Search before asking

• I had searched in the issues and found no similar feature requirement.

Description

This is the current conditions for a RayCluster in steady state:

  status:
    availableWorkerReplicas: 1
    conditions:
    - lastTransitionTime: "2024-08-09T19:16:09Z"
      message: ""
      reason: HeadPodRunningAndReady
      status: "True"
      type: HeadPodReady
    - lastTransitionTime: "2024-08-09T19:16:09Z"
      message: Only check head after all Ray Pods are initially ready
      reason: HeadPodRunningAndReady
      status: "True"
      type: RayClusterReady

I find it a bit confusing that both the HeadPodReady and RayClusterReady condition have the same reason HeadPodRunningAndReady. After the initialization of the cluster, they are semantically the same.

I propose that we change the RayClusterReady condition to something like RayClusterProvisioned or RayClusterInitialized which only represents the first time all pods are ready and does not track head Pod readiness afterwards. This reduces the overlap between these two conditions and makes it clearer which condition should be checked.

Use case

No response

Related issues

No response

Are you willing to submit a PR?

• Yes I am willing to submit a PR!