You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@yunikorn.apache.org by "Manikandan R (Jira)" <ji...@apache.org> on 2023/09/12 05:12:00 UTC

[jira] [Created] (YUNIKORN-1973) Queue Limits configuration at same level should not compare each other

Manikandan R created YUNIKORN-1973:
--------------------------------------

             Summary: Queue Limits configuration at same level should not compare each other
                 Key: YUNIKORN-1973
                 URL: https://issues.apache.org/jira/browse/YUNIKORN-1973
             Project: Apache YuniKorn
          Issue Type: Bug
          Components: core - scheduler
            Reporter: Manikandan R
            Assignee: Manikandan R
             Fix For: 1.4.0


Limits configured for queues at same level should not be compared against each other. As long as it lesser than its immediate or ancestor parent limit settings, it should be allowed.

 

For example, In below configuration, user "test" of queue "parent2" is greater than user "test" of queue "parent1". This setting should be allowed, but it is throwing the error.

 
partitions:
    - name: default
      queues:
        - name: root
          parent: true
          submitacl: '*'
          queues:
            - name: parent1
              parent: true
              resources:
                guaranteed:
                    memory: 250G
                    nvidia.com/gpu: "250"
                    vcore: "250"
                max:
                    memory: 1000G
                    nvidia.com/gpu: "1000"
                    vcore: "1000"
              limits:
                - limit: ""
                  users:
                    - test
                  maxresources:
                    cpu: "200"
            - name: parent2
              parent: true
              resources:
                guaranteed:
                    memory: 250G
                    nvidia.com/gpu: "250"
                    vcore: "250"
                max:
                    memory: 1000G
                    nvidia.com/gpu: "1000"
                    vcore: "1000"
              limits:
                - limit: ""
                  users:
                    - test
                  maxresources:
                    cpu: "210"



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@yunikorn.apache.org
For additional commands, e-mail: issues-help@yunikorn.apache.org