命中更多的分片，但是缺点是无法实现范围区分；

3.组合片键：数据库中没有比较合适的键值供片键选择，或者是打算使用的片键基数太小（即变化少如星期只有7天可变化），可以选另一个字段使用组合片键，甚至可以添加冗余字段来组合；

4.标签片键：数据存储在指定的分片服务器上，可以为分片添加tag标签，然后指定相应的tag，比如让10.*.*.*(T)出现在shard0000上，11.*.*.*(Q)出现在shard0001或shard0002上，就可以使用tag让均衡器指定分发；

四、案例：mongodb分片结合复制集高效存储

实验环境：

192.168.100.101 config.linuxfan.cn	192.168.100.102 shard1.linuxfan.cn	192.168.100.103 shard2.linuxfan.cn
Mongos：27025	mongos：27025	mongos：27025
config（configs）：27017	shard（shard1）：27017	shard（shard2）：27017
config（configs）：27018	shard（shard1）：27018	shard（shard2）：27018
config（configs）：27019	shard（shard1）：27019	shard（shard2）：27019

实验步骤：

在所有节点安装mongodb服务：
创建config节点的三个实例：
配置config节点的configs复制集；
配置config节点的mongos进程；
创建shard1节点的三个实例：
配置shard1节点的shard1复制集；
配置shard1节点的mongs进程；
创建shard2节点的三个实例：
配置shard2节点的shard1复制集；
配置shard2节点的mongs进程；
选择任意节点的mongos进程配置分片；
配置开启testdb数据库和table1集合的分片；
创建上述步骤开启的数据库和集合测试是否分片；
配置开启testdb2数据库和table1集合的分片；
创建上述步骤开启的数据库和集合测试是否分片；
扫描某个集合的分片情况；
配置开启testdb7数据库和hehe集合的分片（实现在一个集合中的多个document进行分片，通过设置散列片键）；
创建上述步骤开启的数据库和集合测试是否分片；
扫描某个集合的分片情况进行验证；
在192.168.100.102和192.168.100.103上登录mongos节点查看上述配置，发现已经同步；
在192.168.100.102和192.168.100.103上登录复制集primary节点查看上述配置，发现已经存在各自的分片；
在192.168.100.102上关闭shard1复制集的primary节点，测试mongos访问数据依然没有问题，实现了复制集的高可用；

在所有节点安装mongodb服务：

192.168.100.101、192.168.100.102、192.168.100.103：

[[email protected] ~]# tar zxvf mongodb-linux-x86_64-rhel70-3.6.3.tgz

[[email protected] ~]# mv mongodb-linux-x86_64-rhel70-3.6.3 /usr/local/mongodb

[[email protected] ~]# echo "export PATH=/usr/local/mongodb/bin:\$PATH" >>/etc/profile

[[email protected] ~]# source /etc/profile

[[email protected] ~]# ulimit -n 25000

[[email protected] ~]# ulimit -u 25000

[[email protected] ~]# echo 0 >/proc/sys/vm/zone_reclaim_mode

[[email protected] ~]# sysctl -w vm.zone_reclaim_mode=0

[[email protected] ~]# echo never >/sys/kernel/mm/transparent_hugepage/enabled

[[email protected] ~]# echo never >/sys/kernel/mm/transparent_hugepage/defrag

[[email protected] ~]# cd /usr/local/mongodb/bin/

[[email protected] bin]# mkdir {../mongodb1,../mongodb2,../mongodb3}

[[email protected] bin]# mkdir ../logs

[[email protected] bin]# touch ../logs/mongodb{1..3}.log

[[email protected] bin]# chmod 777 ../logs/mongodb*

创建config节点的三个实例：

192.168.100.101:

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb1.conf

bind_ip=192.168.100.101

port=27017

dbpath=/usr/local/mongodb/mongodb1/

logpath=/usr/local/mongodb/logs/mongodb1.log

logappend=true

fork=true

maxConns=5000

replSet=configs

#replication name

configsvr=true

END

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb2.conf

bind_ip=192.168.100.101

port=27018

dbpath=/usr/local/mongodb/mongodb2/

logpath=/usr/local/mongodb/logs/mongodb2.log

logappend=true

fork=true

maxConns=5000

replSet=configs

configsvr=true

END

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb3.conf

bind_ip=192.168.100.101

port=27019

dbpath=/usr/local/mongodb/mongodb3/

logpath=/usr/local/mongodb/logs/mongodb3.log

logappend=true

fork=true

maxConns=5000

replSet=configs

configsvr=true

END

[[email protected] bin]# cd

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb1.conf

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb2.conf

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb3.conf

[[email protected] ~]# netstat -utpln |grep mongod

tcp 0 0 192.168.100.101:27019 0.0.0.0:* LISTEN 2271/mongod

tcp 0 0 192.168.100.101:27017 0.0.0.0:* LISTEN 2440/mongod

tcp 0 0 192.168.100.101:27018 0.0.0.0:* LISTEN 1412/mongod

[[email protected] ~]# echo -e "/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/mongodb1.conf \n/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/mongodb2.conf\n/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/mongodb3.conf">>/etc/rc.local

[[email protected] ~]# chmod +x /etc/rc.local

[[email protected] ~]# cat <<END >>/etc/init.d/mongodb

#!/bin/bash

INSTANCE=\$1

ACTION=\$2

case "\$ACTION" in

'start')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf;;

'stop')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf --shutdown;;

'restart')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf --shutdown

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf;;

esac

END

[[email protected] ~]# chmod +x /etc/init.d/mongodb

配置config节点的configs复制集；

[[email protected] ~]# mongo --port 27017 --host 192.168.100.101

> cfg={"_id":"configs","members":[{"_id":0,"host":"192.168.100.101:27017"},{"_id":1,"host":"192.168.100.101:27018"},{"_id":2,"host":"192.168.100.101:27019"}]}

> rs.initiate(cfg)

configs:PRIMARY> rs.status()

{

"set" : "configs",

"date" : ISODate("2018-04-24T18:53:44.375Z"),

"myState" : 1,

"term" : NumberLong(1),

"configsvr" : true,

"heartbeatIntervalMillis" : NumberLong(2000),

"optimes" : {

"lastCommittedOpTime" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"readConcernMajorityOpTime" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"appliedOpTime" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"durableOpTime" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

}

"members" : [

{

"_id" : 0,

"name" : "192.168.100.101:27017",

"health" : 1,

"state" : 1,

"stateStr" : "PRIMARY",

"uptime" : 6698,

"optime" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T18:53:40Z"),

"electionTime" : Timestamp(1524590293, 1),

"electionDate" : ISODate("2018-04-24T17:18:13Z"),

"configVersion" : 1,

"self" : true

{

"_id" : 1,

"name" : "192.168.100.101:27018",

"health" : 1,

"state" : 2,

"stateStr" : "SECONDARY",

"uptime" : 5741,

"optime" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"optimeDurable" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T18:53:40Z"),

"optimeDurableDate" : ISODate("2018-04-24T18:53:40Z"),

"lastHeartbeat" : ISODate("2018-04-24T18:53:42.992Z"),

"lastHeartbeatRecv" : ISODate("2018-04-24T18:53:43.742Z"),

"pingMs" : NumberLong(0),

"syncingTo" : "192.168.100.101:27017",

"configVersion" : 1

{

"_id" : 2,

"name" : "192.168.100.101:27019",

"health" : 1,

"state" : 2,

"stateStr" : "SECONDARY",

"uptime" : 5741,

"optime" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"optimeDurable" : {

"ts" : Timestamp(1524596020, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T18:53:40Z"),

"optimeDurableDate" : ISODate("2018-04-24T18:53:40Z"),

"lastHeartbeat" : ISODate("2018-04-24T18:53:42.992Z"),

"lastHeartbeatRecv" : ISODate("2018-04-24T18:53:43.710Z"),

"pingMs" : NumberLong(0),

"syncingTo" : "192.168.100.101:27017",

"configVersion" : 1

}

"ok" : 1,

"operationTime" : Timestamp(1524596020, 1),

"$gleStats" : {

"lastOpTime" : Timestamp(0, 0),

"electionId" : ObjectId("7fffffff0000000000000001")

"$clusterTime" : {

"clusterTime" : Timestamp(1524596020, 1),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

configs:PRIMARY> show dbs

admin 0.000GB

config 0.000GB

local 0.000GB

configs:PRIMARY> exit

配置config节点的mongos进程；

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongos.conf

bind_ip=192.168.100.101

port=27025

logpath=/usr/local/mongodb/logs/mongos.log

fork=true

maxConns=5000

configdb=configs/192.168.100.101:27017,192.168.100.101:27018,192.168.100.101:27019

END

注：mongos的configdb参数只能指定一个（复制集中的primary）或多个（复制集中的全部节点）；

[[email protected] bin]# touch ../logs/mongos.log

[[email protected] bin]# chmod 777 ../logs/mongos.log

[[email protected] bin]# mongos -f /usr/local/mongodb/bin/mongos.conf

about to fork child process, waiting until server is ready for connections.

forked process: 1562

child process started successfully, parent exiting

[[email protected] ~]# netstat -utpln |grep mongo

tcp 0 0 192.168.100.101:27019 0.0.0.0:* LISTEN 1601/mongod

tcp 0 0 192.168.100.101:27020 0.0.0.0:* LISTEN 1345/mongod

tcp 0 0 192.168.100.101:27025 0.0.0.0:* LISTEN 1822/mongos

tcp 0 0 192.168.100.101:27017 0.0.0.0:* LISTEN 1437/mongod

tcp 0 0 192.168.100.101:27018 0.0.0.0:* LISTEN 1541/mongod

创建shard1节点的三个实例：

192.168.100.102:

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb1.conf

bind_ip=192.168.100.102

port=27017

dbpath=/usr/local/mongodb/mongodb1/

logpath=/usr/local/mongodb/logs/mongodb1.log

logappend=true

fork=true

maxConns=5000

replSet=shard1

#replication name

shardsvr=true

END

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb2.conf

bind_ip=192.168.100.102

port=27018

dbpath=/usr/local/mongodb/mongodb2/

logpath=/usr/local/mongodb/logs/mongodb2.log

logappend=true

fork=true

maxConns=5000

replSet=shard1

shardsvr=true

END

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb3.conf

bind_ip=192.168.100.102

port=27019

dbpath=/usr/local/mongodb/mongodb3/

logpath=/usr/local/mongodb/logs/mongodb3.log

logappend=true

fork=true

maxConns=5000

replSet=shard1

shardsvr=true

END

[[email protected] bin]# cd

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb1.conf

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb2.conf

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb3.conf

[[email protected] ~]# netstat -utpln |grep mongod

tcp 0 0 192.168.100.101:27019 0.0.0.0:* LISTEN 2271/mongod

tcp 0 0 192.168.100.101:27017 0.0.0.0:* LISTEN 2440/mongod

tcp 0 0 192.168.100.101:27018 0.0.0.0:* LISTEN 1412/mongod

[[email protected] ~]# chmod +x /etc/rc.local

[[email protected] ~]# cat <<END >>/etc/init.d/mongodb

#!/bin/bash

INSTANCE=\$1

ACTION=\$2

case "\$ACTION" in

'start')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf;;

'stop')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf --shutdown;;

'restart')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf --shutdown

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf;;

esac

END

[[email protected] ~]# chmod +x /etc/init.d/mongodb

配置shard1节点的shard1复制集；

[[email protected] ~]# mongo --port 27017 --host 192.168.100.102

>cfg={"_id":"shard1","members":[{"_id":0,"host":"192.168.100.102:27017"},{"_id":1,"host":"192.168.100.102:27018"},{"_id":2,"host":"192.168.100.102:27019"}]}

> rs.initiate(cfg)

{ "ok" : 1 }

shard1:PRIMARY> rs.status()

{

"set" : "shard1",

"date" : ISODate("2018-04-24T19:06:53.160Z"),

"myState" : 1,

"term" : NumberLong(1),

"heartbeatIntervalMillis" : NumberLong(2000),

"optimes" : {

"lastCommittedOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"readConcernMajorityOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"appliedOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"durableOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

}

"members" : [

{

"_id" : 0,

"name" : "192.168.100.102:27017",

"health" : 1,

"state" : 1,

"stateStr" : "PRIMARY",

"uptime" : 6648,

"optime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T19:06:50Z"),

"electionTime" : Timestamp(1524590628, 1),

"electionDate" : ISODate("2018-04-24T17:23:48Z"),

"configVersion" : 1,

"self" : true

{

"_id" : 1,

"name" : "192.168.100.102:27018",

"health" : 1,

"state" : 2,

"stateStr" : "SECONDARY",

"uptime" : 6195,

"optime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDurable" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T19:06:50Z"),

"optimeDurableDate" : ISODate("2018-04-24T19:06:50Z"),

"lastHeartbeat" : ISODate("2018-04-24T19:06:52.176Z"),

"lastHeartbeatRecv" : ISODate("2018-04-24T19:06:52.626Z"),

"pingMs" : NumberLong(0),

"syncingTo" : "192.168.100.102:27017",

"configVersion" : 1

{

"_id" : 2,

"name" : "192.168.100.102:27019",

"health" : 1,

"state" : 2,

"stateStr" : "SECONDARY",

"uptime" : 6195,

"optime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDurable" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T19:06:50Z"),

"optimeDurableDate" : ISODate("2018-04-24T19:06:50Z"),

"lastHeartbeat" : ISODate("2018-04-24T19:06:52.177Z"),

"lastHeartbeatRecv" : ISODate("2018-04-24T19:06:52.626Z"),

"pingMs" : NumberLong(0),

"syncingTo" : "192.168.100.102:27017",

"configVersion" : 1

}

"ok" : 1

}

shard1:PRIMARY> show dbs

admin 0.000GB

config 0.000GB

local 0.000GB

shard1:PRIMARY> exit

配置shard1节点的mongs进程；

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongos.conf

bind_ip=192.168.100.102

port=27025

logpath=/usr/local/mongodb/logs/mongos.log

fork=true

maxConns=5000

configdb=configs/192.168.100.101:27017,192.168.100.101:27018,192.168.100.101:27019

END

[[email protected] bin]# touch ../logs/mongos.log

[[email protected] bin]# chmod 777 ../logs/mongos.log

[[email protected] bin]# mongos -f /usr/local/mongodb/bin/mongos.conf

about to fork child process, waiting until server is ready for connections.

forked process: 1562

child process started successfully, parent exiting

[[email protected] ~]# netstat -utpln| grep mongo

tcp 0 0 192.168.100.102:27019 0.0.0.0:* LISTEN 1098/mongod

tcp 0 0 192.168.100.102:27020 0.0.0.0:* LISTEN 1125/mongod

tcp 0 0 192.168.100.102:27025 0.0.0.0:* LISTEN 1562/mongos

tcp 0 0 192.168.100.102:27017 0.0.0.0:* LISTEN 1044/mongod

tcp 0 0 192.168.100.102:27018 0.0.0.0:* LISTEN 1071/mongod

创建shard2节点的三个实例：

192.168.100.103:

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb1.conf

bind_ip=192.168.100.103

port=27017

dbpath=/usr/local/mongodb/mongodb1/

logpath=/usr/local/mongodb/logs/mongodb1.log

logappend=true

fork=true

maxConns=5000

replSet=shard2

#replication name

shardsvr=true

END

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb2.conf

bind_ip=192.168.100.103

port=27018

dbpath=/usr/local/mongodb/mongodb2/

logpath=/usr/local/mongodb/logs/mongodb2.log

logappend=true

fork=true

maxConns=5000

replSet=shard2

shardsvr=true

END

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongodb3.conf

bind_ip=192.168.100.103

port=27019

dbpath=/usr/local/mongodb/mongodb3/

logpath=/usr/local/mongodb/logs/mongodb3.log

logappend=true

fork=true

maxConns=5000

replSet=shard2

shardsvr=true

END

[[email protected] bin]# cd

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb1.conf

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb2.conf

[[email protected] ~]# mongod -f /usr/local/mongodb/bin/mongodb3.conf

[[email protected] ~]# netstat -utpln |grep mongod

tcp 0 0 192.168.100.101:27019 0.0.0.0:* LISTEN 2271/mongod

tcp 0 0 192.168.100.101:27017 0.0.0.0:* LISTEN 2440/mongod

tcp 0 0 192.168.100.101:27018 0.0.0.0:* LISTEN 1412/mongod

[[email protected] ~]# chmod +x /etc/rc.local

[[email protected] ~]# cat <<END >>/etc/init.d/mongodb

#!/bin/bash

INSTANCE=\$1

ACTION=\$2

case "\$ACTION" in

'start')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf;;

'stop')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf --shutdown;;

'restart')

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf --shutdown

/usr/local/mongodb/bin/mongod -f /usr/local/mongodb/bin/"\$INSTANCE".conf;;

esac

END

[[email protected] ~]# chmod +x /etc/init.d/mongodb

配置shard2节点的shard2复制集；

[[email protected] ~]# mongo --port 27017 --host 192.168.100.103

>cfg={"_id":"shard2","members":[{"_id":0,"host":"192.168.100.103:27017"},{"_id":1,"host":"192.168.100.103:27018"},{"_id":2,"host":"192.168.100.103:27019"}]}

> rs.initiate(cfg)

{ "ok" : 1 }

shard2:PRIMARY> rs.status()

{

"set" : "shard2",

"date" : ISODate("2018-04-24T19:06:53.160Z"),

"myState" : 1,

"term" : NumberLong(1),

"heartbeatIntervalMillis" : NumberLong(2000),

"optimes" : {

"lastCommittedOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"readConcernMajorityOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"appliedOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"durableOpTime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

}

"members" : [

{

"_id" : 0,

"name" : "192.168.100.103:27017",

"health" : 1,

"state" : 1,

"stateStr" : "PRIMARY",

"uptime" : 6648,

"optime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T19:06:50Z"),

"electionTime" : Timestamp(1524590628, 1),

"electionDate" : ISODate("2018-04-24T17:23:48Z"),

"configVersion" : 1,

"self" : true

{

"_id" : 1,

"name" : "192.168.100.103:27018",

"health" : 1,

"state" : 2,

"stateStr" : "SECONDARY",

"uptime" : 6195,

"optime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDurable" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T19:06:50Z"),

"optimeDurableDate" : ISODate("2018-04-24T19:06:50Z"),

"lastHeartbeat" : ISODate("2018-04-24T19:06:52.176Z"),

"lastHeartbeatRecv" : ISODate("2018-04-24T19:06:52.626Z"),

"pingMs" : NumberLong(0),

"syncingTo" : "192.168.100.103:27017",

"configVersion" : 1

{

"_id" : 2,

"name" : "192.168.100.103:27019",

"health" : 1,

"state" : 2,

"stateStr" : "SECONDARY",

"uptime" : 6195,

"optime" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDurable" : {

"ts" : Timestamp(1524596810, 1),

"t" : NumberLong(1)

"optimeDate" : ISODate("2018-04-24T19:06:50Z"),

"optimeDurableDate" : ISODate("2018-04-24T19:06:50Z"),

"lastHeartbeat" : ISODate("2018-04-24T19:06:52.177Z"),

"lastHeartbeatRecv" : ISODate("2018-04-24T19:06:52.626Z"),

"pingMs" : NumberLong(0),

"syncingTo" : "192.168.100.103:27017",

"configVersion" : 1

}

"ok" : 1

}

shard2:PRIMARY> show dbs

admin 0.000GB

config 0.000GB

local 0.000GB

shard2:PRIMARY> exit

配置shard2节点的mongs进程；

[[email protected] bin]# cat <<END >>/usr/local/mongodb/bin/mongos.conf

bind_ip=192.168.100.103

port=27025

logpath=/usr/local/mongodb/logs/mongos.log

fork=true

maxConns=5000

configdb=configs/192.168.100.101:27017,192.168.100.101:27018,192.168.100.101:27019

END

[[email protected] bin]# touch ../logs/mongos.log

[[email protected] bin]# chmod 777 ../logs/mongos.log

[[email protected] bin]# mongos -f /usr/local/mongodb/bin/mongos.conf

about to fork child process, waiting until server is ready for connections.

forked process: 1562

child process started successfully, parent exiting

[[email protected] ~]# netstat -utpln |grep mongo

tcp 0 0 192.168.100.103:27019 0.0.0.0:* LISTEN 1095/mongod

tcp 0 0 192.168.100.103:27020 0.0.0.0:* LISTEN 1122/mongod

tcp 0 0 192.168.100.103:27025 0.0.0.0:* LISTEN 12122/mongos

tcp 0 0 192.168.100.103:27017 0.0.0.0:* LISTEN 1041/mongod

tcp 0 0 192.168.100.103:27018 0.0.0.0:* LISTEN 1068/mongod

选择任意节点的mongos进程配置分片；

192.168.100.101(随意选择mongos进行设置分片，三台mongos会同步任何一台的操作)：

[[email protected] ~]# mongo --port 27025 --host 192.168.100.101

mongos> use admin;

switched to db admin

mongos> sh.status() ##查看分片的状态

--- Sharding Status ---

sharding version: {

"_id" : 1,

"minCompatibleVersion" : 5,

"currentVersion" : 6,

"clusterId" : ObjectId("5adf66d7518b3e5b3aad4e77")

}

shards:

active mongoses:

"3.6.3" : 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 }

mongos>

sh.addShard("shard1/192.168.100.102:27017,192.168.100.102:27018,192.168.100.102:27019")

##创建shard1第一个分片

{

"shardAdded" : "shard1",

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1524598580, 9),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1524598580, 9)

}

mongos> sh.addShard("shard2/192.168.100.103:27017,192.168.100.103:27018,192.168.100.103:27019")

##创建shard2第二个分片

{

"shardAdded" : "shard2",

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1524598657, 7),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1524598657, 7)

}

mongos> sh.status()

--- Sharding Status ---

sharding version: {

"_id" : 1,

"minCompatibleVersion" : 5,

"currentVersion" : 6,

"clusterId" : ObjectId("5adf66d7518b3e5b3aad4e77")

}

shards:

{ "_id" : "shard1", "host" : "shard1/192.168.100.102:27017,192.168.100.102:27018,192.168.100.102:27019", "state" : 1 }

{ "_id" : "shard2", "host" : "shard2/192.168.100.103:27017,192.168.100.103:27018,192.168.100.103:27019", "state" : 1 }

active mongoses:

"3.6.3" : 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 }

注：目前配置服务、路由服务、分片服务、副本集服务都已经串联起来了，但我们的目的是希望插入数据，数据能够自动分片。连接在mongos上，准备让指定的数据库、指定的集合分片生效。

注：configs复制集内无法创建数据，shard1和shard2复制集内可以创建数据；

配置开启testdb数据库和table1集合的分片；

[[email protected] ~]# mongo --port 27025 --host 192.168.100.101

mongos> use admin

mongos> sh.enableSharding("testdb") ##开启数据库的分片

{

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1524599672, 13),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1524599672, 13)

mongos> sh.status()

--- Sharding Status ---

sharding version: {

"_id" : 1,

"minCompatibleVersion" : 5,

"currentVersion" : 6,

"clusterId" : ObjectId("5adf66d7518b3e5b3aad4e77")

}

shards:

{ "_id" : "shard1", "host" : "shard1/192.168.100.102:27017,192.168.100.102:27018,192.168.100.102:27019", "state" : 1 }

{ "_id" : "shard2", "host" : "shard2/192.168.100.103:27017,192.168.100.103:27018,192.168.100.103:27019", "state" : 1 }

active mongoses:

"3.6.3" : 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:

shard1 1

{ "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : shard1 Timestamp(1, 0)

{ "_id" : "testdb", "primary" : "shard2", "partitioned" : true }

mongos> db.runCommand({shardcollection:"testdb.table1", key:{_id:1}}); ##开启数据库中集合的分片，同时会创建该数据库该集合

{

"collectionsharded" : "testdb.table1",

"collectionUUID" : UUID("883bb1e2-b218-41ab-8122-6a5cf4df5e7b"),

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1524601471, 14),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1524601471, 14)

}

创建上述步骤开启的数据库和集合测试是否分片；

[[email protected] ~]# mongo --port 27025 --host 192.168.100.101

mongos> use testdb; ##创建数据库中的记录，测试分片情况；

mongos> for(i=1;i<=10000;i++){db.table1.insert({"id":i,"name":"huge"})};

WriteResult({ "nInserted" : 1 })

mongos> show collections

table1

mongos> db.table1.count()

10000

mongos> sh.status()

--- Sharding Status ---

sharding version: {

"_id" : 1,

"minCompatibleVersion" : 5,

"currentVersion" : 6,

"clusterId" : ObjectId("5adf66d7518b3e5b3aad4e77")

}

shards:

{ "_id" : "shard1", "host" : "shard1/192.168.100.102:27017,192.168.100.102:27018,192.168.100.102:27019", "state" : 1 }

{ "_id" : "shard2", "host" : "shard2/192.168.100.103:27017,192.168.100.103:27018,192.168.100.103:27019", "state" : 1 }

active mongoses:

"3.6.3" : 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:

shard1 1

{ "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : shard1 Timestamp(1, 0)

{ "_id" : "testdb", "primary" : "shard2", "partitioned" : true }

testdb.table1

shard key: { "_id" : 1 }

unique: false

balancing: true

chunks:

shard2 1

{ "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : shard2 Timestamp(1, 0)

配置开启testdb2数据库和table1的分片；

[[email protected] ~]# mongo --port 27025 --host 192.168.100.101

mongos> use admin ##开启testdb2的分片

switched to db admin

mongos> sh.enableSharding("testdb2")

{

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1524602371, 7),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1524602371, 7)

}

mongos> db.runCommand({shardcollection:"testdb2.table1", key:{_id:1}});

##开启testdb2下table1集合的分片

创建上述步骤开启的数据库和集合测试是否分片；

[[email protected] ~]# mongo --port 27025 --host 192.168.100.101

mongos> use testdb2 ##创建数据进行测试testdb2.table1的分片情况

switched to db testdb2

mongos> for(i=1;i<=10000;i++){db.table1.insert({"id":i,"name":"huge"})};

WriteResult({ "nInserted" : 1 })

mongos> sh.status()

--- Sharding Status ---

sharding version: {

"_id" : 1,

"minCompatibleVersion" : 5,

"currentVersion" : 6,

"clusterId" : ObjectId("5adf66d7518b3e5b3aad4e77")

}

shards:

{ "_id" : "shard1", "host" : "shard1/192.168.100.102:27017,192.168.100.102:27018,192.168.100.102:27019", "state" : 1 }

{ "_id" : "shard2", "host" : "shard2/192.168.100.103:27017,192.168.100.103:27018,192.168.100.103:27019", "state" : 1 }

active mongoses:

"3.6.3" : 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:

shard1 1

{ "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : shard1 Timestamp(1, 0)

{ "_id" : "testdb", "primary" : "shard2", "partitioned" : true }

testdb.table1

shard key: { "_id" : 1 }

unique: false

balancing: true

chunks:

shard2 1

{ "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : shard2 Timestamp(1, 0)

{ "_id" : "testdb2", "primary" : "shard1", "partitioned" : true }

testdb2.table1

shard key: { "_id" : 1 }

unique: false

balancing: true

chunks:

shard1 1

{ "_id" : { "$minKey" : 1 } } -->> { "_id" : { "$maxKey" : 1 } } on : shard1 Timestamp(1, 0)

扫描某个集合的分片情况；

[[email protected] ~]# mongo --port 27025 --host 192.168.100.101

mongos> use testdb2

switched to db testdb2

mongos> db.table1.stats() ##查看集合的分片情况

{

"sharded" : true,

"capped" : false,

"ns" : "testdb2.table1",

"count" : 10000,

"size" : 490000,

"storageSize" : 167936,

"totalIndexSize" : 102400,

"indexSizes" : {

"_id_" : 102400

"avgObjSize" : 49,

"nindexes" : 1,

"nchunks" : 1,

"shards" : {

"shard1" : {

"ns" : "testdb2.table1",

"size" : 490000,

"count" : 10000,

"avgObjSize" : 49,

"storageSize" : 167936,

"capped" : false,

"wiredTiger" : {

"metadata" : {

"formatVersion" : 1

"creationString" :

...

配置开启testdb7数据库和hehe集合的分片（实现在一个集合中的多个document进行分片，通过设置散列片键）；

mongos> db.hehe.ensureIndex({"id":"hashed"}) ##创建hash索引，名称为hehe，以此作为散列片键分片使用，设置键名为id，作为分片的依据

{

"raw" : {

"shard2/192.168.100.103:27017,192.168.100.103:27018,192.168.100.103:27019" : {

"createdCollectionAutomatically" : true,

"numIndexesBefore" : 1,

"numIndexesAfter" : 2,

"ok" : 1

}

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1534192213, 2),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1534192213, 2)

}

mongos> sh.enableSharding("testdb7"); ##开启testdb7数据库的分片

{

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1534192235, 6),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1534192235, 6)

}

mongos> sh.shardCollection("testdb7.hehe",{"id":"hashed"}) ##开启数据库中hehe集合的分片，采用id键名作为分片依据；

{

"collectionsharded" : "testdb7.hehe",

"collectionUUID" : UUID("03ae881d-9cd2-4445-81a4-781ec9bacc44"),

"ok" : 1,

"$clusterTime" : {

"clusterTime" : Timestamp(1534192301, 20),

"signature" : {

"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),

"keyId" : NumberLong(0)

}

"operationTime" : Timestamp(1534192301, 12)

}

创建上述步骤开启的数据库和集合测试是否分片；

mongos> use testdb7

switched to db testdb7

mongos> for(i=1;i<=1000;i++){db.hehe.insert({"id":i,"name":"huge"})};

WriteResult({ "nInserted" : 1 })

扫描某个集合的分片情况进行验证；

mongos> db.hehe.stats()

{

"sharded" : true,

"capped" : false,

"ns" : "testdb7.hehe",

"count" : 1000,

"size" : 49000,

"storageSize" : 40960,

"totalIndexSize" : 73728,

"indexSizes" : {

"_id_" : 32768,

"id_hashed" : 40960

"avgObjSize" : 49,

"nindexes" : 2,

"nchunks" : 4,

"shards" : {

"shard1" : {

"ns" : "testdb7.hehe",

"size" : 23520,

"count" : 480,

"avgObjSize" : 49,

"storageSize" : 20480,

"capped" : false,

"wiredTiger" : {

"metadata" : {

"formatVersion" : 1

...

"shard2" : {

"ns" : "testdb7.hehe",

"size" : 25480,

"count" : 520,

"avgObjSize" : 49,

"storageSize" : 20480,

"capped" : false,

"wiredTiger" : {

"metadata" : {

"formatVersion" : 1

"creationString" :

...

在192.168.100.102和192.168.100.103上登录mongos节点查看上述配置，发现已经同步；
在192.168.100.102和192.168.100.103上登录复制集primary节点查看上述配置，发现已经存在各自的分片；
在192.168.100.102上关闭shard1复制集的primary节点，测试mongos访问数据依然没有问题，实现了复制集的高可用；

Mongodb分片结合复制集,在实现不同数据库分片的基础上，实现集合中document的分片（使用片键）

相关推荐