MongoDB修复config配置节点

生产环境的一套MongoDB发生故障，其中一个config节点的磁盘损坏。
修复流程如下：

1. # 登录到config库，查看config的节点状态
   
2. # 可以看到_id为2的节点状态是不可访问的
   
3. [root@mongos-mq ~]# /opt/mongodb-linux-x86_64-3.2.10/bin/mongo --port 10000
   
4. MongoDB shell version: 3.2.10
   
5. connecting to: 127.0.0.1:10000/test
   
6. config_rs:PRIMARY> use admin
   
7. switched to db admin
   
8. config_rs:PRIMARY> db.auth("root", "xxx")
   
9. 1
   
10. config_rs:PRIMARY> rs.status()
    
11. {
    
12.     "set" : "config_rs",
    
13.     "date" : ISODate("2018-05-30T04:53:21.081Z"),
    
14.     "myState" : 1,
    
15.     "term" : NumberLong(1),
    
16.     "configsvr" : true,
    
17.     "heartbeatIntervalMillis" : NumberLong(2000),
    
18.     "members" : [
    
19.         {
    
20.             "_id" : 0,
    
21.             "name" : "192.168.210.208:10000",
    
22.             "health" : 1,
    
23.             "state" : 1,
    
24.             "stateStr" : "PRIMARY",
    
25.             "uptime" : 1112804,
    
26.             "optime" : {
    
27.                 "ts" : Timestamp(1527656000, 1),
    
28.                 "t" : NumberLong(1)
    
29.             },
    
30.             "optimeDate" : ISODate("2018-05-30T04:53:20Z"),
    
31.             "electionTime" : Timestamp(1526543249, 2),
    
32.             "electionDate" : ISODate("2018-05-17T07:47:29Z"),
    
33.             "configVersion" : 3,
    
34.             "self" : true
    
35.         },
    
36.         {
    
37.             "_id" : 1,
    
38.             "name" : "192.168.210.209:10000",
    
39.             "health" : 1,
    
40.             "state" : 2,
    
41.             "stateStr" : "SECONDARY",
    
42.             "uptime" : 1111358,
    
43.             "optime" : {
    
44.                 "ts" : Timestamp(1527656000, 1),
    
45.                 "t" : NumberLong(1)
    
46.             },
    
47.             "optimeDate" : ISODate("2018-05-30T04:53:20Z"),
    
48.             "lastHeartbeat" : ISODate("2018-05-30T04:53:20.836Z"),
    
49.             "lastHeartbeatRecv" : ISODate("2018-05-30T04:53:19.329Z"),
    
50.             "pingMs" : NumberLong(0),
    
51.             "syncingTo" : "192.168.210.208:10000",
    
52.             "configVersion" : 3
    
53.         },
    
54.         {
    
55.             "_id" : 2,
    
56.             "name" : "192.168.210.207:10000",
    
57.             "health" : 0,
    
58.             "state" : 8,
    
59.             "stateStr" : "(not reachable/healthy)",
    
60.             "uptime" : 0,
    
61.             "optime" : {
    
62.                 "ts" : Timestamp(0, 0),
    
63.                 "t" : NumberLong(-1)
    
64.             },
    
65.             "optimeDate" : ISODate("1970-01-01T00:00:00Z"),
    
66.             "lastHeartbeat" : ISODate("2018-05-30T04:53:20.850Z"),
    
67.             "lastHeartbeatRecv" : ISODate("2018-05-26T09:07:27.318Z"),
    
68.             "pingMs" : NumberLong(0),
    
69.             "lastHeartbeatMessage" : "Connection refused",
    
70.             "configVersion" : -1
    
71.         }
    
72.     ],
    
73.     "ok" : 1
    
74. }
    
75. 
    
76. # 在故障节点上面重新安装MongoDB，创建数据文件目录
    
77. [root@tomcat-207 opt]# tar xfz mongodb-linux-x86_64-3.2.10.tgz
    
78. [root@tomcat-207 opt]# mkdir -p /data/mongodb_data/config_rs1_10000
    
79. 
    
80. # 配置参数文件
    
81. [root@tomcat-207 opt]# cat /etc/mongodb/config_rs1_10000.conf
    
82. systemLog:
    
83.     destination: file
    
84.     path: /data/mongodb_data/config_rs1_10000/mongod.log
    
85. processManagement:
    
86.     fork: true
    
87. net:
    
88.     bindIp: 0.0.0.0
    
89.     port: 10000
    
90. security:
    
91.     keyFile: /data/mongodb_data/keyfile
    
92. storage:
    
93.     dbPath: /data/mongodb_data/config_rs1_10000
    
94.     directoryPerDB: true
    
95. operationProfiling:
    
96.     slowOpThresholdMs: 100
    
97.     mode: slowOp
    
98. replication:
    
99.     replSetName: config_rs
    
100. sharding:
     
101.     clusterRole: configsvr
     
102. 
     
103. # 从其他config节点拷贝过来keyfile文件
     
104. [root@mongos-mq opt]# scp /data/mongodb_data/keyfile root@192.168.210.207:/data/mongodb_data/
     
105. 
     
106. # 启动Mongo
     
107. [root@mongos-mq opt]# /opt/mongodb-linux-x86_64-3.2.10/bin/mongod -f /etc/mongodb/config_rs1_10000.conf
     
108. about to fork child process, waiting until server is ready for connections.
     
109. forked process: 3744
     
110. child process started successfully, parent exiting
     
111. 
     
112. # 查看config节点状态
     
113. # 所有的config节点都恢复正常
     
114. config_rs:PRIMARY> rs.status()
     
115. {
     
116.     "set" : "config_rs",
     
117.     "date" : ISODate("2018-05-30T05:29:40.103Z"),
     
118.     "myState" : 1,
     
119.     "term" : NumberLong(1),
     
120.     "configsvr" : true,
     
121.     "heartbeatIntervalMillis" : NumberLong(2000),
     
122.     "members" : [
     
123.         {
     
124.             "_id" : 0,
     
125.             "name" : "192.168.210.208:10000",
     
126.             "health" : 1,
     
127.             "state" : 1,
     
128.             "stateStr" : "PRIMARY",
     
129.             "uptime" : 1114983,
     
130.             "optime" : {
     
131.                 "ts" : Timestamp(1527658178, 4),
     
132.                 "t" : NumberLong(1)
     
133.             },
     
134.             "optimeDate" : ISODate("2018-05-30T05:29:38Z"),
     
135.             "electionTime" : Timestamp(1526543249, 2),
     
136.             "electionDate" : ISODate("2018-05-17T07:47:29Z"),
     
137.             "configVersion" : 3,
     
138.             "self" : true
     
139.         },
     
140.         {
     
141.             "_id" : 1,
     
142.             "name" : "192.168.210.209:10000",
     
143.             "health" : 1,
     
144.             "state" : 2,
     
145.             "stateStr" : "SECONDARY",
     
146.             "uptime" : 1113537,
     
147.             "optime" : {
     
148.                 "ts" : Timestamp(1527658178, 4),
     
149.                 "t" : NumberLong(1)
     
150.             },
     
151.             "optimeDate" : ISODate("2018-05-30T05:29:38Z"),
     
152.             "lastHeartbeat" : ISODate("2018-05-30T05:29:39.535Z"),
     
153.             "lastHeartbeatRecv" : ISODate("2018-05-30T05:29:39.941Z"),
     
154.             "pingMs" : NumberLong(0),
     
155.             "syncingTo" : "192.168.210.208:10000",
     
156.             "configVersion" : 3
     
157.         },
     
158.         {
     
159.             "_id" : 2,
     
160.             "name" : "192.168.210.207:10000",
     
161.             "health" : 1,
     
162.             "state" : 2,
     
163.             "stateStr" : "SECONDARY",
     
164.             "uptime" : 1588,
     
165.             "optime" : {
     
166.                 "ts" : Timestamp(1527658178, 4),
     
167.                 "t" : NumberLong(1)
     
168.             },
     
169.             "optimeDate" : ISODate("2018-05-30T05:29:38Z"),
     
170.             "lastHeartbeat" : ISODate("2018-05-30T05:29:39.535Z"),
     
171.             "lastHeartbeatRecv" : ISODate("2018-05-30T05:29:39.691Z"),
     
172.             "pingMs" : NumberLong(0),
     
173.             "syncingTo" : "192.168.210.209:10000",
     
174.             "configVersion" : 3
     
175.         }
     
176.     ],
     
177.     "ok" : 1
     
178. }

来自 “ ITPUB博客 ” ，链接：http://blog.itpub.net/26506993/viewspace-2155380/，如需转载，请注明出处，否则将追究法律责任。

转载于:http://blog.itpub.net/26506993/viewspace-2155380/