Kubernetes has become the de-facto platform to deploy and manage stateless applications. It is also making strides into deploying and managing stateful applications. Managing stateful applications is more than provisioning and relocating containers on failures. Stateful application deal with data gravity, IP persistence and almost always mandate procedures to protect data. Hadoop is one of the biggest applications and required careful resource planning. The architecture of Hadoop makes assumptions on the compute and storage layers such as local disks, having separate racks for fault tolerance. In addition to this Hadoop is a platform and not a single application. Every component in the stack are designed differently. MySQL is a SQL database, datanodes are replicated, name nodes are active, passive, zookeeper is a consensus system with latency requirements. Designing a common infrastructure that can accommodate all these asks is a big challenge. With this talk, Ravikumar Alluboyina, will shed light on some of the challenges and proposes solutions to address them.