[
https://issues.apache.org/jira/browse/AIRFLOW-2971?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16945954#comment-16945954
]
Colin Campbell commented on AIRFLOW-2971:
-----------------------------------------
I don't think the server health check really resolves this, since it can't be
used as a health check for scheduler pods in a k8s environment. This feature
would be useful for command based health/liveness checks.
https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/#define-a-liveness-command
> Health check command for scheduler
> ----------------------------------
>
> Key: AIRFLOW-2971
> URL: https://issues.apache.org/jira/browse/AIRFLOW-2971
> Project: Apache Airflow
> Issue Type: Bug
> Components: scheduler
> Reporter: Jon Davies
> Priority: Major
>
> As part of a Kubernetes deployment of Airflow, I would like to define an exec
> command based health check for the Airflow scheduler:
> -
> https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-probes/
> ...the webserver is simple as all that needs is checking that the HTTP port
> is available. For the scheduler, it would be neat to have a command such as:
> airflow scheduler health
> That returned OK and exit 0/NOT OK and a non-zero value when it cannot reach
> the database for instance.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
