July 2025 - Mailman-users - lists.mailman3.org

The RequestOptions.auto_parse_form_urlencoded option is deprecated.
by Odhiambo Washington July 16, 2025

July 16, 2025

Hello Admins, The subject appears as part of the output of `systemctl status mailman3`. Anyone fixed it? -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 In an Internet failure case, the #1 suspect is a constant: DNS. "Oh, the cruft.", egrep -v '^$|^.*#' ¯\_(ツ)_/¯ :-) [How to ask smart questions: http://www.catb.org/~esr/faqs/smart-questions.html]

2 2

New Lists Appear in HyperKitty Archives, But No Mail Data is Shown
by Asif Muzammil July 14, 2025

July 14, 2025

Hello, I have been troubleshooting an issue with our Mailman 3 archives. I have made a key observation that has helped narrow down the problem. While a newly created list appears in the HyperKitty archives, no email data is shown inside that list. *Problem:* When I create a new mailing list in Mailman Core and then visit the HyperKitty archives, the list is correctly displayed. However, when I send an email to that list, no mail data (messages) are shown inside it. The message count for the list remains at zero in the archives UI. *Expected Behavior:* I expect that when a new list is created, it should appear in the archives, and any emails sent to it should also be correctly displayed within that list. *System Environment:* - *Mailman Version:* Mailman 3.x - *Archiver:* HyperKitty - *Web Interface:* Postorius - *Database:* PostgreSQL - *Setup:* Mailman Core (/opt/mailman/venv) and the web components (venv_hk) are running in separate virtual environments on the same server. *Troubleshooting Steps Taken:* 1. *List Synchronization:* Creating a new list in Mailman Core immediately makes it visible in the HyperKitty archives, which confirms that the mailmanweb service is correctly communicating with the hyperkitty_mailinglist table. 2. *Email Injection Test:* I used the mailman shell to inject a test email directly into a list. The inject_message function executed without errors. 3. *Configuration Check:* I have verified that the base_url in the Mailman archiver configuration (mailman-hyperkitty.cfg) and the MAILMAN_ARCHIVER_KEY in the HyperKitty settings file (settings.py) are correctly set and match. Could this be due to an API key or permissions issue preventing the data from being written, even though the tables exist? Any advice on what to check next would be extremely helpful. Thanks

4 9

Re: RFC-2369 List-Unsubscribe header missing
by Stephen J. Turnbull July 14, 2025

July 14, 2025

Hi Paul Glad you got that sorted! I wonder why those filters were in there? Not really the kind of thing I'd expect Postfix to be doing out of the box! Looks like you didn't send to the list (at least, I got my copy via personal mail, not the list). Replying for the benefit of Mailman-Users (nothing new to you here). Sparks, Paul writes: > Postfix was filtering the headers. I was not familiar with this > postfix functionality. > > For others who might see this, the filter lines were in /etc/postfix/header_checks > /^List-Unsubscribe:/ IGNORE > /^List-Subscribe:/ IGNORE > > How are you capturing mail at postfix? > > For capturing the email at postifx I used the following to defer SMTP sending. > postconf -e defer_transports=smtp; postfix reload That swooshing sound is that concept returning to hover over my head. > # send the email to the server > mailq > # Use ID from mailq to examine the email using postcat > postcat -vq C23D82B60024 > # Re-enable SMTP sending > postconf -e defer_transports=; postfix reload; postfix flush That's potentially useful in a lot of situations! Steve -- GNU Mailman consultant (installation, migration, customization) Sirius Open Source https://www.siriusopensource.com/ Software systems consulting in Europe, North America, and Japan

1 0

RFC-2369 List-Unsubscribe header missing
by paul.sparks＠us.idemia.com July 12, 2025

July 12, 2025

Thanks to Mark Sapiro and others who have answered questions on this list. It has made my job easier. I'm hoping someone can help with another issue I'm seeing. I'm setting up a new mailman 3 server and doing some test posts to a list. This server is intended only as an announcement list with limited members allowed to post. The resulting emails coming from the list do not include the List-Unsubscribe header although other "List-XXX" headers are included. I'm trying to figure out why the header is missing. It seems clear from the code that the header should be included. Here is partial text of the email source as received from the list. I've tried viewing in Outlook.com, Thunderbird, and even captured the email text at postfix prior to sending to look for the missing header. My domain has been replaced with "example.com" throughout. This is from the postfix capture ________ ... To: "test1(a)example.com" <test1(a)example.com> Thread-Topic: Post 8 test Thread-Index: AQHb8oBBzPBOYgrDxUOIiAxhvUzthA== Date: Fri, 11 Jul 2025 16:24:38 +0000 Accept-Language: en-US Content-Language: en-US msip_labels: authentication-results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=example.com; MIME-Version: 1.0 Message-ID-Hash: FWHZCRE4EA6IFP4BZTPIF23DKAKVRMDT X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop; banned-address; emergency; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header X-Content-Filtered-By: Mailman/MimeDel 3.3.10 From: "Somewhere1, SW" <test1(a)example.com> X-Mailman-Version: 3.3.10 Precedence: list Reply-To: no-reply(a)example.com Subject: [Test1] Post 8 test List-Id: "Somewhere1, SW" <test1.example.com> List-Help: <mailto:test1-request@example.com?subject=help> List-Owner: <mailto:test1-owner@example.com> List-Post: NO Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Message-Id: <20250711163343.017C31DBB79(a)example.com> Body text ... __________ Details on my test setup: # Mailman version. __version__ = '3.3.10' OS: Amazon Linux 2023 (RHEL 9 compatible) MTA: postfix-3.7.2 I am using the following code with mailman shell to configure the list. include_rfc2369_headers is left at the default value of True (verified via mailman shell) ____________ # Description value must be passed as a parameter def configure_list(ml, description): #ml.accept_these_nonmembers = allowed_to_post ml.admin_immed_notify = False ml.advertised = False ml.allow_list_posts = False ml.anonymous_list = True ml.archive_policy = ArchivePolicy.never ml.bounce_info_stale_after = datetime.timedelta(days=90) ml.bounce_notify_owner_on_disable = False ml.bounce_notify_owner_on_removal = False ml.bounce_score_threshold = 3 ml.bounce_you_are_disabled_warnings = 0 ml.default_member_action = Action.discard ml.default_nonmember_action = Action.discard ml.description = description ml.digests_enabled = False ml.filter_content = True ml.first_strip_reply_to = True ml.forward_auto_discards = False ml.max_days_to_hold = 1 ml.max_message_size = 0 ml.nntp_prefix_subject_too = False ml.pass_extensions = [] ml.pass_types = ['text/plain','multipart/alternative'] ml.reply_goes_to_list = ReplyToMunging.explicit_header ml.reply_to_address = "no-reply(a)example.com" ml.require_explicit_destination = False ml.respond_to_post_requests = False # Allow select members to post to list for mbr in ml.members.members: if str(mbr.address) in allowed_to_post: mbr.moderation_action = Action.defer print('end',mbr.address,mbr.moderation_action) _______________ /etc/mailman3/mailman.cfg file: # /etc/mailman3/mailman.cfg # These settings override the defaults in the following 2 config files # /opt/mailman/lib/python3.9/site-packages/mailman/config/schema.cfg # /opt/mailman/lib/python3.9/site-packages/mailman/config/mailman.cfg [paths.custom] var_dir: /data/lib/mailman pid_file: /data/lib/mailman/pid/master.pid lock_file: /data/lib/mailman/pid/master.lock [mailman] layout: custom # This address is the "site owner" address. Certain messages which must be # delivered to a human, but which can't be delivered to a list owner (e.g. a # bounce from a list owner), will be sent to this address. It should point to # a human. site_owner: ml-owner(a)example.com [database] class: mailman.database.postgresql.PostgreSQLDatabase url: postgresql:///mailman # No archiving enabled [archiver.prototype] enable: no [archiver.hyperkitty] enable: no [shell] history_file: $var_dir/history.py [mta] verp_confirmations: yes verp_personalized_deliveries: yes verp_delivery_interval: 1 [passwords] # When Mailman generates them, this is the default length of passwords. password_length: 15 [logging.debug] path: debug.log level: debug ______________ With debug enabled I see the rfc-2369 pipeline being run when processing the post to the list. Jul 11 17:51:32 2025 (444088) [IncomingRunner] starting oneloop Jul 11 17:51:32 2025 (444088) [IncomingRunner] processing filebase: 1752256291.9136803+bfd305a28151ce750dde5114b413b351fa4d1284 Jul 11 17:51:32 2025 (444088) [IncomingRunner] processing onefile Jul 11 17:51:32 2025 (444090) [OutgoingRunner] starting oneloop Jul 11 17:51:32 2025 (444090) [OutgoingRunner] ending oneloop: 0 Jul 11 17:51:32 2025 (444097) [DigestRunner] starting oneloop Jul 11 17:51:32 2025 (444097) [DigestRunner] ending oneloop: 0 Jul 11 17:51:32 2025 (444088) [IncomingRunner] finishing filebase: 1752256291.9136803+bfd305a28151ce750dde5114b413b351fa4d1284 Jul 11 17:51:32 2025 (444088) [IncomingRunner] doing periodic Jul 11 17:51:32 2025 (444088) [IncomingRunner] committing transaction Jul 11 17:51:32 2025 (444088) [IncomingRunner] checking short circuit Jul 11 17:51:32 2025 (444088) [IncomingRunner] ending oneloop: 1 Jul 11 17:51:32 2025 (444088) [IncomingRunner] starting oneloop Jul 11 17:51:32 2025 (444088) [IncomingRunner] ending oneloop: 0 Jul 11 17:51:32 2025 (444087) [CommandRunner] starting oneloop Jul 11 17:51:32 2025 (444087) [CommandRunner] ending oneloop: 0 Jul 11 17:51:32 2025 (444091) [PipelineRunner] starting oneloop Jul 11 17:51:32 2025 (444091) [PipelineRunner] processing filebase: 1752256292.373131+a372784a3ff7a76a662ed1c62c224a53a8008b8b Jul 11 17:51:32 2025 (444091) [PipelineRunner] processing onefile Jul 11 17:51:32 2025 (444096) [VirginRunner] starting oneloop Jul 11 17:51:32 2025 (444096) [VirginRunner] ending oneloop: 0 Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: validate-authenticity Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: mime-delete Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: tagger Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: member-recipients Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: avoid-duplicates Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: cleanse Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: cleanse-dkim Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: cook-headers Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: subject-prefix Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: rfc-2369 Jul 11 17:51:32 2025 (444091) RFC 2369 process Jul 11 17:51:32 2025 (444091) RFC 2369 process continue 1 Jul 11 17:51:32 2025 (444091) RFC 2369 process headers extend Jul 11 17:51:32 2025 (444091) RFC 2369 process header for loop Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: to-archive Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: to-digest Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: to-usenet Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: after-delivery Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: acknowledge Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: dmarc Jul 11 17:51:32 2025 (444091) <20250711175131.B8CEC1DA51B(a)example.com> pipeline default-posting-pipeline processing: to-outgoing Jul 11 17:51:32 2025 (444091) [PipelineRunner] finishing filebase: 1752256292.373131+a372784a3ff7a76a662ed1c62c224a53a8008b8b Jul 11 17:51:32 2025 (444091) [PipelineRunner] doing periodic Jul 11 17:51:32 2025 (444091) [PipelineRunner] committing transaction Jul 11 17:51:32 2025 (444091) [PipelineRunner] checking short circuit Jul 11 17:51:32 2025 (444091) [PipelineRunner] ending oneloop: 1 Jul 11 17:51:32 2025 (444091) [PipelineRunner] starting oneloop Jul 11 17:51:32 2025 (444091) [PipelineRunner] ending oneloop: 0 Jul 11 17:51:32 2025 (444085) [ArchiveRunner] starting oneloop Jul 11 17:51:32 2025 (444085) [ArchiveRunner] ending oneloop: 0 Jul 11 17:51:33 2025 (444090) [OutgoingRunner] starting oneloop Jul 11 17:51:33 2025 (444090) [OutgoingRunner] processing filebase: 1752256292.8885148+5e5c4a68eacfead13561df616fe7ca19e1d1d182 Jul 11 17:51:33 2025 (444090) [OutgoingRunner] processing onefile Jul 11 17:51:33 2025 (444097) [DigestRunner] starting oneloop Jul 11 17:51:33 2025 (444097) [DigestRunner] ending oneloop: 0 Jul 11 17:51:33 2025 (444090) [outgoing] <function deliver at 0x7f21ed7b1310>: <175225629286.444091.14942604217243398224(a)ip-10-0-0-168.us-gov-west-1.compute.internal> Jul 11 17:51:33 2025 (444088) [IncomingRunner] starting oneloop _____________ I tried inserting some log.debug calls into lib/python3.9/site-packages/mailman/handlers/rfc_2369.py to see where headers were being removed and added. I modified: log = logging.getLogger('mailman.debug') and inserted log.debug throughout the process function and one in the top level. ___________ ... @public @implementer(IHandler) class RFC2369: """Add the RFC 2369 List-* headers.""" name = 'rfc-2369' description = _('Add the RFC 2369 List-* headers.') log.debug('RFC 2369 processing') def process(self, mlist, msg, msgdata): """See `IHandler`.""" process(mlist, msg, msgdata) __________ When running with debug enabled, I see this top level message many times at the top of the debug.log file during mailman startup, but not after that. I don't see any of my other debug messages. _____________ Jul 11 16:03:36 2025 (409226) RFC 2369 processing Jul 11 16:03:40 2025 (409244) RFC 2369 processing Jul 11 16:03:41 2025 (409250) RFC 2369 processing Jul 11 16:03:49 2025 (409271) RFC 2369 processing Jul 11 16:03:50 2025 (409271) [DigestRunner] starting oneloop Jul 11 16:03:50 2025 (409271) [DigestRunner] ending oneloop: 0 Jul 11 16:03:50 2025 (409261) RFC 2369 processing Jul 11 16:03:50 2025 (409261) [BounceRunner] starting oneloop Jul 11 16:03:50 2025 (409261) [BounceRunner] ending oneloop: 0 Jul 11 16:03:50 2025 (409264) RFC 2369 processing Jul 11 16:03:50 2025 (409267) RFC 2369 processing Jul 11 16:03:51 2025 (409262) RFC 2369 processing Jul 11 16:03:51 2025 (409262) [CommandRunner] starting oneloop Jul 11 16:03:51 2025 (409262) [CommandRunner] ending oneloop: 0 Jul 11 16:03:51 2025 (409266) RFC 2369 processing Jul 11 16:03:51 2025 (409266) [PipelineRunner] starting oneloop Jul 11 16:03:51 2025 (409266) [PipelineRunner] ending oneloop: 0 ... __________ Any suggestions on where else I can look? Any thoughts on why my debug calls are not showing in the logs?

2 1

Displaying addresses in emails from a Mailman 3 installation
by Steve Brown July 5, 2025

July 5, 2025

In a list I administer, the From line in an incoming email shows [Poster's Name] via [List Name] <address for posting a reply> . I use Outlook as my client, but at least some other clients behave similarly. However, I have at least one subscriber who uses Mozilla's Thunderbird as his client, and periodically the format changes to something else, most recently to just [List Name <address for posting a reply> . I suspect this occurs from some combination of the way Thunderbird parses the header and the way it populates some kind of address book that makes autofill work, but I'm not sure. Is this a known problem? Is there an easy fix that will be stable? In the past, putting an association in his address book and removing others worked, but it didn't hold and he is frustrated.

3 12