Mailing List Archive

autoreply bug + enhancement request
Hi

DBMail 3.1.17, native autoreplies

i just wondered about a reply to "debrapascalibonaro.com@mydomain" while
"mydomain" is the postfix setting - the reply to header is just a domain
- could we skip try reply to invalid addresses without @?

From: Orgasmic Birth <dpascali@optonline.net>
Reply-to: debrapascalibonaro.com
____________________________________________

BTW:

DBMail take some headers into account to not send native replies
while "X-Auto-Response-Suppress: All" is missing here and used by
facebook as example
____________________________________________

i see in the database a reply to XING while the original message has
these headers - uhm - i thought wer already supress in that case?

From: XING <mailrobot@xing.com>
Reply-to: "Keine Antwortadresse " <no.reply@xing.com>
Auto-Submitted: auto-generated
____________________________________________

wouldn't it make sense to make the reply-supression headers
configuarebale instead hardcoded?
Re: autoreply bug + enhancement request [ In reply to ]
On 30-10-14 11:23, Reindl Harald wrote:
> Hi
>
> DBMail 3.1.17, native autoreplies
>
> i just wondered about a reply to "debrapascalibonaro.com@mydomain" while
> "mydomain" is the postfix setting - the reply to header is just a domain
> - could we skip try reply to invalid addresses without @?
>
> From: Orgasmic Birth <dpascali@optonline.net>
> Reply-to: debrapascalibonaro.com

I understand your problem, but debrapascalibonaro.com really is a valid
address, albeit a local one.



> ____________________________________________
>
> BTW:
>
> DBMail take some headers into account to not send native replies
> while "X-Auto-Response-Suppress: All" is missing here and used by
> facebook as example
> ____________________________________________


X- headers are totally ad-hoc. Maybe we could setup a configuration
setting to define a regex or list of regexes to skip auto replies.

src/dm_message.c, send_reply would be the starting point.

>
> i see in the database a reply to XING while the original message has
> these headers - uhm - i thought wer already supress in that case?
>
> From: XING <mailrobot@xing.com>
> Reply-to: "Keine Antwortadresse " <no.reply@xing.com>
> Auto-Submitted: auto-generated
> ____________________________________________
>
> wouldn't it make sense to make the reply-supression headers
> configuarebale instead hardcoded?

Absolutely. The only suppression that is currently done is for
Precedence: <bulk|list>, and to prevent a mail-loop between dbmail and
itself.


--
________________________________________________________________
Paul J Stevens pjstevns @ gmail, twitter, github, linkedin
www.nfg.nl/info@nfg.nl/+31.85.877.99.97
_______________________________________________
DBmail mailing list
DBmail@dbmail.org
http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail
Re: autoreply bug + enhancement request [ In reply to ]
Am 02.11.2014 um 14:40 schrieb Paul J Stevens:
>
>
> On 30-10-14 11:23, Reindl Harald wrote:
>> Hi
>>
>> DBMail 3.1.17, native autoreplies
>>
>> i just wondered about a reply to "debrapascalibonaro.com@mydomain" while
>> "mydomain" is the postfix setting - the reply to header is just a domain
>> - could we skip try reply to invalid addresses without @?
>>
>> From: Orgasmic Birth <dpascali@optonline.net>
>> Reply-to: debrapascalibonaro.com
>
> I understand your problem, but debrapascalibonaro.com really is a valid
> address, albeit a local one.

hmm - can we agree that in context of autoreplies unqualified addresses
don't really count? :-)

>> ____________________________________________
>>
>> BTW:
>>
>> DBMail take some headers into account to not send native replies
>> while "X-Auto-Response-Suppress: All" is missing here and used by
>> facebook as example
>> ____________________________________________
>
>
> X- headers are totally ad-hoc

yes and no, "Precedence" is also a pseudo-standard

http://msdn.microsoft.com/en-us/library/ee219609%28v=exchg.80%29.aspx

> Maybe we could setup a configuration
> setting to define a regex or list of regexes to skip auto replies.
>
> src/dm_message.c, send_reply would be the starting point.
>>
>> i see in the database a reply to XING while the original message has
>> these headers - uhm - i thought wer already supress in that case?
>>
>> From: XING <mailrobot@xing.com>
>> Reply-to: "Keine Antwortadresse " <no.reply@xing.com>
>> Auto-Submitted: auto-generated
>> ____________________________________________
>>
>> wouldn't it make sense to make the reply-supression headers
>> configuarebale instead hardcoded?
>
> Absolutely. The only suppression that is currently done is for
> Precedence: <bulk|list>, and to prevent a mail-loop between dbmail and
> itself

on the other hand dbmail has a table to prevent loops without any
headers - wouldn't it make sense to just add the few known pseudo
standard headers in the existing code and save the burden make it
configureable that way?

to prevent loops the null-sender-*envelope* (<>) would be the
standard-way on most sites (the dbmail table itself is a good additional
thing)

that below should cover almost anything

Precedence: junk
Precedence: bulk
Precedence: list
Auto-Submitted: auto-generated
X-Auto-Response-Suppress: AutoReply
X-Auto-Response-Suppress: All
X-Auto-Response-Suppress: OOF

http://msdn.microsoft.com/en-us/library/ee219609%28v=exchg.80%29.aspx
http://stackoverflow.com/questions/154718/precedence-header-in-email
http://tools.ietf.org/html/rfc3834#section-5
Re: autoreply bug + enhancement request [ In reply to ]
Am 08.11.2014 um 22:32 schrieb Reindl Harald:
> on the other hand dbmail has a table to prevent loops without any
> headers - wouldn't it make sense to just add the few known pseudo
> standard headers in the existing code and save the burden make it
> configureable that way?
>
> to prevent loops the null-sender-*envelope* (<>) would be the
> standard-way on most sites (the dbmail table itself is a good additional
> thing)
>
> that below should cover almost anything
>
> Precedence: junk
> Precedence: bulk
> Precedence: list
> Auto-Submitted: auto-generated
> X-Auto-Response-Suppress: AutoReply
> X-Auto-Response-Suppress: All
> X-Auto-Response-Suppress: OOF
>
> http://msdn.microsoft.com/en-us/library/ee219609%28v=exchg.80%29.aspx
> http://stackoverflow.com/questions/154718/precedence-header-in-email
> http://tools.ietf.org/html/rfc3834#section-5

uhm - and could we also skip "noreply@" and "do_not_reply@" independent
of headers? at least "noreply" as local part is known for many years

noreply@facebookmail.com
do_not_reply@itunes.com