Re: [Technical Errata Reported] RFC7231 (5806)

[Amos Jeffries <[e-mail]>]

On 12/08/19 5:49 pm, Willy Tarreau wrote:
> On Sun, Aug 11, 2019 at 07:27:30PM -0700, RFC Errata System wrote:
>> The following errata report has been submitted for RFC7231,
>> "Hypertext Transfer Protocol (HTTP/1.1): Semantics and Content".
>>
>> --------------------------------------
>> You may review the report below and at:
>> https://www.rfc-editor.org/errata/eid5806
>>
>> --------------------------------------
>> Type: Technical
>> Reported by: Anders Kaseorg <[e-mail]>
>>
>> Section: 4.3.7
>>
>> Original Text
>> -------------
>> A server MUST generate a Content-Length field with a value of "0" if no
>> payload body is to be sent in the response.
>>
>> Corrected Text
>> --------------
>> If no payload body is to be sent in the response, a server MUST
>> generate a status code of 204 (No Content) or a Content-Length field
>> with a value of "0" (but not both).
>>
>> Notes
>> -----
>> The original text contradicts RFC 7230 §3.3.2: "A server MUST NOT send a
>> Content-Length header field in any response with a status code of 1xx
>> (Informational) or 204 (No Content)", unless the intention was to disallow
>> all 204 responses to OPTIONS requests, which I assume it was not.
> 
> Indeed, it seems valid to me.

Ditto.

> With this said, I don't know how common
> it is to respond with 204 to an OPTIONS request given that 204 is
> reportedly cacheable by default (6.3.5) while OPTIONS is said not to
> be. Thus more confusion may arise on this point as well.
> 

That should not be an issue since the 204 caching is explicitly "unless
otherwise indicated by the method definition".

Amos