Skip to content

Error with hyperpod start-job command #98

@mvinci12

Description

@mvinci12

Error when running hyperpod start-job command from hyperpod fsdp workshop example:

root@7e04bfc5b4f7:/hyperpod/projects/test-workshop/awsome-distributed-training/3.test_cases/pytorch/FSDP/kubernetes# hyperpod start-job --config-file ./hpcli-fsdp.yaml
2025-06-06 16:47:42 - hyperpod_cli.validators.job_validator - ERROR - Scheduler type is 'SageMaker' however cannot find namespace 'aws-hyperpod' managed by SageMaker. Please ensure namespace exists and you have 'get' access to it.

I've tried default and kubeflow namespaces as well. The only namespaces it works in are the namespaces created from making the Team Quota allocations in Task Governance workshop example.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions