Re: Addition to <draft-ietf-mailext-smtpas-00.txt>

[michael.derrico@xxxxxxxxxxxx (Michael D'Errico)]

> > Shouldn't the error code be 421?
> 
> This is far from clear. One way of looking at it is that this is an
> application, not a connection timeout -- the mail application is the thing
> that's timing out and reporting the error -- and as such the proper code would
> be a new 45X value.

I hadn't thought of it that way, but it does make sense.  I was merely looking 
at RFC 821 (sections 4.2 and 4.3) which shows that 421 is the only valid 4xy 
code for all commands, and RFC 1123 suggests using only the codes listed in 
821.  If you want to add "453 Timeout waiting for command" I'd go along with 
that.

> The bottom line is that the first digit is really all that matters, and the
> rest of it is pretty much up in the air depending on how you read the rules. I
> think using any 4YX code would be fine, but if 421 makes you happy, I'm all for
> that!

I never even check the 2nd or 3rd digit, so I'm happy with 421 or 453 or ....

> > (There's also a reference to using 220 as a reply code to EXPN and VRFY in
> > section 2.2.3 which should be 250.)
> 
> The second digit of a success code is even less meaningful than that for an
> error code. However, in this case the rules do seem clear (this is definitiely
> from the mail system, not the connection) and 250 is therefore the proper
> value.

Again, I was just trying to make sure this document is consistent with 821.

Mike