consumer: remove pods for deleted workflows

closes reanahub#437

Author: Vladyslav Moisieienkov

reana_workflow_controller/consumer.py

@@ -31,7 +31,7 @@
     build_unique_component_name,
 )
 from reana_db.database import Session
-from reana_db.models import Job, JobCache, Workflow, RunStatus
+from reana_db.models import Job, JobCache, Workflow, RunStatus, JobStatus
 from sqlalchemy.exc import SQLAlchemyError
 from sqlalchemy.orm.attributes import flag_modified
 
@@ -118,11 +118,21 @@ def on_message(self, body, message):
                     )
             elif workflow and workflow.status not in ALIVE_STATUSES:
                 logging.warning(
-                    f"Event for not alive workflow {workflow.id_} with DB status {workflow.status} received:\n"
-                    f"{body}\nIgnoring..."
+                    f"Event for not alive workflow {workflow.id_} with DB status {workflow.status} received."
                 )
+                try:
+                    _delete_workflow_batch_pod(workflow)
+                    logging.info(
+                        f"Remove batch-pod for not alive {workflow.id_} workflow."
+                    )
+                except REANAWorkflowControllerError as exception:
+                    logging.error(
+                        f"Could not clean up not alive workflow {workflow.id_} batch pod for workflow."
+                        f" Error: {exception}"
+                    )
+                _delete_workflow_jobs(workflow)
             else:
-                logging.warning(
+                logging.error(
                     f"Event for workflow {workflow_uuid} that doesn't exist in DB received:\n"
                     f"{body}\nIgnoring..."
                 )
@@ -162,7 +172,7 @@ def _update_workflow_status(workflow, status, logs):
 
             if RunStatus.should_cleanup_job(status):
                 try:
-                    _delete_workflow_job(workflow)
+                    _delete_workflow_batch_pod(workflow)
                 except REANAWorkflowControllerError as exception:
                     logging.error(
                         f"Could not clean up workflow job for workflow {workflow.id_}."
@@ -281,7 +291,7 @@ def _update_job_cache(msg):
     Session.add(cached_job)
 
 
-def _delete_workflow_job(workflow: Workflow) -> None:
+def _delete_workflow_batch_pod(workflow: Workflow) -> None:
     job_name = build_unique_component_name("run-batch", workflow.id_)
     try:
         current_k8s_batchv1_api_client.delete_namespaced_job(
@@ -295,6 +305,36 @@ def _delete_workflow_job(workflow: Workflow) -> None:
         )
 
 
+def _delete_workflow_jobs(workflow: Workflow) -> None:
+    """
+    Delete jobs that belong to workflow. Ignore Kubernetes API errors.
+    If job is successfully deleted, update its status to JobStatus.stopped.
+    """
+    jobs = Session.query(Job).filter(
+        Job.workflow_uuid == workflow.id_,
+    )
+    for job in jobs:
+        job_name = build_unique_component_name("run-job", job.id_)
+        if job.status in [
+            JobStatus.running,
+            JobStatus.queued,
+            JobStatus.created,
+        ]:
+            try:
+                current_k8s_batchv1_api_client.delete_namespaced_job(
+                    name=job_name,
+                    namespace=REANA_RUNTIME_KUBERNETES_NAMESPACE,
+                    propagation_policy="Background",
+                )
+                job.status = JobStatus.stopped
+                Session.add(job)
+            except ApiException as e:
+                logging.error(
+                    f"run-job pod {job_name} for {workflow.id_} could not be deleted. Error: {e}"
+                )
+    Session.commit()
+
+
 def _get_workflow_engine_pod_logs(workflow: Workflow) -> str:
     try:
         pods = current_k8s_corev1_api_client.list_namespaced_pod(