The only thing I see is that it could|should have a CLIENT perspective
Client writers MUST pay close attention to the timeout specified in
section 6.1 of RFC 5321.
Because the problem is mostly with the clients not following the 10
minute recommendation. I don't know if it should be mentioned but...
.... Clients using anything less than the recommended 10 minutes
could promote retransmission issues.
That is not in any way promoting that servers can take its sweet time
with 10 minute delays.
Tony Hansen wrote:
Hmmm, if the following were added to 4409bis somewhere, I don't think it
should affect going from Draft => Full:
It has been observed that MUAs have in existing practice not been
allowing a long time for the Submit Server to respond to
the final <CRLF>.<CRLF>.
Submit Server writers MUST pay close attention to the Server
timeout specified in section 6.1 of RFC 5321, where it says:
... a receiver-SMTP MUST seek to minimize the time
required to respond to the final <CRLF>.<CRLF> end of
MUAs and other Submit clients SHOULD wait at least the normal TCP/IP
timeout lengths of time after sending the final <CRLF>.<CRLF>, if not
the full timeout specified in section 188.8.131.52.6 of RFC 5321.
or something along those lines.
On 8/11/2010 12:41 PM, Dave CROCKER wrote:
On 8/11/2010 9:19 AM, Tony Hansen wrote:
I'm wondering if, instead of changing the 5321 guidance, there needs
to be text
added to 4409bis about timeouts being shorter than those normally
used by SMTP.
Any useful guidance should be made wherever it is helpful. This issue
is certainly in the realm of 'useful'.
My note, here, isn't about that. It's a 'process' question, meant
mostly for academic consideration:
Is this the sort of change that is appropriate for going from Draft
I would have thought that it was too technical and substantive and
that, at the least, the doc would have to cycle at, perhaps, Draft.
To repeat: this is meant purely as an academic exercise. I'm curious.