Deployment using light weight git server
Image credit : https://www.plesk.com/blog/business-industry/devops-cycle-2-automated-deployment-operations-monitoring/

Deployment using light weight git server

In the world of deployment of platforms, this one is a bit different

In this post, I will be talking about my experience with deployment of an execution platform geared towards running machine learning models.

In my current role, my responsibility is to examine multiple scenarios possible for deployment of the platform which will be running the machine learning models. One of the scenarios which is frequently discussed is of that when the customer does not have a VCS repository in their premises. How do I bootstrap my platform onto his Kubernetes system.

Our primary weapon is Argo CD. What Argo CD does is that it watches a manifest GitOps repo for changes and then syncs the deployed application to its desired state based on the state of the manifest. The primary requirement here is a manifest repo. Now in case the customer does not have the repo, what do you do?.

I started looking at solutions to docker-ize lightweight git server into a container. One of the inspirational repos which I went through is https://github.com/jkarlosb/git-server-docker

However I had to make this work as a service in Kubernetes which ARGO CD could watch via its svc address. My work at https://github.com/samuel-sujith/gitasservice-k8s is intended to make a lightweight git server run as a service in Kubernetes. This will hold the manifest repo which ARGO CD can then deploy on to the target cluster.

Git : https://github.com/samuel-sujith/gitasservice-k8s

Medium : https://medium.com/@samuel.sujith/deployment-using-light-weight-git-server-as-service-in-kubernetes-36e52bd313d0

Please feel free to reach out to me www.garudax.id/in/samuelsujith for any questions or comments.

Please feel free to raise issues on the git.

Credits:

https://github.com/jkarlosb/git-server-docker

To view or add a comment, sign in

More articles by Sujith S.

  • Inspiration

    Many a times, we hear stories of people around us who cross insurmountable odds to finally realize their passion and do…

    6 Comments
  • Non-deterministic programming

    Coding for machine learning. How does it differ from the code that I have always done, COBOL, Java or setting up rules…

    2 Comments
  • Data science and deployment

    Back in the old days when I was a Mainframe programmer, when I used to code, I always knew how the end user would see…

    5 Comments

Explore content categories