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

[MM3-users] Re: Hyperkitty performance problem

by Mark Sapiro

On 12/31/24 06:07, monochromec via Mailman-users wrote: > Further RCA reveals the fact that the archiver runner tries to archive the same message over and over again - hence the overhead during pipeline processing. The total # of attempts reaches well into five digits for a single message. This is not directly related to the pipeline runner issues but they may have the same underlying cause. All the pipeline runner does is put the message in the archive queue. It's the archive runner that is doing all these retries. However, the pipeline runner in processing the rfc_2369.py handler is also communicating with hyperkitty to get the values for the List-Archive: and Archived-At: headers. You need to focus on the communication between mailman_hyperkitty and HyperKitty. If there are significant delays there, it could explain your pipeline runner issue. Btw, I see https://gitlab.com/mailman/mailman/-/issues/1189 is also from you. I wondered at the time why you were trying to increase the number of pipeline runners which is not something that would normally be required. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

1 year, 7 months

[MM3-users] Re: Linking to a forum thread from a (news) website.

by Mark Sapiro

On 9/25/19 2:29 AM, Paul Arenson wrote: > SOLUTION > I figured out that all replies are listed under Re:, so if the first comment does not have the Re:, then it gets cut off. ... > QUESTION > Am I correct that if I want to create a thread and link to it, but make sure the first note is included, I need to make the thread title: Re: Something? Re: has nothing to do with it. Replies are not threaded by subject. Replies are threaded by In-Reply-To: and References: headers in the reply. These headers are generated by the sending MUA (mail client) when replying to a message. If you send a 'new' message without those headers, it will be a new thread regardless of whether or not the Subject: matches with or without the Re:. Also, there are broken MUAs that don't properly generate In-Reply-To: and References: headers and replies from these MUAs will never be properly threaded. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

6 years, 10 months

[MM3-users] Re: Cannot start lmtp runner due to error while attempting to bind on address ('127.0.0.1', 8024): address already in use

by Mark Sapiro

On 12/2/19 5:04 AM, actionmystique(a)gmail.com wrote: > > Of course the port is already in use: it is used by dovecot. > ``` > tcp6 0 0 ::1:8024 :::* LISTEN 0 1317360 129620/dovecot > tcp 0 0 127.0.0.1:8024 0.0.0.0:* LISTEN 0 1316324 129620/dovecot > ``` > I must be missing something. What is Dovecot listening for on the lmtp port? If this is Postfix communicating with Dovecot as the LDA or for address verification and using TCP port 8024, you have at least two choices. You can have Postfix/Dovecot communicate via a unix socket instead or you can configure Mailman to use a different port. See <https://mailman.readthedocs.io/en/latest/src/mailman/docs/mta.html> which says in part: lmtp_port is fairly arbitrary (there is no standard port for LMTP). Use any port convenient for your site. “8024” is as good as any, unless another service is using it. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

6 years, 8 months

[MM3-users] Re: Aborted connection message in log in every 5 minutes

by Mark Sapiro

On 1/30/24 10:46, andrea.janna.unipd(a)gmail.com wrote: > Same issue at my site. > I'm using mailman3 3.3.8 shipped with Debian 12. Then your first line of support should be Debian. See <https://wiki.list.org/x/12812344> > To debug the problem I stopped apache, cron, postfix and mailman3-web services. > But I still get the mariadb aborted connection warnings every 5 minutes. The only thing that runs every 5 minutes is `mailman gatenews`. The issue may be with that. You could try running it manually via ``` _MAILMAN_GATENEWS_NNTP=yes;sudo -u list mailman gatenews ``` ... > The regular mailman log doesn't show anything and I'm not sure about how to enable an appropriate debug log for mailman. Mailman has a number of virtual logs, most of which are written to the file mailman.log. You can enable debug level for those with ``` [logging.template] level:debug ``` See <https://gitlab.com/mailman/mailman/-/blob/master/src/mailman/config/schema.…> for more info on logging. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

2 years, 6 months

[MM3-users] Re: Trouble Installing Mailman3

by Mark Sapiro

On 6/3/25 08:22, Henry Hartley via Mailman-users wrote: > > (venv) mailman@wz-prd-tpmd-1:~$ pip -v --trusted-host pypi.org install wheel mailman psycopg2-binary > Using pip 24.0 from /opt/mailman/venv/lib/python3.12/site-packages/pip (python 3.12) > ERROR: Could not find a version that satisfies the requirement wheel (from versions: none) > ERROR: No matching distribution found for wheel > > The same error shows up if I try to install just mailman or psycopg2-binary, except the word "wheel" is replace with whichever package I've requested. I don't understand why this is occurring. All those packages exist in pypi.org https://pypi.org/project/wheel/ https://pypi.org/project/mailman/ https://pypi.org/project/psycopg2-binary/ Also, pypi.org has a valid ssl certificate signed by GlobalSign nv-sa If I create and activate a new virtualenv named zzz, and do ``` (zzz) mark@msapiro:~$ pip -v install wheel mailman psycopg2-binary ``` I get a successful installation. I am guessing that there is something in your local network that's redirecting connects to pypi.org to somewhere else. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

1 year, 2 months

[MM3-users] Re: Can't subscribe on new site/domain

by Mark Sapiro

On 10/9/25 5:02 AM, Flo via Mailman-users wrote: > Hi, > > I added a new site in Django: lists.example.com > > Usually I click on 'view site' and I can add a Domain in Postorios. In > this case I get an error message and I need to add a Domain in Django. > Then I can go to 'view site'. > > I can create a mailing list there. However when I send an email to be > subscribed, I get the following in mailman.log (and nothing else happens): > > > Oct 09 13:46:24 2025 (110452) deque: do_confirm_verify > Traceback (most recent call last): ... > File "/usr/lib/python3/dist-packages/mailman/model/domain.py", line > 177, in __getitem__ > raise KeyError(mail_host) > KeyError: 'lists.example.com' It appears you don't have a domain in Mailman core with mail_host = 'lists.example.com'. Do you see one in Postorius' domains view? If not, perhaps you need to add it. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

10 months, 1 week

[MM3-users] Re: hyperkitty not archiving new mail

by Mark Sapiro

On 8/6/26 11:51, Jaldhar H. Vyas via Mailman-users wrote: > Hello all, > > Mailman 3.3.10 on Debian 13. > > I recently upgraded from a well-running installation of Mailman2. Importing the existing mailing lists went well but I'm having problems with hyperkitty. First it was complaining that I didn't have a valid SSL certificate for localhost. I rectified that but in the mean time I ended up with some .pck files in /var/lib/mailman3/archives/hyperkitty/spool . Inspecting these with mailman-wrapper qfile reveals that they are messages which have been sent since the import. How do I reindex just those messages? I tried update_index and rebuild_index but apart from taking forever, they don't seem to help. They will be retried the next time a post is sent to HyperKitty to be archived. Assuming you fixed the underlying issue that caused them to fail, they will be archived at that time. If archiving continues to fail, there should be error messages logged in mailman.log with more info. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

2 weeks, 1 day

[MM3-users] Re: sqlalchemy.orm.exc.MultipleResultsFound: Multiple rows were found for one_or_none()

by Mark Sapiro

On 7/24/20 10:34 AM, Shashikanth Komandoor wrote: > Hi Team, > > Today, I found an error message like "Multiple rows were found > for one_or_none()" when a hold message is approved for a list. Until > yesterday the mails were accepted and delivered fine. We have seen this issue before. See <https://gitlab.com/mailman/mailman/-/issues/657>. There are also multiple threads on it in the archive, e.g., <https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/thread/…>. We don't know how it happens, but the issue is that the `listarchiver` table in the database has multiple `mhonarc` entries for the same `mailing_list_id`. You can see the problem with the query select * from listarchiver where name = 'mhonarc'; There will be two entries with the same `mailing_list_id` and deleting one of them will fix this and you can run `mailman unshunt` to reprocess the shunted messages. The above referenced issue contains a script that can be run to remove the extra entry. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

6 years

[MM3-users] Re: Hyperkitty Recommended mySQL Database Setup & Archive Importing

by Mark Sapiro

On 12/9/20 6:39 PM, matthew(a)alberti.us wrote: > > For core, the connection string seems to accept a tailing > "?charset=utf8&use_unicode=1" per > https://mailman.readthedocs.io/en/stable/src/mailman/docs/database.html So > for core, maybe just updating the connection string to include > "?charset=utfmb4&use_unicode=1" will just work? The link lists utf8 as the > recommend encoding... maybe that should be updated? I'm in the process of doing that update. > Seems the takeaway is that, for mysql databases only, the use of utf8mb4 is > recommended, and it needs to be explicitly identified in the database > settings for BOTH core and web, and probably somehow at database creation > time? It's not as critical for core because barring things like 4-byte utf-8 unicode graphics in list configurations, core doesn't store anything like that in its tables. The major issue is with HyperKitty where messages have emojis and other 4-byte utf-8 encodings in message subjects and bodies. -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

5 years, 8 months

[MM3-users] Re: Is there a way of recording bounce messages?

by Mark Sapiro

On 9/21/23 17:30, Thomas Ward wrote: > > MS365's bounce handling is weird and they do not include a delivery report in the payload, instead they take the log data and append it as a text payload portion to the message. It's annoying and problematic when debugging (and likely triggers the mailman2 equivalent of "unprocessable bounce message"). > > If Andrew can provide the bounce message as an example in raw form (even off list) we can dissect it - myself probably more rapidly as I am fluent in reading them at this point from weeks worth of dealing with them - and that will probably yield the info you are looking for. It appears however that Mailman does recognize the DSN returned by MS365 as the user's delivery is being disabled by bounce. The issue is whatever DSN is returned to the list-bounces address is not attached to the notice sent to the list owner about the disable and it should be. -- 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

Search results for query "sapiro"