Imagine you are a Site Reliability Engineer (SRE) that is tasked with examining if an application deployment in your cluster is enabled for high availability. Where would you start? What are the main aspects you should analyze?
In this article, we argue that most transient failures can be mitigated by leveraging a high availability (HA) solution that encompasses four main pillars.
We consider a simple multi-tiered application, and examine what it would take to enable it for high availability.