set endpoints via extra config

Author: adwk67

stacks/jupyterhub-keycloak/jupyterhub.yaml

@@ -14,10 +14,6 @@ options:
       GenericOAuthenticator:
         client_id: jupyterhub
         client_secret: jupyterhubjupyterhub
-        oauth_callback_url: http://172.19.0.4:31095/hub/oauth_callback
-        authorize_url: https://172.19.0.4:31093/realms/demo/protocol/openid-connect/auth
-        token_url: https://172.19.0.4:31093/realms/demo/protocol/openid-connect/token
-        userdata_url: https://172.19.0.4:31093/realms/demo/protocol/openid-connect/userinfo
         username_claim: preferred_username
         scope:
           - openid
@@ -27,6 +23,16 @@ options:
       CACERT: /etc/ssl/certs/ca-certificates.crt
       CERT: /etc/ssl/certs/ca-certificates.crt
       CURLOPT_CAINFO: /etc/ssl/certs/ca-certificates.crt
+      KEYCLOAK_NODEPORT_URL:
+        valueFrom:
+          configMapKeyRef:
+            name: keycloak-address
+            key: keycloakAddress
+      KEYCLOAK_NODE_IP:
+        valueFrom:
+          configMapKeyRef:
+            name: keycloak-address
+            key: keycloakNodeIp
     extraVolumes:
       - name: tls-ca-cert
         ephemeral:
@@ -100,6 +106,22 @@ options:
           )
 
         c.KubeSpawner.after_pod_created_hook = after_pod_created_hook
+      03-set-endpoints: |
+        import os
+        from oauthenticator.generic import GenericOAuthenticator
+
+        keycloak_url = os.getenv("KEYCLOAK_NODEPORT_URL")
+        if not keycloak_url:
+            raise ValueError("KEYCLOAK_NODEPORT_URL environment variable not set")
+
+        keycloak_node_ip = os.getenv("KEYCLOAK_NODE_IP")
+        if not keycloak_node_ip:
+            raise ValueError("KEYCLOAK_NODE_IP environment variable not set")
+
+        c.GenericOAuthenticator.oauth_callback_url: f"http://{keycloak_node_ip}:31095/hub/oauth_callback"
+        c.GenericOAuthenticator.authorize_url = f"https://{keycloak_url}/realms/demo/protocol/openid-connect/auth"
+        c.GenericOAuthenticator.token_url = f"https://{keycloak_url}/realms/demo/protocol/openid-connect/token"
+        c.GenericOAuthenticator.userdata_url = f"https://{keycloak_url}/realms/demo/protocol/openid-connect/userinfo"
     service:
       type: NodePort
   proxy:

stacks/jupyterhub-keycloak/keycloak.yaml

@@ -74,7 +74,8 @@ spec:
                     metadata:
                       name: keycloak-address
                     data:
-                      address: "$KEYCLOAK_HOSTNAME:$KEYCLOAK_PORT"
+                      keycloakAddress: "$KEYCLOAK_HOSTNAME:$KEYCLOAK_PORT"
+                      keycloakNodeIp: "$KEYCLOAK_HOSTNAME"
               EOF
                 fi
 

stacks/jupyterhub-keycloak/process-s3.ipynb

@@ -43,33 +43,12 @@
     "NAMESPACE = os.environ.get(\"NAMESPACE\", \"default\")\n",
     "POD_NAME = os.environ.get(\"HOSTNAME\", f\"jupyter-{os.environ.get('USER', 'default')}-{NAMESPACE}\")\n",
     "\n",
-    "# works with python-3.11 notebook image\n",
-    "#EXECUTOR_IMAGE = \"oci.stackable.tech/sdp/spark-k8s:3.5.0-stackable24.3.0\" \n",
-    "\n",
-    "# jars differ in size, 17.0.12 vs. 17.0.13, 3.11.10 vs. 3.11.9\n",
-    "#SerializableBuffer conflict\n",
-    "#EXECUTOR_IMAGE = \"oci.stackable.tech/sdp/spark-k8s:3.5.2-stackable24.11.1\" \n",
-    "\n",
-    "# java, jars match\n",
-    "#Python in worker has different version (3, 10) than that in driver 3.11, PySpark cannot run with different minor versions. \n",
-    "#EXECUTOR_IMAGE = \"apache/spark:3.5.2-java17-python3\" \n",
-    "\n",
-    "# java, jars match\n",
-    "#Python in worker has different version (3, 10) than that in driver 3.11, PySpark cannot run with different minor versions. \n",
-    "#EXECUTOR_IMAGE = \"spark:3.5.2-scala2.12-java17-python3-ubuntu\" \n",
-    "\n",
-    "#Python in worker has different version (3, 12) than that in driver 3.11, PySpark cannot run with different minor versions.\n",
-    "#EXECUTOR_IMAGE = \"bitnami/spark:3.5.2\"\n",
-    "\n",
-    "# custom image with python 3.11 - works!\n",
-    "# based off: spark:3.5.2-scala2.12-java17-ubuntu\n",
-    "# see: \n",
     "EXECUTOR_IMAGE = \"spark:3.5.2-python311\" \n",
     "\n",
     "spark = (\n",
     "    SparkSession.builder\n",
     "    .master(f'k8s://https://{os.environ[\"KUBERNETES_SERVICE_HOST\"]}:{os.environ[\"KUBERNETES_SERVICE_PORT\"]}')\n",
-    "    .appName(\"process-s3-data\")\n",
+    "    .appName(\"process-s3-data-\"{os.environ.get('USER', 'default')}-{NAMESPACE})\n",
     "    .config(\"spark.kubernetes.container.image\", EXECUTOR_IMAGE)\n",
     "    .config(\"spark.kubernetes.container.image.pullPolicy\", \"IfNotPresent\")\n",
     "    .config(\"spark.kubernetes.namespace\", NAMESPACE)\n",
@@ -80,8 +59,6 @@
     "    .config(\"spark.executor.instances\", \"1\")\n",
     "    .config(\"spark.executor.memory\", \"1g\")\n",
     "    .config(\"spark.executor.cores\", \"1\")\n",
-    "    # bitnami. See https://github.com/bitnami/containers/issues/52698#issuecomment-2275913474\n",
-    "    #.config(\"spark.executorEnv.LD_PRELOAD\", \"/opt/bitnami/common/lib/libnss_wrapper.so\")\n",
     "    .config(\"spark.hadoop.fs.s3a.endpoint\", \"http://minio:9000/\")\n",
     "    .config(\"spark.hadoop.fs.s3a.path.style.access\", \"true\")\n",
     "    .config(\"spark.hadoop.fs.s3a.access.key\", minio_user)\n",