Sync kubeflow training operator manifests v1.5.0 rc.0 (#2232)

* readme: Use new link for training-operator Don't use the tf-operator but training-operator Signed-off-by: Kimonas Sotirchos <kimwnasptd@arrikto.com> * hack: Update sync script for Training Operator Signed-off-by: Kimonas Sotirchos <kimwnasptd@arrikto.com> * Update kubeflow/training-operator manifests from v1.5.0-rc.0
kubeflow · Jun 30, 2022 · 6fce75e · 6fce75e
1 parent 7a74ea2
commit 6fce75e
Show file tree

Hide file tree

Showing 11 changed files with 8,605 additions and 3,873 deletions.
diff --git a/README.md b/README.md
@@ -41,7 +41,7 @@ This repo periodically syncs all official Kubeflow components from their respect
 
 | Component | Local Manifests Path | Upstream Revision |
 | - | - | - |
-| Training Operator | apps/training-operator/upstream | [v1.4.0](https://github.com/kubeflow/tf-operator/tree/v1.4.0/manifests) |
+| Training Operator | apps/training-operator/upstream | [v1.5.0-rc.0](https://github.com/kubeflow/training-operator/tree/v1.5.0-rc.0/manifests) |
 | Notebook Controller | apps/jupyter/notebook-controller/upstream | [v1.5.0](https://github.com/kubeflow/kubeflow/tree/v1.5.0/components/notebook-controller/config) |
 | Tensorboard Controller | apps/tensorboard/tensorboard-controller/upstream | [v1.5.0](https://github.com/kubeflow/kubeflow/tree/v1.5.0/components/tensorboard-controller/config) |
 | Central Dashboard | apps/centraldashboard/upstream | [v1.5.0](https://github.com/kubeflow/kubeflow/tree/v1.5.0/components/centraldashboard/manifests) |

diff --git a/apps/training-operator/upstream/base/crds/kubeflow.org_mpijobs.yaml b/apps/training-operator/upstream/base/crds/kubeflow.org_mpijobs.yaml
diff --git a/apps/training-operator/upstream/base/crds/kubeflow.org_mxjobs.yaml b/apps/training-operator/upstream/base/crds/kubeflow.org_mxjobs.yaml
diff --git a/apps/training-operator/upstream/base/crds/kubeflow.org_pytorchjobs.yaml b/apps/training-operator/upstream/base/crds/kubeflow.org_pytorchjobs.yaml
diff --git a/apps/training-operator/upstream/base/crds/kubeflow.org_tfjobs.yaml b/apps/training-operator/upstream/base/crds/kubeflow.org_tfjobs.yaml
diff --git a/apps/training-operator/upstream/base/crds/kubeflow.org_xgboostjobs.yaml b/apps/training-operator/upstream/base/crds/kubeflow.org_xgboostjobs.yaml
diff --git a/apps/training-operator/upstream/base/deployment.yaml b/apps/training-operator/upstream/base/deployment.yaml
@@ -40,12 +40,14 @@ spec:
               port: 8081
             initialDelaySeconds: 15
             periodSeconds: 20
+            timeoutSeconds: 3
           readinessProbe:
             httpGet:
               path: /readyz
               port: 8081
-            initialDelaySeconds: 5
-            periodSeconds: 10
+            initialDelaySeconds: 10
+            periodSeconds: 15
+            timeoutSeconds: 3
           resources:
             limits:
               cpu: 100m

diff --git a/apps/training-operator/upstream/overlays/kubeflow/kubeflow-training-roles.yaml b/apps/training-operator/upstream/overlays/kubeflow/kubeflow-training-roles.yaml
@@ -22,6 +22,7 @@ rules:
   - apiGroups:
       - kubeflow.org
     resources:
+      - mpijobs
       - tfjobs
       - pytorchjobs
       - mxjobs
@@ -37,6 +38,7 @@ rules:
   - apiGroups:
       - kubeflow.org
     resources:
+      - mpijobs/status
       - tfjobs/status
       - pytorchjobs/status
       - mxjobs/status
@@ -55,6 +57,7 @@ rules:
   - apiGroups:
       - kubeflow.org
     resources:
+      - mpijobs
       - tfjobs
       - pytorchjobs
       - mxjobs
@@ -66,6 +69,7 @@ rules:
   - apiGroups:
       - kubeflow.org
     resources:
+      - mpijobs/status
       - tfjobs/status
       - pytorchjobs/status
       - mxjobs/status

diff --git a/apps/training-operator/upstream/overlays/kubeflow/kustomization.yaml b/apps/training-operator/upstream/overlays/kubeflow/kustomization.yaml
@@ -6,5 +6,4 @@ resources:
   - kubeflow-training-roles.yaml
 images:
   - name: kubeflow/training-operator
-    newName: public.ecr.aws/j1r0q0g6/training/training-operator
-    newTag: "174e8813666951ded505daf334a37f60fd50c18d"
+    newTag: "v1-e1434f6"
diff --git a/apps/training-operator/upstream/overlays/standalone/kustomization.yaml b/apps/training-operator/upstream/overlays/standalone/kustomization.yaml
@@ -6,5 +6,4 @@ resources:
   - namespace.yaml
 images:
   - name: kubeflow/training-operator
-    newName: public.ecr.aws/j1r0q0g6/training/training-operator
-    newTag: "174e8813666951ded505daf334a37f60fd50c18d"
+    newTag: "v1-e1434f6"
diff --git a/hack/sync-training-operator-manifests.sh b/hack/sync-training-operator-manifests.sh
@@ -55,8 +55,15 @@ cp $SRC_DIR/manifests $DST_DIR -r
 
 echo "Successfully copied all manifests."
 
+echo "Updating README..."
+SRC_TXT="\[.*\](https://github.com/kubeflow/training-operator/tree/.*/manifests)"
+DST_TXT="\[$COMMIT\](https://github.com/kubeflow/training-operator/tree/$COMMIT/manifests)"
+
+sed -i "s|$SRC_TXT|$DST_TXT|g" ${MANIFESTS_DIR}/README.md
+
 # DEV: Comment out these commands when local testing
 echo "Committing the changes..."
 cd $MANIFESTS_DIR
 git add apps
+git add README.md
 git commit -m "Update kubeflow/training-operator manifests from ${COMMIT}"