一、环境
1$ cat /etc/redhat-release 2CentOS Linux release 7.0.1406 (Core) 3$ uname -a 4Linux zhaopin-2-201 3.10.0-123.el7.x86_64 #1 SMP Mon Jun 30 12:09:22 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux 5$ mongo 6MongoDB shell version: 3.0.6 7connecting to: test 8rs0:PRIMARY> rs.status(); 9{ 10 "set" : "rs0", 11 "date" : ISODate("2015-09-28T07:00:05.507Z"), 12 "myState" : 1, 13 "members" : [ 14 { 15 "_id" : 0, 16 "name" : "172.30.2.201:27017", 17 "health" : 1, 18 "state" : 1, 19 "stateStr" : "PRIMARY", 20 "uptime" : 83, 21 "optime" : Timestamp(1443423600, 1), 22 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 23 "electionTime" : Timestamp(1443423535, 2), 24 "electionDate" : ISODate("2015-09-28T06:58:55Z"), 25 "configVersion" : 3, 26 "self" : true 27 }, 28 { 29 "_id" : 1, 30 "name" : "172.30.2.203:27017", 31 "health" : 1, 32 "state" : 2, 33 "stateStr" : "SECONDARY", 34 "uptime" : 44, 35 "optime" : Timestamp(1443423600, 1), 36 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 37 "lastHeartbeat" : ISODate("2015-09-28T07:00:04.918Z"), 38 "lastHeartbeatRecv" : ISODate("2015-09-28T07:00:05.042Z"), 39 "pingMs" : 0, 40 "syncingTo" : "172.30.2.201:27017", 41 "configVersion" : 3 42 }, 43 { 44 "_id" : 2, 45 "name" : "172.30.2.202:27017", 46 "health" : 1, 47 "state" : 5, 48 "stateStr" : "STARTUP2", 49 "uptime" : 4, 50 "optime" : Timestamp(0, 0), 51 "optimeDate" : ISODate("1970-01-01T00:00:00Z"), 52 "lastHeartbeat" : ISODate("2015-09-28T07:00:04.918Z"), 53 "lastHeartbeatRecv" : ISODate("2015-09-28T07:00:04.926Z"), 54 "pingMs" : 0, 55 "configVersion" : 3 56 } 57 ], 58 "ok" : 1 59}
二、单节点故障
1.primary节点故障
1）关闭primay节点
1rs0:PRIMARY> use admin; 2switched to db admin 3rs0:PRIMARY> db.shutdownServer(); 42015-09-28T15:00:51.828+0800 I NETWORK DBClientCursor::init call() failed 5server should be down... 62015-09-28T15:00:51.830+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed 72015-09-28T15:00:51.831+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) ok 82015-09-28T15:00:51.831+0800 I NETWORK DBClientCursor::init call() failed 9> 10bye
2）查看集群状态
1$ mongo 2MongoDB shell version: 3.0.6 3connecting to: test 4rs0:SECONDARY> rs.status(); 5{ 6 "set" : "rs0", 7 "date" : ISODate("2015-09-28T07:01:28.818Z"), 8 "myState" : 2, 9 "members" : [ 10 { 11 "_id" : 0, 12 "name" : "172.30.2.201:27017", 13 "health" : 0, 14 "state" : 8, 15 "stateStr" : "(not reachable/healthy)", 16 "uptime" : 0, 17 "optime" : Timestamp(0, 0), 18 "optimeDate" : ISODate("1970-01-01T00:00:00Z"), 19 "lastHeartbeat" : ISODate("2015-09-28T07:01:27.006Z"), 20 "lastHeartbeatRecv" : ISODate("2015-09-28T07:00:50.935Z"), 21 "pingMs" : 0, 22 "lastHeartbeatMessage" : "Failed attempt to connect to 172.30.2.201:27017; couldn't connect to server 172.30.2.201:27017 (172.30.2.201), connection attempt failed", 23 "configVersion" : -1 24 }, 25 { 26 "_id" : 1, 27 "name" : "172.30.2.203:27017", 28 "health" : 1, 29 "state" : 1, 30 "stateStr" : "PRIMARY", 31 "uptime" : 87, 32 "optime" : Timestamp(1443423600, 1), 33 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 34 "lastHeartbeat" : ISODate("2015-09-28T07:01:26.963Z"), 35 "lastHeartbeatRecv" : ISODate("2015-09-28T07:01:27.078Z"), 36 "pingMs" : 0, 37 "electionTime" : Timestamp(1443423653, 1), 38 "electionDate" : ISODate("2015-09-28T07:00:53Z"), 39 "configVersion" : 3 40 }, 41 { 42 "_id" : 2, 43 "name" : "172.30.2.202:27017", 44 "health" : 1, 45 "state" : 2, 46 "stateStr" : "SECONDARY", 47 "uptime" : 90, 48 "optime" : Timestamp(1443423600, 1), 49 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 50 "configVersion" : 3, 51 "self" : true 52 } 53 ], 54 "ok" : 1 55}
发现集群进行了自动切换，把172.30.2.202：27017变为了primary
3）启动原来的primary
1$ sudo /opt/mongodb/bin/mongod --config /data/mongodb/conf/db0/mongodb.conf 2about to fork child process, waiting until server is ready for connections. 3forked process: 25738 4child process started successfully, parent exiting 5$ mongo 6MongoDB shell version: 3.0.6 7connecting to: test 8rs0:PRIMARY> rs.status(); 9{ 10 "set" : "rs0", 11 "date" : ISODate("2015-09-28T07:02:24.312Z"), 12 "myState" : 1, 13 "members" : [ 14 { 15 "_id" : 0, 16 "name" : "172.30.2.201:27017", 17 "health" : 1, 18 "state" : 2, 19 "stateStr" : "SECONDARY", 20 "uptime" : 13, 21 "optime" : Timestamp(1443423600, 1), 22 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 23 "lastHeartbeat" : ISODate("2015-09-28T07:02:23.189Z"), 24 "lastHeartbeatRecv" : ISODate("2015-09-28T07:02:22.873Z"), 25 "pingMs" : 0, 26 "configVersion" : 3 27 }, 28 { 29 "_id" : 1, 30 "name" : "172.30.2.203:27017", 31 "health" : 1, 32 "state" : 1, 33 "stateStr" : "PRIMARY", 34 "uptime" : 185, 35 "optime" : Timestamp(1443423600, 1), 36 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 37 "electionTime" : Timestamp(1443423653, 1), 38 "electionDate" : ISODate("2015-09-28T07:00:53Z"), 39 "configVersion" : 3, 40 "self" : true 41 }, 42 { 43 "_id" : 2, 44 "name" : "172.30.2.202:27017", 45 "health" : 1, 46 "state" : 2, 47 "stateStr" : "SECONDARY", 48 "uptime" : 143, 49 "optime" : Timestamp(1443423600, 1), 50 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 51 "lastHeartbeat" : ISODate("2015-09-28T07:02:23.103Z"), 52 "lastHeartbeatRecv" : ISODate("2015-09-28T07:02:22.990Z"), 53 "pingMs" : 0, 54 "configVersion" : 3 55 } 56 ], 57 "ok" : 1 58}
发现原来的primary自动切为了secondary
2.secondary节点故障
1）关闭secondary节点
1$ mongo 2MongoDB shell version: 3.0.6 3connecting to: test 4rs0:SECONDARY> use admin; 5switched to db admin 6rs0:SECONDARY> db.shutdownServer(); 72015-09-28T15:04:39.064+0800 I NETWORK DBClientCursor::init call() failed 8server should be down... 92015-09-28T15:04:39.066+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed 102015-09-28T15:04:39.067+0800 W NETWORK Failed to connect to 127.0.0.1:27017, reason: errno:111 Connection refused 112015-09-28T15:04:39.067+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) failed failed couldn't connect to server 127.0.0.1:27017 (127.0.0.1), connection attempt failed 122015-09-28T15:04:39.070+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed 132015-09-28T15:04:39.070+0800 W NETWORK Failed to connect to 127.0.0.1:27017, reason: errno:111 Connection refused 142015-09-28T15:04:39.070+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) failed failed couldn't connect to server 127.0.0.1:27017 (127.0.0.1), connection attempt failed 15> 16bye
2）查看集群状态
1$ mongo 2MongoDB shell version: 3.0.6 3connecting to: test 4rs0:PRIMARY> rs.status(); 5{ 6 "set" : "rs0", 7 "date" : ISODate("2015-09-28T07:05:12.140Z"), 8 "myState" : 1, 9 "members" : [ 10 { 11 "_id" : 0, 12 "name" : "172.30.2.201:27017", 13 "health" : 1, 14 "state" : 2, 15 "stateStr" : "SECONDARY", 16 "uptime" : 180, 17 "optime" : Timestamp(1443423600, 1), 18 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 19 "lastHeartbeat" : ISODate("2015-09-28T07:05:11.265Z"), 20 "lastHeartbeatRecv" : ISODate("2015-09-28T07:05:10.951Z"), 21 "pingMs" : 0, 22 "configVersion" : 3 23 }, 24 { 25 "_id" : 1, 26 "name" : "172.30.2.203:27017", 27 "health" : 1, 28 "state" : 1, 29 "stateStr" : "PRIMARY", 30 "uptime" : 353, 31 "optime" : Timestamp(1443423600, 1), 32 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 33 "electionTime" : Timestamp(1443423653, 1), 34 "electionDate" : ISODate("2015-09-28T07:00:53Z"), 35 "configVersion" : 3, 36 "self" : true 37 }, 38 { 39 "_id" : 2, 40 "name" : "172.30.2.202:27017", 41 "health" : 0, 42 "state" : 8, 43 "stateStr" : "(not reachable/healthy)", 44 "uptime" : 0, 45 "optime" : Timestamp(0, 0), 46 "optimeDate" : ISODate("1970-01-01T00:00:00Z"), 47 "lastHeartbeat" : ISODate("2015-09-28T07:05:11.226Z"), 48 "lastHeartbeatRecv" : ISODate("2015-09-28T07:04:37.055Z"), 49 "pingMs" : 0, 50 "lastHeartbeatMessage" : "Failed attempt to connect to 172.30.2.202:27017; couldn't connect to server 172.30.2.202:27017 (172.30.2.202), connection attempt failed", 51 "configVersion" : -1 52 } 53 ], 54 "ok" : 1 55}
可见单个secondary节点故障对集群没有影响
3）再启动secondary
1$ sudo /opt/mongodb/bin/mongod --config /data/mongodb/conf/db0/mongodb.conf 2about to fork child process, waiting until server is ready for connections. 3forked process: 49507 4child process started successfully, parent exiting 5$ mongo 6MongoDB shell version: 3.0.6 7connecting to: test 8rs0:SECONDARY> rs.status(); 9{ 10 "set" : "rs0", 11 "date" : ISODate("2015-09-28T07:06:41.733Z"), 12 "myState" : 2, 13 "members" : [ 14 { 15 "_id" : 0, 16 "name" : "172.30.2.201:27017", 17 "health" : 1, 18 "state" : 2, 19 "stateStr" : "SECONDARY", 20 "uptime" : 12, 21 "optime" : Timestamp(1443423600, 1), 22 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 23 "lastHeartbeat" : ISODate("2015-09-28T07:06:40.999Z"), 24 "lastHeartbeatRecv" : ISODate("2015-09-28T07:06:41.233Z"), 25 "pingMs" : 0, 26 "lastHeartbeatMessage" : "could not find member to sync from", 27 "configVersion" : 3 28 }, 29 { 30 "_id" : 1, 31 "name" : "172.30.2.203:27017", 32 "health" : 1, 33 "state" : 1, 34 "stateStr" : "PRIMARY", 35 "uptime" : 12, 36 "optime" : Timestamp(1443423600, 1), 37 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 38 "lastHeartbeat" : ISODate("2015-09-28T07:06:40.999Z"), 39 "lastHeartbeatRecv" : ISODate("2015-09-28T07:06:41.360Z"), 40 "pingMs" : 0, 41 "electionTime" : Timestamp(1443423653, 1), 42 "electionDate" : ISODate("2015-09-28T07:00:53Z"), 43 "configVersion" : 3 44 }, 45 { 46 "_id" : 2, 47 "name" : "172.30.2.202:27017", 48 "health" : 1, 49 "state" : 2, 50 "stateStr" : "SECONDARY", 51 "uptime" : 13, 52 "optime" : Timestamp(1443423600, 1), 53 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 54 "configVersion" : 3, 55 "self" : true 56 } 57 ], 58 "ok" : 1 59}
重新启动后又重新连上了集群
三、多节点故障
1.primary和secondary节点同时故障
1）停掉一个secondary节点
1$ mongo 2MongoDB shell version: 3.0.6 3connecting to: test 4rs0:SECONDARY> use admin; 5switched to db admin 6rs0:SECONDARY> db.shutdownServer(); 72015-09-28T15:10:43.049+0800 I NETWORK DBClientCursor::init call() failed 8server should be down... 92015-09-28T15:10:43.051+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed 102015-09-28T15:10:43.052+0800 W NETWORK Failed to connect to 127.0.0.1:27017, reason: errno:111 Connection refused 112015-09-28T15:10:43.052+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) failed failed couldn't connect to server 127.0.0.1:27017 (127.0.0.1), connection attempt failed 122015-09-28T15:10:43.055+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed 132015-09-28T15:10:43.055+0800 W NETWORK Failed to connect to 127.0.0.1:27017, reason: errno:111 Connection refused 142015-09-28T15:10:43.055+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) failed failed couldn't connect to server 127.0.0.1:27017 (127.0.0.1), connection attempt failed
2）停掉primary节点
$ mongoMongoDB shell version: 3.0.6connecting to: testrs0:PRIMARY> use admin;switched to db adminrs0:PRIMARY> db.shutdownServer();2015-09-28T15:10:53.069+0800 I NETWORK DBClientCursor::init call() failedserver should be down...2015-09-28T15:10:53.072+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed2015-09-28T15:10:53.073+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) ok2015-09-28T15:10:53.073+0800 I NETWORK DBClientCursor::init call() failed2015-09-28T15:10:53.076+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed2015-09-28T15:10:53.076+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) ok2015-09-28T15:10:53.888+0800 I NETWORK Socket recv() errno:104 Connection reset by peer 127.0.0.1:270172015-09-28T15:10:53.888+0800 I NETWORK SocketException: remote: 127.0.0.1:27017 error: 9001 socket exception [RECV_ERROR] server [127.0.0.1:27017]2015-09-28T15:10:53.888+0800 I NETWORK DBClientCursor::init call() failed
3）查看集群状态
1$ mongo 2MongoDB shell version: 3.0.6 3connecting to: test 4rs0:SECONDARY> rs.status(); 5{ 6 "set" : "rs0", 7 "date" : ISODate("2015-09-28T07:12:10.946Z"), 8 "myState" : 2, 9 "members" : [ 10 { 11 "_id" : 0, 12 "name" : "172.30.2.201:27017", 13 "health" : 1, 14 "state" : 2, 15 "stateStr" : "SECONDARY", 16 "uptime" : 600, 17 "optime" : Timestamp(1443423600, 1), 18 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 19 "configVersion" : 3, 20 "self" : true 21 }, 22 { 23 "_id" : 1, 24 "name" : "172.30.2.203:27017", 25 "health" : 0, 26 "state" : 8, 27 "stateStr" : "(not reachable/healthy)", 28 "uptime" : 0, 29 "optime" : Timestamp(0, 0), 30 "optimeDate" : ISODate("1970-01-01T00:00:00Z"), 31 "lastHeartbeat" : ISODate("2015-09-28T07:12:10.008Z"), 32 "lastHeartbeatRecv" : ISODate("2015-09-28T07:10:51.422Z"), 33 "pingMs" : 0, 34 "lastHeartbeatMessage" : "Failed attempt to connect to 172.30.2.203:27017; couldn't connect to server 172.30.2.203:27017 (172.30.2.203), connection attempt failed", 35 "configVersion" : -1 36 }, 37 { 38 "_id" : 2, 39 "name" : "172.30.2.202:27017", 40 "health" : 0, 41 "state" : 8, 42 "stateStr" : "(not reachable/healthy)", 43 "uptime" : 0, 44 "optime" : Timestamp(0, 0), 45 "optimeDate" : ISODate("1970-01-01T00:00:00Z"), 46 "lastHeartbeat" : ISODate("2015-09-28T07:12:09.477Z"), 47 "lastHeartbeatRecv" : ISODate("2015-09-28T07:10:41.112Z"), 48 "pingMs" : 0, 49 "lastHeartbeatMessage" : "Failed attempt to connect to 172.30.2.202:27017; couldn't connect to server 172.30.2.202:27017 (172.30.2.202), connection attempt failed", 50 "configVersion" : -1 51 } 52 ], 53 "ok" : 1 54}
只剩下一个secondary节点，集群变得不可用了
4）解决方案
重新配置：
1rs0:SECONDARY> cfg={_id:"rs0", members:[ {_id:0,host:"172.30.2.201:27017"}] } 2{ 3 "_id" : "rs0", 4 "members" : [ 5 { 6 "_id" : 0, 7 "host" : "172.30.2.201:27017" 8 } 9 ] 10} 11rs0:SECONDARY> rs.reconfig(cfg, {force:true}); 12{ "ok" : 1 } 13rs0:PRIMARY> rs.status(); 14{ 15 "set" : "rs0", 16 "date" : ISODate("2015-09-28T07:14:09.350Z"), 17 "myState" : 1, 18 "members" : [ 19 { 20 "_id" : 0, 21 "name" : "172.30.2.201:27017", 22 "health" : 1, 23 "state" : 1, 24 "stateStr" : "PRIMARY", 25 "uptime" : 719, 26 "optime" : Timestamp(1443423600, 1), 27 "optimeDate" : ISODate("2015-09-28T07:00:00Z"), 28 "electionTime" : Timestamp(1443424428, 1), 29 "electionDate" : ISODate("2015-09-28T07:13:48Z"), 30 "configVersion" : 71840, 31 "self" : true 32 } 33 ], 34 "ok" : 1 35}
此时就变成了单primary节点，可以提供读写服务，然后再制作secondary节点
2.两个secondary节点故障
1）故障前状态
1rs0:PRIMARY> rs.status(); 2{ 3 "set" : "rs0", 4 "date" : ISODate("2015-09-28T07:16:06.571Z"), 5 "myState" : 1, 6 "members" : [ 7 { 8 "_id" : 0, 9 "name" : "172.30.2.201:27017", 10 "health" : 1, 11 "state" : 1, 12 "stateStr" : "PRIMARY", 13 "uptime" : 836, 14 "optime" : Timestamp(1443424538, 1), 15 "optimeDate" : ISODate("2015-09-28T07:15:38Z"), 16 "electionTime" : Timestamp(1443424534, 1), 17 "electionDate" : ISODate("2015-09-28T07:15:34Z"), 18 "configVersion" : 71842, 19 "self" : true 20 }, 21 { 22 "_id" : 1, 23 "name" : "172.30.2.202:27017", 24 "health" : 1, 25 "state" : 2, 26 "stateStr" : "SECONDARY", 27 "uptime" : 31, 28 "optime" : Timestamp(1443424538, 1), 29 "optimeDate" : ISODate("2015-09-28T07:15:38Z"), 30 "lastHeartbeat" : ISODate("2015-09-28T07:16:06.230Z"), 31 "lastHeartbeatRecv" : ISODate("2015-09-28T07:16:06.089Z"), 32 "pingMs" : 0, 33 "configVersion" : 71842 34 }, 35 { 36 "_id" : 2, 37 "name" : "172.30.2.203:27017", 38 "health" : 1, 39 "state" : 2, 40 "stateStr" : "SECONDARY", 41 "uptime" : 26, 42 "optime" : Timestamp(1443424538, 1), 43 "optimeDate" : ISODate("2015-09-28T07:15:38Z"), 44 "lastHeartbeat" : ISODate("2015-09-28T07:16:06.229Z"), 45 "lastHeartbeatRecv" : ISODate("2015-09-28T07:16:06.233Z"), 46 "pingMs" : 0, 47 "configVersion" : 71842 48 } 49 ], 50 "ok" : 1 51}
2）停掉两个secondary节点
在两个secondary节点分别执行：
1$ mongo 2MongoDB shell version: 3.0.6 3connecting to: test 4rs0:SECONDARY> use admin; 5switched to db admin 6rs0:SECONDARY> db.shutdownServer(); 72015-09-28T15:18:11.114+0800 I NETWORK DBClientCursor::init call() failed 8server should be down... 92015-09-28T15:18:11.117+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed 102015-09-28T15:18:11.118+0800 W NETWORK Failed to connect to 127.0.0.1:27017, reason: errno:111 Connection refused 112015-09-28T15:18:11.118+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) failed failed couldn't connect to server 127.0.0.1:27017 (127.0.0.1), connection attempt failed 122015-09-28T15:18:11.121+0800 I NETWORK trying reconnect to 127.0.0.1:27017 (127.0.0.1) failed 132015-09-28T15:18:11.121+0800 W NETWORK Failed to connect to 127.0.0.1:27017, reason: errno:111 Connection refused 142015-09-28T15:18:11.121+0800 I NETWORK reconnect 127.0.0.1:27017 (127.0.0.1) failed failed couldn't connect to server 127.0.0.1:27017 (127.0.0.1), connection attempt failed 15> 16bye
3）查看集群状态
1rs0:SECONDARY> rs.status(); 2{ 3 "set" : "rs0", 4 "date" : ISODate("2015-09-28T07:19:09.196Z"), 5 "myState" : 2, 6 "members" : [ 7 { 8 "_id" : 0, 9 "name" : "172.30.2.201:27017", 10 "health" : 1, 11 "state" : 2, 12 "stateStr" : "SECONDARY", 13 "uptime" : 1019, 14 "optime" : Timestamp(1443424538, 1), 15 "optimeDate" : ISODate("2015-09-28T07:15:38Z"), 16 "configVersion" : 71842, 17 "self" : true 18 }, 19 { 20 "_id" : 1, 21 "name" : "172.30.2.202:27017", 22 "health" : 0, 23 "state" : 8, 24 "stateStr" : "(not reachable/healthy)", 25 "uptime" : 0, 26 "optime" : Timestamp(0, 0), 27 "optimeDate" : ISODate("1970-01-01T00:00:00Z"), 28 "lastHeartbeat" : ISODate("2015-09-28T07:19:08.371Z"), 29 "lastHeartbeatRecv" : ISODate("2015-09-28T07:18:10.147Z"), 30 "pingMs" : 0, 31 "lastHeartbeatMessage" : "Failed attempt to connect to 172.30.2.202:27017; couldn't connect to server 172.30.2.202:27017 (172.30.2.202), connection attempt failed", 32 "configVersion" : -1 33 }, 34 { 35 "_id" : 2, 36 "name" : "172.30.2.203:27017", 37 "health" : 0, 38 "state" : 8, 39 "stateStr" : "(not reachable/healthy)", 40 "uptime" : 0, 41 "optime" : Timestamp(0, 0), 42 "optimeDate" : ISODate("1970-01-01T00:00:00Z"), 43 "lastHeartbeat" : ISODate("2015-09-28T07:19:08.350Z"), 44 "lastHeartbeatRecv" : ISODate("2015-09-28T07:18:34.298Z"), 45 "pingMs" : 0, 46 "lastHeartbeatMessage" : "Failed attempt to connect to 172.30.2.203:27017; couldn't connect to server 172.30.2.203:27017 (172.30.2.203), connection attempt failed", 47 "configVersion" : -1 48 } 49 ], 50 "ok" : 1 51}
可见剩下的primary节点自动变为了secondary节点，集群变得不可用了
4）解决方案
1rs0:SECONDARY> cfg={_id:"rs0", members:[ {_id:0,host:"172.30.2.201:27017"}] } 2{ 3 "_id" : "rs0", 4 "members" : [ 5 { 6 "_id" : 0, 7 "host" : "172.30.2.201:27017" 8 } 9 ] 10} 11rs0:SECONDARY> rs.reconfig(cfg, {force:true}); 12{ "ok" : 1 } 13rs0:PRIMARY> rs.status(); 14{ 15 "set" : "rs0", 16 "date" : ISODate("2015-09-28T07:20:08.099Z"), 17 "myState" : 1, 18 "members" : [ 19 { 20 "_id" : 0, 21 "name" : "172.30.2.201:27017", 22 "health" : 1, 23 "state" : 1, 24 "stateStr" : "PRIMARY", 25 "uptime" : 1078, 26 "optime" : Timestamp(1443424538, 1), 27 "optimeDate" : ISODate("2015-09-28T07:15:38Z"), 28 "electionTime" : Timestamp(1443424795, 1), 29 "electionDate" : ISODate("2015-09-28T07:19:55Z"), 30 "configVersion" : 127342, 31 "self" : true 32 } 33 ], 34 "ok" : 1 35}
处理方法和上面的相同，也是强制将剩下的secondary节点配置为单primary节点