On 9/2/11 11:18 PM, ned+ietf-smtp(_at_)mrochek(_dot_)com wrote:
>
>>> -----Original Message-----
>>> From: Frank Ellermann
[mailto:hmdmhdfmhdjmzdtjmzdtzktdkztdjz(_at_)gmail(_dot_)com]
>>> Sent: Friday, September 02, 2011 12:50 PM
>>> To: Murray S. Kucherawy
>>> Cc: ietf-smtp(_at_)imc(_dot_)org
>>> Subject: Re: Received: and sending IP
>>>
>>> IIRC many servers note the client (sender) IP as comment
>>> only if it does not match any IP of the EHLO name, but I
>>> guess that is not your question.
>> Yes, it's a comment. But still others show the name (if any) and the IP (in
all cases).
>> The question is really: Why a comment and not an additional preposition
clause?
> As I understand it, the feeling was that introducing a new formal clause
> was likely to break extant parsers for this stuff.
>
> I personally did not and do not agree with this choice, but the consensus
> went the other way, and now the installed base is kinda difficult to ignore.
Yep. I have at least two customers which run core business applications
in which they use (home-grown) code that parses the Received lines, to
determine a date/time timestamp. I'm sure there are many others out
there. Changing the format of the Received lines will cause problems
here and there, no doubt.
To the point where if we were to support such a change at this point, it would
be optional, off by default, and I doubt we'd recommend enabling it.
Ned