Skip to content

Conversation

@hussein-awala
Copy link
Member

This PR fixes a bug introduced by #32627, where we try to get the state from the database when it's None in the executor, and for some reason, it's None in the database(maybe the task is cleared or its state is set manually to None...); in this case, we cannot convert it to TaskInstanceState, and we should use None as before the mentioned PR.

closes: #35888

@boring-cyborg boring-cyborg bot added area:providers provider:cncf-kubernetes Kubernetes (k8s) provider related issues labels Nov 27, 2023
@hussein-awala hussein-awala merged commit 623f989 into apache:main Nov 27, 2023
@tirkarthi
Copy link
Contributor

Thanks @hussein-awala for the quick fix. We will try to test this patch in next Kubernetes provider release since this missed the current RC provider candidate.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

area:providers provider:cncf-kubernetes Kubernetes (k8s) provider related issues

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Infinite loop on scheduler when kubernetes state event is None along with state in database also None

4 participants