How we automated the handling of MySQL primary failovers in DBaaS
[In part I of this blog series we saw an overview of our DBaaS design. In this blog (part II) we discuss the automated MySQL primary failover design in some detail.] We use the hashicorp consul and a custom consul watcher service, along with the orchestrator service, to inject proxySQL route changes in the event of a MySQL primary failure in a shard. Consul cluster