Cloud formation for MLflow stack

Question

I'm trying to follow MLflow command line documentation to make a server on Fargate but I face this error can anyone please guide me to solve it thanks:

error: User: arn:aws:sts::567791023873:assumed-role/AmazonSageMaker-ExecutionRole-20230615T154712/SageMaker is not authorized to perform: cloudformation:DescribeStacks on resource: arn:aws:cloudformation:us-east-1:567791023873:stack/CDKToolkit/* because no identity-based policy allows the cloudformation:DescribeStacks action

Accepted Answer

Hi,

I am assuming you are following the steps mentioned on this [blog](https://aws.amazon.com/blogs/machine-learning/managing-your-machine-learning-lifecycle-with-mlflow-and-amazon-sagemaker/). It appears the execution role (shown in bold) doesn't have the required permissions.

mlflow.sagemaker.deploy(
    mode='create',
    app_name=endpoint_name,
    model_uri=model_uri,
    image_url=image_uri,
    execution_role_arn=**role**,
    instance_type='ml.m5.xlarge',
    instance_count=1,
    region_name=region
)

I would suggest finding the role in IAM->Roles and adding the cloudformation:DescribeStacks action as a valid permission to that role.

Thanks,
Rama

Cloud formation for MLflow stack

Relevant content