Role of Helm repository in CI/CD pipeline using Jenkins and Kubernetes

Question

I am trying to learn the implementation of CI/CD pipeline using Jenkins and Kubernetes for my Spring Boot microservice deployment.

I following the following links for references,

1. https://dzone.com/articles/easily-automate-your-cicd-pipeline-with-jenkins-he
2. https://medium.com/jfrogplatform/easily-automate-your-ci-cd-pipeline-with-jenkins-helm-and-kubernetes-c96283c25701

Here I am finding the way that to use Kubernetes Helm chart for simplifying application deployment. In the aboev link it showing that, to use one Helm repository along with Docker registry (I am planning to use Dockerhub.com).

Confusion

Here my confusion is that, If we are using Helm chart with Kubernetes, then why to use Helm repository for my CI/CD pipeline?

Answer 1

You could work with your helm charts unpackaged, effectively deploying them from source. You don't necessarily have to package them and host them in a repo. You probably will want to package the charts and host them in a repo if you have multiple apps/teams consuming the charts and perhaps building their own charts using them and/or you want to be able to have different versions of a chart running in different places.

Answer 2

You will need to use Helm repository to store your Helm charts.

More details you can find in The Chart Repository Guide document.

Answer 3

You can use Helm to create generic charts that can be used to deployed hundreds of microservices. This allows simplifying the complexity of Kubernetes deployments by using a chart to define all the required objects (deployment, service, HPA, ingress, config map etc) and allows simple customization. You can read more how this is working for us in this blog post (full disclosure: I wrote this post).

Anyway, this does not sound to me like your use case - if you have a small number of APIs using Helm has some overhead, and not worth it for a small number of APIs.