You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@submarine.apache.org by "Che-Kuang Chu (Jira)" <ji...@apache.org> on 2021/08/02 14:13:00 UTC
[jira] [Updated] (SUBMARINE-954) Submarine experiment and notebook
pods are still alive after the server is down
[ https://issues.apache.org/jira/browse/SUBMARINE-954?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Che-Kuang Chu updated SUBMARINE-954:
------------------------------------
Description:
After the submarine-server is down, we can find that the experiment and notebook pods are still alive in the cluster.
Reproduce steps:
# helm install submarine ./helm-charts/submarine
# access to the workbench by port forwarding
# add notebook or experiment from the workbench
# helm uninstall submarine
# kubectl get pod, and we can find that the pods are still in the cluster
To fix the bug, we can maybe add onwerReferences to the created resources.
Reference:
* ownerReference: *[https://kubernetes.io/docs/concepts/overview/working-with-objects/owners-dependents/]*
was:
After the submarine-server is down, we can find that the experiment and notebook pods are still alive in the cluster.
Reproduce steps:
# helm install submarine ./helm-charts/submarine
# access to the workbench by port forwarding
# add notebook or experiment from the workbench
# helm uninstall submarine
# kubectl get pod, and we can find that the pods are still in the cluster
To fix the bug, we can maybe add onwerReferences to the created resources.
Reference:
* ownerReference: *[https://kubernetes.io/docs/concepts/overview/working-with-objects/owners-dependents/]***
> Submarine experiment and notebook pods are still alive after the server is down
> -------------------------------------------------------------------------------
>
> Key: SUBMARINE-954
> URL: https://issues.apache.org/jira/browse/SUBMARINE-954
> Project: Apache Submarine
> Issue Type: Bug
> Components: Backend Server
> Reporter: Che-Kuang Chu
> Assignee: Che-Kuang Chu
> Priority: Major
>
> After the submarine-server is down, we can find that the experiment and notebook pods are still alive in the cluster.
> Reproduce steps:
> # helm install submarine ./helm-charts/submarine
> # access to the workbench by port forwarding
> # add notebook or experiment from the workbench
> # helm uninstall submarine
> # kubectl get pod, and we can find that the pods are still in the cluster
> To fix the bug, we can maybe add onwerReferences to the created resources.
> Reference:
> * ownerReference: *[https://kubernetes.io/docs/concepts/overview/working-with-objects/owners-dependents/]*
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@submarine.apache.org
For additional commands, e-mail: dev-help@submarine.apache.org