My redis sentinel is not working properly. As when my master (15...143) went down, one of the slaves elected as master but it is showing as failover-abort-no-good-slave error, why it is showing as Nogoodslave to promote from slaves to master. This results in selecting the old master (15...143) itself. Can anyone help?
+new-epoch 44 +try-failover master mymaster 15...143 6379 +vote-for-leader 466614318b614075783e046bafb66d9ae59adcde 44 <---------slave1 myid 9612c233dbbe3b43f919a98d563ce90154d7c503 voted for 466614318b614075783e046bafb66d9ae59adcde 44 +elected-leader master mymaster 15...143 6379 +failover-state-select-slave master mymaster 15...143 6379 -failover-abort-no-good-slave master mymaster 15...143 6379 Next failover delay: I will not start a failover before Thu Jan 25 12:48:50 2024
13...213:26379> SENTINEL get-master-addr-by-name mymaster
- "15...143" <-----old master IP is showing
- "6379"
Configuration details
=======================
redis1 15...143 8ce8bc1209aab6ba4b2e0479611b82e13ddead0e 172...106 < ----------------master redis2 13...54 466614318b614075783e046bafb66d9ae59adcde 172...187 <-----------------slave1 redis3 13...213 9612c233dbbe3b43f919a98d563ce90154d7c503 172...118 <-----------------slave2
Sentinel configuration in master
=============================== bind 0.0.0.0 sentinel monitor mymaster 15...143 6379 2 sentinel down-after-milliseconds mymaster 5000 sentinel failover-timeout mymaster 60000 sentinel auth-pass mymaster masterpassword protected-mode no
sentinel known-replica mymaster 13...213 6379 sentinel known-replica mymaster 13...54 6379 sentinel known-sentinel mymaster 172...118 26379 9612c233dbbe3b43f919a98d563ce90154d7c503 sentinel known-sentinel mymaster 172...187 26379 466614318b614075783e046bafb66d9ae59adcde
Sentinel configuration in slave1
================================
bind 0.0.0.0 sentinel monitor mymaster 15...143 6379 2 sentinel down-after-milliseconds mymaster 5000 sentinel failover-timeout mymaster 60000 sentinel auth-pass mymaster masterpassword sentinel known-replica mymaster 13...213 6379 sentinel known-replica mymaster 13...54 6379 sentinel known-sentinel mymaster 172...118 26379 9612c233dbbe3b43f919a98d563ce90154d7c503 sentinel known-sentinel mymaster 172...106 26379 8ce8bc1209aab6ba4b2e0479611b82e13ddead0e
Sentinel configuration in slave2
================================ bind 0.0.0.0 sentinel monitor mymaster 15...143 6379 2 sentinel down-after-milliseconds mymaster 5000 sentinel failover-timeout mymaster 60000 sentinel auth-pass mymaster masterpassword
sentinel known-replica mymaster 13...213 6379 sentinel known-replica mymaster 13...54 6379 sentinel known-sentinel mymaster 172..187 26379 466614318b614075783e046bafb66d9ae59adcde sentinel known-sentinel mymaster 172...106 26379 8ce8bc1209aab6ba4b2e0479611b82e13ddead0e
Now i am getting the error like in between the slaves, one is voted as master but again when i am checking the get-mymaster command, it is show the old down master, not the new one. and will show as nogoodslave as error
===================
In slave2 server
13...213:26379> SENTINEL get-master-addr-by-name mymaster
- "15...143" <-----old master IP is showing
- "6379"
In sentinal logs
+new-epoch 42 +vote-for-leader 466614318b614075783e046bafb66d9ae59adcde 42 <-------------slave1 myid Next failover delay: I will not start a failover before Thu Jan 25 12:44:49 2024
--------------------------------------------------** In slave1 server
redis-cli -h 13...54 -p 26379 13...54:26379> SENTINEL get-master-addr-by-name mymaster
- "15...143" <------------------old master IP
- "6379"
In sentinal logs
30299:X 25 Jan 2024 12:46:49.847 # +new-epoch 44 30299:X 25 Jan 2024 12:46:49.847 # +try-failover master mymaster 15.206.167.143 6379 30299:X 25 Jan 2024 12:46:49.851 # +vote-for-leader 466614318b614075783e046bafb66d9ae59adcde 44 <----------------------------slave1 myid 30299:X 25 Jan 2024 12:46:49.861 # 9612c233dbbe3b43f919a98d563ce90154d7c503 voted for 466614318b614075783e046bafb66d9ae59adcde 44 30299:X 25 Jan 2024 12:46:49.934 # +elected-leader master mymaster 15.206.167.143 6379 30299:X 25 Jan 2024 12:46:49.934 # +failover-state-select-slave master mymaster 15.206.167.143 6379 30299:X 25 Jan 2024 12:46:50.035 # -failover-abort-no-good-slave master mymaster 15.206.167.143 6379 30299:X 25 Jan 2024 12:46:50.106 # Next failover delay: I will not start a failover before Thu Jan 25 12:48:50 2024
===================
I need other slaves to get promoted as master when master is failed. Please help