Search results for "Postorius" - Mailman-users

Re: message awaits moderator approval

by Eggert Ehmke

I think I solved that. The background might be of interest for other admins, and for the developers. I had set up some new mailing lists on that server. I also had deleted an old mailing list for a domain that is no longer hosted on that server. To delete the old mailing list, I used the methods provided by postorius. Now it turned out that there were several relicts of the old list in the database, in particular in the template table. I removed them using direct SQL commands. Now my new mailing lists work as expected. Is this a bug in postorius? I think when I delete a mailing list, all relicts should be deleted from the database. Am Mittwoch, 21. Dezember 2022, 14:46:56 CET schrieb Eggert Ehmke via Mailman- users: > When I send a mail to a list that exceeds the max size, I get the above > answer. However, the moderator of that list does not get a message, and > there is no entry in the list of hold back messages. > > This should be a configuration problem, on another list on the same server > this works fine. I cannot find any differences in the settings. > > > > _______________________________________________ > Mailman-users mailing list -- mailman-users(a)mailman3.org > To unsubscribe send an email to mailman-users-leave(a)mailman3.org > https://lists.mailman3.org/mailman3/lists/mailman-users.mailman3.org/ > Archived at: > https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message > /AIYSQM3XBDBQOJTOJQVXRT5PGOVUIJ73/ > > This message sent to eggert(a)eehmke.de

3 years, 5 months

request to held messages yields set of email addresses

by Stephen J. Turnbull

Thomas Krichel writes: > File "/usr/lib/python3.12/json/encoder.py", line 180, in default > raise TypeError(f'Object of type {o.__class__.__name__} ' > TypeError: Object of type set is not JSON serializable This appears to be related to https://gitlab.com/mailman/mailman/-/work_items/668. There's a patch at the end that may help. I can't promise it will. But I have assigned the issue to myself for later followup. > I suspect that to get to the bottom of this problem, I need to use > the low-level postorious functions that actually access the parse > the held message data. No, I think you want to make sure that the REST API doesn't pass a set to the JSON encoder. I suppose that there are a bunch of "things" there that might be "the same", and the developer who wrote the code wanted to compress the representation a bit. But they didn't realize that sets won't be serialized by the stdlib json module. I doubt very much that Postorius cares: it probably just treats that object as a sequence. I don't understand why sets (like tuples and lists) don't just encode to JSON sequences, and let the deserializer decide which model is best for the container, but that's life, I guess. Regards, Steve -- GNU Mailman consultant (installation, migration, customization) Sirius Open Source https://www.siriusopensource.com/ Software systems consulting in Europe, North America, and Japan

2 months

Container Images for Mailman 3.2 release

by Abhilash Raj

Hi All, I have updated and pushed releases for Container Images with 3.2 release of the suite. They are available at: docker.io/maxking/mailman-web:0.2 docker.io/maxking/mailman-core:0.2 A similar copy of both is also available at quay: quay.io/maxking/mailman-web:0.2 quay.io/maxking/mailman-core:0.2 The releases contain the following versions of components: mailman-core: - Mailman Core 3.2.0 - Mailman-hyperkitty 1.1.0 mailman-web: - Postorius 1.2.1 - Hyperkitty 1.2.0 - Django-mailman3 1.2.0 - MailmanClient 3.2.0 The upgrade process from 0.1 release should be relative painless, except for the full-text index, which needs to be rebuilt due to the migration from Python 2 to 3 of the full-text indexing library whoosh. You can run the following commands to do so, after the upgrade: $ docker exec -it mailman-web bash bash # python manage.py rebuild_index If you are not using the default docker-compose.yaml file provided in the repo, you need to add an extra environment variable for mailman-web container, POSTORIUS_TEMPLATE_BASE_URL, which defaults to `http://mailman-web:8000`. It should be the URL where your mailman-web container is listening on with HTTP (port 8000 by default). The rolling releases are not being rolled out right everyday now, because of broken build scripts. I'd recommend everyone use the new stable releases that I mentioned above. Going forward, I'll try to release stable versions more often, so that there isn't any need for rolling version, but fixing the rolling releases is on my TODO. Documentation for these images is available at: https://github.com/maxking/docker-mailman -- thanks, Abhilash Raj (maxking)

7 years, 10 months

Re: after last upgrade I've got a CSRF failed verification

by Mark Sapiro

On 10/26/22 04:33, Guillermo Hernandez (Oldno7) via Mailman-users wrote: > > I did it via pip and all seems to be working, but when I try to discard > or accept any message via postorius it shows a Forbidden 403 error CSRF > verification failed. > > In my settings.py I have commented out these variables: > > # Other security settings > # SECURE_SSL_REDIRECT = True > # If you set SECURE_SSL_REDIRECT to True, make sure the > SECURE_REDIRECT_EXEMPT > # contains at least this line: > # SECURE_REDIRECT_EXEMPT = [ > # "archives/api/mailman/.*", # Request from Mailman. > # ] > # SESSION_COOKIE_SECURE = True > # SECURE_CONTENT_TYPE_NOSNIFF = True > # SECURE_BROWSER_XSS_FILTER = True > # CSRF_COOKIE_SECURE = True > # CSRF_COOKIE_HTTPONLY = True > # X_FRAME_OPTIONS = 'DENY' > > I've been messing with them whith no results. It's like it lacked some > value in some template... It's not an issue with the code per se or the templates. These updates are all installed on mail.python.org and lists.mailman3.org and are working fine. After upgrading did you run django-admin collectstatic --clear django-admin compress django admin migrate If you have and it doesn't help, you probably need to add CSRF_TRUSTED_ORIGINS to your settings.py. See https://docs.djangoproject.com/en/4.1/ref/settings/#csrf-trusted-origins and note the differences between Django <4.0 and >=4.0. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

3 years, 7 months

Re: Mailman3 via Docker and NGINX

by Abhilash Raj

On Wed, Dec 19, 2018, at 8:18 AM, Mark Sapiro wrote: > On 12/19/18 7:03 AM, brian(a)emwd.com wrote: > > I have Mailman3 installed on CentOS 7 using maxking's docker image. I believe the installation went fine and running "curl http://172.19.199.3:8000/postorius/lists/" produces the correct output. I also will be using exim and nginx with this installation. > ... > > location / { > > include uwsgi_params; > > uwsgi_pass 172.19.199.3:8000; Change this to use port 8080 instead of 8000. 8000 is for HTTP protocol and 8080 is for uwsgi protocol. See the docs[1]. [1]: https://github.com/maxking/docker-mailman#setting-up-your-web-server > > uwsgi_read_timeout 300; > > proxy_set_header Host $host; > > proxy_set_header X-Forwarded-For $remote_addr; > > } > > } > > > Since you can curl http://172.19.199.3:8000/postorius/lists/ and get the > correct output, uwsgi or whatever your wsgi process is is working at > that level, and I don't think you should be invoking uwsgi in nginx. I > think you should be using proxy_pass. > > See > <https://wiki.list.org/DOC/Mailman%203%20installation%20experience?action=At…> > for the nginx config used on lists.mailman3.org. > > -- > Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, > San Francisco Bay Area, California better use your sense - B. Dylan > _______________________________________________ > Mailman-users mailing list -- mailman-users(a)mailman3.org > To unsubscribe send an email to mailman-users-leave(a)mailman3.org > https://lists.mailman3.org/mailman3/lists/mailman-users.mailman3.org/ > -- thanks, Abhilash Raj (maxking)

7 years, 6 months

Re: spanish localization

by Victoriano Giralt

On Fri, 2021-02-12 at 12:24 -0800, Mark Sapiro wrote: > > If the manage.py in the directory in which you ran this is the > correct one for your Mailman Django project, I don't know why it > didn't work. Me neither :-( That's why I asked, and probably Guillermo, as well. > Is your Mailman installed in a virtualenv? If so, was the virtualenv > activated? No virtualenv. Mine is deployed using system python on a CentOS 7 system, toghether with another in house developed low requirements Django project, both served by uWSGI vassals. And then, all of a sudden, while I was writting this, it dawned on me! Postorius, Hyperkitty and django-mailman are installed with pip, so, the end up in the site-packages directory (usually /usr/local/lib/python3.x/site-packages), but, normally, django projects are deployed somewhere in the web server hierarchy (/var/ww/django in my case, due to old personal tastes), and applications properly defined in the settings.py for the project. So, the message files are out of reach for manage.py. Changing to /usr/local/lib/python3.x/site-packages/X (with X being postorius, hyperkitty and django_mailman3) and using django-admin compilemessages all works! Thanks for pushing my thoughts in the right direction! -- Victoriano Giralt Innovation Director Digital Transformation Vicerectorate University of Malaga +34952131415 SPAIN ================================================================== Note: signature.asc is the electronic signature of present message A: Yes. > Q: Are you sure ? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting annoying in email ?

5 years, 4 months

Re: Patch/kludge suggestion and a question on adding accepted addresses

by Barry Warsaw

On Apr 23, 2017, at 10:33 PM, David Krantz wrote: >I found out that I had a broken email in one of my held messages >queues). That gave a HTTP 500 error although that should be fixed . >After some digging I found out that where the error occured. > >I'd suggest adding a catch for AttributeError similar to the one for >KeyError in _HeldMessageBase in rest/post_moderation.py. It solved my >immediate problem at least. I run the version from git with the tag >263711b3b7988c7c474e3b8952dc7e0f02f04fa2 Can you file a bug and attach either the problematic message or the traceback? >The message causing this was probably one where I had sent a minimal >email via telnet to the SMTP port and forgot the MessageID-header or >something similarily silly. But it prevented me from seeing the held >messages-list in postorius. Did you talk to your MTA's SMTP port, to Mailman's LMTP port, or some other process? It matters because LMTP is the way messages get from the upstream SMTP daemon (e.g. Postfix, Exim) into Mailman, and the LMTP server rejects any messages that don't have a Message-ID. >I have another question: In the old mailman you could add accepted >senders without adding them to the list. Is that possible in Mailman >3/Postorius? You do this by creating a "nonmember" subscription for the sender, and then setting that member's moderation_action to Action.accept. Cheers, -Barry

9 years, 1 month

Re: spanish localization

by Victoriano Giralt

On Fri, 2021-02-12 at 12:36 -0800, Mark Sapiro wrote: > > Why do you say the .mo files are old. The timestamps are all Feb 3 > 02:47 which is when the package was built for PyPI. Because my brain goes faster than my fingers and I say the reverse of what I mean. I meant .mo files, all .po where properly dated, and I reversed the meaning of my thoughs while typing. Thanks for brining my confussion to my (and everybody's) attention, and my appologies for the noise. > When things are changed on weblate, we get automatically generated > merge requests to update our source. We don't always merge them > immediately, but we certainly do merge all the outstanding ones > before making a release. I'm 95% certain that I edited messages after Nov 7th in the Spanish translation of both mailman, postorius and hyperkitty, but the .po file is dated Nov 7h. Actually, I did a review of outstanding strings when Abhilash annonced the RC for the current version. That is why I was bit surprised. But then, it may be me and my faint memory. And, of course, I'm very grateful and happy with all the work that is being done on Mailman3. -- Victoriano Giralt Innovation Director Digital Transformation Vicerectorate University of Malaga +34952131415 SPAIN ================================================================== Note: signature.asc is the electronic signature of present message A: Yes. > Q: Are you sure ? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting annoying in email ?

5 years, 4 months

Re: How to remove Held Unsubscriptions

by Jered Floyd

What version are you running? It looks like this was added in Postorius 1.3.5. To confirm a request for unsubscription before that version requires using the shell or REST API, as described in https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/thread/… Copying for convenience: # mailman shell -l listname(a)domain.com Welcome to the GNU Mailman shell Use commit() to commit changes. Use abort() to discard changes since the last commit. Exit with ctrl+D does an implicit commit() but exit() does not. The variable 'm' is the list(a)example.com mailing list >>> pendingsdb = getUtility(IPendings) >>> for token, data in pendingsdb.find(m, pend_type='unsubscription'): ... if data['token_owner'] == 'moderator': ... pendingsdb.confirm(token) ... [data about the request will be displayed here] >>> commit() --Jered ----- On Jan 5, 2024, at 12:47 PM, Mark imark(a)posteo.net wrote: > No joy. There is no menu entry for that page ( > ".../unsubscription_requests") and when entering the URL manually I get: > > Page not found > This page either doesn't exist, or it moved somewhere else. > > The menu item and page for ".../subscription_requests" display okay: > > Subscriptions pending approval (0) > There are currently no subscription requests for this list. > > > > > > On 2024-01-06 04:35, Jered Floyd wrote: >> I see these under "Subscription requests > Unsubscription requests", >> or at the URI (matching your sample) >> >> https://lists.domain.com/postorius/lists/listname.domain.com/unsubscription… >> >> Does this menu entry appear for you? >> >> --Jered >> >> ----- On Jan 5, 2024, at 12:30 PM, Mark imark(a)posteo.net wrote: >> >>> Hi >>> >>> As a moderator I receive an email each day for "Held Unsubscriptions" >>> for a subscriber that was unsubscribed by the Owner sometime ago >>> (using >>> "Mass operations"). >>> >>> +++++++++++++++++++++++++++++++ >>> The listname(a)domain.com list has 1 moderation requests waiting. >>> >>> Held Unsubscriptions: >>> User: bart(a)example.com >>> >>> Please attend to this at your earliest convenience. >>> +++++++++++++++++++++++++++++++ >>> >>> There is nothing showing under "Subscription requests > Pending >>> approval". >>> >>> Could anyone tell me how I can stop these email notifications? I have >>> access to the mailman shell and the databases. >>> >>> Thanks, >>> Mark >>> >>> >>> _______________________________________________ >>> Mailman-users mailing list -- mailman-users(a)mailman3.org >>> To unsubscribe send an email to mailman-users-leave(a)mailman3.org >>> https://lists.mailman3.org/mailman3/lists/mailman-users.mailman3.org/ >>> Archived at: >>> https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message… >>> >>> This message sent to jered(a)convivian.com >> _______________________________________________ >> Mailman-users mailing list -- mailman-users(a)mailman3.org >> To unsubscribe send an email to mailman-users-leave(a)mailman3.org >> https://lists.mailman3.org/mailman3/lists/mailman-users.mailman3.org/ >> Archived at: >> https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message… >> > > This message sent to imark(a)posteo.net

2 years, 5 months

Urgent help needed...

by Darren Smith

Hello, This morning our Mailman 3/Hyperkitty/Postorius installation has become unresponsive. When looking at the machine, postgres is pegging all 4 CPUs on the machine. We did some digging in postgres, and here is the query that is causing problems: "SELECT hyperkitty_email.date FROM hyperkitty_email WHERE hyperkitty_email.mailinglist_id = 'fooey(a)rootsweb.com' ORDER BY hyperkitty_email.date ASC LIMIT 1;" We have found that it doesn't matter what mailinglist_id is being used. We have also found that this might be an issue with the LIMIT 1 - our investigation is showing that using LIMIT 1 in postgres makes things execute very differently than if you use a limit 2 or more, and that this is probably what is going on. Basically, every time this gets called, it pegs an entire CPU for 20-30 seconds. It appears to be called a LOT. So our users are getting 504 gateway timeouts right now. What has changed: We have been in the process of importing our 40 million archived emails from our old mailman 2 archives. Questions: 1) We have tried doing a grep in the codebase - mailman3, hyperkitty, postorius, django - basically anything that might be calling this code, but we can't find it. Does anyone know what generates this query? 2) We are positing that it might be our archive imports that are causing this problem. Does anyone know of any sort of limit/threshold that might cause this to go haywire? Any help you can give, any direction you can point us would be useful. Right now our lists are just down for our users. Thank you, Darren

8 years

Mailman-users search results for query "Postorius"