[prev in list] [next in list] [prev in thread] [next in thread] 

List:       postfix-users
Subject:    Re: null reverse-path
From:       "Rask Ingemann Lambertsen" <rask-postfix () kampsax ! k-net ! dk>
Date:       2000-06-30 19:03:24
[Download RAW message or body]

Den 30-Jun-00 15:34:01 skrev Ronny Haryanto fĝlgende om "null reverse-path (was Re: reject_unknown_sender_domain flawed?)":
>On 30-Jun-2000, Lars Hecking wrote:
>> > > MAIL FROM:<>
>> > This is not a bug. This is a feature :)
>>  This is not a feature. It is required. See RFC822/1123.

>Umm, did you mean 821?

>I was just debating about this in a mailing list. I think server-SMTP
>MUST support null reverse-path, and by support I mean it should not be
>rejected. This debate forced me to read section 3.6 of RFC 821 and
>sections 5.2.9 and 5.3.3 of RFC 1123 over and over trying to
>understand it (English is not my first language).
[cut]

   It is my clear understanding of the RFCs that the server-SMTP MUST
accept a null reverse-path in all cases where it would have accepted a
non-null reverse-path. It is also my understading that this affects
the following RCPT and DATA commands as well as MAIL itself. So that if

MAIL FROM:<sender@domain.com>
RCPT TO:<recipient@domain.com>
DATA
blabla
.

results in a 2xx response, then it would also be a RFC 821/1123 violation
if

MAIL FROM:<>
RCPT TO:<recipient@domain.com>
DATA
blabla
.

does not also result in a 2xx response. I.e.

MAIL FROM:<>
250 Ok
RCPT TO:<recipient@domain.com>
550 MAIL FROM:<> no accepted
RSET
250 Ok
MAIL FROM:<sender@domain.com>
250 Ok
RCPT TO:<recipient@domain.com>
250 Ok

is also a violation of RFC 821/1123.

Regards,

/ŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻTŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻŻ\
| Rask Ingemann Lambertsen       | E-mail: mailto:rask@kampsax.k-net.dk  |
| A4000, 896 kkeys/s (RC5-64)    | "ThrustMe" on XPilot, ARCnet and IRC  |
|   To err is human, but to really mess things up you need a computer.   |

[prev in list] [next in list] [prev in thread] [next in thread]