Re: [yam] Ambiguities in RFC 5321

["John R Levine" <johnl at taugh.com>]

> > I think we can split hairs about whether it means "no SMTP service at all"
> > vs. "no SMTP service for you", although it's easy enough to define so that 
> > it doesn't matter and the client does the same thing either way.
>
> This leads down the slippery slope of needing to define a long list of 
> differential reasons the service does not exist:

Indeed, which is why I suggested we adjust the wording so it's clear what 
the client is supposed to do, regardless of the reason.  The leading 
candidates appear to be:

a) Total failure, bounce whatever you were planning to deliver

b) Host failure, forget this one but try other MXes

Note that in the original scenario of a host that is getting mail by 
mistake, due to A record fallback, these are equivalent.

Regards,
John Levine, johnl at taugh.com, Taughannock Networks, Trumansburg NY
"I dropped the toothpaste", said Tom, crestfallenly.