leodengzx 2020-05-05
副本集是什么 - MongoDB复制是将数据同步在多个服务器的过程。 -复制提供了数据的冗余备份,并在多个服务器上存储数据副本,提高了数据的可用性, 并可以保证数据的安全性。 -复制还允许您从硬件故障和服务中断中恢复数据
副本集工作过程 mongodb的复制至少需要两个节点。其中一个是主节点,负责处理客户端请求, 其余的都是从节点,负责复制主节点上的数据。 mongodb各个节点常见的搭配方式为: 一主一从、一主多从。 主节点记录在其上的所有操作oplog,从节点定期轮询主节点获取这些操作, 然后对自己的数据副本执行这些操作,从而保证从节点的数据与主节点一致。
Replica Sets复制集 MongoDB在1.6版本对开发了新功能replica set,这比之前的replication功能要强大一些 ,增加了故障 自动切换和自动修复成员节点,各个DB之间数据完全一致, 大大降低了维护成功。 使用replica set故障切换完全自动。 Replica Sets的结构类似一个集群,完全可以把它当成一个集群,因为它确实与集群实现 的作用是一样的: 如果其中一个节点出现故障,其他节点马上会将业务 接管过来而无须停机操作
Master- Slave主从复制 实现数据同步只需要在某一台服务器启动时加上"master"参数,以指明此服务器的角色 是primary ;另-台服务器加上"-slave‘和" -source参数,以指明此服务器的角色是slave。 .主从复制的优点如下: 从服务器可以执行查询工作,降低主服务器访问压力。-在从服务器执行备份, 避免备份期间锁定主服务器的数据。 当主服务器出现故障时,可以快速切换到从服务器,减少宕机时间。
====================================================================
具体要求:
准备3台mongodb服务器
验证副本集配置 准备三台虚拟机,配置mongodb副本集,ip分别为192.158.4.51, 192.168.4.52, 192.168.4.53其中一个是主节点,负责处理客户端请求,其余的都是从节点, 负责复制主节点上的数据, 实现存储数据副本,提高了数据的可用性,
运行服务
启动服务时,指定主机所在副本集名称 副本集成员间使用形同的副本集名称
[ ~]# mkdir -p /data/db [ ~]# mongod --bind_ip 192.168.200.102 --port 27061 --logpath=/var/l og/mongod.log --replSet rs1 & //指定副本集名称 [ ~]# [ ~]# netstat -lnpt|grep mongod tcp 0 0 192.168.200.102:27078 0.0.0.0:* LISTEN 6097/mongod tcp 0 0 192.168.200.102:27061 0.0.0.0:* LISTEN 6276/mongod [ ~]# jobs [1]+ 运行中 mongod --bind_ip 192.168.200.102 --port 27061 --logpat h=/var/log/mongod.log --replSet rs1 &
步骤一:创建mongodb副本集
1)三台主机安装mongodb(以4.51为例)
[ ~]# cd mongodb/ [ mongodb]# mkdir /usr/local/mongodb [ mongodb]# cd /usr/local/mongodb [ mongodb]# cp -r \ /root/mongodb/mongodb-linux-x86_64-rhel70-3.6.3/bin/ . [ mongodb]# ls bin [ mongodb]# mkdir etc [ mongodb]# mkdir log [ mongodb]# mkdir -p data/db [ mongodb]# vim etc/mongodb.conf dbpath=/usr/local/mongodb/data/db/ logpath=/usr/local/mongodb/log/mongodb.log logappend=true fork=true bind_ip=192.168.4.51 port=27077 replSet=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
2)设置PATH变量
[ mongodb]# vim /etc/profile export PATH=/usr/local/mongodb/bin:$PATH [ mongodb]# source /etc/profile
3)由于启动和停止服务名字太长,可以起一个别名
给停止服务起一个别名 [ mongodb]# alias cmdb=‘mongod --dbpath=/usr/local/mongodb/data/db/ --shutdown‘ 给启动服务起一个别名 [ mongodb]# alias smdb=‘mongod -f /usr/local/mongodb/etc/mongodb.conf‘
4)启动服务并连接
[ ~]# smdb about to fork child process, waiting until server is ready for connections. forked process: 5656 child process started successfully, parent exiting [ ~]# mongo --host 192.168.4.51 --port 27077 MongoDB shell version v3.6.3 connecting to: mongodb://192.168.4.51:27077/ MongoDB server version: 3.6.3
5)配置集群信息,任意一台都可以,在这里在51上面操作
> rs1_config = { //rs1_config随便起变量名,要记住 _id:"rs1", //必须为rs1这个,三台主机集群名,配置文件里面写的是这个 members:[ {_id:0,host:"192.168.4.51:27077"}, //_id值随意,host值固定 {_id:1,host:"192.168.4.52:27078"}, {_id:2,host:"192.168.4.53:27079"} ] }; //回车,出现下面情况为成功 { "_id" : "rs1", "members" : [ { "_id" : 0, "host" : "192.168.4.51:27077" }, { "_id" : 1, "host" : "192.168.4.52:27078" }, { "_id" : 2, "host" : "192.168.4.53:27079" } ] } >
6)初始化Replica Sets环境
> rs.initiate(rs1_config) { "ok" : 1, "operationTime" : Timestamp(1538187475, 1), "$clusterTime" : { "clusterTime" : Timestamp(1538187475, 1), "signature" : { "hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="), "keyId" : NumberLong(0) } } } rs1:SECONDARY> rs1:PRIMARY> //提示PRIMARY,51为主
工作过程: Bully算法(有最新操作为主、或者成为主的时间短) 选举 心跳 同步 rs1_config = { _id:"rs1", members:[ {_id:0,host:"192.168.4.51:27077",priority:0}, {_id:1,host:"192.168.4.52:27078",priority:3}, {_id:2,host:"192.168.4.53:27079",priority:2} ] };
7)在52和53上面查看
[ ~]# mongo --host 192.168.4.52 --port 27078 MongoDB shell version v3.6.3 connecting to: mongodb://192.168.4.52:27078/ MongoDB server version: 3.6.3 ... ... rs1:SECONDARY> //提示SECONDARY,52为从 rs1:SECONDARY> rs1:SECONDARY> [ ~]# mongo --host 192.168.4.53 --port 27079 MongoDB shell version v3.6.3 connecting to: mongodb://192.168.4.53:27079/ MongoDB server version: 3.6.3 ... ... rs1:SECONDARY> //提示SECONDARY,53为从 rs1:SECONDARY>
注意:如果初始化错误,重启服务登陆之后重新设置变量,之后再重新初始化
8)查看状态信息
rs1:PRIMARY> rs.status() ... "members" : [ { "_id" : 0, "name" : "192.168.4.51:27077", "health" : 1, "state" : 1, "stateStr" : "PRIMARY", "uptime" : 2295, ... ... }, { "_id" : 1, "name" : "192.168.4.52:27078", "health" : 1, "state" : 2, "stateStr" : "SECONDARY", "uptime" : 384, ... ... }, { "_id" : 2, "name" : "192.168.4.53:27079", "health" : 1, "state" : 2, "stateStr" : "SECONDARY", ... ...
9)查看是否是master库
rs1:PRIMARY> rs .isMaster( ) { "hosts" : [ "192.168.4.51:27077", "192.168.4.52:27078", "192.168.4.53:27079" ], "setName" : "rs1", "setVersion" : 1, "ismaster" : true, //主库 "secondary" : false, "primary" : "192.168.4.51:27077", "me" : "192.168.4.51:27077", ... ...
10)验证副本集,同步数据验证(51上面写数据)
rs1:PRIMARY> use gamedb2 switched to db gamedb2 rs1:PRIMARY> db.a.save({name:"yaya",age:75,em:""}) WriteResult({ "nInserted" : 1 })
52上面查看
[ ~]# mongo --host 192.168.4.52 --port 27078 rs1:SECONDARY> db.getMongo().setSlaveOk() //允许从库查看数据 rs1:SECONDARY> show dbs //有gamedb2库 admin 0.000GB config 0.000GB ddsdb 0.000GB gamedb2 0.000GB local 0.000GB test 0.000GB
步骤三:切换主库验证
1)自动切换主库验证
关闭51
[ ~]# cmdb //之前设置的别名 killing process with pid: 5656
查看52和53
[ ~]# mongo --host 192.168.4.52 --port 27078 MongoDB shell version v3.6.3 connecting to: mongodb://192.168.4.52:27078/ MongoDB server version: 3.6.3 ... ... rs1:PRIMARY> //52为主 rs1:PRIMARY> [ ~]# mongo --host 192.168.4.53 --port 27079 MongoDB shell version v3.6.3 connecting to: mongodb://192.168.4.53:27079/ MongoDB server version: 3.6.3 ... ... rs1:SECONDARY> //53为从 启动51,启动后不会再变为主,会成为52的从 [ ~]# smdb about to fork child process, waiting until server is ready for connections. forked process: 6598 child process started successfully, parent exiting rs1:SECONDARY> rs.isMaster() { "hosts" : [ "192.168.4.51:27077", "192.168.4.52:27078", "192.168.4.53:27079" ], "setName" : "rs1", "setVersion" : 1, "ismaster" : false, "secondary" : true, "primary" : "192.168.4.52:27078", "me" : "192.168.4.51:27077", ...
==================================
基于MongoDB环境完成下列练习:
插入文档
查询文档
更新文档
删除文档
步骤一:管理文档
1)把系统用户信息/etc/passwd存储到mdb库下的user集合里
rs1:PRIMARY> use mdb switched to db mdb rs1:PRIMARY> db.user.save({name:"yaya",password:"x",uid:9999,gid:9999,comment:"",homdir:"/home",shell:"/bin/bash"}) WriteResult({ "nInserted" : 1 }) rs1:PRIMARY> exit bye [ ~]# mongoexport --host 192.168.4.52 --port 27078 -d mdb -c user -f name,password,uid,gid,comment,homdir,shell --type=csv -o /tmp/user.csv 2018-09-29T11:04:14.967+0800 connected to: 192.168.4.52:27078 2018-09-29T11:04:14.968+0800 exported 1 record [ ~]# cp /etc/passwd /tmp [ ~]# sed -i ‘s/:/,/g‘ /tmp/passwd [ ~]# sed -i ‘$r /tmp/passwd‘ /tmp/user.csv [ ~]# mongoimport --host 192.168.4.52 --port 27078 -d mdb -c user --headerline --type=csv /tmp/user.csv 2018-09-29T11:06:08.355+0800 connected to: 192.168.4.52:27078 2018-09-29T11:06:08.363+0800 imported 41 documents
2)查看
[ ~]# mongo --host 192.168.4.52 --port 27078 rs1:PRIMARY> use mdb switched to db mdb rs1:PRIMARY> db.user.findOne() { "_id" : ObjectId("5baeeb37ce3cc5539aa21f38"), "name" : "yaya", "password" : "x", "uid" : 9999, "gid" : 9999, "comment" : "", "homdir" : "/home", "shell" : "/bin/bash" }
db.user.find({条件},{定义显示的字段}) #指定查询条件并指定显示的字段
0不显示 1显示
rs1:PRIMARY> db.user.find() { "_id" : ObjectId("5baeeb37ce3cc5539aa21f38"), "name" : "yaya", "password" : "x", "uid" : 9999, "gid" : 9999, "comment" : "", "homdir" : "/home", "shell" : "/bin/bash" } ... Type "it" for more //出现这个按it,默认出现20行
查看每行的name字段
rs1:PRIMARY> db.user.find({},{name:1}) { "_id" : ObjectId("5baeeb37ce3cc5539aa21f38"), "name" : "yaya" } ... ... { "_id" : ObjectId("5baeec2001805180a1011843"), "name" : "rpc" } Type "it" for more
不看_id字段
rs1:PRIMARY> db.user.find({},{_id:0}) { "name" : "yaya", "password" : "x", "uid" : 9999, "gid" : 9999, "comment" : "", "homdir" : "/home", "shell" : "/bin/bash" } ... ... { "name" : "rpc", "password" : "x", "uid" : 32, "gid" : 32, "comment" : "Rpcbind Daemon", "homdir" : "/var/lib/rpcbind", "shell" : "/sbin/nologin" } Type "it" for more
不看_id那一列,看name那一列
rs1:PRIMARY> db.user.find({},{_id:0,name:1}) { "name" : "yaya" } ... ... { "name" : "rpc" } Type "it" for more
查看以a开头的name字段
rs1:PRIMARY> db.user.find({name:/^a/},{_id:0}) { "name" : "adm", "password" : "x", "uid" : 3, "gid" : 4, "comment" : "adm", "homdir" : "/var/adm", "shell" : "/sbin/nologin" } { "name" : "abrt", "password" : "x", "uid" : 173, "gid" : 173, "comment" : "", "homdir" : "/etc/abrt", "shell" : "/sbin/nologin" } { "name" : "avahi", "password" : "x", "uid" : 70, "gid" : 70, "comment" : "Avahi mDNS/DNS-SD Stack", "homdir" : "/var/run/avahi-daemon", "shell" : "/sbin/nologin" }
显示查询结果的前一行
limit 数字
rs1:PRIMARY> db.user.find({name:/^a/},{_id:0}).limit (1) { "name" : "adm", "password" : "x", "uid" : 3, "gid" : 4, "comment" : "adm", "homdir" : "/var/adm", "shell" : "/sbin/nologin" }
显示name字段以a开头的第一行
rs1:PRIMARY> db.user.findOne({name:/^a/},{_id:0,name:1,shell:1,uid:1}) { "name" : "adm", "uid" : 3, "shell" : "/sbin/nologin" }
跳过几行显示 (2行)
skip 数字
rs1:PRIMARY> db.user.find({name:/^a/},{_id:0,name:1,shell:1}).skip (2) { "name" : "avahi", "shell" : "/sbin/nologin" }
默认升序排序
sort 字段名
rs1:PRIMARY> db.user.find({name:/^a/},{_id:0,name:1,shell:1,uid:1}).sort({uid:1}) { "name" : "adm", "uid" : 3, "shell" : "/sbin/nologin" } { "name" : "avahi", "uid" : 70, "shell" : "/sbin/nologin" } { "name" : "abrt", "uid" : 173, "shell" : "/sbin/nologin" }
降序排序
rs1:PRIMARY> db.user.find({name:/^a/},{_id:0,name:1,shell:1,uid:1}).sort({uid:-1}) { "name" : "abrt", "uid" : 173, "shell" : "/sbin/nologin" } { "name" : "avahi", "uid" : 70, "shell" : "/sbin/nologin" } { "name" : "adm", "uid" : 3, "shell" : "/sbin/nologin" }
显示name字段以a开头和uid为3的所有行
rs1:PRIMARY> db.user.find({name:/^a/,uid:3},{_id:0,name:1,shell:1,uid:1}) { "name" : "adm", "uid" : 3, "shell" : "/sbin/nologin" }
3)条件判断的表示方式
$in 在…里
rs1:PRIMARY> db.user.find({uid:{$in:[1,6,9]}}) //uid的为1或者6或者9的匹配 { "_id" : ObjectId("5baeec2001805180a1011833"), "name" : "bin", "password" : "x", "uid" : 1, "gid" : 1, "comment" : "bin", "homdir" : "/bin", "shell" : "/sbin/nologin" } { "_id" : ObjectId("5baeec2001805180a1011838"), "name" : "shutdown", "password" : "x", "uid" : 6, "gid" : 0, "comment" : "shutdown", "homdir" : "/sbin", "shell" : "/sbin/shutdown" }
$nin 不在…里
rs1:PRIMARY> db.user.find({uid:{$nin:[1,6,9]}},{_id:0,name:1,uid:1}) { "name" : "yaya", "uid" : 9999 } ... ... { "name" : "saslauth", "uid" : 996 } Type "it" for more
$or 条件满足任意其中一个即可
rs1:PRIMARY> db.user.find({$or:[{name:"root"},{uid:1}]},{_id:0,name:1,uid:1}) { "name" : "root", "uid" : 0 } { "name" : "bin", "uid" : 1 }
4)正则匹配,以a开头的name字段
rs1:PRIMARY> db.user.find({name:/^a/},{_id:0,name:1,uid:1}) { "name" : "adm", "uid" : 3 } { "name" : "abrt", "uid" : 173 } { "name" : "avahi", "uid" : 70 }
objectId对象类型 使用12字节的存储空间每个字节2位十六进制数字共计24位字符串.
0 1 2 3 4 5 6 7 8 9 10 11
时间戳 机器 PID 计数器
5)数值比较
$lt(小于) $lte(小于等于) $gt(大于) $gte(大于等于) $ne(不等于)
rs1:PRIMARY> db.user.find({uid:{$gte:10,$lte:40}},{_id:0,name:1,uid:1}) { "name" : "operator", "uid" : 11 } { "name" : "games", "uid" : 12 } { "name" : "ftp", "uid" : 14 } { "name" : "rpc", "uid" : 32 } { "name" : "rpcuser", "uid" : 29 } { "name" : "ntp", "uid" : 38 }
匹配null:可以匹配没有的字段,也可以检查这个字段有没有
rs1:PRIMARY> db.user.save({name:null,uid:null}) WriteResult({ "nInserted" : 1 }) rs1:PRIMARY> db.user.find({name:null}) { "_id" : ObjectId("5baef385f9f3bf625ea1dbd6"), "name" : null, "uid" : null } rs1:PRIMARY> db.user.find({shell:null}) //表示此条文档没有shell字段 { "_id" : ObjectId("5baef385f9f3bf625ea1dbd6"), "name" : null, "uid" : null }
6)save和insert的区别
相同点:当集合不存在时创建集合,并插入记录
不同点:
save() _id字段值已经存在时,修改文档字段值
insert() _id字段值已经存在时,放弃修改文档字段值
rs1:PRIMARY> db.t1.save({name:"bob",age:19}) rs1:PRIMARY> db.t1.insert({name:"bob",age:19}) rs1:PRIMARY> db.t1.save({_id:7,name:"bob",age:19}) WriteResult({ "nMatched" : 0, "nUpserted" : 1, "nModified" : 0, "_id" : 7 }) rs1:PRIMARY> db.t1.find() ... ... { "_id" : 7, "name" : "bob", "age" : 19 } rs1:PRIMARY> db.t1.save({_id:7,name:"tom",age:19}) //把上一条的记录直接修改 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.t1.find() ... ... { "_id" : 7, "name" : "tom", "age" : 19 } rs1:PRIMARY> db.t1.insert({_id:8,name:"tom",age:19}) //可以存上 WriteResult({ "nInserted" : 1 }) rs1:PRIMARY> db.t1.insert({_id:8,name:"tom",age:19}) //存不上 WriteResult({ "nInserted" : 0, "writeError" : { "code" : 11000, "errmsg" : "E11000 duplicate key error collection: mdb.t1 index: _id_ dup key: { : 8.0 }" } })
7)插入多行文档
rs1:PRIMARY> db.t1.insertMany([{name:"xiaojiu"},{name:"laoshi"}]) { "acknowledged" : true, "insertedIds" : [ ObjectId("5baef526f9f3bf625ea1dbd9"), ObjectId("5baef526f9f3bf625ea1dbda") ] } rs1:PRIMARY> db.t1.find() ... ... { "_id" : ObjectId("5baef526f9f3bf625ea1dbd9"), "name" : "xiaojiu" } { "_id" : ObjectId("5baef526f9f3bf625ea1dbda"), "name" : "laoshi" }
8)update修改
>db.user.find.update({条件},{修改的字段}) rs1:PRIMARY> db.user.update({name:"root"},{password:"XXX"}) //如果这一列不写完整,这一行除了password这一行,这一列的其他值都没有了 相当于删除(要写完整) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.t1.find({name:"root"}) rs1:PRIMARY> db.user.find({name:"root"}) //没有东西,除了password:"XXX"
9)$set 条件匹配时,修改指定字段的值(局部修改)
rs1:PRIMARY> db.user.update({name:"adm"},{$set:{password:"AAA"}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.user.find({name:"adm"}) //还存在 { "_id" : ObjectId("5baeec2001805180a1011835"), "name" : "adm", "password" : "AAA", "uid" : 3, "gid" : 4, "comment" : "adm", "homdir" : "/var/adm", "shell" : "/sbin/nologin" } rs1:PRIMARY> db.user.update({name:/^r/},{$set:{password:"FFF"}}) //默认修改匹配条件的第一行 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.user.update({name:/^a/},{$set:{password:"FFF"}},false,true) //改匹配到的所有 WriteResult({ "nMatched" : 3, "nUpserted" : 0, "nModified" : 3 })
10)$unset 删除与条件匹配文档的字段
rs1:PRIMARY> db.user.update({name:"sync"},{$unset:{password:1}}) //删除password字段 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
11)数组
rs1:PRIMARY> db.user.insert({name:"bob",like:["a","b","c","d","e","f",]}) WriteResult({ "nInserted" : 1 })
$pop
删除数组末尾一个元素,1删除最后一个,-1删除第一个
rs1:PRIMARY> db.user.update({name:"bob"},{$pop:{like:1}}) //删除匹配的第一条的最后一个 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.user.update({name:"bob"},{$pop:{like:-1}}) //删除匹配的第一条的第一个 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
$push 向数组中添加新元素
rs1:PRIMARY> db.user.update({name:"bob"},{$push:{like:"Z"}}) //默认添加到最后 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.user.update({name:"bob"},{$push:{like:"W"}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.user.find({name:"bob"}) { "_id" : ObjectId("5baef7b2034891a205de2959"), "name" : "bob", "like" : [ "b", "c", "d", "e", "Z", "W" ] }
$addToSet 避免重复添加
rs1:PRIMARY> db.user.update({name:"bob"},{$addToSet:{like:"W"}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 0 }) rs1:PRIMARY> db.user.find({name:"bob"}) { "_id" : ObjectId("5baef7b2034891a205de2959"), "name" : "bob", "like" : [ "b", "c", "d", "e", "Z", "W" ] }
$pull 删除数组里的指定元素,若有两个bob可以用_id值定义把name:"bob"换成id值
rs1:PRIMARY> db.user.update({name:"bob"},{$pull:{like:"c"}}) WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.user.find({name:"bob"}) { "_id" : ObjectId("5baef7b2034891a205de2959"), "name" : "bob", "like" : [ "b", "d", "e", "Z", "W" ] } rs1:PRIMARY> db.user.update({"_id":ObjectId("5afc1a717eff45e9cfc57ed3")},{$push:{like:"S"}}) WriteResult({ "nMatched" : 0, "nUpserted" : 0, "nModified" : 0 }) rs1:PRIMARY>
$inc 条件匹配时,字段值自加或自减
rs1:PRIMARY> db.user.update({uid:{$lte:10}},{$inc:{uid:2}}) //设置字段值自加2,默认改第一行 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 }) rs1:PRIMARY> db.user.update({uid:{$lte:10}},{$inc:{uid:2}},false,true) //设置字段值自加2,false,true改全部 WriteResult({ "nMatched" : 8, "nUpserted" : 0, "nModified" : 8 }) rs1:PRIMARY> db.user.update({uid:{$lte:10}},{$inc:{uid:-1}}) //负数时是自减1,默认改第一行 WriteResult({ "nMatched" : 1, "nUpserted" : 0, "nModified" : 1 })
12)删除文档
remove()与drop()的区别
remove()删除文档时不删除索引
drop()删除集合的时候同时删除索引
rs1:PRIMARY> db.t1.remove({}) WriteResult({ "nRemoved" : 6 }) rs1:PRIMARY> db.user.remove({name:"/^a/"}) //删除以a开头的记录 WriteResult({ "nRemoved" : 0 }) rs1:PRIMARY> db.t1.drop() //删除集合t1 true