Managing container images with Helm across multiple clusters can be challenging. Any slight adjustments in image, values, or configuration can lead to issues, and charts become unwieldy quickly. I'm overwhelmed with environment overrides, tagging, and versioning, and I previously tried using Chainguard for image auditing and building, but it didn't suit our requirements—it felt too heavy and inflexible for our setup. I'm seeking suggestions for lighter or more adaptable solutions that can scale effectively, including workflows and tools. Any ideas?
4 Answers
I've heard a lot of good feedback about ArgoCD with ApplicationSets too; people seem to find it quite effective.
Yoke combined with ArgoCD has been a pretty solid setup for us. It avoids the Helm templating altogether and uses the GitOps pull model of ArgoCD, which is nice and straightforward!
We’ve been having great success with ArgoCD and Helm! I’m curious about your specific issues with Helm, though, because I haven’t encountered any problems with image tag updates breaking the chart. Maybe looking at how you're organizing your repositories could help—it allows us to update each cluster independently without hassle.
Have you considered using ArgoCD with ApplicationSets alongside Kustomize? It could really streamline your process!
Exactly like this! I've been doing it for ages, and I wouldn't know how to manage it differently!