SageMaker can't access S3 when creating a labeling job

0

I have an IAM role with the following permissions:

  • AmazonSageMaker-ExecutionPolicy-20221112T214872
  • AmazonS3FullAccess
  • AmazonSageMakerFullAccess
  • AmazonSageMakerCanvasFullAccess

However when I try to create a labeling job in SageMaker and I click Create Labeling Job with:

  • Manual data setup
  • Copy-paste the IAM ARN
  • Click browse S3

I don't see the files in the bucket. When I enter the s3:// path to the .jsonl file, I get this:

NetworkingError: Network Failure - The S3 bucket 'cvat.mouseover.ai' you entered in Input dataset location cannot be reached. Either the bucket does not exist, or you do not have permission to access it. If the bucket does not exist, update Input dataset location with a new S3 URI. If the bucket exists, give the IAM entity you are using to create this labeling job permission to read and write to this S3 bucket, and try your request again.

I also have CORS enabled for the bucket and full public read access, just in case. Policy simulator confirms full access of the role to S3.

1 Answer
0

Hello,

Thank you for using AWS SageMaker.

From the error, I understand that the labeling job is not able to establish a network with S3 bucket that you have mentioned manually for Input dataset location. I was able to successfully replicate the same error under my account, the reason behind this type of error was that the my labeling job and S3 bucket were in 2 different region. Once the I provided S3 bucket from similar region my issue was resolved. So I'd recommend you to provide S3 bucket from the same the region were you are trying to create a labeling job or vice versa. At this moment SageMaker labeling job doesn't support cross region S3 bucket, so you need to have both the resource in same region to create labeling job successfully.

After following the above recommendation, if you still observe same error message, I'd recommend you to reach out to AWS Support along with all the resource details and error that you observe for further investigation. To open a support case with AWS using the link: https://console.aws.amazon.com/support/home?#/case/create

AWS
SUPPORT ENGINEER
answered a year ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions