MongoDB服务器-白红宇

MongoDB服务器

阅读量：3945 次

发布时间：2019-05-24

本文共 14696 字，大约阅读时间需要 48 分钟。

MongoDB服务器1）在主机192.168.4.51上部署MongoDB服务MongoDB：介于关系数据库和非关系数据库之间的产品一款基于分布式文件存储的数据库，旨在为WEB应用提供可扩展的高性能数据存储解决方案将数据存储为一个文档（类似于JSON对象），数据结构由键值（key=>value）对组成支持丰富的查询表达，可以设置任何属性的索引支持副本集，分片

[root@mongodb51~]# cd mongodb/[root@mongodb51mongodb]# lsmongodb-linux-x86_64-rhel70-3.6.3.tgz[root@mongodb51mongodb]# tar -xfmongodb-linux-x86_64-rhel70-3.6.3.tgz[root@mongodb51mongodb]# mkdir /usr/local/mongodb[root@mongodb51mongodb]#cd /usr/local/mongodb[root@mongodb51mongodb]#cp -r /root/mongodb/mongodb-linux-x86_64-rhel70-3.6.3  /bin/ .[root@mongodb51mongodb]# lsbin[root@mongodb51mongodb]# mkdir etc//创建存放配置文件的目录[root@mongodb51mongodb]#mkdir log /创建存放日志文件的目录[root@mongodb51mongodb]# mkdir -p data/db//创建存放数据库的目录db，必须为这个名字[root@mongodb51mongodb]#vim etc/mongodb.condbpath=/usr/local/mongodb/data/db/ //指定数据库目录logpath=/usr/local/mongodb/log/mongodb.log //指定日志文件logappend=true //以追加的方式记录日志信息fork=true //服务以守护进程的方式运行

设置PATH变量

[ root@mongodb51mongodb] # v im /etc/profileexport PATH=/usr/local/mongodb/bin:$PATH[ root@mongodb51mongodb] # source /etc/profile

启动服务

[ root@mongodb51mongodb] # mongod - f /usr/local/mongodb/etc/mongodb.confabout to fork child process, waiting until server is ready for connections.forked process: 28001child process started successfully , parent exiting //启动成功

验证配置，默认端口为27017

[ root@mongodb51mongodb] # ls /usr/local/mongodb/data/db/[ root@mongodb51mongodb] # ps - C mongodPID TTY TIME CMD28001? 00:00:01mongod[ root@mongodb51mongodb] # netstat - antup | grep mongodtcp 0 0 127.0.0.1:27017 0.0.0.0:* LISTEN 28001/mongod

连接服务

[ root@mongodb51mongodb] # mongo //默认本地连接，没有密码MongoDB shell version v3.6.3connecting to: mongodb://127.0.0.1:27017MongoDB server version: 3.6.3....................> show dbs //显示已有的库admin 0.000GBconfig 0.000GBlocal 0.000GB> db //查看当前所在的库test> exitbye

停止服务

[ root@mongodb51mongodb] # mongod - - dbpath=/usr/local/mongodb/data/db/ - - shutdown //指定数据库目录停止服务，不然停止不了killing process with pid: 28001

由于启动和停止服务名字太长，可以起一个别名

给停止服务起一个别名

[ root@mongodb51mongodb] # alias cmdb='mongod - - dbpath=/usr/local/mongodb/data/db/ - - shutdown'

给启动服务起一个别名

[ root@mongodb51mongodb] # alias smdb='mongod - f /usr/local/mongodb/etc/mongodb.conf'

修改配置文件，使用ip和端口连接服务

[ root@mongodb51mongodb] # v im /usr/local/mongodb/etc/mongodb.confbind_ip=192.168.4.51 //在原先的基础上面加上这两个，指定ipport=27077 //指定端口号[ root@mongodb51mongodb] # smdb //启动服务，之前设置过别名about to fork child process, waiting until server is ready for connections.forked process: 28240child process started successfully , parent exiting[ root@mongodb51mongodb] # ps - C mongodPID TTY TIME CMD28240 ? 00:00:01mongod[ root@mongodb51mongodb] # netstat - antup | grep mongodtcp 0 0 192.168.4.51:27077 0.0.0.0:* LISTEN 28240/mongod

连接服务

[ root@mongodb51mongodb] # mongo -- host 192.168.4.51- - port 27077 //成功

MongoDB常用管理操作

数据库管理

命令格式：
show dbs 查看已有的库
db 显示当前所在的库
use 库名切换库，若库不存在延时创建库
show collections 或 show tables 查看库下已有集合
db.dropDatabase() 删除当前所在的库
查看

> show dbs //查看已有的库admin 0.000GBconfig 0.000GBlocal 0.000GB

创建，切换

> db //显示当前所在的库test> use ddsdb //切换库，若库不存在的话 会自动延时创建库switched to db ddsdb > db ddsdb

集合管理

命令格式：

show collections 或 show tables 查看集合
db.集合名.drop() 删除集合
db.集合名.save({’’,’’}) 创建集合，集合不存在时，创建并添加文档

> db.t1.save( { name: "y ay a",age:60,addr: "hebei",email: "y ay a@163.com"} )WriteResult( { "nInserted" : 1} )> db.t1.save( { name: "y ay a",age:70} )WriteResult( { "nInserted" : 1} )> show tablest1

查看集合里的所有内容

> db.t1.save( { name: "y ay a",age:70} )WriteResult( { "nInserted" : 1} )> db.t1.find( ){ "_id" : ObjectId( "5badf1b420cdd1574b851f12") , "name" : "y ay a", "age" : 60, "addr" : "hebei", "email" : "y ay a@163.com" }{ "_id" : ObjectId( "5badf21820cdd1574b851f13") , "name" : "y ay a", "age" : 70 }>

查看第一行的内容

> db.t1.findOne( ){    "_id" : ObjectId( "5badf1b420cdd1574b851f12") ,    "name" : "y ay a",    "age" : 60,    "addr" : "hebei",    "email" : "y ay a@163.com"}>

文档管理

命令格式：

db.集合名.find() db.集合名.count() db.集合名.insert({“name”:”jim”}) db.集合名.find(条件)
db.集合名.findOne() 返回查询一条文档
db.集合名.remove({}) 删除所有文档
db.集合名.remove({条件}) 删除与条件匹配的所有文档

> db.t1.save( { name: "xm",age:18,"addr": "hn","email": "xm.qq.com","like": "nicai"} )WriteResult( { "nInserted" : 1} )> db.t1.count( )3> db.t1.find( { name: "xm"} ){ "_id" : ObjectId( "5badf4bc20cdd1574b851f14") , "name" : "xm", "age" : 18, "addr" : "hn", "email" : "xm.qq.com", "like" : "nicai" }

删除与条件匹配的所有文档

> db.t1.remove( { name: "y ay a"} )WriteResult( { "nRemoved" : 2 } )

删除所有文档

> db.t1.remove( { } )WriteResult( { "nRemoved" : 1} )

插入记录

> db.col.insert({ title: 'MongoDB 教程',    description: 'MongoDB 是一个 Nosql 数据库',    by : 'MongoDB中文网',    url: 'http://www.mongodb.org.cn',    tags: [ 'mongodb', 'database', 'NoSQL'] ,    likes: 100})WriteResult( { "nInserted" : 1} )

删除记录

> db.col.remove( { 'title':'MongoDB 教程'} )WriteResult( { "nRemoved" : 1} )

基本数据类型

null：用于表示空值或者不存在的字段，{“x”:null}

> db.t1.save( { name: "bob",work:null} )WriteResult( { "nInserted" : 1} )> db.t1.find( ){ "_id" : ObjectId( "5badf71520cdd1574b851f16") , "name" : "bob", "work" : null }> db.t1.save( { _id:9,name: "jerry ",work:null} )//_id默认自己创建出来，按一定的规律生成WriteResult( { "nMatched" : 0, "nUpserted" : 1, "nModified" : 0, "_id" : 9 } )> db.t1.find( ) { "_id" : ObjectId( "5badf71520cdd1574b851f16") , "name" : "bob", "work" : null } { "_id" : 9, "name" : "jerry ", "work" : null }

布尔值：布尔类型有两个值true和false，{“x”:true}

> db.t1.save( { name: "zhangsan",addr: "shanghai",marry : "true"} )WriteResult( { "nInserted" : 1} )> db.t1.save( { name: "lisi",age:35,addr: "beijing",marry : "false"} )WriteResult( { "nInserted" : 1} )> db.t1.find( ){ "_id" : ObjectId( "5badf71520cdd1574b851f16") , "name" : "bob", "work" : null }{ "_id" : 9, "name" : "jerry ", "work" : null }{ "_id" : ObjectId( "5badf7b020cdd1574b851f17") , "name" : "zhangsan", "addr" : "shanghai", "marry " : "true" }{ "_id" : ObjectId( "5badf7b720cdd1574b851f18") , "name" : "lisi", "age" : 35, "addr" : "beijing", "marry " : "false" }>

数值：shell默认使用64为浮点型数值，{“x”：3.14}或{“x”：3}，对于整型值，可以使用NumberInt（4字节符号整数）或NumberLong（8字节符号整数），{“x”:NumberInt(“3”)}{“x”:NumberLong(“3”)}

> db.t1.save( { name: "alice",woker: "gcs",pay :28888.88} )WriteResult( { "nInserted" : 1} )> db.t1.find( { pay :28888.88} ){ "_id" : ObjectId( "5badf80f20cdd1574b851f19") , "name" : "alice", "woker" : "gcs", "pay " : 28888.88 }> db.t1.save( { name: "lilei",woker: "cxy ",pay :28000} )WriteResult( { "nInserted" : 1} )> db.t1.find( { pay :28000} ){ "_id" : ObjectId( "5badf81f20cdd1574b851f1a") , "name" : "lilei", "woker" : "cxy ", "pay " : 28000 }

字符串：UTF-8字符串都可以表示为字符串类型的数据，{“x”：“呵呵”}

> db.t1.save( { name: "hehe",woker:null,pay :null} )WriteResult( { "nInserted" : 1} )> db.t1.save( { name: "呵呵",woker: "没有",pay : "没有"} )WriteResult( { "nInserted" : 1} )

日期:日期被存储为自新纪元依赖经过的毫秒数，不存储时区，{“x”:new Date()}

> db.t1.save( { name: "bobo",wokertime:new Date( ) } )WriteResult( { "nInserted" : 1} )> db.t1.find( { name: "bobo"} ){ "_id" : ObjectId( "5badf8ff6827555e3fd8680f ") , "name" : "bobo", "wokertime" : ISODate( "2018- 09- 28T09:48:47.437Z") }>

正则表达式：查询时，使用正则表达式作为限定条件，语法与JavaScript的正则表达式相同，{“x”😕[abc]/}

> db.t1.save( { procname: "php",code:/abc$/} )WriteResult( { "nInserted" : 1} )> db.t1.save( { procname: "php",code:/[ a- z] /} )WriteResult( { "nInserted" : 1} )> db.t1.save( { procname: "java",code:/[ a- z] /} )WriteResult( { "nInserted" : 1} )> db.t1.find( )..................{ "_id" : ObjectId( "5badf93b6827555e3fd86810") , "procname" : "php", "code" : /abc$/ }{ "_id" : ObjectId( "5badf9426827555e3fd86811") , "procname" : "php", "code" : /[ a- z] / }{ "_id" : ObjectId( "5badf9496827555e3fd86812") , "procname" : "java", "code" : /[ a- z] / }

数组：数据列表或数据集可以表示为数组，{“x”： [“a“，“b”,”c”]}，一个字段有多个值

> db.t1.save( { name: "jerry ",email:[ "plj@163.com","lij@y ahoo.net","lij@tedu"] } )WriteResult( { "nInserted" : 1} )> db.t1.find( { name: "jerry "} ){ "_id" : 9, "name" : "jerry ", "work" : null } { "_id" : ObjectId( "5badf9976827555e3fd86813") , "name" : "jerry ", "email" : [ "plj@163.com", "lij@y ahoo.net", "lij@tedu" ] } >

内嵌文档：文档可以嵌套其他文档，被嵌套的文档作为值来处理，{“x”:{“y”:3 }}

> db.t1.save( { book: { zuozhe: "dmy ",bname: "y uweizhidao",jiaqian:69,version:2.0 } } )WriteResult( { "nInserted" : 1} )> db.t1.find( )................{ "_id" : ObjectId( "5badfa4a6827555e3fd86814") , "book" : { "zuozhe" : "dmy ", "bname" : "y uweizhidao", "jiaqian" : 69, "version" : 2 } }

对象id：对象id是一个12字节的字符串，是文档的唯一标识，{“x”: objectId() }

> db.t1.save( { stunum:ObjectId( ) ,name: "y ay a",class: "nsd"} )WriteResult( { "nInserted" : 1} )> db.t1.find( { name: "y ay a"} ){ "_id" : ObjectId( "5badfa966827555e3fd86816") , "stunum" : ObjectId( "5badfa966827555e3fd86815") , "name" : "y ay a", "class" : "nsd" }//自己定义一个id字段，一般情况下都用内置的id字段，相当于my sql里的primary key auto_increment

二进制数据：二进制数据是一个任意字节的字符串。它不能直接在shell中使用。如果要将非utf-字符保存到数据库中，二进制数据是唯一的方式

代码：查询和文档中可以包括任何JavaScript代码，{“x”:function(){/…/}}

> db.t1.save( { lname: "html",codecript:function( ) { /*...*/} } )WriteResult( { "nInserted" : 1} )> db.t1.save( { lname: "html",codecript:function( ) { /*abc
*/} } )WriteResult( { "nInserted" : 1} )> db.t1.find( )........................{ "_id" : ObjectId( "5badfd626827555e3fd86817") , "lname" : "html", "codecript" : { "code" : "function ( ) { /*...*/} " } }{ "_id" : ObjectId( "5badfd6a6827555e3fd86818") , "lname" : "html", "codecript" : { "code" : "function ( ) { /*abc
*/} " } }>

数据导入导出/备份/恢复

数据备份与恢复

数据备份

[ root@mongodb51~] # mongodump - - host 192.168.4.51- - port 27077//不指定备份哪个库，默认备份所有，不指定目录，自动生成dump目录，备份的数据在这个里面2018- 09- 28T18:14:12.585+0800 writing admin.sy stem.version to2018- 09- 28T18:14:12.586+0800 done dumping admin.sy stem.version ( 1document)2018- 09- 28T18:14:12.586+0800 writing ddsdb.t1to2018- 09- 28T18:14:12.586+0800 writing test.t1to2018- 09- 28T18:14:12.586+0800 writing ddsdb.col to2018- 09- 28T18:14:12.587+0800 done dumping ddsdb.t1( 17 documents)2018- 09- 28T18:14:12.588+0800 done dumping test.t1( 1document)2018- 09- 28T18:14:12.588+0800 done dumping ddsdb.col ( 0 documents)[ root@mongodb51~] # lsdnmp[ root@mongodb51~] # bsondump dump/ddsdb/t1.bson //查看bson文件内容{ "_id":{ "$oid": "5badf71520cdd1574b851f16"} ,"name": "bob","work":null}{ "_id":9.0,"name": "jerry ","work":null}....................{ "_id":{ "$oid": "5badfd626827555e3fd86817"} ,"lname": "html","codecript":{ "$code": "function ( ) { /*...*/} "} } { "_id":{ "$oid": "5badfd6a6827555e3fd86818"} ,"lname": "html","codecript":{ "$code": "function ( ) { /*\u003chtml\u003e\u003ch1\u003eabc\u003c/h1 2018- 09- 28T18:15:45.948+0800 17 objects found

备份时指定备份的库和备份目录

[ root@mongodb51~] # mongodump - - host 192.168.4.51- - port 27077 - d ddsdb - o /root/bbsdb2018- 09- 28T18:23:30.389+0800 writing ddsdb.t1to2018- 09- 28T18:23:30.389+0800 writing ddsdb.col to2018- 09- 28T18:23:30.391+0800 done dumping ddsdb.t1( 17 documents)2018- 09- 28T18:23:30.391+0800 done dumping ddsdb.col ( 0 documents)//- d备哪个库，- o指定备份的目录，备份bbsdb库里的所有到/root/bbsdb

只备份ddsdb库里的集合t1

[ root@mongodb51~] # mongodump - - host 192.168.4.51- - port 27077 - d ddsdb - c t1- o /root/bbsdb.t2018- 09- 28T18:19:00.210+0800 writing ddsdb.t1to2018- 09- 28T18:19:00.211+0800 done dumping ddsdb.t1( 17 documents)

恢复数据

[ root@mongodb51~] # mongo - - host 192.168.4.51- - port 27077> show tables;colt1> db.t1.remove( { } )WriteResult( { "nRemoved" : 17 } )>exit[ root@mongodb51~] # mongorestore - - host 192.168.4.51- - port 27077 - d ddsdb /root/bbsdb.t/ddsdb///- d ddsdb恢复到数据库的目录，从/root/bbsdb.t1/ddsdb/目录恢复2018- 09- 28T18:26:16.889+0800 the - - db and - - collection args should only be used when restoring from a BSON file. Other uses are deprecated and will not ex2018- 09- 28T18:26:16.890+0800 building a list of collections to restore from /root/bbsdb.t/ddsdb dir2018- 09- 28T18:26:16.891+0800 reading metadata for ddsdb.t1from /root/bbsdb.t/ddsdb/t1.metadata.json2018- 09- 28T18:26:16.891+0800 restoring ddsdb.t1from /root/bbsdb.t/ddsdb/t1.bson2018- 09- 28T18:26:16.893+0800 no indexes to restore2018- 09- 28T18:26:16.893+0800 finished restoring ddsdb.t1( 17 documents)2018- 09- 28T18:26:16.893+0800 done

数据的导入导出

导出

用csv的格式导出

[ root@mongodb51~] # mongoexport - - host 192.168.4.51- - port 27077 - d ddsdb - c t1- f name - - ty pe=csv - o /root/lig1.csv//导出csv格式，必须要指定导出的字段名 ，导出name字段2018- 09- 28T18:29:24.653+0800 connected to: 192.168.4.51:270772018- 09- 28T18:29:24.654+0800 exported 17 records[ root@mongodb51~] # cat lig1.csvnamebobjerryzhangsanlisialicelileihehe呵呵bobojerryy ay a[ root@mongodb51~] # mongoexport - - host 192.168.4.51- - port 27077 - d ddsdb - c t1- q '{ name: "bob"} ' - f name,age - - ty pe=csv - o /root/lig2//从ddsdb的它1里导出名字为bob的name字段和age字段2018- 09- 28T18:31:25.627+0800 connected to: 192.168.4.51:270772018- 09- 28T18:31:25.628+0800 exported 1record

用json的格式导出

[ root@mongodb51~] # mongoexport - - host 192.168.4.51- - port 27077 - d ddsdb - c t1- - ty pe=json - o /root/lig3.json//导出json格式2018- 09- 28T18:33:13.349+0800 connected to: 192.168.4.51:270772018- 09- 28T18:33:13.350+0800 exported 17 records[ root@mongodb51~] # mongoexport - - host 192.168.4.51- - port 27077 - d ddsdb - c t1- f name - - ty pe=json - o /root/lig4.json//指定列名导出，导出name字段2018- 09- 28T18:33:35.914+0800 connected to: 192.168.4.51:270772018- 09- 28T18:33:35.915+0800 exported 17 records

导入

[ root@mongodb51~] # mongo - - host 192.168.4.51- - port 27077> use ddsdbswitched to db ddsdb> show tables;colt1> db.t1.remove( { } )WriteResult( { "nRemoved" : 17 } )> exit

用json的格式导入：表里要没有数据，不然导入不成功

[ root@mongodb51~] # mongoimport - - host 192.168.4.51- - port 27077 - d ddsdb - c t1- - ty pe=json /root/lig3.json2018- 09- 28T18:35:22.341+0800 connected to: 192.168.4.51:270772018- 09- 28T18:35:22.343+0800 imported 17 documents[ root@mongodb51~] # mongo - - host 192.168.4.51- - port 27077> use ddsdbswitched to db ddsdb> db.t1.count( { } )17

用csv的格式导入：表里可以有数据

[ root@mongodb51~] # mongoimport - - host 192.168.4.51- - port 27077 - d ddsdb - c t1 - - headerline - - ty pe=csv /root/lig1.csv//必须指定文件的列名，不然不成功 - f和- - headerline不能一起用 - - headerline：把第一行的字段隐藏即去掉文件列的标题name，不然标题也会导进去，导入2018- 09- 28T18:37:36.778+0800 connected to: 192.168.4.51:270772018- 09- 28T18:37:36.779+0800 imported 11documents

创建mongodb副本集

三台主机安装mongodb[ root@mongodb51~] # cd mongodb/[ root@mongodb51mongodb] # mkdir /usr/local/mongodb[ root@mongodb51mongodb] # cd /usr/local/mongodb[ root@mongodb51mongodb] # cp - r /root/mongodb/mongodb- linux- x86_64- rhel70- 3.6.3/bin/ .[ root@mongodb51mongodb] # lsbin[ root@mongodb51mongodb] # mkdir etc[ root@mongodb51mongodb] # mkdir log[ root@mongodb51mongodb] # mkdir - p data/db[ root@mongodb51mongodb] # v im etc/mongodb.confdbpath=/usr/local/mongodb/data/db/logpath=/usr/local/mongodb/log/mongodb.loglogappend=truefork=truebind_ip=192.168.4.51port=27077replSet=rs1//加入到副本集，rs1名字随便起，想知道谁和我在一个副本集里，三台机器的名字一样,都写rs1