Mailing List Archive: SPF: SRS

SRS and Callbacks

Index | Next | Previous | View Threaded

csweeney at osubucks

Jan 14, 2007, 9:17 PM

Post #1 of 6 (8158 views)
Permalink

SRS and Callbacks

OK I have to ask for help now. I have a user that is unable to send
email because of what I consider broken call back. When this particular
place does call back its coming from a real email address instead of a
<> null address. In this case when you send email to
something [at] epronar it calls back with
sender_verification [at] sandiegort Is there away to get around this
broken call back problem?

Thanks
Chris

---------------------------- Original Message
---------------------------- Subject: Returned mail: see transcript for
details From: "Mail Delivery Subsystem"
<MAILER-DAEMON [at] stewie> Date: Sun, January 14, 2007 9:03 am
To: rsweeney [at] ohiowebhosting
--------------------------------------------------------------------------
The original message was received at Sun, 14 Jan 2007 09:02:56 -0500
from webmail.osubucks.org [70.60.69.214] ----- The following addresses
had permanent fatal errors ----- <e-PROTalk [at] eProNAR> (reason: 554
5.1.7 <SRS0=x/HgM3dy=GX=ohiowebhosting.com=rsweeney [at] osubucks>:
Sender address rejected:
unde...hiowebhosting.com=rsweeney [at] osubucks>... SRS only supported in
DSN! (in reply to RCPT TO command)) ----- Transcript of session follows
----- ... while talking to mx1.icsandiego.com.:

>>> >>> DATA
>>>
<<< 554 5.1.7 <SRS0=x/HgM3dy=GX=ohiowebhosting.com=rsweeney [at] osubucks>:
Sender address rejected: undeliverable address: host
test.osubucks.org[70.60.69.212] said: 550 5.7.6
<SRS0=x/HgM3dy=GX=ohiowebhosting.com=rsweeney [at] osubucks>... SRS only
supported in DSN! (in reply to RCPT TO command) 554 5.0.0 Service
unavailable
<<< 554 5.5.1 Error: no valid recipients

>>> >>> RSET
>>>
<<< 421 4.7.0 mx3.sandiegort.com Error: too many errors

-- This message has been scanned for viruses and dangerous content by
MailScanner, and is believed to be clean.

------------------------------------------------------------------------

Reporting-MTA: dns; stewie.osubucks.org
Arrival-Date: Sun, 14 Jan 2007 09:02:56 -0500

Final-Recipient: RFC822; e-PROTalk [at] eProNAR
Action: failed
Status: 5.1.7
Remote-MTA: DNS; mx1.icsandiego.com
Diagnostic-Code: SMTP; 554 5.1.7 <SRS0=x/HgM3dy=GX=ohiowebhosting.com=rsweeney [at] osubucks>: Sender address rejected: undeliverable address: host test.osubucks.org[70.60.69.212] said: 550 5.7.6 <SRS0=x/HgM3dy=GX=ohiowebhosting.com=rsweeney [at] osubucks>... SRS only supported in DSN! (in reply to RCPT TO command)
Last-Attempt-Date: Sun, 14 Jan 2007 09:03:10 -0500

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1129

Attachments:

smime.p7s (5.07 KB)

stuart at bmsi

Jan 14, 2007, 9:29 PM

Post #2 of 6 (7693 views)
Permalink

Re: SRS and Callbacks [In reply to]

On Mon, 15 Jan 2007, Chris Sweeney wrote:

> OK I have to ask for help now. I have a user that is unable to send
> email because of what I consider broken call back. When this particular
> place does call back its coming from a real email address instead of a
> <> null address. In this case when you send email to

Yup. That is real broken. Unclear on the concept, even.

> something [at] epronar it calls back with
> sender_verification [at] sandiegort Is there away to get around this
> broken call back problem?

Pymilter has a feature to work around such braindamage. It is called
banned_users, and whenever the local part is one of the banned_users,
the MAIL FROM is considered equivalent to <>, including requiring
a valid SRS sig in RCPT. I usually include 'postmaster' in the
banned_users, and you would include 'sender_verification'. Of course,
this means that postmaster [at] anydomai can't send me email directly.
This is usually not a problem because postmaster is normally redirected to
some other user, who responds from that account.

self.is_bounce = (f == '<>' or t[0].lower() in banned_users)

'Banned_users' is maybe not the best name for the config option, but
that is in effect what it does.

--
Stuart D. Gathman <stuart [at] bmsi>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1129

csweeney at osubucks

Jan 14, 2007, 10:24 PM

Post #3 of 6 (7695 views)
Permalink

Re: SRS and Callbacks [In reply to]

do you know if there is an RFC that states the proper way to do callbacks?

Stuart D. Gathman wrote:
> On Mon, 15 Jan 2007, Chris Sweeney wrote:
>
>
>> OK I have to ask for help now. I have a user that is unable to send
>> email because of what I consider broken call back. When this particular
>> place does call back its coming from a real email address instead of a
>> <> null address. In this case when you send email to
>>
>
> Yup. That is real broken. Unclear on the concept, even.
>
>
>> something [at] epronar it calls back with
>> sender_verification [at] sandiegort Is there away to get around this
>> broken call back problem?
>>
>
> Pymilter has a feature to work around such braindamage. It is called
> banned_users, and whenever the local part is one of the banned_users,
> the MAIL FROM is considered equivalent to <>, including requiring
> a valid SRS sig in RCPT. I usually include 'postmaster' in the
> banned_users, and you would include 'sender_verification'. Of course,
> this means that postmaster [at] anydomai can't send me email directly.
> This is usually not a problem because postmaster is normally redirected to
> some other user, who responds from that account.
>
> self.is_bounce = (f == '<>' or t[0].lower() in banned_users)
>
> 'Banned_users' is maybe not the best name for the config option, but
> that is in effect what it does.
>
>

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1129

Attachments:

smime.p7s (5.07 KB)

stuart at bmsi

Jan 15, 2007, 9:05 AM

Post #4 of 6 (7707 views)
Permalink

Re: SRS and Callbacks [In reply to]

On Mon, 15 Jan 2007, Chris Sweeney wrote:

> do you know if there is an RFC that states the proper way to do callbacks?

Its called RFC2821. Callbacks are supposed to verify MAIL FROM. MAIL FROM
is *only* for use with a DSN, i.e. MAIL FROM of <>.

Now there might be a callback trying to verify header From (a dubious
practice), but it would not be using your SRS signed MAIL FROM.

--
Stuart D. Gathman <stuart [at] bmsi>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1129

csweeney at osubucks

Jan 15, 2007, 2:24 PM

Post #5 of 6 (7697 views)
Permalink

Re: SRS and Callbacks [In reply to]

Thanks

Stuart D. Gathman wrote:
> On Mon, 15 Jan 2007, Chris Sweeney wrote:
>
>
>> do you know if there is an RFC that states the proper way to do callbacks?
>>
>
> Its called RFC2821. Callbacks are supposed to verify MAIL FROM. MAIL FROM
> is *only* for use with a DSN, i.e. MAIL FROM of <>.
>
> Now there might be a callback trying to verify header From (a dubious
> practice), but it would not be using your SRS signed MAIL FROM.
>
>

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1129

Attachments:

smime.p7s (5.07 KB)

stuart at bmsi

Jan 15, 2007, 6:19 PM

Post #6 of 6 (7705 views)
Permalink

Re: SRS and Callbacks [In reply to]

On Mon, 15 Jan 2007, Stuart D. Gathman wrote:

> > something [at] epronar it calls back with
> > sender_verification [at] sandiegort Is there away to get around this
> > broken call back problem?
>
>
> self.is_bounce = (f == '<>' or t[0].lower() in banned_users)
>
> 'Banned_users' is maybe not the best name for the config option, but
> that is in effect what it does.

FYI, my banned_users is currently:

banned_users = mailer-daemon, clamav, postmaster

"sender_verification" would be an obvious addition...

--
Stuart D. Gathman <stuart [at] bmsi>
Business Management Systems Inc. Phone: 703 591-0911 Fax: 703 591-6154
"Confutatis maledictis, flammis acribus addictis" - background song for
a Microsoft sponsored "Where do you want to go from here?" commercial.

-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/?list_id=1129

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads