

This eliminates the need to setup builders with different OS dependent libraries and utilities.
#Continous delvery workflowusing docker and kubernetes software
Minimal Builder Setup – Containers are used to build the software and build artifacts are delivered as container images.By monitoring the utilization, we are able to leverage the most out of our test infrastructure. This enables us to get the optimal utilization of the underlying nodes in the Kubernetes cluster. Efficient Utilization of Resources – Kubernetes provides built-in scheduling of pods (containers).

We use TeamCity as our build management server. And yes you read that correctly, we dogfood our own product by running our services on top of the very product we deliver to our customers.

The lifecycle (deployment, monitoring, upgrade) of these clusters is performed by our Platform9 Managed Kubernetes product. To do this, an automated CI/CD process is critical.Īt our infrastructure level, we have a collection of Kubernetes clusters on which we deploy some of our key services. Considering our customers run their production workloads on our products, it is imperative that we cannot compromise on the quality and reliability of our products – while maintaining the agility of our software development process. This post provides a recipe on how we achieve CI/CD with Kubernetes.Īt Platform9, we build scalable, distributed SaaS solutions, which means releasing often and with very high confidence in the product quality. Kubernetes, along with container technologies, has many built-in constructs and capabilities that simplify building this pipeline. A key aspect of achieving this is having a Continuous Integration/Continuous Deployment (CI/CD) pipeline in place to validate and deploy these services. Most of the today’s IT dependent businesses are adopting the DevOps friendly practice of building agile and highly reliable services.
