[ 
https://issues.apache.org/jira/browse/YUNIKORN-2009?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Wilfred Spiegelenburg resolved YUNIKORN-2009.
---------------------------------------------
    Fix Version/s: 1.4.0
       Resolution: Fixed

Due to the way the LimitRange controller handles default limits the current 
placeholder create could break. In YUNIKORN-1908 limits for extended resources 
were added. To support LimitRange with defaults set we need to add limits to 
all resources, not just extended resources.

Otherwise the LimitRange controller could update placeholder pods and leave 
them with an unschedulable config.

committed, thank you for the changes

> Placeholder pods should be created with limits
> ----------------------------------------------
>
>                 Key: YUNIKORN-2009
>                 URL: https://issues.apache.org/jira/browse/YUNIKORN-2009
>             Project: Apache YuniKorn
>          Issue Type: Improvement
>          Components: shim - kubernetes
>            Reporter: Wei Huang
>            Assignee: Wei Huang
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 1.4.0
>
>
> In some k8s env, namespace is enabled with {*}{{LimitRange}}{*}:
> {code:java}
> apiVersion: v1
> kind: LimitRange
> metadata:
>   name: default
> spec:
>   limits:
>   - default:
>       cpu: 100m
>       ephemeral-storage: 10Gi
>       memory: 128Mi
>     defaultRequest:
>       cpu: 100m
>       ephemeral-storage: 1Gi
>       memory: 128Mi
>     type: Container{code}
> However, shim code spawns placeholder pods with requests only, which cause 
> the placeholder pods rejected by API server, and hence the whole Job gets 
> into pending:
> {code:java}
> 2023-09-29T13:19:17.851-0700    ERROR   cache/placeholder_manager.go:99 
> failed to create placeholder pod    {"error": "Pod 
> \"tg-foo-task-group-foo-gang-0\" is invalid: 
> spec.containers[0].resources.requests: Invalid value: \"1\": must be less 
> than or equal to cpu limit"}{code}
> I'd propose setting the same amount of request to the placeholder Pod's 
> limits, so system's LimitRange won't auto-set lower limits value.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to