I have been occasionally getting the same type of error, when timer executed integration jobs run in OpenText BizLink / BizManager...

A couple of differences I'm seeing between the OP's issue and mine is that the connection settings never change. ever.

com.amazonaws.services.s3.model.AmazonS3Exception: Bad Request (Service: Amazon S3; Status Code: 400; Error Code: 400 Bad Request; Request ID: null; Proxy: xXxXxXx.xXx.XxXxX.com), S3 Extended Request ID: null]. After executing [0] of [0] retries with delay of [0] second(s).

We may get this error at 6AM during a file push process, it fails, and when I re-execute the push it is successful.

Due to the intermittency I started to expect this was some type of internet connectivity issue.

And there's the oddity that there were no retries, though I have specified 2 retries with a 600 second delay. So it's like it's bypassing the additional retries because perhaps it's never reaching AWS from our HTTP Proxy host which handles the connectivity?