task(RHOAIENG-26590): Report RayJob status via SDK

.github/workflows/coverage-badge.yaml

@@ -4,7 +4,7 @@ name: Coverage Badge
 
 on:
   push:
-    branches: [ main ]
+    branches: [ main, ray-jobs-feature ]
 
 jobs:
   report:

.github/workflows/e2e_tests.yaml

@@ -6,6 +6,7 @@ on:
     branches:
       - main
       - 'release-*'
+      - ray-jobs-feature
     paths-ignore:
       - 'docs/**'
       - '**.adoc'

.github/workflows/unit-tests.yml

@@ -2,9 +2,9 @@ name: Python Tests
 
 on:
   pull_request:
-    branches: [ main ]
+    branches: [ main, ray-jobs-feature ]
   push:
-    branches: [ main ]
+    branches: [ main, ray-jobs-feature ]
 
 jobs:
   unit-tests:

demo-notebooks/additional-demos/hf_interactive.ipynb

@@ -114,7 +114,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "cluster.up()"
+    "cluster.apply()"
    ]
   },
   {

demo-notebooks/additional-demos/local_interactive.ipynb

@@ -79,7 +79,7 @@
    },
    "outputs": [],
    "source": [
-    "cluster.up()"
+    "cluster.apply()"
    ]
   },
   {

demo-notebooks/additional-demos/ray_job_client.ipynb

@@ -78,7 +78,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()\n",
+    "cluster.apply()\n",
     "cluster.wait_ready()"
    ]
   },

demo-notebooks/guided-demos/0_basic_ray.ipynb

@@ -98,7 +98,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()"
+    "cluster.apply()"
    ]
   },
   {

demo-notebooks/guided-demos/1_cluster_job_client.ipynb

@@ -89,7 +89,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()\n",
+    "cluster.apply()\n",
     "cluster.wait_ready()"
    ]
   },

demo-notebooks/guided-demos/2_basic_interactive.ipynb

@@ -96,7 +96,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()\n",
+    "cluster.apply()\n",
     "cluster.wait_ready()"
    ]
   },

demo-notebooks/guided-demos/notebook-ex-outputs/0_basic_ray.ipynb

@@ -94,7 +94,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()"
+    "cluster.apply()"
    ]
   },
   {

demo-notebooks/guided-demos/notebook-ex-outputs/1_cluster_job_client.ipynb

@@ -78,7 +78,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()\n",
+    "cluster.apply()\n",
     "cluster.wait_ready()"
    ]
   },

demo-notebooks/guided-demos/notebook-ex-outputs/2_basic_interactive.ipynb

@@ -84,7 +84,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()\n",
+    "cluster.apply()\n",
     "cluster.wait_ready()"
    ]
   },

demo-notebooks/guided-demos/preview_nbs/0_basic_ray.ipynb

@@ -94,7 +94,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()"
+    "cluster.apply()"
    ]
   },
   {

demo-notebooks/guided-demos/preview_nbs/1_cluster_job_client.ipynb

@@ -78,7 +78,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()\n",
+    "cluster.apply()\n",
     "cluster.wait_ready()"
    ]
   },

demo-notebooks/guided-demos/preview_nbs/2_basic_interactive.ipynb

@@ -84,7 +84,7 @@
    "outputs": [],
    "source": [
     "# Bring up the cluster\n",
-    "cluster.up()\n",
+    "cluster.apply()\n",
     "cluster.wait_ready()"
    ]
   },

pyproject.toml

@@ -29,6 +29,14 @@ cryptography = "43.0.3"
 executing = "1.2.0"
 pydantic = "< 2"
 ipywidgets = "8.1.2"
+odh-kuberay-client = {version = "0.0.0.dev40", source = "testpypi"}
+
+[[tool.poetry.source]]
+name = "pypi"
+
+[[tool.poetry.source]]
+name = "testpypi"
+url = "https://test.pypi.org/simple/"
 
 [tool.poetry.group.docs]
 optional = true

src/codeflare_sdk/__init__.py

@@ -10,6 +10,7 @@
     AWManager,
     AppWrapperStatus,
     RayJobClient,
+    RayJob,
 )
 
 from .common.widgets import view_clusters

src/codeflare_sdk/common/widgets/test_widgets.py

@@ -28,7 +28,7 @@
 @patch.dict(
     "os.environ", {"JPY_SESSION_NAME": "example-test"}
 )  # Mock Jupyter environment variable
-def test_cluster_up_down_buttons(mocker):
+def test_cluster_apply_down_buttons(mocker):
     mocker.patch("kubernetes.client.ApisApi.get_api_versions")
     mocker.patch(
         "kubernetes.client.CustomObjectsApi.get_cluster_custom_object",
@@ -45,36 +45,38 @@ def test_cluster_up_down_buttons(mocker):
     ) as MockCheckbox, patch("ipywidgets.Output"), patch("ipywidgets.HBox"), patch(
         "ipywidgets.VBox"
     ), patch.object(
-        cluster, "up"
-    ) as mock_up, patch.object(
+        cluster, "apply"
+    ) as mock_apply, patch.object(
         cluster, "down"
     ) as mock_down, patch.object(
         cluster, "wait_ready"
     ) as mock_wait_ready:
         # Create mock button & CheckBox instances
-        mock_up_button = MagicMock()
+        mock_apply_button = MagicMock()
         mock_down_button = MagicMock()
         mock_wait_ready_check_box = MagicMock()
 
         # Ensure the mock Button class returns the mock button instances in sequence
         MockCheckbox.side_effect = [mock_wait_ready_check_box]
-        MockButton.side_effect = [mock_up_button, mock_down_button]
+        MockButton.side_effect = [mock_apply_button, mock_down_button]
 
         # Call the method under test
-        cf_widgets.cluster_up_down_buttons(cluster)
+        cf_widgets.cluster_apply_down_buttons(cluster)
 
         # Simulate checkbox being checked or unchecked
         mock_wait_ready_check_box.value = True  # Simulate checkbox being checked
 
         # Simulate the button clicks by calling the mock on_click handlers
-        mock_up_button.on_click.call_args[0][0](None)  # Simulate clicking "Cluster Up"
+        mock_apply_button.on_click.call_args[0][0](
+            None
+        )  # Simulate clicking "Cluster Apply"
         mock_down_button.on_click.call_args[0][0](
             None
         )  # Simulate clicking "Cluster Down"
 
-        # Check if the `up` and `down` methods were called
+        # Check if the `apply` and `down` methods were called
         mock_wait_ready.assert_called_once()
-        mock_up.assert_called_once()
+        mock_apply.assert_called_once()
         mock_down.assert_called_once()
 
 

src/codeflare_sdk/common/widgets/widgets.py

@@ -271,19 +271,19 @@ def display_widgets(self):
         )
 
 
-def cluster_up_down_buttons(
+def cluster_apply_down_buttons(
     cluster: "codeflare_sdk.ray.cluster.cluster.Cluster",
 ) -> widgets.Button:
     """
-    The cluster_up_down_buttons function returns two button widgets for a create and delete button.
+    The cluster_apply_down_buttons function returns two button widgets for a create and delete button.
     The function uses the appwrapper bool to distinguish between resource type for the tool tip.
     """
     resource = "Ray Cluster"
     if cluster.config.appwrapper:
         resource = "AppWrapper"
 
-    up_button = widgets.Button(
-        description="Cluster Up",
+    apply_button = widgets.Button(
+        description="Cluster Apply",
         tooltip=f"Create the {resource}",
         icon="play",
     )
@@ -298,13 +298,13 @@ def cluster_up_down_buttons(
     output = widgets.Output()
 
     # Display the buttons in an HBox wrapped in a VBox which includes the wait_ready Checkbox
-    button_display = widgets.HBox([up_button, delete_button])
+    button_display = widgets.HBox([apply_button, delete_button])
     display(widgets.VBox([button_display, wait_ready_check]), output)
 
-    def on_up_button_clicked(b):  # Handle the up button click event
+    def on_apply_button_clicked(b):  # Handle the apply button click event
         with output:
             output.clear_output()
-            cluster.up()
+            cluster.apply()
 
             # If the wait_ready Checkbox is clicked(value == True) trigger the wait_ready function
             if wait_ready_check.value:
@@ -315,7 +315,7 @@ def on_down_button_clicked(b):  # Handle the down button click event
             output.clear_output()
             cluster.down()
 
-    up_button.on_click(on_up_button_clicked)
+    apply_button.on_click(on_apply_button_clicked)
     delete_button.on_click(on_down_button_clicked)
 
 

src/codeflare_sdk/ray/__init__.py

@@ -4,6 +4,13 @@
     RayJobClient,
 )
 
+from .rayjobs import (
+    RayJob,
+    RayJobDeploymentStatus,
+    CodeflareRayJobStatus,
+    RayJobInfo,
+)
+
 from .cluster import (
     Cluster,
     ClusterConfiguration,

src/codeflare_sdk/ray/cluster/build_ray_cluster.py

@@ -136,6 +136,7 @@ def build_ray_cluster(cluster: "codeflare_sdk.ray.cluster.Cluster"):
                 "enableIngress": False,
                 "rayStartParams": {
                     "dashboard-host": "0.0.0.0",
+                    "dashboard-port": "8265",
                     "block": "true",
                     "num-gpus": str(head_gpu_count),
                     "resources": head_resources,
@@ -245,6 +246,7 @@ def get_labels(cluster: "codeflare_sdk.ray.cluster.Cluster"):
     """
     labels = {
         "controller-tools.k8s.io": "1.0",
+        "ray.io/cluster": cluster.config.name,  # Enforced label always present
     }
     if cluster.config.labels != {}:
         labels.update(cluster.config.labels)

src/codeflare_sdk/ray/cluster/cluster.py

@@ -20,8 +20,12 @@
 
 from time import sleep
 from typing import List, Optional, Tuple, Dict
+import copy
 
-from ray.job_submission import JobSubmissionClient
+from ray.job_submission import JobSubmissionClient, JobStatus
+import time
+import uuid
+import warnings
 
 from ...common.kubernetes_cluster.auth import (
     config_check,
@@ -43,7 +47,7 @@
     AppWrapperStatus,
 )
 from ...common.widgets.widgets import (
-    cluster_up_down_buttons,
+    cluster_apply_down_buttons,
     is_notebook,
 )
 from kubernetes import client
@@ -57,7 +61,6 @@
 from kubernetes.client.rest import ApiException
 
 from kubernetes.client.rest import ApiException
-import warnings
 
 CF_SDK_FIELD_MANAGER = "codeflare-sdk"
 
@@ -88,7 +91,7 @@ def __init__(self, config: ClusterConfiguration):
             self.resource_yaml = self.create_resource()
 
         if is_notebook():
-            cluster_up_down_buttons(self)
+            cluster_apply_down_buttons(self)
 
     def get_dynamic_client(self):  # pragma: no cover
         return DynamicClient(get_api_client())
@@ -760,6 +763,7 @@ def get_cluster(
         head_extended_resource_requests=head_extended_resources,
         worker_extended_resource_requests=worker_extended_resources,
     )
+
     # Ignore the warning here for the lack of a ClusterConfiguration
     with warnings.catch_warnings():
         warnings.filterwarnings(

src/codeflare_sdk/ray/cluster/test_cluster.py

@@ -758,5 +758,11 @@ def custom_side_effect(group, version, namespace, plural, **kwargs):
 
 # Make sure to always keep this function last
 def test_cleanup():
-    os.remove(f"{aw_dir}test-all-params.yaml")
-    os.remove(f"{aw_dir}aw-all-params.yaml")
+    # Remove files only if they exist
+    test_file = f"{aw_dir}test-all-params.yaml"
+    if os.path.exists(test_file):
+        os.remove(test_file)
+
+    aw_file = f"{aw_dir}aw-all-params.yaml"
+    if os.path.exists(aw_file):
+        os.remove(aw_file)

src/codeflare_sdk/ray/rayjobs/__init__.py

@@ -0,0 +1,2 @@
+from .rayjob import RayJob
+from .status import RayJobDeploymentStatus, CodeflareRayJobStatus, RayJobInfo