i have a mysql statefulset, database on Kubernetes cluster.
mysql-clusterip----> mysql-0(master)<------(mysql-headless-svc)
mysql-clusterip-----> mysql-1(read replica)
mysql-clusterip-----> mysql-2(read replica)
so i have 2 endpoint mysql-0.mysql-headless-svc for writting to the master only, and mysq-clusterip for read request to the all the sts pods.
now how do i configure the backend?do i initiate 2 database connection depending on the request comming?i guess this is not efficient process.
i was thinking about proxysql but i can not find any good documentation which explains me this thing nicely.what is the production grade strategy?
