Case 1 : Master and Slave are not in correct position.
(Assume 7001 is the master but it is in slave now. )
#redis-cli -a Password -c -p 7001 cluster failover
Case 2 : Slave is promoted to Master while Master is still active.
In this case, we have to remove the wrong node from cluster and add it back as slave
(Asssume the wrong port/service is redis_7005
#redic-cli -a Password --cluster del-node master-node-ip:port id-of-node-to-be-del
#systemctl stop redis_7005
#rm -rf /var/lib/redis/7005/*
#system start redis_7005
#redis-cli -a Password --cluster add-node master-node-ip:port ip-of-slave-node:port --cluster-slave
Then you can use the following command to check it
#redis-cli -a Password -c -p 7001 cluster nodes
沒有留言:
發佈留言