CodeFlare Operator Installation

CodeFlare Operator Installation

0. Pre-reqs:

0.1 Assumes you have an OpenShift Cluster

0.2 Assumes you have ODH installed in opendatahub namespace. (Or you're using local integration and not ODH)

0.3 Assumes you've deployed the notebook image in your Open Data Hub namespace. For example:

podman login -u kubeadmin -p $(oc whoami -t) $(oc registry info) --tls-verify=false
export ODH_NAMESPACE=opendatahub
oc apply -f https://raw.githubusercontent.com/project-codeflare/codeflare-sdk/main/custom-nb-image/imagestream.yaml -n $ODH_NAMESPACE

1. Operator Installation

1.1 Using OperatorHub on your OpenShift console, install CodeFlare Operator to openshift-operators

1.2 Using the UI, go to CodeFlare Operator you just deployed and deploy MCAD operator using these step:

In the Provided APIs area, Click on Create instance of the MCAD item

2. Install KubeRay Operator

Old Method:

# KubeRay
oc create -k "github.com/ray-project/kuberay/ray-operator/config/crd?ref=v0.3.0"
helm install kuberay-operator --namespace ray-system --create-namespace $(curl -s https://api.github.com/repos/ray-project/kuberay/releases/tags/v0.3.0 | grep '"browser_download_url":' | sort | grep -om1 'https.*helm-chart-kuberay-operator.*tgz')

New Method:

oc apply -f https://github.com/opendatahub-io-contrib/odh-contrib-manifests/blob/main/ray/ray-minimal-kfdef.yaml