mongodb安装配置及副本集的操作
一、Mongdb3.0安装
官网下载最新稳定版本的mongodb软体
https://www.mongodb.org/downloads
解压软体
[root@test-weidangkou service]# tar zxvf mongodb-linux-x86_64-rhel62-3.0.3.tgz mongodb-linux-x86_64-rhel62-3.0.3/README mongodb-linux-x86_64-rhel62-3.0.3/THIRD-PARTY-NOTICES mongodb-linux-x86_64-rhel62-3.0.3/GNU-AGPL-3.0 mongodb-linux-x86_64-rhel62-3.0.3/bin/mongodump mongodb-linux-x86_64-rhel62-3.0.3/bin/mongorestore mongodb-linux-x86_64-rhel62-3.0.3/bin/mongoexport mongodb-linux-x86_64-rhel62-3.0.3/bin/mongoimport mongodb-linux-x86_64-rhel62-3.0.3/bin/mongostat mongodb-linux-x86_64-rhel62-3.0.3/bin/mongotop mongodb-linux-x86_64-rhel62-3.0.3/bin/bsondump mongodb-linux-x86_64-rhel62-3.0.3/bin/mongofiles mongodb-linux-x86_64-rhel62-3.0.3/bin/mongooplog mongodb-linux-x86_64-rhel62-3.0.3/bin/mongoperf mongodb-linux-x86_64-rhel62-3.0.3/bin/mongod mongodb-linux-x86_64-rhel62-3.0.3/bin/mongos mongodb-linux-x86_64-rhel62-3.0.3/bin/mongo
创建mongdb数据存放路径及日志存放路径
[root@test-weidangkou service]# mkdir -p /date/{mongodbdata,mongodb_logs}
建立mongod账号，授权数据目录
[root@test-weidangkou date]# useradd mongod [root@test-weidangkou date]# chown -R mongod.mongod /date/mongodbdata/
创建mongod.conf配置文件，只需要以下几个选项即可
[root@test-weidangkou date]#cat /etc/mongod.conf dbpath = /data/mongodata #mongodb数据存放路径 logpath = /data/mongodb_logs/mongodb.log #mongodb日志存放路径 httpinterface = true #mongodb管理端口默认+1000：280 fork = true #后台启动
二、启动Mongodb初始化数据
(1)第一种直接在命令行中指定路径和日志
/usr/local/service/mongodb-linux-x86_64-rhel62-3.0.3/bin/mongod --dbpath=/date/mongodbdate/ --logpath=/date/mongodb_logs/mongodb.log --fork about to fork child process, waiting until server is ready for connections. forked process: 5238 child process started successfully, parent exiting 端口已经起来了： [root@test-weidangkou mongodb_logs]# lsof -i :27017 COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME mongod 5238 root 6u IPv4 41939 0t0 TCP *:27017 (LISTEN) 初始化由于要同步数据，建立本地的一些数据文件。会启动比较慢。
(2)第二种方式启动：通过加载配置文件启动（推荐）
[root@test-weidangkou mongodbdate]# /usr/local/service/mongodb-linux-x86_64-rhel62-3.0.3/bin/mongod -f /etc/mongod.conf about to fork child process, waiting until server is ready for connections. forked process: 5382 child process started successfully, parent exiting
进入终端查看数据库
[root@test-weidangkou mongodb_logs]# /usr/local/service/mongodb-linux-x86_64-rhel62-3.0.3/bin/mongo MongoDB shell version: 3.0.3 connecting to: test Welcome to the MongoDB shell. For interactive help, type "help". For more comprehensive documentation, see http://docs.mongodb.org/ Questions? Try the support group http://groups.google.com/group/mongodb-user Server has startup warnings: 2015-06-30T17:37:53.882+0800 I CONTROL [initandlisten] ** WARNING: You are running this process as the root user, which is not recommended. 2015-06-30T17:37:53.882+0800 I CONTROL [initandlisten] 2015-06-30T17:37:53.905+0800 I CONTROL [initandlisten] 2015-06-30T17:37:53.905+0800 I CONTROL [initandlisten] ** WARNING: /sys/kernel/mm/transparent_hugepage/enabled is 'always'. 2015-06-30T17:37:53.905+0800 I CONTROL [initandlisten] ** We suggest setting it to 'never' 2015-06-30T17:37:53.905+0800 I CONTROL [initandlisten] 2015-06-30T17:37:53.905+0800 I CONTROL [initandlisten] ** WARNING: /sys/kernel/mm/transparent_hugepage/defrag is 'always'. 2015-06-30T17:37:53.906+0800 I CONTROL [initandlisten] ** We suggest setting it to 'never' 2015-06-30T17:37:53.906+0800 I CONTROL [initandlisten] > show dbs; local 0.078GB
启动报错会出现以下：
[root@test-weidangkou mongodbdate]# /usr/local/service/mongodb-linux-x86_64-rhel62-3.0.3/bin/mongod -f /etc/mongod.conf about to fork child process, waiting until server is ready for connections. forked process: 5298
检查原因：
第一：由于之前没有正常的退出Mongodb服务，需要删除mongod.lock文件，再次启动恢复正常。
第二：配置文件有问题，需要检查配置文件是否有特殊字符，导致无法解析
三、mongodb副本集配置
配置mongodb的副本集务必要做好以下几点：
1、各个节点能够正常访问mongodb的socket端口，iptables要放行
2、SELINUX必须要关闭掉，以免禁止端口复制
3、各个服务器节点要做好解析/etc/hosts
4、由于副本集是通过Heartbeat心跳线连接到，务必同步好各个节点的时间
小注：本人由于之前没有注意到这些，被坑了好久。哈哈！！！
mongodb复制过程及概念
（1）主节点将数据修改操作保存至oplog,oplog：大小固定的文件，存储在local数据库
（2）初始同步（inital sync)
节点没有任何数据
节点丢失副本复制历史
：克隆所有的数据库
：应用数据集的所有改变：复制oplog并应用于本地
：为所有collection的构建索引
（3）回滚后追赶（post-rollback catch-up)
(4)切分块迁移（shareding chunk migrations)
local:存放了副本集的所有元素和oplog;用于存储oplog的是一个名为oplog.rs的collection;
oplog.rs的大小依赖于OS及文件系统；但可以自定义其大小oplogsize
1、环境配置
192.168.254.7 samba
192.168.254.229 git
192.168.254.9 weidangkou
3t台机器的配置文件修改vim /etc/mongodb.conf，增加两个选项
replIndexPrefetch = _id_only #索引的ID号复制 replSet = 9tong1 #副本集设置成9tong1
为了测试证明复制的数据，可以先在主节点上导入点数据，测试环境的数据如下：
[root@samba mongodata]# ls 9tong.0 9tong.1 9tong.2 9tongAPI.0 9tongAPI.1 9tongAPI.ns 9tong.ns base.0 base.1 base.ns journal local.0 local.ns mongod.lock storage.bson
2、初始化
> show dbs;
2015-07-01T11:38:18.324+0800 E QUERY Error: listDatabases failed:{ "note" : "from execCommand", "ok" : 0, "errmsg" : "not master" }
at Error (<anonymous>)
at Mongo.getDBs (src/mongo/shell/mongo.js:47:15)
at shellHelper.show (src/mongo/shell/utils.js:630:33)
at shellHelper (src/mongo/shell/utils.js:524:36)
at (shellhelp2):1:1 at src/mongo/shell/mongo.js:47
配置好副本集选项启动后，不能直接查看dbs的情况，因为目前还没有初始化不是primary
> rs.initiate()
{
"info2" : "no configuration explicitly specified -- making one",
"me" : "samba:27017",
"ok" : 1
}
9tong1:OTHER>
9tong1:PRIMARY>
初始化后会直接变成9tong1:PRIMARY的标识，可以查看数据库的容量信息
9tong1:PRIMARY> show dbs;
9tong 0.453GB
9tongAPI 0.203GB
base 0.203GB
local 2.077GB
3、添加副本节点
9tong1:PRIMARY> rs.add
rs.add( rs.addArb(
9tong1:PRIMARY> rs.add("192.168.254.9") #默认端口不需要加端口号码
9tong1:PRIMARY> rs.add("192.168.254.9")
{ "ok" : 1 }
9tong1:PRIMARY> rs.status()
{
"set" : "9tong1",
"date" : ISODate("2015-07-01T04:04:51.535Z"),
"myState" : 1,
"members" : [
{
"_id" : 0,
"name" : "samba:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 1645,
"optime" : Timestamp(1435723485, 1),
"optimeDate" : ISODate("2015-07-01T04:04:45Z"),
"electionTime" : Timestamp(1435722018, 2),
"electionDate" : ISODate("2015-07-01T03:40:18Z"),
"configVersion" : 2,
"self" : true
},
{
"_id" : 1,
"name" : "192.168.254.9:27017",
"health" : 1,
"state" : 5,
"stateStr" : "STARTUP2",
"uptime" : 6,
"optime" : Timestamp(0, 0),
"optimeDate" : ISODate("1970-01-01T00:00:00Z"),
"lastHeartbeat" : ISODate("2015-07-01T04:04:51.261Z"),
"lastHeartbeatRecv" : ISODate("2015-07-01T04:04:51.300Z"),
"pingMs" : 2,
"syncingTo" : "samba:27017",
"configVersion" : 2
}
],
"ok" : 1
StateStr:STARTUPU2是复制数据追赶主节点的状态模式。
数据同步好后会变成SECONDARY
{
"_id" : 1,
"name" : "192.168.254.9:27017",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 63,
"optime" : Timestamp(1435723485, 1),
"optimeDate" : ISODate("2015-07-01T04:04:45Z"),
"lastHeartbeat" : ISODate("2015-07-01T04:05:47.313Z"),
"lastHeartbeatRecv" : ISODate("2015-07-01T04:05:47.344Z"),
"pingMs" : 1,
"configVersion" : 2
}
],
"ok" : 1
也可以通过rs.conf()查看各个节点状态
9tong1:PRIMARY> rs.conf()
{
"_id" : "9tong1",
"version" : 2,
"members" : [
{
"_id" : 0,
"host" : "samba:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {
},
"slaveDelay" : 0,
"votes" : 1
},
{
"_id" : 1,
"host" : "192.168.254.9:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {
},
"slaveDelay" : 0,
"votes" : 1
}
],
"settings" : {
"chainingAllowed" : true,
"heartbeatTimeoutSecs" : 10,
"getLastErrorModes" : {
},
"getLastErrorDefaults" : {
"w" : 1,
"wtimeout" : 0
}
}
}
9tong1:PRIMARY> use 9tong;
switched to db 9tong
9tong1:PRIMARY> db.user_contacts.find().count();
171639
从节点上查看
9tong1:SECONDARY> show dbs;
9tong 0.203GB
9tongAPI 0.078GB
base 0.078GB
local 6.075GB
test 0.078GB
9tong1:SECONDARY> use 9tong;
switched to db 9tong
9tong1:SECONDARY> db.user_contacts.find().count();
171639
以上证明数据已经同步过来，状态变成SECONDARY
查看OPOLOG状态
9tong1:PRIMARY> db.printReplicationInfo()
configured oplog size: 1962.66015625MB
log length start to end: 3053secs (0.85hrs)
oplog first event time: Wed Jul 01 2015 11:40:18 GMT+0800 (CST) #第一次更新时间
oplog last event time: Wed Jul 01 2015 12:31:11 GMT+0800 (CST) #最后一次更新时间
now: Wed Jul 01 2015 13:30:25 GMT+0800 (CST) #现在时间
9tong1:SECONDARY> db.printReplicationInfo()
configured oplog size: 4469.238250732422MB
log length start to end: 1586secs (0.44hrs)
oplog first event time: Wed Jul 01 2015 12:04:45 GMT+0800 (CST)
oplog last event time: Wed Jul 01 2015 12:31:11 GMT+0800 (CST) #与主节点最近一次时间一 样说明数据同步完成了。
now: Wed Jul 01 2015 13:31:01 GMT+0800 (CST)
4、连接到从节点192.168.254.9查看状态
[root@weidangkou data]# /usr/local/service/mongodb-linux-x86_64-rhel62-3.0.3/bin/mongo
MongoDB shell version: 3.0.3
connecting to: test
Server has startup warnings:
2015-07-01T12:00:01.448+0800 I CONTROL [initandlisten] ** WARNING: You are running this process as the root user, which is not recommended.
2015-07-01T12:00:01.448+0800 I CONTROL [initandlisten]
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten]
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** WARNING: /sys/kernel/mm/transparent_hugepage/enabled is 'always'.
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** We suggest setting it to 'never'
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten]
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** WARNING: /sys/kernel/mm/transparent_hugepage/defrag is 'always'.
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** We suggest setting it to 'never'
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten]
9tong1:SECONDARY>
已经变成副本集的SECONDARY状态。
查看数据库容量信息，会出现以下错误。
9tong1:SECONDARY> show dbs;
2015-07-01T12:14:14.768+0800 E QUERY Error: listDatabases failed:{ "note" : "from execCommand", "ok" : 0, "errmsg" : "not master" }
at Error (<anonymous>)
at Mongo.getDBs (src/mongo/shell/mongo.js:47:15)
at shellHelper.show (src/mongo/shell/utils.js:630:33)
at shellHelper (src/mongo/shell/utils.js:524:36)
at (shellhelp2):1:1 at src/mongo/shell/mongo.js:47
因为此时该节点不属于主节点，也没有默认为从节点。需要运行如下这个命令：
9tong1:SECONDARY> rs.slaveOk()
9tong1:SECONDARY> show dbs;
9tong 0.203GB
9tongAPI 0.078GB
base 0.078GB
local 6.075GB
也可以通过rs.isMaster()查看是否自己属于主节点
9tong1:SECONDARY> rs.isMaster()
{
"setName" : "9tong1",
"setVersion" : 2,
"ismaster" : false,
"secondary" : true,
"hosts" : [
"samba:27017",
"192.168.254.9:27017"
],
"primary" : "samba:27017",
"me" : "192.168.254.9:27017",
"maxBsonObjectSize" : 16777216,
"maxMessageSizeBytes" : 48000000,
"maxWriteBatchSize" : 1000,
"localTime" : ISODate("2015-07-01T04:23:34.493Z"),
"maxWireVersion" : 3,
"minWireVersion" : 0,
"ok" : 1
}
5、添加192.168.254.229成为仲裁节点。仲裁节点只有投票选举权，没有数据复制功能。
9tong1:PRIMARY> rs.addArb("192.168.254.229")
{ "ok" : 1 }
9tong1:PRIMARY> rs.status()
{
"set" : "9tong1",
"date" : ISODate("2015-07-01T04:27:36.132Z"),
"myState" : 1,
"members" : [
{
"_id" : 0,
"name" : "samba:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 3010,
"optime" : Timestamp(1435724813, 1),
"optimeDate" : ISODate("2015-07-01T04:26:53Z"),
"electionTime" : Timestamp(1435722018, 2),
"electionDate" : ISODate("2015-07-01T03:40:18Z"),
"configVersion" : 3,
"self" : true
},
{
"_id" : 1,
"name" : "192.168.254.9:27017",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 1370,
"optime" : Timestamp(1435724813, 1),
"optimeDate" : ISODate("2015-07-01T04:26:53Z"),
"lastHeartbeat" : ISODate("2015-07-01T04:27:35.993Z"),
"lastHeartbeatRecv" : ISODate("2015-07-01T04:27:34.470Z"),
"pingMs" : 0,
"syncingTo" : "samba:27017",
"configVersion" : 3
},
{
"_id" : 2,
"name" : "192.168.254.229:27017",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 42,
"lastHeartbeat" : ISODate("2015-07-01T04:27:35.987Z"),
"lastHeartbeatRecv" : ISODate("2015-07-01T04:27:35.986Z"),
"pingMs" : 0,
"configVersion" : 3
}
],
"ok" : 1
}
6、测试主节点写数据，查看从节点是否有
主节点：
9tong1:PRIMARY> db.classes.insert({class: "ONE",nostu :40})
WriteResult({ "nInserted" : 1 })
9tong1:PRIMARY> show collections;
classes
system.indexes
从节点查看：
9tong1:SECONDARY> db.classes.findOne()
{
"_id" : ObjectId("55936d0f24c674b1e83360d6"),
"class" : "ONE",
"nostu" : 40
}
注意，从节点没有写权限，插入数据后会报错
9tong1:SECONDARY> db.classes.insert({class: "TOW",nostu :50})
WriteResult({ "writeError" : { "code" : undefined, "errmsg" : "not master" } })
7、主从切换测试
要点：副本集的重新选举的影响条件：
心跳信息，heartbeat会每隔几秒检查连接状态信息
优先级，默认都是1，值越高优先级越高。值为0的节点是不能成为主节点的，也不能触发选举。
optime：主节点必须optime更新的时间要快于从节点是最新的，如果这个值小于线上的其他机器， 是不肯能成为主节点的。
网络连接票数超过法定票数一半
网络分区：票数选举
选举机制：
触发选举的事件
新副本集初始化时
从节点联系不到主节点时
主节点“下台”时
收到setpDown()命令时
某从节点有更高的优先级且已经满足其他所有条件
主节点无法联系到副本集的“多数方”
（1）关闭192.168.254.7-SAMBA,主节点的MongoDB服务
rs.status()
{
"set" : "9tong1",
"date" : ISODate("2015-07-01T05:36:43.453Z"),
"myState" : 1,
"members" : [
{
"_id" : 0,
"name" : "samba:27017",
"health" : 0,
"state" : 8,
"stateStr" : "(not reachable/healthy)",
"uptime" : 0,
"optime" : Timestamp(0, 0),
"optimeDate" : ISODate("1970-01-01T00:00:00Z"),
"lastHeartbeat" : ISODate("2015-07-01T05:36:42.029Z"),
"lastHeartbeatRecv" : ISODate("2015-07-01T05:36:14.991Z"),
"pingMs" : 0,
"lastHeartbeatMessage" : "Failed attempt to connect to samba:27017; couldn't connect to server samba:27017 (192.168.254.7), connection attempt failed",
"configVersion" : -1
},
{
"_id" : 1,
"name" : "192.168.254.9:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 5916,
"optime" : Timestamp(1435725071, 2),
"optimeDate" : ISODate("2015-07-01T04:31:11Z"),
"electionTime" : Timestamp(1435728977, 1),
"electionDate" : ISODate("2015-07-01T05:36:17Z"),
"configVersion" : 3,
"self" : true
},
{
"_id" : 2,
"name" : "192.168.254.229:27017",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 4188,
"lastHeartbeat" : ISODate("2015-07-01T05:36:43.328Z"),
"lastHeartbeatRecv" : ISODate("2015-07-01T05:36:42.799Z"),
"pingMs" : 0,
"configVersion" : 3
}
],
"ok" : 1
}
停止后，10秒左右的时间，明显的看到原来是主节点的SAMBA，监控状态变为0，1表示正常,原先192.168.254.9从SECONDARY变成了PRIMARY。
（2）更改192.168.254.9的优先级。使它变为高优先级成为主节点
9tong1:PRIMARY> cfg=rs.conf() #保存当前状态
{
"_id" : "9tong1",
"version" : 3,
"members" : [
{
"_id" : 0,
"host" : "samba:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {
},
"slaveDelay" : 0,
"votes" : 1
},
{
"_id" : 1,
"host" : "192.168.254.9:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {
},
"slaveDelay" : 0,
"votes" : 1
},
{
"_id" : 2,
"host" : "192.168.254.229:27017",
"arbiterOnly" : true,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {
},
"slaveDelay" : 0,
"votes" : 1
}
],
"settings" : {
"chainingAllowed" : true,
"heartbeatTimeoutSecs" : 10,
"getLastErrorModes" : {
},
"getLastErrorDefaults" : {
"w" : 1,
"wtimeout" : 0
}
}
}
9tong1:PRIMARY> cfg.members[1].priority=2 #更改ID号为1的优先级。
2
9tong1:PRIMARY> rs.reconfig(cfg) #使配置生效
{ "ok" : 1 }
9tong1:PRIMARY>
9tong1:PRIMARY>
2015-07-01T14:01:15.000+0800 I NETWORK DBClientCursor::init call() failed
2015-07-01T14:01:15.002+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed
2015-07-01T14:01:15.002+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) ok
9tong1:SECONDARY>
9tong1:SECONDARY>
可以看到原来主节点SAMBA，自动降级为SECONDARY,从节点自动升级到PRIMARY
[root@weidangkou data]# /usr/local/service/mongodb-linux-x86_64-rhel62-3.0.3/bin/mongo
MongoDB shell version: 3.0.3
connecting to: test
Server has startup warnings:
2015-07-01T12:00:01.448+0800 I CONTROL [initandlisten] ** WARNING: You are running this process as the root user, which is not recommended.
2015-07-01T12:00:01.448+0800 I CONTROL [initandlisten]
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten]
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** WARNING: /sys/kernel/mm/transparent_hugepage/enabled is 'always'.
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** We suggest setting it to 'never'
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten]
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** WARNING: /sys/kernel/mm/transparent_hugepage/defrag is 'always'.
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten] ** We suggest setting it to 'never'
2015-07-01T12:00:01.449+0800 I CONTROL [initandlisten]
9tong1:PRIMARY>