Update Charmed Katib Operators + CI to 0.12

.github/workflows/test-charmed-katib.yaml

@@ -15,7 +15,9 @@ jobs:
 
       - name: Install dependencies
         run: |
-          sudo apt-get install python3-setuptools
+          set -eux
+          sudo apt update
+          sudo apt install python3-setuptools
           sudo pip3 install black flake8
 
       - name: Check black
@@ -34,19 +36,18 @@ jobs:
 
       - uses: balchua/microk8s-actions@v0.2.2
         with:
-          channel: "1.20/stable"
+          channel: "1.21/stable"
           addons: '["dns", "storage", "rbac"]'
 
       - name: Install dependencies
         run: |
           set -eux
           sudo apt update
           sudo apt install -y python3-pip
-          sudo snap install charm --classic
           sudo snap install juju --classic
-          sudo snap install juju-helpers --classic
+          sudo snap install juju-bundle --classic
           sudo snap install juju-wait --classic
-          sudo pip3 install charmcraft==1.0.0
+          sudo pip3 install charmcraft==1.3.1
 
       - name: Build Docker images
         run: |
@@ -62,20 +63,20 @@ jobs:
           done
 
       - name: Deploy Katib
+        env:
+          CHARMCRAFT_DEVELOPER: "1"
         run: |
           set -eux
           cd operators/
           git clone git://git.launchpad.net/canonical-osm
           cp -r canonical-osm/charms/interfaces/juju-relation-mysql mysql
           sg microk8s -c 'juju bootstrap microk8s uk8s'
           juju add-model kubeflow
-          juju bundle deploy -b bundle-edge.yaml --build
+          juju bundle deploy --build --destructive-mode --serial
           juju wait -wvt 300
 
       - name: Test Katib
-        run: |
-          set -eux
-          kubectl apply -f examples/v1beta1/hp-tuning/random.yaml
+        run: kubectl apply -f examples/v1beta1/hp-tuning/random.yaml
 
       - name: Get pod statuses
         run: kubectl get all -A
@@ -86,25 +87,32 @@ jobs:
         if: failure()
 
       - name: Get katib-controller workload logs
-        run: kubectl logs --tail 100 -nkubeflow -ljuju-app=katib-controller
+        run: kubectl logs --tail 100 -nkubeflow -lapp.kubernetes.io/name=katib-controller
         if: failure()
 
       - name: Get katib-controller operator logs
-        run: kubectl logs --tail 100 -nkubeflow -ljuju-operator=katib-controller
+        run: kubectl logs --tail 100 -nkubeflow -loperator.juju.is/name=katib-controller
         if: failure()
 
       - name: Get katib-ui workload logs
-        run: kubectl logs --tail 100 -nkubeflow -ljuju-app=katib-ui
+        run: kubectl logs --tail 100 -nkubeflow -lapp.kubernetes.io/name=katib-ui
         if: failure()
 
       - name: Get katib-ui operator logs
-        run: kubectl logs --tail 100 -nkubeflow -ljuju-operator=katib-ui
+        run: kubectl logs --tail 100 -nkubeflow -loperator.juju.is/name=katib-ui
         if: failure()
 
       - name: Get katib-db-manager workload logs
-        run: kubectl logs --tail 100 -nkubeflow -ljuju-app=katib-db-manager
+        run: kubectl logs --tail 100 -nkubeflow -lapp.kubernetes.io/name=katib-db-manager
         if: failure()
 
       - name: Get katib-db-manager operator logs
-        run: kubectl logs --tail 100 -nkubeflow -ljuju-operator=katib-db-manager
+        run: kubectl logs --tail 100 -nkubeflow -loperator.juju.is/name=katib-db-manager
+        if: failure()
+
+      - name: Upload charmcraft logs
+        uses: actions/upload-artifact@v2
+        with:
+          name: charmcraft-logs
+          path: /tmp/charmcraft-log-*
         if: failure()

operators/katib-controller/charmcraft.yaml

@@ -0,0 +1,11 @@
+type: charm
+bases:
+  - build-on:
+    - name: "ubuntu"
+      channel: "20.04"
+    run-on:
+    - name: "ubuntu"
+      channel: "20.04"
+parts:
+  charm:
+    charm-python-packages: [setuptools, pip]

operators/katib-controller/metadata.yaml

@@ -17,7 +17,7 @@ resources:
     type: oci-image
     description: Backing OCI image
     auto-fetch: true
-    upstream-source: docker.io/kubeflowkatib/katib-controller:v1beta1-c6c9172
+    upstream-source: docker.io/kubeflowkatib/katib-controller:v0.12.0
 provides:
   katib-controller:
     interface: http

operators/katib-controller/src/charm.py

@@ -5,12 +5,11 @@
 from subprocess import check_call
 
 import yaml
+from oci_image import OCIImageResource, OCIImageResourceError
 from ops.charm import CharmBase
+from ops.framework import StoredState
 from ops.main import main
 from ops.model import ActiveStatus, MaintenanceStatus
-from ops.framework import StoredState
-
-from oci_image import OCIImageResource, OCIImageResourceError
 
 logger = logging.getLogger(__name__)
 
@@ -62,29 +61,30 @@ def set_pod_spec(self, event):
                                         "namespaces",
                                         "persistentvolumes",
                                         "persistentvolumeclaims",
+                                        "pods",
+                                        "pods/log",
+                                        "pods/status",
                                     ],
                                     "verbs": ["*"],
                                 },
                                 {
-                                    "apiGroups": [""],
-                                    "resources": ["pods", "pods/log", "pods/status"],
+                                    "apiGroups": ["apps"],
+                                    "resources": ["deployments"],
                                     "verbs": ["*"],
                                 },
                                 {
-                                    "apiGroups": ["apps"],
-                                    "resources": ["deployments"],
+                                    "apiGroups": ["rbac.authorization.k8s.io"],
+                                    "resources": [
+                                        "roles",
+                                        "rolebindings",
+                                    ],
                                     "verbs": ["*"],
                                 },
                                 {
                                     "apiGroups": ["batch"],
                                     "resources": ["jobs", "cronjobs"],
                                     "verbs": ["*"],
                                 },
-                                {
-                                    "apiGroups": ["apiextensions.k8s.io"],
-                                    "resources": ["customresourcedefinitions"],
-                                    "verbs": ["create", "get"],
-                                },
                                 {
                                     "apiGroups": ["kubeflow.org"],
                                     "resources": [
@@ -97,24 +97,14 @@ def set_pod_spec(self, event):
                                         "suggestions",
                                         "suggestions/status",
                                         "suggestions/finalizers",
+                                        "tfjobs",
+                                        "pytorchjobs",
+                                        "mpijobs",
+                                        "xgboostjobs",
+                                        "mxjobs",
                                     ],
                                     "verbs": ["*"],
                                 },
-                                {
-                                    "apiGroups": ["kubeflow.org"],
-                                    "resources": ["tfjobs", "pytorchjobs", "mpijobs"],
-                                    "verbs": ["*"],
-                                },
-                                {
-                                    "apiGroups": ["tekton.dev"],
-                                    "resources": ["pipelineruns", "taskruns"],
-                                    "verbs": ["*"],
-                                },
-                                {
-                                    "apiGroups": ["rbac.authorization.k8s.io"],
-                                    "resources": ["roles", "rolebindings"],
-                                    "verbs": ["*"],
-                                },
                             ],
                         }
                     ],

operators/katib-controller/src/crds.yaml

@@ -1,24 +1,31 @@
 ---
-apiVersion: apiextensions.k8s.io/v1beta1
+apiVersion: apiextensions.k8s.io/v1
 kind: CustomResourceDefinition
 metadata:
   name: experiments.kubeflow.org
 spec:
-  additionalPrinterColumns:
-    - JSONPath: .status.conditions[-1:].type
-      name: Type
-      type: string
-    - JSONPath: .status.conditions[-1:].status
-      name: Status
-      type: string
-    - JSONPath: .metadata.creationTimestamp
-      name: Age
-      type: date
   group: kubeflow.org
-  version: v1beta1
   scope: Namespaced
-  subresources:
-    status: {}
+  versions:
+    - name: v1beta1
+      served: true
+      storage: true
+      additionalPrinterColumns:
+        - name: Type
+          type: string
+          jsonPath: .status.conditions[-1:].type
+        - name: Status
+          type: string
+          jsonPath: .status.conditions[-1:].status
+        - name: Age
+          type: date
+          jsonPath: .metadata.creationTimestamp
+      subresources:
+        status: {}
+      schema:
+        openAPIV3Schema:
+          type: object
+          x-kubernetes-preserve-unknown-fields: true
   names:
     kind: Experiment
     singular: experiment
@@ -27,34 +34,40 @@ spec:
       - all
       - kubeflow
       - katib
-
 ---
-apiVersion: apiextensions.k8s.io/v1beta1
+apiVersion: apiextensions.k8s.io/v1
 kind: CustomResourceDefinition
 metadata:
   name: suggestions.kubeflow.org
 spec:
-  additionalPrinterColumns:
-    - JSONPath: .status.conditions[-1:].type
-      name: Type
-      type: string
-    - JSONPath: .status.conditions[-1:].status
-      name: Status
-      type: string
-    - JSONPath: .spec.requests
-      name: Requested
-      type: string
-    - JSONPath: .status.suggestionCount
-      name: Assigned
-      type: string
-    - JSONPath: .metadata.creationTimestamp
-      name: Age
-      type: date
   group: kubeflow.org
-  version: v1beta1
   scope: Namespaced
-  subresources:
-    status: {}
+  versions:
+    - name: v1beta1
+      served: true
+      storage: true
+      additionalPrinterColumns:
+        - name: Type
+          type: string
+          jsonPath: .status.conditions[-1:].type
+        - name: Status
+          type: string
+          jsonPath: .status.conditions[-1:].status
+        - name: Requested
+          type: string
+          jsonPath: .spec.requests
+        - name: Assigned
+          type: string
+          jsonPath: .status.suggestionCount
+        - name: Age
+          type: date
+          jsonPath: .metadata.creationTimestamp
+      subresources:
+        status: {}
+      schema:
+        openAPIV3Schema:
+          type: object
+          x-kubernetes-preserve-unknown-fields: true
   names:
     kind: Suggestion
     singular: suggestion
@@ -63,28 +76,34 @@ spec:
       - all
       - kubeflow
       - katib
-
 ---
-apiVersion: apiextensions.k8s.io/v1beta1
+apiVersion: apiextensions.k8s.io/v1
 kind: CustomResourceDefinition
 metadata:
   name: trials.kubeflow.org
 spec:
-  additionalPrinterColumns:
-    - JSONPath: .status.conditions[-1:].type
-      name: Type
-      type: string
-    - JSONPath: .status.conditions[-1:].status
-      name: Status
-      type: string
-    - JSONPath: .metadata.creationTimestamp
-      name: Age
-      type: date
   group: kubeflow.org
-  version: v1beta1
   scope: Namespaced
-  subresources:
-    status: {}
+  versions:
+    - name: v1beta1
+      served: true
+      storage: true
+      additionalPrinterColumns:
+        - name: Type
+          type: string
+          jsonPath: .status.conditions[-1:].type
+        - name: Status
+          type: string
+          jsonPath: .status.conditions[-1:].status
+        - name: Age
+          type: date
+          jsonPath: .metadata.creationTimestamp
+      subresources:
+        status: {}
+      schema:
+        openAPIV3Schema:
+          type: object
+          x-kubernetes-preserve-unknown-fields: true
   names:
     kind: Trial
     singular: trial

operators/katib-controller/src/early-stopping.json

@@ -1,6 +1,5 @@
 {
   "medianstop": {
-    "image": "docker.io/kubeflowkatib/earlystopping-medianstop",
-    "imagePullPolicy": "Always"
+    "image": "docker.io/kubeflowkatib/earlystopping-medianstop:v0.12.0"
   }
 }

operators/katib-controller/src/metrics-collector-sidecar.json

@@ -1,12 +1,12 @@
 {
   "StdOut": {
-    "image": "docker.io/kubeflowkatib/file-metrics-collector"
+    "image": "docker.io/kubeflowkatib/file-metrics-collector:v0.12.0"
   },
   "File": {
-    "image": "docker.io/kubeflowkatib/file-metrics-collector"
+    "image": "docker.io/kubeflowkatib/file-metrics-collector:v0.12.0"
   },
   "TensorFlowEvent": {
-    "image": "docker.io/kubeflowkatib/tfevent-metrics-collector",
+    "image": "docker.io/kubeflowkatib/tfevent-metrics-collector:v0.12.0",
     "resources": {
       "limits": {
         "memory": "1Gi"

operators/katib-controller/src/pytorchJobTemplate.yaml

@@ -1,4 +1,4 @@
-apiVersion: kubeflow.org/v1
+apiVersion: "kubeflow.org/v1"
 kind: PyTorchJob
 spec:
   pytorchReplicaSpecs: