database - AWS Elasticache Redis failover -
i using redis on elasticache node application , today node went down means our app stopped working. took 20 minutes new node provisioned.
from reading documentation seems can set cluster automatically promotes slave primary in case of failure. big gotcha seems have set client write primary node , read slave nodes.
this means in case of failure, have reconfigure app point newly created 'read' nodes. takes few minutes slave promoted primary.
is there no way set if primary fails, slave automatically take on read/write operations?
i'm not storing data in redis , low read/write operations, required run app (live video sessions!).
if can't have seamless failover in redis, there can use provides functionality? i'm hoping don't have move traditional dbms works need able handle failure well.
thanks
multi az's should automatically switch on minimal downtime. once have created 1 of these instances, endpoint cluster. amazon point dns entry proper failover node, , handle promotion of slave, if master instances dies.
Comments
Post a Comment