Re: WGLC p1: Tear-down

[Willy Tarreau <[e-mail]>]

On Mon, Apr 29, 2013 at 08:33:41PM +0100, Ben Niven-Jenkins wrote:
> Section 6.6 of p1 states:
> 
>    A server that sends a close connection option MUST initiate a
>    lingering close of the connection after it sends the response
>    containing close.  The server MUST NOT process any further requests
>    received on that connection.
> 
>    A client that receives a close connection option MUST cease sending
>    requests on that connection and close the connection after reading
>    the response message containing the close; if additional pipelined
>    requests had been sent on the connection, the client SHOULD assume
>    that they will not be processed by the server.
> 
> The last sentence can be interpreted one of two ways:
> 1) The client SHOULD assume the additional pipelined requests will NOT be processed by the server and therefore can happily re-try them knowing the server has not processed the previous ones.
> 
> 2) The client SHOULD NOT assume the additional pipelined requests will be processed (which implies the client simply can not know whether the server has processed them or not).
> 
> As the client has no way of knowing whether the server may have processed them or not (e.g. the client may be talking to a proxy that has already relayed the pipelined requests to the origin and done so before the proxy was aware that it wanted to close the connection on this response) I would suggest rewording the last sentence quoted above:
> 
> OLD:
>    the client SHOULD assume that they will not be processed by the server.
> NEW:
>    the client SHOULD NOT assume that they will be processed by the server.

+1 that's perfectly true.

Willy