sentinel.conf配置

阅读目录

• 开始
• 1.常用的配置
• 2.官方完整的配置

回到顶部

1.常用的配置

1. port 26379
2. # sentinel announce-ip <ip>
3. # sentinel announce-port <port>
4. dir /tmp
5.  
6. ################################# master001 #################################
7. sentinel monitor master001 192.168.110.101 6379 2
8. # sentinel auth-pass <master-name> <password>
9. sentinel down-after-milliseconds master001 30000
10. sentinel parallel-syncs master001 1
11. sentinel failover-timeout master001 180000
12. # sentinel notification-script <master-name> <script-path>
13. # sentinel client-reconfig-script <master-name> <script-path>
14.  
15. # 可以配置多个master节点
16. ################################# master002 #################################

配置文件说明：

1. port :当前Sentinel服务运行的端口

 

2. dir : Sentinel服务运行时使用的临时文件夹

 

3.sentinel monitor master001 192.168.110.101 6379 2:Sentinel去监视一个名为master001的主redis实例，这个主实例的IP地址为本机地址192.168.110.101，端口号为6379，而将这个主实例判断为失效至少需要2个 Sentinel进程的同意，只要同意Sentinel的数量不达标，自动failover就不会执行

 

4.sentinel down-after-milliseconds master001 30000:指定了Sentinel认为Redis实例已经失效所需的毫秒数。当实例超过该时间没有返回PING，或者直接返回错误，那么Sentinel将这个实例标记为主观下线。只有一个 Sentinel进程将实例标记为主观下线并不一定会引起实例的自动故障迁移：只有在足够数量的Sentinel都将一个实例标记为主观下线之后，实例才会被标记为客观下线，这时自动故障迁移才会执行

 

5.sentinel parallel-syncs master001 1：指定了在执行故障转移时，最多可以有多少个从Redis实例在同步新的主实例，在从Redis实例较多的情况下这个数字越小，同步的时间越长，完成故障转移所需的时间就越长

 

6.sentinel failover-timeout master001 180000：如果在该时间（ms）内未能完成failover操作，则认为该failover失败

 

7.sentinel notification-script <master-name> <script-path>：指定sentinel检测到该监控的redis实例指向的实例异常时，调用的报警脚本。该配置项可选，但是很常用

回到顶部

2.官方完整的配置

1. # Example sentinel.conf
3. # port <sentinel-port>
4. # The port that this sentinel instance will run on
5. # sentinel实例运行的端口
6. port 26379
8. # sentinel announce-ip <ip>
9. # sentinel announce-port <port>
10. #
11. # The above two configuration directives are useful in environments where,
12. # because of NAT, Sentinel is reachable from outside via a non-local address.
13. #
14. # When announce-ip is provided, the Sentinel will claim the specified IP address
15. # in HELLO messages used to gossip its presence, instead of auto-detecting the
16. # local address as it usually does.
17. #
18. # Similarly when announce-port is provided and is valid and non-zero, Sentinel
19. # will announce the specified TCP port.
20. #
21. # The two options don't need to be used together, if only announce-ip is
22. # provided, the Sentinel will announce the specified IP and the server port
23. # as specified by the "port" option. If only announce-port is provided, the
24. # Sentinel will announce the auto-detected local IP and the specified port.
25. #
26. # Example:
27. #
28. # sentinel announce-ip 1.2.3.4
30. # dir <working-directory>
31. # Every long running process should have a well-defined working directory.
32. # For Redis Sentinel to chdir to /tmp at startup is the simplest thing
33. # for the process to don't interferer with administrative tasks such as
34. # unmounting filesystems.
35. dir /tmp
37. # sentinel monitor <master-name> <ip> <redis-port> <quorum>
38. # master-name : master Redis Server名称
39. # ip : master Redis Server的IP地址
40. # redis-port : master Redis Server的端口号
41. # quorum : 主实例判断为失效至少需要 quorum 个 Sentinel 进程的同意，只要同意 Sentinel 的数量不达标，自动failover就不会执行
42. #
43. # Tells Sentinel to monitor this master, and to consider it in O_DOWN
44. # (Objectively Down) state only if at least <quorum> sentinels agree.
45. #
46. # Note that whatever is the ODOWN quorum, a Sentinel will require to
47. # be elected by the majority of the known Sentinels in order to
48. # start a failover, so no failover can be performed in minority.
49. #
50. # Slaves are auto-discovered, so you don't need to specify slaves in
51. # any way. Sentinel itself will rewrite this configuration file adding
52. # the slaves using additional configuration options.
53. # Also note that the configuration file is rewritten when a
54. # slave is promoted to master.
55. #
56. # Note: master name should not include special characters or spaces.
57. # The valid charset is A-z 0-9 and the three characters ".-_".
58. #
59. sentinel monitor mymaster 127.0.0.1 6379 2
61. # sentinel auth-pass <master-name> <password>
62. #
63. # Set the password to use to authenticate with the master and slaves.
64. # Useful if there is a password set in the Redis instances to monitor.
65. #
66. # Note that the master password is also used for slaves, so it is not
67. # possible to set a different password in masters and slaves instances
68. # if you want to be able to monitor these instances with Sentinel.
69. #
70. # However you can have Redis instances without the authentication enabled
71. # mixed with Redis instances requiring the authentication (as long as the
72. # password set is the same for all the instances requiring the password) as
73. # the AUTH command will have no effect in Redis instances with authentication
74. # switched off.
75. #
76. # Example:
77. #
78. # sentinel auth-pass mymaster MySUPER--secret-0123passw0rd
80. # sentinel down-after-milliseconds <master-name> <milliseconds>
81. #
82. # Number of milliseconds the master (or any attached slave or sentinel) should
83. # be unreachable (as in, not acceptable reply to PING, continuously, for the
84. # specified period) in order to consider it in S_DOWN state (Subjectively
85. # Down).
86. # 选项指定了 Sentinel 认为Redis实例已经失效所需的毫秒数。当实例超过该时间没有返回PING，或者直接返回错误， 那么 Sentinel 将这个实例标记为主观下线（subjectively down，简称 SDOWN ）
87. #
88. # Default is 30 seconds.
89. sentinel down-after-milliseconds mymaster 30000
91. # sentinel parallel-syncs <master-name> <numslaves>
92. #
93. # How many slaves we can reconfigure to point to the new slave simultaneously
94. # during the failover. Use a low number if you use the slaves to serve query
95. # to avoid that all the slaves will be unreachable at about the same
96. # time while performing the synchronization with the master.
97. # 选项指定了在执行故障转移时， 最多可以有多少个从Redis实例在同步新的主实例， 在从Redis实例较多的情况下这个数字越小，同步的时间越长，完成故障转移所需的时间就越长。
98. sentinel parallel-syncs mymaster 1
100. # sentinel failover-timeout <master-name> <milliseconds>
101. #
102. # Specifies the failover timeout in milliseconds. It is used in many ways:
103. #
104. # - The time needed to re-start a failover after a previous failover was
105. # already tried against the same master by a given Sentinel, is two
106. # times the failover timeout.
107. #
108. # - The time needed for a slave replicating to a wrong master according
109. # to a Sentinel current configuration, to be forced to replicate
110. # with the right master, is exactly the failover timeout (counting since
111. # the moment a Sentinel detected the misconfiguration).
112. #
113. # - The time needed to cancel a failover that is already in progress but
114. # did not produced any configuration change (SLAVEOF NO ONE yet not
115. # acknowledged by the promoted slave).
116. #
117. # - The maximum time a failover in progress waits for all the slaves to be
118. # reconfigured as slaves of the new master. However even after this time
119. # the slaves will be reconfigured by the Sentinels anyway, but not with
120. # the exact parallel-syncs progression as specified.
121. # 如果在该时间（ms）内未能完成failover操作，则认为该failover失败
122. #
123. # Default is 3 minutes.
124. sentinel failover-timeout mymaster 180000
126. # SCRIPTS EXECUTION
127. #
128. # sentinel notification-script and sentinel reconfig-script are used in order
129. # to configure scripts that are called to notify the system administrator
130. # or to reconfigure clients after a failover. The scripts are executed
131. # with the following rules for error handling:
132. #
133. # If script exits with "1" the execution is retried later (up to a maximum
134. # number of times currently set to 10).
135. #
136. # If script exits with "2" (or an higher value) the script execution is
137. # not retried.
138. #
139. # If script terminates because it receives a signal the behavior is the same
140. # as exit code 1.
141. #
142. # A script has a maximum running time of 60 seconds. After this limit is
143. # reached the script is terminated with a SIGKILL and the execution retried.
145. # NOTIFICATION SCRIPT
146. #
147. # sentinel notification-script <master-name> <script-path>
148. #
149. # Call the specified notification script for any sentinel event that is
150. # generated in the WARNING level (for instance -sdown, -odown, and so forth).
151. # This script should notify the system administrator via email, SMS, or any
152. # other messaging system, that there is something wrong with the monitored
153. # Redis systems.
154. #
155. # The script is called with just two arguments: the first is the event type
156. # and the second the event description.
157. #
158. # The script must exist and be executable in order for sentinel to start if
159. # this option is provided.
160. # 指定sentinel检测到该监控的redis实例指向的实例异常时，调用的报警脚本。该配置项可选，但是很常用。
161. #
162. # Example:
163. #
164. # sentinel notification-script mymaster /var/redis/notify.sh
166. # CLIENTS RECONFIGURATION SCRIPT
167. #
168. # sentinel client-reconfig-script <master-name> <script-path>
169. #
170. # When the master changed because of a failover a script can be called in
171. # order to perform application-specific tasks to notify the clients that the
172. # configuration has changed and the master is at a different address.
173. #
174. # The following arguments are passed to the script:
175. #
176. # <master-name> <role> <state> <from-ip> <from-port> <to-ip> <to-port>
177. #
178. # <state> is currently always "failover"
179. # <role> is either "leader" or "observer"
180. #
181. # The arguments from-ip, from-port, to-ip, to-port are used to communicate
182. # the old address of the master and the new address of the elected slave
183. # (now a master).
184. #
185. # This script should be resistant to multiple invocations.
186. #
187. # Example:
188. #
189. # sentinel client-reconfig-script mymaster /var/redis/reconfig.sh