I have been working on ClickHouse and Kubernetes for several years at my current company, Altinity. Earlier this year we found out about Argo CD from one of our customers and have been learning how to use it to stand up analytic database stacks.
In case you are not familiar with Argo CD, it provides a way to define Kubernetes applications, store them in GitHub, and deploy/update them in live Kubernetes clusters. It's addictive once you get used to it. Based on the experience, I wrote a tutorial on how to set up ClickHouse clusters on Kubernetes.
Bring up ClickHouse on Kubernetes with Argo CD
One of the things we picked up on quickly is that Argo is very powerful to use for building stacks, but there's a learning curve. It's extremely helpful to see working Argo CD code to understand how to use it effectively. We are planning to build out the project the article references into a proper blueprint for analytic stacks.
I would love to hear what people think of this tutorial and more generally how you are building stacks on Kubernetes. Are you using Argo CD? If not, what are you using? Flux? Terraform? Mega-scripts of death?
Let the games begin!
Top comments (0)