[
https://issues.apache.org/jira/browse/YUNIKORN-966?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17523284#comment-17523284
]
ted commented on YUNIKORN-966:
------------------------------
Hi [~yuchaoran2011],
Thank you for the explanation.
Yes, I think that the code is already behaving as expected without implementing
this JIRA.
I don't understand SparkApp CRD user name.
I would like some clarifications. Thank you.
When I look at this: kubectl describe pods spark-pi-driver
I found that can observe the spark user here:
{code:java}
Environment:
SPARK_USER: ted{code}
Then I read the api-docs of spark-on-k8s-operator and there is a parameter in
it: proxyUser
I found that when proxyUser is set, SPARK_USER will be the same as proxyUser
When proxyUser is not set, SPARK_USER will be root.
So proxyUser should be the user name of SparkApp CRD?
Then now yunikorn has no way to read proxyUser, now we have to find a way to
read proxyUser, and then use proxyUser as the first choice.
> Retrieve the username from the SparkApp CRD
> -------------------------------------------
>
> Key: YUNIKORN-966
> URL: https://issues.apache.org/jira/browse/YUNIKORN-966
> Project: Apache YuniKorn
> Issue Type: Sub-task
> Components: shim - kubernetes
> Reporter: Chaoran Yu
> Assignee: ted
> Priority: Minor
>
> Currently the shim only looks at the pods to get the value of the label
> yunikorn.apache.org/username. When the Spark operator plugin is enabled, we
> should look at the SparkApp CRD for the label.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@yunikorn.apache.org
For additional commands, e-mail: issues-h...@yunikorn.apache.org
