1 Answer
- Newest
- Most votes
- Most comments
0
It usually occurs when the code is trying to use a custom S3 connector or Hadoop libraries as it can conflict with libraries that we use for other S3 backend systems. One possible solution is checking the S3 sink example at reference [1] and compare how the dependencies are shaded in that application to yours, to see if there are any discrepancies there.
Reference: [1] - https://github.com/aws-samples/amazon-managed-service-for-apache-flink-examples/tree/main/java/S3Sink
Relevant content
asked 2 years ago

Thank you for your response! The source of the issue was this line:
FileSystem.initialize(config, null)It is required for local testing but for AWS runtime it breaks the runtime configuration from the AWS side.