Managing State in Containers

Managing the state of an application is one of the key aspects when architecting any software solution, regardless of whether it is a monolith desktop application or a complex, distributed system hosted in a cloud environment. Even if most of your services in the system are stateless, some part of your system will be stateful, for example, a cloud-hosted NoSQL database or a dedicated service you have implemented yourself. And if you are aiming at good scalability for your design, you have to ensure that the storage for your stateful services scales appropriately. In these terms, services or applications hosted in Docker containers are no different – you need to manage the state, especially if you want the data to be persisted on container restarts or failures.

In this chapter, we will provide you with a better understanding of how state can be persisted in Docker containers running on Windows and how these concepts relate to data persistence in Kubernetes applications. You will learn about the concepts of volumes and bind mounts and how they can be used in order to share state between containers and the container host.

This chapter covers the following topics:

• Mounting local volumes for stateful applications
• Using remote/cloud storage for container storage
• Running clustered solutions inside containers