原文链接：https://blog.csdn.net/jacobxian/article/details/103634866

在mongodb副本集集群中,若是丢失了n/2+1个节点,也就是失去了多数节点时候,在集群内无法投票表决剩下的节点谁能当主库,那就需要强制其中一个节点为主库.
下面是操作步骤:
1.执行cfg=rs.conf()
2.需要设置需要的_id为主库，
cfg.members=[cfg.members[3]]
3.执行集群重新设置
4.rs.reconfig(cfg, {force: true});
5.rs.status()

下面是详细操作步骤,该集群是四个节点的副本集,其中三个节点已经丢失,使剩下的节点为主节点,单实例运行.
执行cfg=rs.conf(),获取集群的节点信息:

cms:SECONDARY> cfg=rs.conf()
{
"_id" : "cms",
"version" : 32948,
"protocolVersion" : NumberLong(1),
"members" : [
{
"_id" : 0,
"host" : "192.168.1.193:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {

},
"slaveDelay" : NumberLong(0),
"votes" : 1
},
{
"_id" : 1,
"host" : "192.168.1.195:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {

},
"slaveDelay" : NumberLong(0),
"votes" : 1
},
{
"_id" : 2,
"host" : "192.168.1.198:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {

},
"slaveDelay" : NumberLong(0),
"votes" : 1
},
{
"_id" : 3,
"host" : "192.168.1.199:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {

},
"slaveDelay" : NumberLong(0),
"votes" : 1
}
],
"settings" : {
"chainingAllowed" : true,
"heartbeatIntervalMillis" : 2000,
"heartbeatTimeoutSecs" : 10,
"electionTimeoutMillis" : 10000,
"getLastErrorModes" : {

},
"getLastErrorDefaults" : {
"w" : 1,
"wtimeout" : 0
},
"replicaSetId" : ObjectId("59cab0d69b86b9e815cfc02b")
}
}

下面指定199为主节点

cms:SECONDARY> cfg.members=[cfg.members[3]]
[
{
"_id" : 3,
"host" : "192.168.1.199:27017",
"arbiterOnly" : false,
"buildIndexes" : true,
"hidden" : false,
"priority" : 1,
"tags" : {

},
"slaveDelay" : NumberLong(0),
"votes" : 1
}
]

cms:SECONDARY> rs.reconfig(cfg, {force: true})
{ "ok" : 1 }

查看更改后的集群状态,操作完成,变成了单实例节点.然后通过添加从节点来实现副本集

cms:SECONDARY> rs.status()
{
"set" : "cms",
"date" : ISODate("2018-09-04T01:54:12.445Z"),
"myState" : 1,
"term" : NumberLong(6),
"heartbeatIntervalMillis" : NumberLong(2000),
"members" : [
{
"_id" : 3,
"name" : "192.168.1.199:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 3152420,
"optime" : {
"ts" : Timestamp(1536026052, 24),
"t" : NumberLong(6)
},
"optimeDate" : ISODate("2018-09-04T01:54:12Z"),
"infoMessage" : "could not find member to sync from",
"electionTime" : Timestamp(1536026047, 1),
"electionDate" : ISODate("2018-09-04T01:54:07Z"),
"configVersion" : 137682,
"self" : true
}
],
"ok" : 1
}

添加新的节点:

config = { _id:"cms", members:
[
{_id:0,host:"192.168.1.193:27017"},
{_id:1,host:"192.168.1.194:27017"},
{_id:2,host:"192.168.1.198:27017"},
{_id:3,host:"192.168.1.199:27017"}
]
}

rs.reconfig(config);

然后在新节点启动mongod进程,必须要把新节点数据文件清空.
节点启动以后,会自动拉取主库数据,然后创建索引.
mongod --fork --config /data0/mongodb/mongod.conf