tanka VS kubevela

I would recommend implementing a similar API to Grafana Tanka: https://tanka.dev

When you "synthesise", the returned value should be an array or an object.

1. If it's an object, check if it has an `apiVersion` and `kind` key. If it does, yield that as a kubernetes object and do not recurse.

Maybe you'd like jsonnet: https://jsonnet.org/

I find it particularly useful for configurations that often have repeated boilerplate, like ansible playbooks or deploying a bunch of "similar-but" services to kubernetes (with https://tanka.dev).

Dhall is also quite interesting, with some tradeoffs: https://dhall-lang.org/

A few years ago I did a small comparison by re-implementing one of my simpler ansible playbooks: https://github.com/retzkek/ansible-dhall-jsonnet

- validation is often impractical (at least identifying exactly where the error is… I’m looking at you Helm!)

Unrelated to OP, but you can leverage Tanka to extend helm charts with functionality not provided by upstream.

https://tanka.dev/

Although jsonette might be considered more complex Tanka is a great alternative for k8s config management.

At Grafana Labs we're using jsonnet at scale, while being a powerful functional language it is also excellent for rendering JSON/YAML config. We have developed Tanka[0] to work with Kubernetes, for other purposes I can recommend this course[1] (authored by me).

[0] https://tanka.dev/

[1] https://jsonnet-libs.github.io/jsonnet-training-course/

If you're hitting the limits of Kustomize, maybe look at Tanka as well.

Yes, try Tanka.

yes. basically. and this is a path that multiple people are trying to solve. e.g. AWS CDK8s, https://tanka.dev/, etc

Compose would be awesome.

http://tanka.dev

(Note I work for Grafana Labs who fund Tanka and use it for all production config)

Finally, I think the OAM model offers one possible future. Take a look at projects like KubeVela and Crossplane. These allow you to compose your own custom abstraction layer. The developer creates a simple CRD called "Application" and this is translated into ths k8s or even off-cluster resources. Problem right now is the complexity is transferred onto guys configuring the platform..... I want to see more "out of the box" implementations.

Gotcha, thanks! We already run ArgoCD but having devs write raw manifests feels so low-level when it’s usually the same combo of configmaps, ingresses, services, deployments… Maybe this is more in the direction of what I’m looking for? 🤔 https://kubevela.io

Note: On that topic, I'm keeping a close eye on the Open Application Model and the kubevela projects. I think it’ll help write a representation of an application and its components that we can validate the structure of our code against, and generate documentation from it. Not a complete solution to the problem, but it'll help with certain parts of architecture documentation

Also, kubevela is getting ready to connect to clusterpedia https://github.com/kubevela/kubevela/issues/4237,

https://kubevela.io/ is an alternative that has been around longer, exposes CUE naturally, and builds on open standards for app packs. I'd recommend this open-source product.