[Top] [All Lists]

Re: SNPP - A Clarification

1993-09-06 10:26:32
First, you may note that I'm copying John Klensin.  He's an area director
for applications, under which this probably falls, and I'm not.  I took
the initial contact with you to help off-load some work from him and because
I'm interested in the topic.  But I want to make sure he is fully in the
loop on this.  Your note doesn't contain anything that feels "private" so
i'm including him, but don't want to take liberties too far.

Oh, I'm sorry.  I meant to copy that to the list, but did not.  Let me
see if I can find a copy of it, and I'll forward it.

To quote an old joke, that means that we are mostly just haggling price.


    Therefore, if the IXO/TAP gateway did not like the pager "PIN" 
    number, the only way to be notified of failure is via email.  

I'll note that I'd expect the pin number to be in the email address, so
that that particular rejection could be in response to the RCPT-TO exchange.

No!  You can't do that, because the PIN can't even be presented to the
IXO until *after* the DATA exchange.  The packet going from the gateway
to the IXO is made up of PIN+MESSAGE.  You have to have the message in
order to ship the PIN across.  You should think of the exchange, rather
than validating a PIN, the IXO terminal tells you "whether the page
made it or not" and a reason.  Therefore, the confirmation *must*
take place after the final CRLF-.-CRLF (in the DATA segment).

How much of a protocol change would this take to SMTP to allow
immediate verification of receipt by the IXO terminal?  How long
would it take to "standardize?"

<Prev in Thread] Current Thread [Next in Thread>