The controversy here is the fact that the API Gateway limits the upload resulting in having to engineer a workaround workflow using s3 and triggers (even if thi

The controversy here is the fact that the API Gateway limits the upload resulting in having to engineer a workaround workflow using s3 and triggers (even if this is the serverless way) when all you want to do is upload a file. A POST call with an octet-octet stream. Let http handle resume. But you can’t and you end up going around the side door, when all you really want is client_body_max_size The sarcasm of correctness yet playing down its complexity is entirely my own. We used to be able to do things easily.