So I am very aware of this other thread that asks the same question: Configure Firehose so it writes only one record per S3 object?

However, that was two years ago and Amazon is constantly adding/changing things. Is this answer still valid or is there now a way to configure firehose to do this?

CodePudding user response：

Sadly there is not. It still writes entire content of its buffer to s3. You would have to setup lambda transaction for the records and do the writing yourself using the lambda function.