Skip to content

AWS_REGION is ignored #357

@msusta

Description

@msusta

There seems to be something in the code that ignores AWS region setting provided from ENV - you have to provide --region option.

Observed behaviour:

(hyperpod-cli) msusta@yuki:~/hyperpod-cli/hyperpod-poc-4$ echo $AWS_REGION

(hyperpod-cli) msusta@yuki:~/hyperpod-cli/hyperpod-poc-4$ AWS_REGION=us-east-2 AWS_PROFILE=test hyp list-cluster
[]
(hyperpod-cli) msusta@yuki:~/hyperpod-cli/hyperpod-poc-4$ AWS_PROFILE=test hyp list-cluster --region us-east-2
Updated context arn:aws:eks:us-east-2:123456789012:cluster/sagemaker-hyperpod-poc-4-12345678-eks in /tmp/kubeconfig_hyperpod-poc-4
[
    {
        "Cluster": "hyperpod-poc-4",
        "Instances": [
            {
                "InstanceType": "ml.c6i.4xlarge",
                "TotalNodes": 2,
                "AcceleratorDevicesAvailable": "N/A",
                "NodeHealthStatus=Schedulable": 2,
                "DeepHealthCheckStatus=Passed": "N/A"
            }
        ]
    }
]

Expected behaviour:
I can use the usual Boto ENV to supply information to the CLI. As I jump between many environments it's easier to handle ENV setup with other tools rather than having to provide them with every command.

Versions:

hyp version: 3.4.0
hyperpod-pytorch-job-template version: 1.2.0
hyperpod-custom-inference-template version: 1.2.0
hyperpod-jumpstart-inference-template version: 1.1.1

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