Closed MustaphaU closed 3 months ago
Thanks for reaching out. The start_document_analysis command makes a call to the StartDocumentAnalysis API. That error is coming from the API. The error has come up a few times before in other issues, for example https://github.com/boto/boto3/issues/2653. As mentioned there (and also in a Stack Overflow post) an invalid S3 bucket name could cause the error.
Can you verify the the S3 bucket name you're using is valid? And can you also confirm that the IAM role you're using has the necessary permissions to use the Textract service and access the S3 bucket where the document is stored?
If still seeing an issue please provide the complete code snippet you're using to reproduce the issue.
Thanks @tim-finnigan
The error was an incorrect role arn specification. I was passing the policy ARN
of the SNS permissions attached to the textract role rather than the textract role ARN
itself.
This issue is now closed. Comments on closed issues are hard for our team to see. If you need more assistance, please open a new issue that references this one.
Describe the bug
When I try to run an async job for document analysis by following the guideline in Performing Asynchronous Operation with Textract, I get the error:
Expected Behavior
The result of of the document analysis i.e the tables and form extracts
Current Behavior
The log:
Reproduction Steps
Follow the guidelines in Performing Asynchronous Operation with Textract and run the python script.
Possible Solution
No response
Additional Information/Context
No response
SDK version used
Name: boto3 Version: 1.34.117
Environment details (OS name and version, etc.)
Windows 11 pro