Hi, I want to know that when i have 1st redis cluster run as 3 master and 3 slave, they are complete redis cluster. Later, i run 2nd redis cluster that also run as 3 master and 3 slave. This time , i have 2 complete redis cluster, 1st and 2nd redis clsuter have their 3 number master alone, total is 6 master. But i want to the 2nd join into the 1st redis cluster , and the master keep the 1st redis cluster master, the 2nd all redis become slave. How can i process this ? Only can i reset the 2nd all redis and join into 1st redis ? Do you have good way for this case?
thanks.
Comment From: Jarvif
Actually , i donot want to reset my redis cluster by hand, because it is dangerours. Please tell me the way if you have good way for this case.
Comment From: madolson
What exactly are you trying to accomplish? Reseting and moving is definitely the recommended way to do this, but is the second cluster serving live traffic? Can you provision new hardware to augment the first cluster, move all the traffic, and then simply retire the second cluster?
Comment From: Jarvif
已收到，谢谢。
Comment From: Jarvif
yeah, my micro service will connect to the 1st redis cluster, and the same micro service also will connect to the 2nd redis cluster. At the first , 1st and 2nd micro service will connect to their redis cluster(1st redis clusrer and 2nd redis cluster). But when the micor service run keepalive , the 2nd micro service will not work. And the 1st micro service and 1st redis cluster should be the leader .The 2nd redis cluster will run as the 1st redis cluster as slave. It also can mean , i want to the 'cluster meet' can join the redis cluster as slave. Can you tell me how can i do it ? Now , the 2nd redis may become the leader, can the 1st redis will be the slave , and the 1st redis cluster will be empty(it not my expect!)
Comment From: madolson
There isn't any way to do what you are asking for today. You can't merge two functioning clusters. You would either need to take cluster 2 out of service and redirect all of the traffic to cluster 1, then do the resets and merge them as new nodes, or allocate new resources into cluster 1, and then decommission cluster 2.
Comment From: Jarvif
ok, thanks.