Added local_interactive to cluster-configuration doc

Author: ChristianZaccaria

docs/cluster-configuration.md

@@ -23,8 +23,10 @@ cluster = Cluster(ClusterConfiguration(
     instascale=False, # Default False
     machine_types=["m5.xlarge", "g4dn.xlarge"],
     ingress_domain="example.com" # Default None, Mandatory for Vanilla Kubernetes Clusters - ingress_domain is ignored on OpenShift Clusters as a route is created.
+    local_interactive=False, # Default False
 ))
 ```
+Note: On OpenShift, the `ingress_domain` is only required when `local_interactive` is enabled. - This may change soon.
 
 Upon creating a cluster configuration with `mcad=True` an appwrapper will be created featuring the Ray Cluster and any Routes, Ingresses or Secrets that are needed to be created along side it.<br>
 From there a user can call `cluster.up()` and `cluster.down()` to create and remove the appwrapper thus creating and removing the Ray Cluster.
@@ -36,7 +38,7 @@ The Ray Cluster and service will be created by KubeRay directly and the other co
 To create a Ray Cluster using the CodeFlare SDK in a Vanilla Kubernetes environment an `ingress_domain` must be passed in the Cluster Configuration.
 This is used for the creation of the Ray Dashboard and Client ingresses.
 
-`ingress_options` can be passed to create a custom Ray Dashboard ingress, `ingress_domain` is still a required variable for the Client ingress.
+`ingress_options` can be passed to create a custom Ray Dashboard ingress, `ingress_domain` is still a required variable for the Client route/ingress.
 An example of `ingress_options` would look like this.
 
 ```

tests/e2e/mnist_raycluster_sdk_auth.py

@@ -1,5 +1,6 @@
 import sys
 import os
+import requests
 
 from time import sleep
 
@@ -9,6 +10,63 @@
 from codeflare_sdk.job.jobs import DDPJobDefinition
 from codeflare_sdk.cluster.auth import TokenAuthentication
 
+
+def submit_ddp_job(cluster, cluster_token, server_url):
+    auth = TokenAuthentication(token=cluster_token, server=server_url, skip_tls=True)
+    auth.login()
+
+    jobdef = DDPJobDefinition(
+        name="mnist",
+        script="mnist.py",
+        scheduler_args={"requirements": "requirements.txt"},
+    )
+    job = jobdef.submit(cluster)
+
+    done = False
+    time = 0
+    timeout = 900
+    while not done:
+        status = job.status()
+        if is_terminal(status.state):
+            break
+        if not done:
+            print(status)
+            if timeout and time >= timeout:
+                raise TimeoutError(f"job has timed out after waiting {timeout}s")
+            sleep(5)
+            time += 5
+
+    print(f"Job has completed: {status.state}")
+    print(job.logs())
+    cluster.down()
+
+    if not status.state == AppState.SUCCEEDED:
+        exit(1)
+    else:
+        exit(0)
+
+
+def submit_job_post_request(cluster, dashboard_url):
+    jobdata = {
+        "entrypoint": "python mnist.py",
+        "runtime_env": {"working_dir": "workdir", "pip": "requirements.txt"},
+    }
+    try:
+        response = requests.post(
+            dashboard_url + "/api/jobs/", verify=False, json=jobdata
+        )
+        cluster.down()
+        if response.status_code == 403:
+            exit(0)
+        else:
+            response.raise_for_status()
+            exit(1)
+
+    except Exception as e:
+        print(f"An unexpected error occurred: {e}")
+        exit(1)
+
+
 namespace = sys.argv[1]
 ray_image = os.getenv("RAY_IMAGE")
 isAuthLoginEnable = os.getenv("OAUTH_LOGIN")
@@ -33,10 +91,6 @@
     )
 )
 
-if isAuthLoginEnable == "true":
-    auth = TokenAuthentication(token=cluster_token, server=server_url, skip_tls=True)
-    auth.login()
-
 cluster.up()
 
 cluster.status()
@@ -47,34 +101,8 @@
 
 cluster.details()
 
-jobdef = DDPJobDefinition(
-    name="mnist",
-    script="mnist.py",
-    scheduler_args={"requirements": "requirements.txt"},
-)
-job = jobdef.submit(cluster)
-
-done = False
-time = 0
-timeout = 900
-while not done:
-    status = job.status()
-    if is_terminal(status.state):
-        break
-    if not done:
-        print(status)
-        if timeout and time >= timeout:
-            raise TimeoutError(f"job has timed out after waiting {timeout}s")
-        sleep(5)
-        time += 5
-
-print(f"Job has completed: {status.state}")
-
-print(job.logs())
-
-cluster.down()
-
-if not status.state == AppState.SUCCEEDED:
-    exit(1)
+if isAuthLoginEnable == "true":
+    submit_ddp_job(cluster, cluster_token, server_url)
 else:
-    exit(0)
+    dashboard_url = cluster.cluster_dashboard_uri()
+    submit_job_post_request(cluster, dashboard_url)

tests/e2e/mnist_raycluster_sdk_auth_test.go

@@ -17,11 +17,8 @@ limitations under the License.
 package e2e
 
 import (
-	"bytes"
 	"testing"
-	"time"
 
-	"github.com/onsi/gomega"
 	. "github.com/onsi/gomega"
 	. "github.com/project-codeflare/codeflare-common/support"
 	mcadv1beta1 "github.com/project-codeflare/multi-cluster-app-dispatcher/pkg/apis/controller/v1beta1"
@@ -33,7 +30,7 @@ import (
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 )
 
-// This test covers the Ray Cluster creation authentication functionality of openshift_oauth without login and assert the warning  only authorised users can submit jobs
+// This test covers the Ray Cluster creation authentication functionality of openshift_oauth without login and assert the Unauthorized 403 error
 func TestMNISTRayClusterSDKWithoutAuthLogin(t *testing.T) {
 	test := With(t)
 	test.T().Parallel()
@@ -44,6 +41,7 @@ func TestMNISTRayClusterSDKWithoutAuthLogin(t *testing.T) {
 
 	// Create a namespace
 	namespace := test.NewTestNamespace()
+
 	// Test configuration
 	config := CreateConfigMap(test, namespace.Name, map[string][]byte{
 		// SDK script
@@ -69,7 +67,7 @@ func TestMNISTRayClusterSDKWithoutAuthLogin(t *testing.T) {
 			Resources: []string{"rayclusters", "rayclusters/status"},
 		},
 		{
-			Verbs:     []string{"get", "create", "delete", "list", "patch", "update"},
+			Verbs:     []string{"get", "list", "create", "update", "delete"},
 			APIGroups: []string{"route.openshift.io"},
 			Resources: []string{"routes"},
 		},
@@ -83,13 +81,7 @@ func TestMNISTRayClusterSDKWithoutAuthLogin(t *testing.T) {
 	// Create cluster wide RBAC, required for SDK OpenShift Oauth check
 	clusterPolicyRules := []rbacv1.PolicyRule{
 		{
-			Verbs:         []string{"get", "list"},
-			APIGroups:     []string{"config.openshift.io"},
-			Resources:     []string{"ingresses"},
-			ResourceNames: []string{"cluster"},
-		},
-		{
-			Verbs:     []string{"create", "update"},
+			Verbs:     []string{"create", "update", "delete"},
 			APIGroups: []string{""},
 			Resources: []string{"services", "serviceaccounts"},
 		},
@@ -108,6 +100,11 @@ func TestMNISTRayClusterSDKWithoutAuthLogin(t *testing.T) {
 			APIGroups: []string{"authentication.k8s.io"},
 			Resources: []string{"tokenreviews"},
 		},
+		{
+			Verbs:     []string{"get"},
+			APIGroups: []string{""},
+			Resources: []string{"pods"},
+		},
 	}
 	sa := CreateServiceAccount(test, namespace.Name)
 	role := CreateRole(test, namespace.Name, policyRules)
@@ -210,10 +207,16 @@ func TestMNISTRayClusterSDKWithoutAuthLogin(t *testing.T) {
 	// Setup the codeflare-sdk inside the pod associated to the created job
 	SetupCodeflareSDKInsidePod(test, namespace, job.Name)
 
-	test.T().Logf("Checking pod logs contains insecure requests warning")
-	// Assert the job pod as warning of insecurerequests
-	assertPodLogs(test, namespace, job.Name, "InsecureRequestWarning: Unverified HTTPS request is being made to host", 3*time.Minute)
+	test.T().Logf("Waiting for Job %s/%s to complete", job.Namespace, job.Name)
+	test.Eventually(Job(test, job.Namespace, job.Name), TestTimeoutLong).Should(
+		Or(
+			WithTransform(ConditionStatus(batchv1.JobComplete), Equal(corev1.ConditionTrue)),
+			WithTransform(ConditionStatus(batchv1.JobFailed), Equal(corev1.ConditionTrue)),
+		))
 
+	// Assert the job has completed successfully
+	test.Expect(GetJob(test, job.Namespace, job.Name)).
+		To(WithTransform(ConditionStatus(batchv1.JobComplete), Equal(corev1.ConditionTrue)))
 }
 
 // This test covers the Ray Cluster creation authentication functionality with openshift_oauth login and check for Ray job is successfull
@@ -264,12 +267,6 @@ func TestMNISTRayClusterSDKWithAuthLogin(t *testing.T) {
 
 	// Create cluster wide RBAC, required for SDK OpenShift Oauth check
 	clusterPolicyRules := []rbacv1.PolicyRule{
-		{
-			Verbs:         []string{"get", "list"},
-			APIGroups:     []string{"config.openshift.io"},
-			Resources:     []string{"ingresses"},
-			ResourceNames: []string{"cluster"},
-		},
 		{
 			Verbs:     []string{"create", "update", "delete"},
 			APIGroups: []string{""},
@@ -414,26 +411,3 @@ func TestMNISTRayClusterSDKWithAuthLogin(t *testing.T) {
 	test.Expect(GetJob(test, job.Namespace, job.Name)).
 		To(WithTransform(ConditionStatus(batchv1.JobComplete), Equal(corev1.ConditionTrue)))
 }
-
-func assertPodLogs(test Test, namespace *corev1.Namespace, labelName string, expectedLogSubstring string, timeout time.Duration) {
-
-	// Get pod
-	pod := GetPod(test, namespace, labelName)
-
-	startTime := time.Now()
-	for {
-		// Get the current pod logs
-		podLogs := GetPodLogs(test, pod, corev1.PodLogOptions{})
-
-		// Check if the expected log substring is present
-		if bytes.Contains(podLogs, []byte(expectedLogSubstring)) {
-			test.Expect(podLogs).To(gomega.ContainSubstring(expectedLogSubstring))
-			break
-		}
-
-		if time.Since(startTime) > timeout {
-			test.T().Fatal("Timeout waiting for expected Incorrect Requests Warning log not found!")
-		}
-		time.Sleep(5 * time.Second)
-	}
-}