MongoDB的副本集（Replica Set）是一个高可用性、可扩展性和冗余性的数据库解决方案。它能够确保数据库的高可用性，同时保障了数据的安全性。在本文中，我们将探讨如何在一个已经包含三个数据节点的副本集集群中，添加一个仲裁节点，并同时删除原先的一个数据节点。
1、副本集的基本概念
MongoDB副本集由多个数据节点+若干个仲裁节点（可以没有）组成，其中一个数据节点被选举为主节点(Primary），其余节点为次要节点（Secondary）。在故障时，副本集会自动进行主节点的切换，确保服务的高可用性。副本集中的仲裁节点（Arbiter）不存储数据，仅用于选举过程，帮助副本集进行决策。
例如，当前配置一个3节点的数据节点集群，组件集群命令如下：
> use admin
> rs.initiate( {_id: "test1",members: [{ _id: 0, host: "192.168.122.36:27017" },{ _id: 1, host: "192.168.122.36:27018" },{ _id: 2, host: "192.168.122.36:27019" } ] })
查看状态：
test1:PRIMARY> use admin
switched to db admin
test1:PRIMARY> rs.status()
{
"set" : "test1",
"date" : ISODate("2023-10-23T02:31:02.345Z"),
"myState" : 1,
"term" : NumberLong(1),
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"heartbeatIntervalMillis" : NumberLong(2000),
"majorityVoteCount" : 2,
"writeMajorityCount" : 2,
"optimes" : {
"lastCommittedOpTime" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"lastCommittedWallTime" : ISODate("2023-10-23T02:30:54.919Z"),
"readConcernMajorityOpTime" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"readConcernMajorityWallTime" : ISODate("2023-10-23T02:30:54.919Z"),
"appliedOpTime" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"durableOpTime" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"lastAppliedWallTime" : ISODate("2023-10-23T02:30:54.919Z"),
"lastDurableWallTime" : ISODate("2023-10-23T02:30:54.919Z")
},
"lastStableRecoveryTimestamp" : Timestamp(1698028224, 3),
"lastStableCheckpointTimestamp" : Timestamp(1698028224, 3),
"electionCandidateMetrics" : {
"lastElectionReason" : "electionTimeout",
"lastElectionDate" : ISODate("2023-10-23T02:30:24.838Z"),
"electionTerm" : NumberLong(1),
"lastCommittedOpTimeAtElection" : {
"ts" : Timestamp(0, 0),
"t" : NumberLong(-1)
},
"lastSeenOpTimeAtElection" : {
"ts" : Timestamp(1698028214, 1),
"t" : NumberLong(-1)
},
"numVotesNeeded" : 2,
"priorityAtElection" : 1,
"electionTimeoutMillis" : NumberLong(10000),
"numCatchUpOps" : NumberLong(0),
"newTermStartDate" : ISODate("2023-10-23T02:30:24.912Z"),
"wMajorityWriteAvailabilityDate" : ISODate("2023-10-23T02:30:25.497Z")
},
"members" : [
{
"_id" : 0,
"name" : "192.168.122.36:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 183,
"optime" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T02:30:54Z"),
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "could not find member to sync from",
"electionTime" : Timestamp(1698028224, 1),
"electionDate" : ISODate("2023-10-23T02:30:24Z"),
"configVersion" : 1,
"self" : true,
"lastHeartbeatMessage" : ""
},
{
"_id" : 1,
"name" : "192.168.122.36:27018",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 47,
"optime" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"optimeDurable" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T02:30:54Z"),
"optimeDurableDate" : ISODate("2023-10-23T02:30:54Z"),
"lastHeartbeat" : ISODate("2023-10-23T02:31:00.889Z"),
"lastHeartbeatRecv" : ISODate("2023-10-23T02:31:01.443Z"),
"pingMs" : NumberLong(0),
"lastHeartbeatMessage" : "",
"syncingTo" : "192.168.122.36:27017",
"syncSourceHost" : "192.168.122.36:27017",
"syncSourceId" : 0,
"infoMessage" : "",
"configVersion" : 1
},
{
"_id" : 2,
"name" : "192.168.122.36:27019",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 47,
"optime" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"optimeDurable" : {
"ts" : Timestamp(1698028254, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T02:30:54Z"),
"optimeDurableDate" : ISODate("2023-10-23T02:30:54Z"),
"lastHeartbeat" : ISODate("2023-10-23T02:31:00.890Z"),
"lastHeartbeatRecv" : ISODate("2023-10-23T02:31:01.446Z"),
"pingMs" : NumberLong(0),
"lastHeartbeatMessage" : "",
"syncingTo" : "192.168.122.36:27017",
"syncSourceHost" : "192.168.122.36:27017",
"syncSourceId" : 0,
"infoMessage" : "",
"configVersion" : 1
}
],
"ok" : 1,
"$clusterTime" : {
"clusterTime" : Timestamp(1698028254, 1),
"signature" : {
"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
"keyId" : NumberLong(0)
}
},
"operationTime" : Timestamp(1698028254, 1)
}此时可以用一个程序验证集群的可用情况：
from pymongo import MongoClient
from pymongo.errors import ConnectionFailure
# MongoDB副本集的连接信息
replica_set = "test1"
mongo_host = ["192.168.122.36:27017", "192.168.122.36:27018", "192.168.122.36:27019"] # 主机和端口号列表
# 尝试连接MongoDB副本集
try:
client = MongoClient(mongo_host, replicaSet=replica_set, serverSelectinotallow=5000)
db = client["testdb"]
print("Successfully connected to MongoDB replica set.")
collection = db['test1']
data = {"b":2}
result = collection.insert_one(data)
except ConnectionFailure:
print("Failed to connect to MongoDB replica set. Please check your connection settings.")运行结果如下：
查看数据库中运行结果：
2、添加节点
要添加一个仲裁节点，我们需要在现有副本集中的任一节点上执行如下操作：
rs.add( { host: "IP:PORT", arbiterOnly:true } )执行这个命令后，MongoDB将会在副本集中添加一个仲裁节点，增加了选主的稳定性，同时不会增加数据存储的压力。具体实际案例如下
test1:PRIMARY> use admin
switched to db admin
test1:PRIMARY> rs.add( { host: "192.168.122.36:27020", arbiterOnly:true } )
{
"ok" : 1,
"$clusterTime" : {
"clusterTime" : Timestamp(1698030806, 1),
"signature" : {
"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
"keyId" : NumberLong(0)
}
},
"operationTime" : Timestamp(1698030806, 1)
}
test1:PRIMARY> rs.status();
{
"set" : "test1",
"date" : ISODate("2023-10-23T03:13:32.511Z"),
"myState" : 1,
"term" : NumberLong(1),
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"heartbeatIntervalMillis" : NumberLong(2000),
"majorityVoteCount" : 3,
"writeMajorityCount" : 3,
"optimes" : {
"lastCommittedOpTime" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"lastCommittedWallTime" : ISODate("2023-10-23T03:13:26.379Z"),
"readConcernMajorityOpTime" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"readConcernMajorityWallTime" : ISODate("2023-10-23T03:13:26.379Z"),
"appliedOpTime" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"durableOpTime" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"lastAppliedWallTime" : ISODate("2023-10-23T03:13:26.379Z"),
"lastDurableWallTime" : ISODate("2023-10-23T03:13:26.379Z")
},
"lastStableRecoveryTimestamp" : Timestamp(1698030805, 1),
"lastStableCheckpointTimestamp" : Timestamp(1698030805, 1),
"electionCandidateMetrics" : {
"lastElectionReason" : "electionTimeout",
"lastElectionDate" : ISODate("2023-10-23T02:30:24.838Z"),
"electionTerm" : NumberLong(1),
"lastCommittedOpTimeAtElection" : {
"ts" : Timestamp(0, 0),
"t" : NumberLong(-1)
},
"lastSeenOpTimeAtElection" : {
"ts" : Timestamp(1698028214, 1),
"t" : NumberLong(-1)
},
"numVotesNeeded" : 2,
"priorityAtElection" : 1,
"electionTimeoutMillis" : NumberLong(10000),
"numCatchUpOps" : NumberLong(0),
"newTermStartDate" : ISODate("2023-10-23T02:30:24.912Z"),
"wMajorityWriteAvailabilityDate" : ISODate("2023-10-23T02:30:25.497Z")
},
"members" : [
{
"_id" : 0,
"name" : "192.168.122.36:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 2733,
"optime" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T03:13:26Z"),
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "",
"electionTime" : Timestamp(1698028224, 1),
"electionDate" : ISODate("2023-10-23T02:30:24Z"),
"configVersion" : 2,
"self" : true,
"lastHeartbeatMessage" : ""
},
{
"_id" : 1,
"name" : "192.168.122.36:27018",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 2597,
"optime" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"optimeDurable" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T03:13:26Z"),
"optimeDurableDate" : ISODate("2023-10-23T03:13:26Z"),
"lastHeartbeat" : ISODate("2023-10-23T03:13:32.393Z"),
"lastHeartbeatRecv" : ISODate("2023-10-23T03:13:32.420Z"),
"pingMs" : NumberLong(1),
"lastHeartbeatMessage" : "",
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "",
"configVersion" : 2
},
{
"_id" : 2,
"name" : "192.168.122.36:27019",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 2597,
"optime" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"optimeDurable" : {
"ts" : Timestamp(1698030806, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T03:13:26Z"),
"optimeDurableDate" : ISODate("2023-10-23T03:13:26Z"),
"lastHeartbeat" : ISODate("2023-10-23T03:13:32.393Z"),
"lastHeartbeatRecv" : ISODate("2023-10-23T03:13:32.445Z"),
"pingMs" : NumberLong(1),
"lastHeartbeatMessage" : "",
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "",
"configVersion" : 2
},
{
"_id" : 3,
"name" : "192.168.122.36:27020",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 6,
"lastHeartbeat" : ISODate("2023-10-23T03:13:32.405Z"),
"lastHeartbeatRecv" : ISODate("2023-10-23T03:13:32.455Z"),
"pingMs" : NumberLong(4),
"lastHeartbeatMessage" : "",
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "",
"configVersion" : 2
}
],
"ok" : 1,
"$clusterTime" : {
"clusterTime" : Timestamp(1698030806, 1),
"signature" : {
"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
"keyId" : NumberLong(0)
}
},
"operationTime" : Timestamp(1698030806, 1)
}此时修改python中程序（插入的数据），运行结果如下：
3、删除数据节点
如果你需要删除一个数据节点，首先，你要确保副本集的健康状态，然后执行如下操作来删除一个Secondary节点。
rs.remove("IP:PORT");执行这个命令后，MongoDB将从副本集中移除该数据节点，副本集会重新进行选举，确保副本集的高可用性。
test1:PRIMARY> rs.remove("192.168.122.36:27019")
{
"ok" : 1,
"$clusterTime" : {
"clusterTime" : Timestamp(1698038888, 1),
"signature" : {
"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
"keyId" : NumberLong(0)
}
},
"operationTime" : Timestamp(1698038888, 1)
}
test1:PRIMARY> rs.status
function() {
return db._adminCommand("replSetGetStatus"); # 选举中
}
test1:PRIMARY> rs.status()
{
"set" : "test1",
"date" : ISODate("2023-10-23T05:28:15.398Z"),
"myState" : 1,
"term" : NumberLong(1),
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"heartbeatIntervalMillis" : NumberLong(2000),
"majorityVoteCount" : 2,
"writeMajorityCount" : 2,
"optimes" : {
"lastCommittedOpTime" : {
"ts" : Timestamp(1698038888, 1),
"t" : NumberLong(1)
},
"lastCommittedWallTime" : ISODate("2023-10-23T05:28:08.335Z"),
"readConcernMajorityOpTime" : {
"ts" : Timestamp(1698038888, 1),
"t" : NumberLong(1)
},
"readConcernMajorityWallTime" : ISODate("2023-10-23T05:28:08.335Z"),
"appliedOpTime" : {
"ts" : Timestamp(1698038888, 1),
"t" : NumberLong(1)
},
"durableOpTime" : {
"ts" : Timestamp(1698038888, 1),
"t" : NumberLong(1)
},
"lastAppliedWallTime" : ISODate("2023-10-23T05:28:08.335Z"),
"lastDurableWallTime" : ISODate("2023-10-23T05:28:08.335Z")
},
"lastStableRecoveryTimestamp" : Timestamp(1698038845, 1),
"lastStableCheckpointTimestamp" : Timestamp(1698038845, 1),
"electionCandidateMetrics" : {
"lastElectionReason" : "electionTimeout",
"lastElectionDate" : ISODate("2023-10-23T02:30:24.838Z"),
"electionTerm" : NumberLong(1),
"lastCommittedOpTimeAtElection" : {
"ts" : Timestamp(0, 0),
"t" : NumberLong(-1)
},
"lastSeenOpTimeAtElection" : {
"ts" : Timestamp(1698028214, 1),
"t" : NumberLong(-1)
},
"numVotesNeeded" : 2,
"priorityAtElection" : 1,
"electionTimeoutMillis" : NumberLong(10000),
"numCatchUpOps" : NumberLong(0),
"newTermStartDate" : ISODate("2023-10-23T02:30:24.912Z"),
"wMajorityWriteAvailabilityDate" : ISODate("2023-10-23T02:30:25.497Z")
},
"members" : [
{
"_id" : 0,
"name" : "192.168.122.36:27017",
"health" : 1,
"state" : 1,
"stateStr" : "PRIMARY",
"uptime" : 10816,
"optime" : {
"ts" : Timestamp(1698038888, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T05:28:08Z"),
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "",
"electionTime" : Timestamp(1698028224, 1),
"electionDate" : ISODate("2023-10-23T02:30:24Z"),
"configVersion" : 3,
"self" : true,
"lastHeartbeatMessage" : ""
},
{
"_id" : 1,
"name" : "192.168.122.36:27018",
"health" : 1,
"state" : 2,
"stateStr" : "SECONDARY",
"uptime" : 10680,
"optime" : {
"ts" : Timestamp(1698038888, 1),
"t" : NumberLong(1)
},
"optimeDurable" : {
"ts" : Timestamp(1698038888, 1),
"t" : NumberLong(1)
},
"optimeDate" : ISODate("2023-10-23T05:28:08Z"),
"optimeDurableDate" : ISODate("2023-10-23T05:28:08Z"),
"lastHeartbeat" : ISODate("2023-10-23T05:28:14.347Z"),
"lastHeartbeatRecv" : ISODate("2023-10-23T05:28:15.388Z"),
"pingMs" : NumberLong(1),
"lastHeartbeatMessage" : "",
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "",
"configVersion" : 3
},
{
"_id" : 3,
"name" : "192.168.122.36:27020",
"health" : 1,
"state" : 7,
"stateStr" : "ARBITER",
"uptime" : 8088,
"lastHeartbeat" : ISODate("2023-10-23T05:28:14.342Z"),
"lastHeartbeatRecv" : ISODate("2023-10-23T05:28:14.369Z"),
"pingMs" : NumberLong(0),
"lastHeartbeatMessage" : "",
"syncingTo" : "",
"syncSourceHost" : "",
"syncSourceId" : -1,
"infoMessage" : "",
"configVersion" : 3
}
],
"ok" : 1,
"$clusterTime" : {
"clusterTime" : Timestamp(1698038888, 1),
"signature" : {
"hash" : BinData(0,"AAAAAAAAAAAAAAAAAAAAAAAAAAA="),
"keyId" : NumberLong(0)
}
},
"operationTime" : Timestamp(1698038888, 1)
}注意：复制集将会短暂的关闭连接并进入选举，选举出一个新的主节点。接口将会自动重连。接口可能将会报错 DBClientCursor::init call() failed 。
