Re: [Mailman-Users] Training Mailman to find email address from NDR
Grant Taylor via Mailman-Users writes: > RFC 3464 has been out for 17 years. I think it's past time that we > stop coddling people that can't conform to it. It's not a matter of coddling nonconformers. The only thing we are likely to be able to do about them is "Sleeping Giants" them on hostmaster Twitter. The only thing that would really hurt the nonconformers is to put them out the airlock. But that would hurt *our* users first, fast, and hard. Only if it became a movement would the large providers even notice. I know, you know that already. But a lot of the folks on this list may not. The fact of the matter is that the troublesome miscreants (Yahoo![1], AOL, Sendgrid) are large, they have lots of "captive" end users who would be very unhappy if any service they use took strong action to prevent arrival of nonconformant mail from those sources, and *our* clients are the list owners, postmasters, and hostmasters who manage sites and lists for those end users. AFAICS we're stuck in a position where the people who do the work have to make a decision whether doing the work is worth it for the project. Mark's been doing this a long time, nobody is more expert on that for Mailman 2. I support his decision: however much I'd like to make things a little bit easier for the list owners, his judgment is better than mine (and I don't know how to do it myself -- learning how would mean a lot more work than it would take for Mark! :-( ) Steve Footnotes: [1] I no longer blame Yahoo! and AOL for "p=reject" given that they were exposed to unimaginably large spam campaigns based on hundreds of millions of leaked address books used to spearphish spam and phishing messages at their users' correspondents. The alternative would have been to shut down their mail systems entirely, which would have hurt nearly a billion people directly, not to mention their correspondents. Something needs to be done about making firms with large stocks of PII accountable for leaking any of it, though, up to and including corporate death on the auction block if it's big enough. -- Associate Professor Division of Policy and Planning Science http://turnbull.sk.tsukuba.ac.jp/ Faculty of Systems and Information Email: turnb...@sk.tsukuba.ac.jp University of Tsukuba Tel: 029-853-5175 Tennodai 1-1-1, Tsukuba 305-8573 JAPAN -- Mailman-Users mailing list Mailman-Users@python.org https://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://wiki.list.org/x/AgA3 Security Policy: http://wiki.list.org/x/QIA9 Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-users/archive%40jab.org
Re: [Mailman-Users] Training Mailman to find email address from NDR
On 5/15/19 10:06 AM, Grant Taylor via Mailman-Users wrote: > On 5/15/19 10:55 AM, Mark Sapiro wrote: > >> The message body is that of a MIME multipart message, but the main >> content type is text/plain instead of >> >> multipart/mixed; boundary="66728b7fa14ce3ed" > > RFC 3464 wants a Content-Type of message/delivery-status. The message/delivery-status part is there as a subpart. I should have said multipart/report; boundary="66728b7fa14ce3ed" rather than multipart/mixed as that is what the outer part's Content-Type should be. -- Mark Sapiro The highway is for gamblers, San Francisco Bay Area, Californiabetter use your sense - B. Dylan -- Mailman-Users mailing list Mailman-Users@python.org https://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://wiki.list.org/x/AgA3 Security Policy: http://wiki.list.org/x/QIA9 Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-users/archive%40jab.org
Re: [Mailman-Users] Training Mailman to find email address from NDR
On 5/15/19 10:55 AM, Mark Sapiro wrote: This message is a disaster. Is this an actual bounce as received? It almost looks like an RFC 3464 compliant DSN except see comment below. ~chuckle~ Agreed. The message body is that of a MIME multipart message, but the main content type is text/plain instead of multipart/mixed; boundary="66728b7fa14ce3ed" RFC 3464 wants a Content-Type of message/delivery-status. so the whole body is just one plain text part and is quoted-printable encoded so that just changing the above Content-Type: won't work because there is quoted-printable encoded content in the sub-parts including sub-part headers. I have seen the other replies in this thread, so I'm not adding much here. Trying to recognize this in Mailman would be a major kludge and not worth the effort. I would argue against hacking Mailman to recognize this as a failed attempt at an RFC 3464 Delivery Status Notification. RFC 3464 has been out for 17 years. I think it's past time that we stop coddling people that can't conform to it. -- Grant. . . . unix || die smime.p7s Description: S/MIME Cryptographic Signature -- Mailman-Users mailing list Mailman-Users@python.org https://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://wiki.list.org/x/AgA3 Security Policy: http://wiki.list.org/x/QIA9 Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-users/archive%40jab.org
Re: [Mailman-Users] Training Mailman to find email address from NDR
On 5/15/19 6:45 AM, Andrew Hodgson wrote: > > Here is the bounce message, I know I can enable VERP and this problem should > go away, but I wanted to see if I could get this working without VERP as we > have a large mail traffic on the list. This message is a disaster. Is this an actual bounce as received? It almost looks like an RFC 3464 compliant DSN except see comment below. > Received: from o198-37-151-26.outbound-mail.sendgrid.net ([198.37.151.26]) > by > samwise.hodgsonfamily.org with esmtps > (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256)(Exim 4.89) (envelope-from > ) id 1hQs8e-Nl-9y for > bcab-boun...@lists.bcab.org.uk; Wed, 15 May 2019 12:32:07 +0100 > Date: Wed, 15 May 2019 11:00:09 + > To: bcab-boun...@lists.bcab.org.uk > From: no-re...@sendgrid.net > Subject: Undelivered Mail Returned to Sender > X-Spam-Score: -0.3 (/) > X-Spam-Report: BAYES_00=-1.9, HELO_STATIC_HOST=-0.001, INVALID_DATE=1.096, > MISSING_MID=0.497, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, > URIBL_BLOCKED=0.001 > MIME-Version: 1.0 > Content-Type: text/plain; charset="utf-8" > Content-Transfer-Encoding: quoted-printable The message body is that of a MIME multipart message, but the main content type is text/plain instead of multipart/mixed; boundary="66728b7fa14ce3ed" so the whole body is just one plain text part and is quoted-printable encoded so that just changing the above Content-Type: won't work because there is quoted-printable encoded content in the sub-parts including sub-part headers. I have seen the other replies in this thread, so I'm not adding much here. Trying to recognize this in Mailman would be a major kludge and not worth the effort. -- Mark Sapiro The highway is for gamblers, San Francisco Bay Area, Californiabetter use your sense - B. Dylan -- Mailman-Users mailing list Mailman-Users@python.org https://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://wiki.list.org/x/AgA3 Security Policy: http://wiki.list.org/x/QIA9 Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-users/archive%40jab.org
Re: [Mailman-Users] Training Mailman to find email address from NDR
Grant Taylor wrote: >On 5/15/19 7:45 AM, Andrew Hodgson wrote: >> Here is the bounce message, I know I can enable VERP and this problem >> should go away, but I wanted to see if I could get this working without >> VERP as we have a large mail traffic on the list. >Did you redact anything significant from the bounce? I changed the email address of the recipient to redac...@yahoo.com (original recipient was in there), and removed the whole digest message. >I don't see the expected boundary="66728b7fa14ce3ed" parameter to the >Content-Type: header. I noticed that as well after I sent the original message. >I'm also surprised that the Content-Type: is text/plain and not >multipart/report. Ok I will relay this onto them. >I don't know enough about Mailman's bounce processing to know if it will >find and process the message/delivery-status or not. I also question >what Mailman would do with a temporary failure, 4.7.0. I don't care at this stage what Mailman would do with the error, I just want it to be attributed to the correct address. >It looks like Yahoo TempFailed the message because they don't like >SendGrid for one reason or another. Yep, pritty poor actually. My IP has better delivery rates than these shared IPs Sendgrid use. Thanks for some pointers anyway, Andrew. -- Mailman-Users mailing list Mailman-Users@python.org https://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://wiki.list.org/x/AgA3 Security Policy: http://wiki.list.org/x/QIA9 Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-users/archive%40jab.org
Re: [Mailman-Users] Training Mailman to find email address from NDR
On 5/15/19 7:45 AM, Andrew Hodgson wrote: Hi, Hi, Here is the bounce message, I know I can enable VERP and this problem should go away, but I wanted to see if I could get this working without VERP as we have a large mail traffic on the list. Did you redact anything significant from the bounce? I don't see the expected boundary="66728b7fa14ce3ed" parameter to the Content-Type: header. I'm also surprised that the Content-Type: is text/plain and not multipart/report. I don't know enough about Mailman's bounce processing to know if it will find and process the message/delivery-status or not. I also question what Mailman would do with a temporary failure, 4.7.0. It looks like Yahoo TempFailed the message because they don't like SendGrid for one reason or another. I'm also somewhat surprised that SendGrid is returning the entire original message instead of just the headers. IMHO that's a good way for a bounced message's content to get trapped by a spam filter. -- Grant. . . . unix || die smime.p7s Description: S/MIME Cryptographic Signature -- Mailman-Users mailing list Mailman-Users@python.org https://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://wiki.list.org/x/AgA3 Security Policy: http://wiki.list.org/x/QIA9 Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-users/archive%40jab.org
