viqFeb 23, 2024
#kubernetes question: our deployment stack is #skaffold plus #helm. We have #OpenEBS deployed, I'm trying to add their monitoring mixin to deployment. But they want to deploy whole kube-prometheus, which we already have. I have a script to generate their manifests and grab just those 4 files we're interested in, and have helm grab those. But here comes the question: how do I tie in the generation with the rest of the deployment?
[1/2]
Show threadviq

- lifecycle hooks of #skaffold. But we don't build any image, sync doesn't seem to do anything, and pre-deploy is too late to be able to see what would get done
- just commit those files to repo, and have a script to update them

Is there some other option I didn't think of?
[2/2]

Feb 23, 2024 at 7:25amWeb
Show threadszamuboyFeb 23, 2024
@viq Hey, viq, using #skaffold and #helm for deploying Kubernetes is a less flexible approach to deploying manifests. Eventually, you will face the rigidity of this technique. I would consider rewriting the deployment pipeline. What you need is yaml files at the end of the day, anyway. A simple #kubectl can deal with that. You can get the yamls from the `helm template` command and post-process them with, e.g., #kustomize—a simple and battle-tested solution.
[1/2]
Show threadszamuboyFeb 23, 2024

@viq I would go for the deployer script if rewriting the deployment pipeline is not an option.

My 2c.
[2/2]

Show threadviqFeb 23, 2024
@szamuboy
Deployer script?
I'm not up for trying to tell everyone that how we've been doing things is wrong and we're to change the tools everyone uses 🤷
Show threadszamuboyFeb 23, 2024
@viq Aaaah, I meant 'updater' script. Your second option is what I pick. Sorry, too it's too early for my brain.
Show threadviqFeb 23, 2024
@szamuboy
Right, yeah, I'm starting also to lean towards local files and a way to update them if needed. Thank you.