[Merged by Bors] - Airflow demo with spark-k8s

demos/airflow-scheduled-job/01-airflow-spark-clusterrole.yaml

@@ -0,0 +1,36 @@
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: airflow-spark-clusterrole
+rules:
+- apiGroups:
+  - spark.stackable.tech
+  resources:
+  - sparkapplications
+  verbs:
+  - create
+  - get
+  - list
+- apiGroups:
+  - airflow.stackable.tech
+  resources:
+  - airflowdbs
+  verbs:
+  - create
+  - get
+  - list
+- apiGroups:
+  - apps
+  resources:
+  - statefulsets
+  verbs:
+  - get
+  - watch
+  - list
+- apiGroups:
+  - ""
+  resources:
+  - persistentvolumeclaims
+  verbs:
+  - list

demos/airflow-scheduled-job/02-airflow-spark-clusterrolebinding.yaml

@@ -0,0 +1,13 @@
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRoleBinding
+metadata:
+  name: airflow-spark-clusterrole-binding
+roleRef:
+  apiGroup: rbac.authorization.k8s.io
+  kind: ClusterRole
+  name: airflow-spark-clusterrole
+subjects:
+- apiGroup: rbac.authorization.k8s.io
+  kind: Group
+  name: system:serviceaccounts

demos/airflow-scheduled-job/03-enable-and-run-spark-dag.yaml

@@ -0,0 +1,26 @@
+---
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: start-pyspark-job
+spec:
+  template:
+    spec:
+      containers:
+        - name: start-pyspark-job
+          image: docker.stackable.tech/stackable/tools:0.2.0-stackable0.3.0
+          # N.B. it is possible for the scheduler to report that a DAG exists, only for the worker task to fail if a pod is unexpectedly
+          # restarted. Additionally, the db-init job takes a few minutes to complete before the cluster is deployed. The wait/watch steps
+          # below are not "water-tight" but add a layer of stability by at least ensuring that the db is initialized and ready and that
+          # all pods are reachable (albeit independent of each other).
+          command: ["bash", "-c", "
+          kubectl wait airflowdb/airflow --for jsonpath='{.status.condition}'=Ready --timeout 600s
+          && kubectl rollout status --watch statefulset/airflow-webserver-default
+          && kubectl rollout status --watch statefulset/airflow-worker-default
+          && kubectl rollout status --watch statefulset/airflow-scheduler-default
+          && curl -i -s --user airflow:airflow http://airflow-webserver-default:8080/api/v1/dags/sparkapp_dag
+          && curl -i -s --user airflow:airflow -H 'Content-Type:application/json' -XPATCH http://airflow-webserver-default:8080/api/v1/dags/sparkapp_dag -d '{\"is_paused\": false}'
+          && curl -i -s --user airflow:airflow -H 'Content-Type:application/json' -XPOST http://airflow-webserver-default:8080/api/v1/dags/sparkapp_dag/dagRuns -d '{}'
+          "]
+      restartPolicy: OnFailure
+  backoffLimit: 20 # give some time for the Airflow cluster to be available

demos/airflow-scheduled-job/04-enable-and-run-date-dag.yaml

@@ -0,0 +1,26 @@
+---
+apiVersion: batch/v1
+kind: Job
+metadata:
+  name: start-date-job
+spec:
+  template:
+    spec:
+      containers:
+        - name: start-date-job
+          image: docker.stackable.tech/stackable/tools:0.2.0-stackable0.3.0
+          # N.B. it is possible for the scheduler to report that a DAG exists, only for the worker task to fail if a pod is unexpectedly
+          # restarted. Additionally, the db-init job takes a few minutes to complete before the cluster is deployed. The wait/watch steps
+          # below are not "water-tight" but add a layer of stability by at least ensuring that the db is initialized and ready and that
+          # all pods are reachable (albeit independent of each other).
+          command: ["bash", "-c", "
+          kubectl wait airflowdb/airflow --for jsonpath='{.status.condition}'=Ready --timeout 600s
+          && kubectl rollout status --watch statefulset/airflow-webserver-default
+          && kubectl rollout status --watch statefulset/airflow-worker-default
+          && kubectl rollout status --watch statefulset/airflow-scheduler-default
+          && curl -i -s --user airflow:airflow http://airflow-webserver-default:8080/api/v1/dags/date_demo
+          && curl -i -s --user airflow:airflow -H 'Content-Type:application/json' -XPATCH http://airflow-webserver-default:8080/api/v1/dags/date_demo -d '{\"is_paused\": false}'
+          && curl -i -s --user airflow:airflow -H 'Content-Type:application/json' -XPOST http://airflow-webserver-default:8080/api/v1/dags/date_demo/dagRuns -d '{}'
+          "]
+      restartPolicy: OnFailure
+  backoffLimit: 20 # give some time for the Airflow cluster to be available

demos/demos-v1.yaml

@@ -7,7 +7,10 @@ demos:
       - airflow
       - job-scheduling
     manifests:
-      - plainYaml: https://raw.githubusercontent.com/stackabletech/stackablectl/main/demos/airflow-scheduled-job/enable-and-run-dag.yaml
+      - plainYaml: https://raw.githubusercontent.com/stackabletech/stackablectl/main/demos/airflow-scheduled-job/01-airflow-spark-clusterrole.yaml
+      - plainYaml: https://raw.githubusercontent.com/stackabletech/stackablectl/main/demos/airflow-scheduled-job/02-airflow-spark-clusterrolebinding.yaml
+      - plainYaml: https://raw.githubusercontent.com/stackabletech/stackablectl/main/demos/airflow-scheduled-job/03-enable-and-run-spark-dag.yaml
+      - plainYaml: https://raw.githubusercontent.com/stackabletech/stackablectl/main/demos/airflow-scheduled-job/04-enable-and-run-date-dag.yaml
   hbase-hdfs-load-cycling-data:
     description: Copy data from S3 bucket to an HBase table
     stackableStack: hdfs-hbase

docs/modules/ROOT/pages/demos/airflow-scheduled-job.adoc

@@ -2,9 +2,9 @@
 
 [NOTE]
 ====
-This guide assumes you already have the demo `airflow-cron-dag` installed.
+This guide assumes you already have the demo `airflow-scheduled-job` installed.
 If you don't have it installed please follow the xref:commands/demo.adoc#_install_demo[documentation on how to install a demo].
-To put it simply you have to run `stackablectl demo install airflow-cron-dag`.
+To put it simply you have to run `stackablectl demo install airflow-scheduled-job`.
 ====
 
 This demo will
@@ -14,8 +14,8 @@ This demo will
 ** *Postgresql*: An open-source database used for Airflow cluster- and job-metadata.
 ** *Redis*: An in-memory data structure store used for queuing Airflow jobs
 ** *Airflow*: An open-source workflow management platform for data engineering pipelines.
-* Mount an Airflow job (referred to as a Directed Acyclic Graph, or DAG) for the cluster to use
-* Enable and schedule the job
+* Mount two Airflow jobs (referred to as Directed Acyclic Graphs, or DAGs) for the cluster to use
+* Enable and schedule the jobs
 * Verify the job status with the Airflow Webserver UI
 
 You can see the deployed products as well as their relationship in the following diagram:
@@ -27,10 +27,11 @@ To list the installed Stackable services run the following command:
 
 [source,console]
 ----
-$ stackablectl services list --all-namespaces
+$ stackablectl services list
  PRODUCT  NAME     NAMESPACE  ENDPOINTS                                  EXTRA INFOS
 
- airflow  airflow  default    webserver-airflow http://172.18.0.4:32754  Admin user: airflow, password: airflow
+ airflow  airflow  default    webserver-airflow http://172.18.0.2:31979  Admin user: airflow, password: airflow
+
 ----
 
 [NOTE]
@@ -42,29 +43,29 @@ In case the product is not ready yet a warning might be shown.
 
 == Airflow Webserver UI
 Superset gives the ability to execute SQL queries and build dashboards.
-Open the `airflow` endpoint `webserver-airflow` in your browser (`http://172.18.0.4:32754` in this case).
+Open the `airflow` endpoint `webserver-airflow` in your browser (`http://172.18.0.2:31979` in this case).
 
 image::demo-airflow-scheduled-job/airflow_1.png[]
 
-Log in with the username `airflow` and password `airflow`. The overview screen shows the DAG that has been mounted during the demo set-up (`date_demo`).
+Log in with the username `airflow` and password `airflow`. The overview screen shows the DAGs that have been mounted during the demo set-up (`date_demo`).
 
 image::demo-airflow-scheduled-job/airflow_2.png[]
 
-There are two things to notice here. The DAG has been enabled, as shown by the slider to the left of the DAG name (DAGs are initially all `paused` and can be activated either manually in the UI or via a REST call, as done in the setup for this demo):
+There are two things to notice here. Both DAGs have been enabled, as shown by the slider to the left of the DAG name (DAGs are initially all `paused` and can be activated either manually in the UI or via a REST call, as done in the setup for this demo):
 
 image::demo-airflow-scheduled-job/airflow_3.png[]
 
-Secondly, the job has been busy, with several runs already logged!
+Secondly, the `date_demo` job has been busy, with several runs already logged! The `sparkapp_dag` has only been run once. This is because they have been defined with different schedules.
 
 image::demo-airflow-scheduled-job/airflow_4.png[]
 
-Clicking on this number will display the individual job runs:
+Clicking on the number under `Runs` will display the individual job runs:
 
 image::demo-airflow-scheduled-job/airflow_5.png[]
 
-The job is running every minute. With Airflow, DAGs can be started manually or scheduled to run when certain conditions are fulfilled: in this case the DAG has been set up to run using a cron table, which is part of the DAG definition.
+The `demo_date` job is running every minute. With Airflow, DAGs can be started manually or scheduled to run when certain conditions are fulfilled: in this case the DAG has been set up to run using a cron table, which is part of the DAG definition.
 
-=== View DAG details
+=== `demo_date` DAG
 
 Let's drill down a bit deeper into this DAG. Click on one of the individual job runs shown in the previous step to display the job details. The DAG is displayed in the form of a graph (this job is so simple that it only has one step, called `run_every_minute`).
 
@@ -86,8 +87,26 @@ To look at the actual DAG code click on `Code`. Here we can see the crontab info
 
 image::demo-airflow-scheduled-job/airflow_10.png[]
 
+=== `sparkapp_dag` DAG
+
+Go back to DAG overview screen. The `sparkapp_dag` job has a scheduled entry of `None` and a last-execution time (`2022-09-19, 07:36:55`). This allows a DAG to be executed exactly once, with neither schedule-based runs nor any https://airflow.apache.org/docs/apache-airflow/stable/dag-run.html?highlight=backfill#backfill[backfill]. The DAG can always be triggered manually again via REST or from within the Webserver UI.
+
+image::demo-airflow-scheduled-job/airflow_11.png[]
+
+By navigating to the graphical overview of the job we can see that DAG has two steps, one to start the job - which runs asynchronously - and another to poll the running job to report on its status.
+
+image::demo-airflow-scheduled-job/airflow_12.png[]
+
+The logs for the first task - `spark-pi-submit` - indicate that it has been started, at which point the task exits without any further information:
+
+image::demo-airflow-scheduled-job/airflow_13.png[]
+
+The second task - `spark-pi-monitor` - polls this job and waits for a final result (in this case: `Success`). In this case, the actual result of the job (a value of `pi`) is logged by Spark in its driver pod, but more sophisticated jobs would persist this in a sink (e.g. a Kafka topic or HBase row) or use the result to trigger subsequent actions.
+
+image::demo-airflow-scheduled-job/airflow_14.png[]
+
 == Summary
 
-This demo showed how a DAG can be made available for Airflow, scheduled, run and then inspected with the Webserver UI.
+This demo showed how DAGs can be made available for Airflow, scheduled, run and then inspected with the Webserver UI.