Search results for "2021" - Mailman-users

Re: Internal server errors after upgrade

by Abhilash Raj

> On Oct 26, 2021, at 10:33 AM, Seth Seeger <seth(a)tofutavern.com> wrote: > > Hello community, > > I'm running the Docker version of MM3. I've just run: > > docker-compose stop > docker-compose pull > docker-compose up -d > > I am now at versions: > > mailman-core: 3.3.4 > mailman-web: 2.2.20 You probably want to look at instructions on updating to newer versions: https://asynchronous.in/docker-mailman/news/#upgrading-to-040-release Also, not sure how you got mailman-web 2.2.20, the latest version is 0.0.5[1] and docker image doesn’t use the mailman-web package at all. [1]: https://pypi.org/project/mailman-web/ > > I am running a 2-core, 4GB RAM VM on Linode. > > The web interface mostly works. However, sometimes/regularly I get a 500 Internal Server Error. In /opt/mailman/core/var/logs/mailman.log I see the errors, but they are usually different. The common thread seems they are always about some sort of database call. Reloading the page is usually successful. > > Did anything change with the latest upgrade in terms of database timeouts? Before this upgrade, I was never getting it. (htop does not show anything unusual.) Can you share the actual errors you are seeing? If there are a lot of them, perhaps two three of them? -- thanks, Abhilash Raj (maxking)

3 years, 8 months

Re: Mailman backend maintenance task

by Eugenio Jordan (external)

Dear Raj. Thanks a lot for your tips. I would like you to please confirm whether you mean I can stop Hiperkitty by just shutting down the Web components instance (actually, a Docker container). Does this mean, therefore, Hiperkitty is to be considered as part of the Web Components? Just for my information, because I'm afraid that both Core and Hiperkitty share the very same database: [database] class: mailman.database.postgresql.PostgreSQLDatabase url: postgres://mailman:mailmanpass@database/mailmandb This is the very same database used by Hyperkitty, so I am afraid that I will have to end up stopping also the Core components, or, anyway, if they cannot access the database "normally" for whatever reason, they will not be delivering the mails to the different distribution lists members and we will have to rely on the external mail servers retries. Could you please confirm this assumption is right? Once more, I really appreciate you kindly took your time to answer and explain about Mailman architecture. Best regards. -----"Abhilash Raj" <[1]maxking(a)asynchronous.in> wrote: ----- To: [2]eugenio.jordan(a)esa.int From: "Abhilash Raj" <[3]maxking(a)asynchronous.in> Date: 08/23/2021 07:31PM Cc: "Mailman users" <[4]mailman-users(a)mailman3.org> Subject: Re: [MM3-users] Mailman backend maintenance task On Sun, Aug 22, 2021, at 10:54 AM, [5]eugenio.jordan(a)esa.int wrote: > Dear all. > > My apologies if the answer to this question is obvious, but I am pretty > newbie for 3 version, and to be very honest I haven't had enough time > yet to have a deep look into the documentation. > > Our customer is currently using PostGRESQL as backend, and we would > like to perform some maintenance tasks, namely running vacuum full, or > at least trying to rebuild hyperkitty_email primary key related index. > We have been asked on the real impact of putting in place such > initiative. Though the latter is related to archiving, I haven't found > a way to stop just Hyperkitty or Django related processes other than > stopping Mailman's core, hence preventing mails addressed to > distribution lists from being delivered, could you please confirm if I > am correct? You can indeed stop Hyperkitty separate from stopping Core. Depending on how your client installed Mailman, there are two separate scripts/processes/commands to start and stop Mailman Core and Web components. You should be able to safely stop Hyperkitty, run maintenance and re-start it. Core will continue to function during this duration, sending out emails and such. The archiving will obviously not happen, but it will be queued and Core we retry to archive that, which should succeed whenever HK comes back up. > > Regarding the former, as far as I have read, the "mappings" lists -> > addresses are stored just in the database, so if we run some kind of > procedure or task like vacuum which will lock exclusively tables, or > want anyway to have the database stopped for a cold backup or whatever, > Mailman willl not work, that is, again the mails addressed to the > distribution lists will not be delivered. Will you please confirm this > point, too? > Both Core and Web components have separate databases. Although, technically speaking, the database name is passed in through configs and you can pass in the same database name to both and since they use separate table names, there is a good chance it would work. Considering that your client has configured separate databases for both, you can run maintenance for the Web database without affecting any operation of Core (other than archiving). Core stores the mappings and data related to MailingLists. While Hyperkitty also has a partial copy of that data in its database that it needs to maintain the archives, that data can be synchronized with a command and is also done routinely using a cron job. A database backup would be a good idea either way. Hope this helps! -- thanks, Abhilash Raj (maxking) This message is intended only for the recipient(s) named above. It may contain p roprietary information and/or protected content. Any unauthorised disclosure, use, retention or dissemination is prohibited. If you have received this e-mail in error, please notify the sender immediately. ESA applies appropri ate organisational measures to protect personal data, in case of data privacy queries, please contact the ESA Data Prot ection Officer (dpo(a)esa.int) References 1. mailto:maxking@asynchronous.in 2. mailto:eugenio.jordan@esa.int 3. mailto:maxking@asynchronous.in 4. mailto:mailman-users@mailman3.org 5. mailto:eugenio.jordan@esa.int

3 years, 10 months

Re: mailman web spamming uwsgi error log.

by Sarah H

from the Django settings the log is here: Make sure that this directory is created or Django will fail on start. LOGGING['handlers']['file']['filename'] = '/opt/mailman/web/logs/mailmanweb.log' but the only new entry in the log is from when I attempted to run the runjobs as the wrong user and the log entries for unable to connect to DB from when the uwsgi log filled the drive and the db process died, which is what originally alerted me to the issue. ERROR 2021-06-28 20:29:46,227 1974245 hyperkitty.lib.utils Failed to update the fulltext index: [Errno 13] Permission denied: 'fulltext_index' Traceback (most recent call last): File "/opt/mailman/venv/lib64/python3.6/site-packages/hyperkitty/lib/utils.py", line 169, in run_with_lock fn(*args, **kwargs) File "/opt/mailman/venv/lib64/python3.6/site-packages/hyperkitty/search_indexes.py", line 118, in update_index update_cmd.update_backend("hyperkitty", "default") File "/opt/mailman/venv/lib64/python3.6/site-packages/haystack/management/commands/update_index.py", line 329, in update_backend last_max_pk=max_pk, File "/opt/mailman/venv/lib64/python3.6/site-packages/haystack/management/commands/update_index.py", line 111, in do_update backend.update(index, current_qs, commit=commit) File "/opt/mailman/venv/lib64/python3.6/site-packages/haystack/backends/whoosh_backend.py", line 262, in update self.setup() File "/opt/mailman/venv/lib64/python3.6/site-packages/haystack/backends/whoosh_backend.py", line 144, in setup os.makedirs(self.path) File "/usr/lib64/python3.6/os.py", line 220, in makedirs mkdir(name, mode) PermissionError: [Errno 13] Permission denied: 'fulltext_index'

4 years

[SOLVED] Re: Re: Help with Newbie error -- I think I deleted my server

by William Oliver

On Fri, 2021-12-24 at 21:12 -0800, Mark Sapiro wrote: > On 12/24/21 8:22 PM, William Oliver wrote: > > > > I clicked on that, and it took me to a page where I could add > > servers. > > So, I added my real domain (libertyfp.org) It successfully added > > it, > > so I decided that I clearly didn't need "example.com" as a server, > > and > > deleted it using the same page. > > > > Immediately, I got an "internal server error" and have now lost all > > web > > interfaces. I tried restarting the daemons and rebooting the box, > > to > > no avail. > > > You have to set SITE_ID in your Django settings to the correct value > for > your site. See > https://docs.mailman3.org/en/latest/faq.html#the-domain-name-displayed-in-h… > > Joy! You are a lifesaver. I had to guess at the ID number, but "2" worked. Just out of curiosity, if I could not have guessed a valid ID number and did not have the admin interface, how could I have figured out what site IDs were available? Thanks. billo

3 years, 6 months

Re: Unable to get mailman3 working on Ubuntu 20.04 LTS

by Shawn Heisey

On 10/27/2021 9:51 PM, John Covici wrote: > I am interested in your docr instructions. I don't use postfix, > instead I use sendmail, but I think I can figure that out, but I have > been having lots of problems installing without docr, and I do have > apache, so this would be great. Send off list if you think that would > not be of general use. The instructions found here are most of what is needed: https://github.com/maxking/docker-mailman/#running Here's the issue where I provided my Apache config: https://github.com/maxking/docker-mailman/issues/525 For each of my mailing lists, I created the list and then imported a mailman 2.1 config. I had problems with the footer being appended to list messages ... it took me a while to figure out how to fix that! You have to delete the footer config file, restart the container, and then if the resulting default footer is insufficent, use the admin web page to redefine it. Instructions for exim and postfix are included on the project page. I haven't used sendmail in many years, so I wouldn't have any idea how to get that working. Even when I was using it, I found the config to be extremely arcane and difficult. Thanks, Shawn

3 years, 8 months

Re: Rejection Reason for refusing a subscription request

by Abhilash Raj

> On Mar 27, 2021, at 3:03 PM, Mark Sapiro <mark(a)msapiro.net> wrote: > > On 3/27/21 2:08 PM, Brian Carpenter wrote: >> I was testing out the process of rejecting an incoming subscription >> request. Choosing the reject option for a moderated subscription request >> does send out a rejection notice but it gives no reason for the >> rejection. > ... >> Is there a place somewhere in Postorius to provide a rejection reasons >> for moderated subscription requests? Does Mailman 3 core provide this >> via its REST api but it is just not revealed in Postorius yet? > > > It's in Postorius since 1.3.3. If in the held message view, you click > the subject of a held message, you get a popup dialog the shows the > message with options to also show headers or the raw message. There is > also a dropdown t=o set the poster's moderation action and a box for a > reject reason if you click the reject button in the dialog. > > Do you not see this? I guess Brian was asking about Subscription Requests. Support for specifying reason when rejecting subscription requests has been added in master and will be a part of next release, 1.3.5 most probably. -- thanks, Abhilash Raj (maxking)

4 years, 3 months

Re: problem with postorius initial setup

by Abhilash Raj

Hi John, > On Mar 16, 2021, at 6:58 AM, John Covici <covici(a)ccs.covici.com> wrote: > > Hi. I am trying to setup postorius, but I have run into a problem. I > am using the example project, except I am using postgresql instead of > sqlite3. I ran python manage.py migrate which was successful. Are you trying to setup locally for development or are you setting up a production installation? > > On running the development server I get: > Exception in thread django-main-thread: > Traceback (most recent call last): > File > "/usr/lib/python3.8/site-packages/django/utils/module_loading.py", > line 20, in import_string > return getattr(module, class_name) > AttributeError: module 'postorius.middleware' has no attribute > ‘APICountingMiddleware' APICountingMiddleware is from a development branch of Postorius, so unless you are installing from git, this won’t be present since it hasn’t been released. If you are actually doing development, you want to install from the Git source tree by running: ``` $ pip install -e . ``` You can skip the `-e` if you want to install from git but not develop. From the top level directory of Postorius (where setup.py is, one level above example_project). If you are trying to setup a production installation, you want settings from https://mailman-web.readthedocs.io/en/latest/settings.html#mailman_web.sett… This is for both Postorius and Hyperkitty. thanks, Abhilash

4 years, 3 months

Re: Strange messages after upgrading Mailman3

by Mark Sapiro

On 12/5/22 08:56, Allan Hansen wrote: > Thank you, Mark, as always. > > We upgraded to the latest versions: > > Mailman Core Version GNU Mailman 3.3.7 (Tom Sawyer) > Mailman Core API Version 3.1 > Mailman Core Python Version 3.9.2 (default, Feb 28, 2021, 17:03:44) [GCC 10.2.1 20210110] > > Also, I don't have any umbrella lists. I need more information about this. As far as I know, this can only happen with messages that are cross posted to multiple lists and held for moderator approval on more than one list. This should have been fixed in Mailman 3.3.6 by <https://gitlab.com/mailman/mailman/-/merge_requests/932>. This should never happen with a post that is delivered to members without being held. Also, I think when this happens to a delivered post, the post in the archive will also have the content lost message. I would like to get as much information as possible about the circumstances when this happens. In particular, was the post held for moderation, was it posted to multiple lists, for how many lists was the content lost. Also, it may help to see the full raw message with the content lost notice. (If there are privacy concerns you can send these to me off list) -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

2 years, 7 months

Apple Mail encoding again

by David Newman

Starting a new thread because I don't know how to reply to this one from November 2021: https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/message… A subscriber's post to a list bounced with a 501 bad encoding error. The subscriber uses Apple Mail on MacOS and PGP-signed his message. The earlier post above suggests the problem could be a bug in either Apple Mail encoding or python's smtp module. To rule out the smtp module, I sent a PGP-signed message using a different MUA (Thunderbird, in my case). Mailman3 accepted this post without complaint, and I reported back to list users that this appears to be an Apple Mail encoding issue. Normally, that'd be the end of it. But this is a list for a bunch of Internet graybeards, including one of the authors of the SMTP protocols and the subscriber is a former IETF chair. He's grumbling that if this really were an Apple Mail issue, wouldn't it be far more widespread? I don't think it would be, because (a) Mailman2 is still far more pervasive than Mailman3 and (b) the combination of PGP signing and Apple Mail on MacOS and Mailman3 is still pretty unusual. Has anyone else hit this encoding issue, other than the post above? And did Mailman2 silently accept this because it was less strict about checking for bad encoding? Thanks. dn

3 years, 2 months

Re: Question about message bounce stats

by Abhilash Raj

On Mon, Aug 30, 2021, at 8:46 AM, Stephen J. Turnbull wrote: > Ruth Ivimey-Cook writes: > > > Does Mailman3 keep such stats, and if so is there any way to get > > access to them, even if only API? I would think it useful to do so > > (and yes, for individual outgoing messages at least for a period). > > In released versions, it currently does not, as far as I know. This > kind of thing should be available from your MTA, of course, and I > think there was a visceral opposition to keeping track of people's > behavior that way from the early days. So those are our excuses. > > However, it's something that comes up occasionally on the list as well > as as a GSoC proposal, and I think there's some code on a branch for > this that Abhilash has been threatening to release. And of course > HyperKitty now publishes stats on posting behavior. So there's really > no reason to oppose it, now it's just a matter of time to implement. I don't think we have any changes since last release related to enhancements in bounce processing, there have been some bug fixes through which occasionally can cause bounce runner to cause some issues by keeping open transactions for quite long. -- thanks, Abhilash Raj (maxking)

3 years, 10 months

Mailman-users search results for query "2021"