SageMaker Groundtruth Named entity recognition Labeling Job Failure

Question

Hi!
I'm unable to create a SageMaker Groundtruth Labeling Job for NER.

Steps:

Created a new bucket.

Created folder for documents.

Created folder for outputs.

Added private work team

Created labeling job - Created new Role for Labeling job.

Have tried  multiple times, always receiving the same error:

```
{
    "labeling-job-name": "Labelling-job-18",
    "event-name": "PRE_HUMAN_LAMBDA_FAILED",
    "event-log-message": "ERROR: Pre-human task Lambda failed for line 1.ClientError: The label categories file located at s3:///20221107/output/Labelling-job-18/annotation-tool/data.json can't be accessed. Make sure it exists in the us-east-1 region, that the role arn:aws:iam:::role/service-role/AmazonSageMaker-ExecutionRole-20221028T185408 has read permissions to it and try your request again."
}
```

How can I fix this? Is the role created in the labelling job missing some kind of lambda policy?

Answer

Hello,

Thank you for using sagemaker Ground-Truth. Usually this exception occurs when the labeling job is not able to access the S3 bucket, Assuming that the bucket is located in the same region and sagemaker full access policy[1] is attached to the role. Please verify below points.

1. Please check if any bucket policy is attached, which is blocking or restricting access[2]
2. Verify if any organizational SCP(service control policy) is setting a permission boundary on this account to prevent S3 access.[3]

Post checking these policies, please try to re-run the job. If you have any difficulty in verifying any of the above mentioned points or if you still run into issue, Please reach out to AWS Support[4] (Sagemaker), along with your issue/use case in detail and share relevant AWS resource names. We will troubleshoot accordingly by verifying relevant job logs.

[1] https://docs.aws.amazon.com/sagemaker/latest/dg/security-iam-awsmanpol.html#security-iam-awsmanpol-AmazonSageMakerFullAccess
[2] https://docs.aws.amazon.com/AmazonS3/latest/userguide/example-bucket-policies.html
[3] https://docs.aws.amazon.com/organizations/latest/userguide/orgs_manage_policies_scps.html
[4]  https://docs.aws.amazon.com/awssupport/latest/user/case-management.html#creating-a-support-casehttps://docs.aws.amazon.com/awssupport/latest/user/case-management.html#creating-a-support-case

SageMaker Groundtruth Named entity recognition Labeling Job Failure

相關內容