dimberman commented on issue #10037:
URL: https://github.com/apache/airflow/issues/10037#issuecomment-692882853
Hmm this is an interesting one.
I think when we made the pod_template_file we did so with the assumption
that users would have ALL specs in the pod_template file and not attempt to
modify further. If you have a pod_template_file that you want to modify before
launching, you can deserialize it using the PodGenerator and then use the `pod`
parameter for the K8sPodOperator.
Something like this
```
from datetime import datetime, timedelta
from airflow.kubernetes.pod_generator import PodGenerator
from airflow.models import DAG
from airflow.contrib.operators.kubernetes_pod_operator import
KubernetesPodOperator
default_args = {
"owner": "test",
"start_date": datetime(2019, 10, 17),
"retries": 1,
"retry_delay": timedelta(minutes=5),
}
pod = PodGenerator.deserialize_model_file("path/to/dags/minimal.yaml")
pod.metadata.namespace="test"
with DAG("bug-report", schedule_interval=None, default_args=default_args) as
dag:
KubernetesPodOperator(
task_id="bug-report-task",
is_delete_operator_pod=False,
in_cluster=True,
get_logs=True,
pod=pod,
)
```
If we documented this use-case better, would this be a reasonable solution
to the problem?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
