OrasCache¶

The CRD “Custom Resource Definition” defines an ORAS Cache (registry)

A CRD is a yaml file that you can apply to your cluster (with the ORAS Operator installed) to ask for an OrasCache to be deployed. Kubernetes has these custom resources to make it easy to automate tasks, and in fact this is the goal of an operator! A Kubernetes operator is conceptually like a human operator that takes your CRD, looks at the cluster state, and does whatever is necessary to get your cluster state to match your request. In the case of the ORAS Operator, this means deploying an OCI Registry as storage (ORAS) in a particular namespace to cache workflow or other artifacts. This document describes the spec of our custom resource definition.

Custom Resource Definition¶

The operator custom resource definition (CRD) currently takes just a few fields!

header¶

The header should specify the name (you’ll need this for your workflow pods) and the namespace (the operator will run in this namespace and can interact with these pods)

apiVersion: cache.converged-computing.github.io/v1alpha1
kind: OrasCache
metadata:
  name: orascache-sample
spec:
...

In the above, we generate it in the “default” namespace and name it “orascache-sample.” Note that all fields described below go under “spec.”

spec¶

image¶

The image to use for the registry, which defaults to the one deployed by oras ghcr.io/oras-project/registry:latest Here is what that might look like (reproducing the default):

spec:
  # We can use all the defaults here (this is a default)
  image: ghcr.io/oras-project/registry:latest

deploy¶

If deploy is set to false, this indicates that you don’t want the operator to deploy a stateful set local registry. This means that you’ll need to provide all jobs / pods with a remote registry field to replace it, and (likely) a secret to pull or push. See orasEnv below for details.

spec:
  deploy: false 

secrets¶

There are several secrets that can be added, if needed.

orasEnv¶

If you deploy a secret to the namespace that you want the ORAS operator to use (providing in the environment to the pod) you can define the name here.

spec:
  secrets: 
    orasEnv: mysecret-name

registryHttp¶

In the case of deploying more than one registry pod (not supported yet), the push secret can be customized.

spec:
  secrets: 
    registryHttp: mysecret

Note that this is not supported yet - likely we would want to add custom volumes (for shared storage) between more than one pod in the stateful set. For now we just need to save small amounts of data and will add this functionality when needed (and the secret will then be relevant).