On Tue, 18 Feb 2003, Greg A. Woods wrote:
> [ On Tuesday, February 18, 2003 at 23:10:16 (+0530), Suresh Ramasubramanian wrote: ]
> > I guess we can wait till after CRLF.CRLF before issuing the 5xx.
>
> No, not "can" -- you absolutely "MUST" wait for the end-of-DATA before
> sending any response code (especially if you hope for it to have any
> effect, let alone the desired effect).
I have to apologise to Greg.
He does sometimes say things that I agree with.
--
Dr. Andrew C. Aitchison Computer Officer, DPMMS, Cambridge
A.C.Aitchison@??? http://www.dpmms.cam.ac.uk/~werdna
