pytorch-operator: Consolidate manifests (#323)

* manifests: Add base/overlays structure

Signed-off-by: Yannis Zarkadas <[email protected]>

* Edit docs and scripts to use new manifests

Signed-off-by: Yannis Zarkadas <[email protected]>

* Delete old manifests

Signed-off-by: Yannis Zarkadas <[email protected]>

Author: yanniszark

README.md

@@ -6,7 +6,7 @@
 
 ## Overview
 
-This repository contains the specification and implementation of `PyTorchJob` custom resource definition. Using this custom resource, users can create and manage PyTorch jobs like other built-in resources in Kubernetes. See [CRD definition](https://github.com/kubeflow/pytorch-operator/blob/master/manifests/crd.yaml)
+This repository contains the specification and implementation of `PyTorchJob` custom resource definition. Using this custom resource, users can create and manage PyTorch jobs like other built-in resources in Kubernetes. See [CRD definition](https://github.com/kubeflow/pytorch-operator/blob/master/manifests/base/crd.yaml)
 
 ## Prerequisites
 

developer_guide.md

@@ -55,7 +55,7 @@ export KUBEFLOW_NAMESPACE=$(your_namespace)
 After the cluster is up, the PyTorch Operator CRD should be created on the cluster.
 
 ```bash
-kubectl create -f ./manifests/crd.yaml
+kubectl create -f ./manifests/base/crd.yaml
 ```
 
 ### Run Operator

manifests/crd.yaml renamed to manifests/base/crd.yaml

@@ -2,14 +2,16 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 namespace: kubeflow
 resources:
-- cluster-role-binding.yaml
-- cluster-role.yaml
+- crd.yaml
+- rbac.yaml
 - deployment.yaml
-- service-account.yaml
 - service.yaml
+- user-clusterroles.yaml
 commonLabels:
+  app: pytorch-operator
   kustomize.component: pytorch-operator
+  app.kubernetes.io/component: pytorch
+  app.kubernetes.io/name: pytorch-operator
 images:
 - name: gcr.io/kubeflow-images-public/pytorch-operator
-  newName: gcr.io/kubeflow-images-public/pytorch-operator
   newTag: vmaster-g518f9c76

manifests/deployment.yaml renamed to manifests/base/deployment.yaml

@@ -1,35 +1,3 @@
-apiVersion: rbac.authorization.k8s.io/v1beta1
-kind: ClusterRole
-metadata:
-  labels:
-    app: pytorch-operator
-  name: pytorch-operator
-rules:
-- apiGroups:
-  - kubeflow.org
-  resources:
-  - pytorchjobs
-  - pytorchjobs/status
-  - pytorchjobs/finalizers
-  verbs:
-  - '*'
-- apiGroups:
-  - apiextensions.k8s.io
-  resources:
-  - customresourcedefinitions
-  verbs:
-  - '*'
-- apiGroups:
-  - ""
-  resources:
-  - pods
-  - services
-  - endpoints
-  - events
-  verbs:
-  - '*'
----
-
 apiVersion: rbac.authorization.k8s.io/v1
 kind: ClusterRole
 metadata:

manifests/pytorch-operator/base/kustomization.yaml renamed to manifests/base/kustomization.yaml

@@ -1,9 +1,12 @@
 apiVersion: kustomize.config.k8s.io/v1beta1
-bases:
+kind: Kustomization
+namespace: kubeflow
+resources:
 - ../../base
 commonLabels:
+  app: pytorch-operator
+  kustomize.component: pytorch-operator
   app.kubernetes.io/component: pytorch
   app.kubernetes.io/name: pytorch-operator
-kind: Kustomization
-resources:
-- application.yaml
+images:
+- name: gcr.io/kubeflow-images-public/pytorch-operator

manifests/podgroup.yaml renamed to manifests/base/podgroup.yaml

@@ -2,14 +2,12 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 namespace: kubeflow
 resources:
-- crd.yaml
+- ../../base
 - namespace.yaml
-- rbac.yaml
-- deployment.yaml
-- service.yaml
 commonLabels:
+  app: pytorch-operator
   kustomize.component: pytorch-operator
+  app.kubernetes.io/component: pytorch
+  app.kubernetes.io/name: pytorch-operator
 images:
 - name: gcr.io/kubeflow-images-public/pytorch-operator
-  newName: 809251082950.dkr.ecr.us-west-2.amazonaws.com/pytorch-operator
-  newTag: "0.1"

manifests/rbac.yaml renamed to manifests/base/rbac.yaml

@@ -33,11 +33,11 @@ aws eks update-kubeconfig --region=${REGION} --name=${CLUSTER_NAME}
 
 echo "Update PyTorch operator manifest with new name and tag"
 #TODO(Jeffwan@): If there's a way to specify context, then we don't need to enter manifests folder
-cd manifests/
+cd manifests/overlays/standalone
 kustomize edit set image gcr.io/kubeflow-images-public/pytorch-operator=${REGISTRY}/${REPO_NAME}:${VERSION}
 
 echo "Installing PyTorch operator manifests"
-kubectl apply -k .
+kustomize build . | kubectl apply -f -
 
 TIMEOUT=30
 until kubectl get pods -n kubeflow | grep pytorch-operator | grep 1/1 || [[ $TIMEOUT -eq 1 ]]; do