Catching Algorithm Error with Retry Policy SageMaker

[karakewe]

I´ve been trying to automize my pipeline with the retry policy, so that it is able to run again if the information from my databases are not fully updated. Raising an error when such database is not updated. The thing is that raising an error in the python script raises an Algorithm Error on AWS SageMaker, but theres no such exception to catch this error on my Retry Policy. I've being trying to simulate a Capacity Error or an Internal Error, but without success.