Search results for "sapiro" - lists.mailman3.org

[MM3-users] Re: Adding members to a mailman3 mailing list

by Mark Sapiro

On 10/15/20 2:07 AM, Onyeibo Oku wrote: > > In the meantime, I moved to option #3 (recommended for production). > I hit a grey area when setting up the virtual environment. At > https://docs.mailman3.org/en/latest/install/virtualenv.html#virtualenv-inst…, > I see: > > $ cd /opt/mailman > $ python3 -m venv venv > > The documentation imply that a 'user' creates the virtual environment > but is not explicit about that 'user'. The user can be any user you want. I use `mailman` on my production installations and `mark` on my development installation. The key is that user should be set as owner of /opt/mailman and then creating the venv and subsequent steps are all done as that user. > (1) Is it better to have a dedicated user (say 'mailman') then use that > while following the docs? I see a lot of docs referring to a 'mailman' > user. Installing mailman3 from distribution repository introduced a > 'mailman' user to my box (with special UID). There has to be reasons > for that. Yes, a dedicated Mailman user is best. > (2) Should the setup proceed with 'root'? (why do I get the feeling > that this is not the design?) No. setup should proceed as the Mailman user. > (3) How about using an already existing local user? What are the > implications? That's OK, but there are potential security implications due to the fact that the Mailman processes running as that user have access to all that user's non-mailman stuff. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

4 years, 10 months

[MM3-users] Re: Moderation -- implicit destination behavior

by Mark Sapiro

On 5/21/25 9:31 AM, Steve Bachinsky wrote: > Hi. We have noticed some posts being held for moderation (even though user has permission to post to list) and Mailman reports: > > moderation_reasons: ['Message has implicit destination’] > > The lists are configured for “require_explicit_destination”. However, the email address of the list is clearly in the CC: header of the email (no alias or subdomain). Is there some other reason why Mailman would hold an email for moderation with the reason of “implicit destination”? No. The rule looks at all the addresses in To:, Cc:, Resent-To: and Resent-Cc: headers using standard Python email library methods, lower cases them any if any lc address matches the list posting address or an acceptable_alias, the rule misses, i.e. doesn't hold the message. There is no other reason. > This moderation is not consistent for same list and poster. The emails come from a large domain with multiple load-balancing email servers. I suspect that it may be something with one or more of the servers being misconfigured, but unable to see anything obvious from the headers and confused why Mailman states the reason to be implicit destination when that does not seem to be the cause. I'm sorry, but I have no idea why this is happening. If the headers of the held message contain the explicit address, I have no clue as to why it is held for that reason. Maybe there is some other reason why the message is held, and the reason is misstated, but I don't know how that could happen. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

3 months, 3 weeks

[MM3-users]Re: Subscribing via email

by Mark Sapiro

On 04/19/2016 11:31 AM, Andrew Hodgson wrote: > Hi, > > I subscribed to this list the other day via email, and I didn't get any email back suggesting I was subscribed. I logged in today via Google accounts, and it said I was subscribed to the list, but all the options for receiving list traffic, my own posts etc weren't filled in - I had to fill them in manually. 1 - You should have received a list welcome message and a 'results of your email commands' message in response to your email subscribe request. I'm Ccing you directly as it appears you can't get any list mail[1]. Visit <https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/> to see the aechives. 2 - The non-filling in of your options is a known issue. See <https://gitlab.com/mailman/mailman/issues/187>. Basically, you get 'default' settings, but they aren't explicitly set for your subscription by the Mailman core. [1] Here's a log message: Apr 6 23:42:22 mailman postfix/smtp[6809]: A9C9B800D9: to=<andrew(a)hodgsonfamily.org>, relay=hodgsonfamily-org.mail.protection.outlook.com[213.199.154.23]:25, delay=5.5, delays=0.05/0.05/0.29/5.1, dsn=5.7.606, status=bounced (host hodgsonfamily-org.mail.protection.outlook.com[213.199.154.23] said: 550 5.7.606 Access denied, banned sending IP [104.239.228.201]. To request removal from this list please visit https://sender.office.com/ and follow the directions. For more information please go to http://go.microsoft.com/fwlink/?LinkID=526653 (AS16012609) (in reply to RCPT TO command)) I have just now gone to https://sender.office.com/ and gone through the delisting process with this result: "Your IP 104.239.228.201 was successfully delisted. This may take up to 30 minutes to take effect." -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

9 years, 4 months

[MM3-users] Re: Upgrading packages

by Mark Sapiro

On 10/13/22 03:49, Eggert Ehmke wrote: > > pip install --upgrade git+https://gitlab.com/mailman/hyperkitty@master > > I noticed that mistune-2.0.0rc1 was upgraded to mistune-2.0.4. > > All packages were upgraded fine. I applied the patch to > venv/lib/python3.9/site-packages/hyperkitty/lib/renderer.py It shouldn't have been necessary to apply any patch. The head of the GitLab hyperkitty master branch is already compatible with mistune 2.0.4 > to change the mistune import line. When I run now > mailman-web migrate > I get: > *CommandError: Conflicting migrations detected; multiple leaf nodes in the migration > graph: (0017_auto_20220628_1132, 0018_alter_emailtemplate_language in postorius).* > *To fix them run 'python manage.py makemigrations --merge'* > > However, when I run the suggested command as: > $ python ./venv/lib/python3.9/site-packages/mailman_web/manage.py makemigrations -- > merge > No conflicts detected to merge. There's something strange here. If you look at the migrations at the head of the gitlab branch at https://gitlab.com/mailman/postorius/-/tree/master/src/postorius/migrations, you'll see that migration 17 is named 0017_alter_emailtemplate_language.py, not 0017_auto_20220628_1132. I suggest comparing the content of your postorius/migrations/ directory with the GitLab head and resolving any differences. > Can I just ignore this behavior? At the moment, I don't dare to carry out these steps on my > production system. Would be nice to shed some light on this. I'm guessing at some point you installed something which required you to manually make this migration and that produced this conflict. Do you have both 0017_alter_emailtemplate_language.py and 0017_auto_20220628_1132 in your postorius/migrations/ directory? If so, probably just removing 0017_auto_20220628_1132 will fix this? -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

2 years, 11 months

[MM3-users] Re: Splitting hosts: mta&core vs postorius

by Mark Sapiro

On 6/21/20 9:38 AM, Ruth Ivimey-Cook wrote: > > > Problem is, if I do that (which I did in the beginning), the core rest > service listens on localhost / 127.0.0.1 (it obviously recognises that > as the , and so cannot be connected to by anything. That is why I set it > to 0.0.0.0 to begin with, and what the documentation implies (fairly > strongly) that this parameter is about. What documentation? I do understand what you are saying and I see the issue. I don't understand why it works for this poster <https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message…> but maybe that's specific to Docker containers. > IMO, the hostname parameter needs to be split up into a "what address > should the rest of the world use to address me" and separately "what > address(es) should I listen on". Multiple listen addresses enable > multihomed hosts to get the right connectivity, too. Try something like this and let us know if it works? If so, we can do something like add a config setting for this. > --- a/src/mailman/rest/gunicorn.py > +++ b/src/mailman/rest/gunicorn.py > @@ -53,7 +53,7 @@ def make_gunicorn_server(): > # We load some options from config.webservice, while other extended options > # for gunicorn can be defined in configuration: section. We also load up > # some logging options since gunicorn sets up it's own loggers. > - host = config.webservice.hostname > + host = '0.0.0.0' > port = int(config.webservice.port) > log_path = os.path.join(config.LOG_DIR, config.logging.http['path']) > options = { -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

5 years, 2 months

[MM3-users] Re: Mails are stuck in Mailman Pipeline Queue

by Mark Sapiro

On 3/26/22 10:28, Shashikanth Komandoor wrote: > Hi Team, > > We have restarted the services as well but no change. Odd. > We tried manually moving the pck file from pipeline queue to > out queue but the file disappeared from out queue but is not getting > delivered. This is expected because if the message hasn't been processed through the pipeline, it has no recipients in the metadata, so the outgoing runner delivered it to no one. Also, it skipped archiving and digesting. > We moved the pck file from pipeline queue to out queue with the > extension .bak still no use. Same as above. > Not sure what to try next. Please help. Mi first guess is the pipeline runner isn't running, but you restarted the services, presumably Mailman core, and that might fix it, but there are various ways the pipeline runner could have died that would cause the master to not restart it. You need to stop Mailman core and then start it rather than just restarting. That may help. Also look in mailman.log. Assuming the pipeline runner died at some point, there should be a log entry and possible an exception with traceback indicating why. You can see which runners are running with something like ``` ps -fwwA|grep runner=|sed s/.*runner=// ``` This should produce a list something like ``` archive:0:1 bounces:0:1 command:0:1 in:0:1 lmtp:0:1 nntp:0:1 out:0:1 pipeline:0:1 rest:0:1 retry:0:1 task:0:1 virgin:0:1 digest:0:1 rest:0:1 rest:0:1 ``` except depending on version, `task` may not be there. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

3 years, 5 months

[MM3-users] Re: django-allauth failures, but only for some users (independent of browser?)

by Mark Sapiro

On 12/6/23 11:00, Jered Floyd wrote: > > These errors are generated to me as the admin, and in the elided environment section SITE_ID is indeed 0, but I'm at a loss to imagine why this happens only sometimes! For me, on a test installation with SITE_ID = 0, this only occurs at https://example.com/accounts/social/connections/ Other URLs such as https://example.com/accounts/password/change/ and https://example.com/accounts/email/ do not throw this exception. Here's the code: ``` def get_current(self, request=None): """ Return the current Site based on the SITE_ID in the project's settings. If SITE_ID isn't defined, return the site with domain matching request.get_host(). The ``Site`` object is cached the first time it's retrieved from the database. """ from django.conf import settings if getattr(settings, "SITE_ID", ""): site_id = settings.SITE_ID return self._get_site_by_id(site_id) elif request: return self._get_site_by_request(request) raise ImproperlyConfigured( 'You\'re using the Django "sites framework" without having ' "set the SITE_ID setting. Create a site in your database and " "set the SITE_ID setting or pass a request to " "Site.objects.get_current() to fix this error." ) ``` I tried patching it from ``` if getattr(settings, "SITE_ID", ""): ``` to ``` if getattr(settings, "SITE_ID", None) is not None: ``` but that's even worse. It throws a bunch of "django.contrib.sites.models.Site.DoesNotExist: Site matching query does not exist." errors. Somehow, get_current is being called without a request object in the cases where it fails. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

1 year, 9 months

[MM3-users] Re: Mailman rewrite the Cc header incorrectly

by Mark Sapiro

On 5/24/20 7:20 PM, Alan So wrote: > Really thanks for your hard work Mark! It is a bit strange that parseaddr behave differently with CRLF and LF. Maybe if the server is a Windows, the behavior is consistent. It might be that on Windows, `mailman inject` would create the message with CRLF line endings. > May I know when or whether will this patch be committed to the repository so that it will be effective in future release of Mailman? Thanks. In the absence of an issue at <https://gitlab.com/mailman/mailman/-/issues>, probably never. <hint> The bottom line is the original mail created by Outlook is defective.[1] While Mailman could defend against this defect, starting to do that opens the possibility that there are other areas of the code that need a similar defense. Have you asked Microsoft to fix Outlook to not create these defective messages? That's the real solution. That said, if an issue is filed, someone might consider adding something along the lines of <https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message…> to the code base. [1] given an address of the form display-name (comment) <addr-spec> in a header, it can be folded either between the display-name and the (comment) or between the (comment) and the <addr-spec>, but if display-name (comment) is quoted as in "display-name (comment)" <addr-spec> "display-name (comment)" is now a phrase within which folding is not allowed and this can only be folded between the "display-name (comment)" and the <addr-spec>. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

5 years, 3 months

[MM3-users] Re: Change language to welcome message in mailman3

by Mark Sapiro

On 7/24/20 2:10 PM, Massimo Zappalà wrote: > Hi everyone, > > My Version installed on Debian buster: > mailman3: 3.2.1 > postorius: 1.2.4 > hyperkitty: 1.2.2 > > All interfaces work well in Italian language, but "Welcome message" is in English. First question: Is the Subject: in English or Italian? If the Subject: is in English you need to compile the Italian message catalog as I described in my reply at <https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message…>. Then, as far as the message body is concerned, it is built from a template. At present, in the Mailman source distribution, the only templates that are translated are French and German You will find the complete set of English templates at /usr/lib/python3/dist-packages/mailman/templates/en/. You have choices. 1) You can create /usr/lib/python3/dist-packages/mailman/templates/it/ and make Italian translations of all the English templates in the it/ directory in files of the same names, and submit a merge request at <https://gitlab.com/mailman/mailman/-/merge_requests> to merge them into the source distribution. If you are not familiar with software development in general and Mailman development in particular, this is probably beyond what you can do. 2) Find Mailman's var/ directory. This is the directory containing archives/, data/, etc/, lists/, locks/, logs/ and perhaps other sub-directories. You then need to create the full path to var/templates/site/it/ and place Italian translations of some or all the templates there. When doing so, you can add site customizations. 3) You can use Postorius to create templates for individual lists. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

5 years, 1 month

[MM3-users] Re: Mailman Core 3.3.2 is now out

by Mark Sapiro

On 11/24/20 7:22 PM, Abhilash Raj wrote: > > I think I should keep in mind next time to use the #<heading-name> like `#3.3.2` pointed out by Torge, instead of `#id` (which is what I see when I hover over the Link in browser so I copied ;). It's more complicated than that. Sometimes the fragment name is the heading and sometimes it is `#idn`. At the moment, neither #3.3.2 nor #3.3.3 work at all as these are #id2 and #id1 respectively. Currently the fragment for 3.3.3 Bugs is #bugs and the fragment for 3.3.2 Bugs is #id3, but presumably prior to the addition of 3.3.3, #bugs was Bugs under 3.3.2, Just as now, #command-line is Command line under 3.3.2, but when a Command line section is added under 3.3.3, that will probably get the command-line id. Thus, the note "Permalink to this headline" es misleading at best because it isn't very permanent fo an evolving document. > You can specify explicit anchor name in .rst too, I think just adding a `.. _news-3.3.2` just above the heading will generate an anchor on the page with #news-3.3.2 (I haven't verified this but I think it will!) which links to the heading. I too think this is correct. I suppose it might be worth doing for NEWS.rst in particular since Sphinx doesn't seem to like fragments like n.n.n. Another way to deal with this would be to always use code names. It seems like a heading `n.n.n` will always get a `#idn` id, but, e.g. `3.3.0 – “Tom Sawyer”` gets `#tom-sawyer` -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

4 years, 9 months

Search results for query "sapiro"