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

[MM3-users] Re: Core, Hyperkitty and Postorius releases

by Odhiambo Washington

On Sat, May 15, 2021 at 12:47 AM Mark Sapiro <mark(a)msapiro.net> wrote: > On 5/14/21 2:16 PM, Odhiambo Washington wrote: > > > > root@gw:/usr/home/wash # ls -al /usr/local/include/sass > > total 104 > > drwxr-xr-x 2 root wheel 512 Apr 16 16:49 . > > drwxr-xr-x 176 root wheel 10240 Apr 19 18:05 .. > > -rw-r--r-- 1 root wheel 2369 Apr 8 11:08 base.h > > -rw-r--r-- 1 root wheel 10742 Apr 8 11:08 context.h > > -rw-r--r-- 1 root wheel 6632 Apr 8 11:08 functions.h > > -rw-r--r-- 1 root wheel 6489 Apr 8 11:08 values.h > > -rw-r--r-- 1 root wheel 198 Apr 8 11:08 version.h > > > > I imagined I needed an executable. > > > > drwxr-xr-x 2 root wheel 512 Apr 16 16:49 sass > > > Those are headers for compiling some sass. Possibly you actually have a > `sassc` installed in which case you may need > > COMPRESS_PRECOMPILERS = ( > ('text/x-scss', 'sassc -t compressed {infile} {outfile}'), > ('text/x-sass', 'sassc -t compressed {infile} {outfile}'), > ) > > This change makes the error I was getting to go away. However, I cannot login to manage lists, even after restarting mailman. The error details are here: https://pastebin.ubuntu.com/p/vM2c5WYyJ4/ -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 "Oh, the cruft.", egrep -v "^$|^.*#" :-)

5 years, 3 months

[MM3-users] Re: REST API authorization failed

by tlhackque

On 10-Jan-22 13:30, Mark Sapiro wrote: > On 1/10/22 1:58 AM, Axel Rau wrote: >>> >>> Anyway, the above is saying that restadmin:??? doesn't match what's >>> configured in Mailman core and reported by `mailman info`. Please >>> check carefully that these are the same in both mailman.cfg >>> admin_pass and settings MAILMAN_REST_API_PASS. >> >> Yes, again with password disclosed: >> - - - > ... >> (mailman_33) [root@lists5 /usr/local/mailman3]# grep admin >> var/etc/mailman.cfg >> admin_user: restadmin >> admin_pass: restpass > > Does `mailman info` show the same password? A common issue that may (or may not) be in play here is leading/trailing whitespace and/or non-spacing control characters. It may be worth making the output (& input) use quotes & escaping, where the input should accept non-spacing controls, leading/trailing whitespace, and non-traditional interior whitespace (e.g. nbsp, numsp, NoBreak, tab) only in escaped form. By escaped form, I mean one or more formats like HTML's  , or Python's at https://python-reference.readthedocs.io/en/latest/docs/str/escapes.html When dealing with user quantities, if it looks like a space, it should be a space. And if you can't see it, it shouldn't be there (using escapes to make the invisible visible is fine). ___________________________________________ Mailman's content filtering has removed the following MIME parts from this message. Replaced multipart/alternative part with first alternative.

4 years, 8 months

[MM3-users] Re: "Unsubscribe all" button

by Stephen J. Turnbull

Mark Sapiro writes: > On 10/20/22 17:32, Philippe B wrote: > > Would it be possible to add a confirmation screen after you click > > on it? Like: WARNING! Clicking "Confirm" will empty the list of > > all its existing members, regardless of what was entered on the > > previous screen. That operation cannot be reversed. Are you > > absolutely sure that you want no one to be on your list anymore? > > There is a confirmation screen although it is not as strongly worded as > you suggest. I'll go check the issue, but let me say I think a confirmation screen is bad UI here. People who know what they're doing shouldn't have to pay the price of a confirmation screen. I know of at least two universities where they "unsubscribe all" from thousands of class mailing lists every term. Of course they automate that pretty quickly when they get to thousands, but it seems plausible to me that such lists for groups that regularly turn over (nearly) 100% of subscribers are a reasonably common use case. Instead, there should be a check for selected users, either disabling the "unsubscribe all" button itself (if we decide to implement more Javascript in the UI) or presenting the confirmation screen if there are any selected users. Also, at least for native speakers, "Unsubscribe everyone" will probably be a better label for the button.

3 years, 10 months

[MM3-users] Re: SpamAssassin and clamav?

by Jeremy Stanley

On 2024-03-15 18:50:32 -0700 (-0700), Marisa Giancarla via Mailman-users wrote: > On 3/15/24 18:40, Mark Sapiro wrote: > > On 3/15/24 17:52, Marisa Giancarla via Mailman-users wrote: > > > Is there a way to have mm3 use SpamAssassin and/or clamav? > > > Marisa > > > > It would require custom code modification to invoke SpamAssassin > > and/or ClamAV from mailman. It is much better to integrate these > > in the MTA. > > Ok sounds good... A big reason this is the preferred approach is that your MTA, if appropriately configured, can scan incoming messages before deciding whether to accept or reject them into its delivery queue. For messages rejected at receipt, this means legitimate senders whose messages were incorrectly misclassified get useful error messages in the NDR generated by their own mail provider, while spoofed spam and malware doesn't generate backscatter to unwilling victims and put your own server at risk of getting listed as a spam source. If Mailman were in charge of checking for spam or malware, its only options would be to silently drop offending messages (providing no feedback to senders of misclassified messages) or reject them (causing your MTA to generate an NDR which often times ends up either part of a coordinated attack against some victim organization or sitting undeliverable in the queue because that address is at an unreachable domain). These are obviously undesirable, even actively harmful, results. -- Jeremy Stanley

2 years, 5 months

[MM3-users] Re: Google Summer of Docs

by Abhilash Raj

On Sun, Mar 24, 2019, at 11:58 PM, Terri Oda wrote: > > > On March 24, 2019 10:51:07 p.m. PDT, "Stephen J. Turnbull" > <turnbull.stephen.fw(a)u.tsukuba.ac.jp> wrote: > >Ralph Seichter writes: > > > * Mark Sapiro: > > > > >> > If you post an issue at > ><https://gitlab.com/mailman/mailman/issues>, > > > > one of our GSOC aspirants may pick it up. > > > > > > Done, although I wonder if any GSOC participant would want to write > > > documentation instead of code... ;-) > > > >For their actual, Google-supported, project, Google explicitly > >excludes documentation. > > > > They do for Google Summer of **Code** but they've just announced Google > Summer of **Docs** which sounds really neat and we might want to > participate. They are calling it Season of Docs, happening in winter I think Sept-Dec AFAIK. Yep, we should participate in that, I sent out an email earlier about this too. > > (It's my personal opinion that our docs have really suffered since I > finished my PhD and stopped using them as a warmup exercise when I was > procrastinating on writing my thesis. *grin*) > _______________________________________________ > 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, 5 months

[MM3-users] Re: Importing config and archives only partially succeeded?

by Eric Broens

`mailman qfile` on the 5 pck files that were in the /out directory looked OK, although I am not sure if I should have looked at something specific. The bak file disappeared as soon as I stopped mailman. When I started mailman again the name of the bak file had changed, and it changed every time I listed the directory.After stopping and starting mailman once more it didn't appear anymore. There are also 6 files in the shunt directory. I don't know what this directory is used for. For now I have disabled the hourly jobs.I will rebuild the search index for each mailinglist separately since the update_index was apparently killed before it finished. Best Regards,Eric On Sunday, December 13, 2020, 06:18:21 PM GMT+1, Mark Sapiro <mark(a)msapiro.net> wrote: On 12/13/20 7:11 AM, Eric Broens via Mailman-users wrote: > Hi Mark, > 1.Regarding the repeated access of /archives/api/mailman/urls seems to be related to mailman not being able to open port 25. > This is weird because other mails have been distributed. > The mailman log shows:Dec 12 22:16:48 2020 (1471) ACCEPT: < message id > > Dec 12 22:16:52 2020 (1474) Cannot connect to SMTP server localhost on port 25 That message is in response to a socket.error exception in attempted delivery. It is logged only once until there is a successful delivery, but the message keeps being retried. > webserver logs:<host > - - [12/Dec/2020:22:16:50 +0100] "GET /archives/api/mailman/urls ?mlist=...&key=*** HTTP/1.1" 200 64 "-" "python-requests/2.25.0" > <host > - - [12/Dec/2020:22:16:52 +0100] "GET /archives/api/mailman/urls?mlist=...&msgid=< message id >&key=*** HTTP/1.1" 200 105 "-" "python-requests/2.25.0" > This last entry is repeated forever (until I stop the mailserver for a few minutes, but later on this happens again for other mails too). Those GETs are part of normal message processing. As I said, they result from core handlers asking HyperKitty for the URL at which the message will be archived so they can add that URL to the Archived-At: header and maybe to a message header or footer. Presumably the one that keeps repeating is the one message/message-id that is throwing the socket.error and being continuously retried. I would stop Mailman and move that one .pck or .bak file out of Mailman's var/queue/out/ directory and then start mailman. You can then examine the queue entry with `mailman qfile` and maybe see what the issue is. > 2..Regarding the mails on the archive page, for most of the lists it is solved now. So probably one of the periodic jobs fixed that.The mailinglist which showed 0 participants 0 subscribers, does show now 0 participants 66 discussions.I have checked the member tables, and the members for this mailinglist are included there. OK. > 3.What I notice now too is that the hourly runjobs tasks don't seem to finish.I would have to check what they exactly do. Can I somehow activate logging for these? > mailman 300042 300016 1 08:00 ? 00:09:18 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 302158 302149 2 09:00 ? 00:09:10 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 304513 304506 2 10:00 ? 00:09:38 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 308631 308620 3 11:00 ? 00:09:29 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 312847 312832 3 12:00 ? 00:09:15 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 317343 317332 4 13:00 ? 00:09:20 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 320079 320072 7 14:00 ? 00:09:08 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 322618 322610 13 15:00 ? 00:08:54 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settingsmailman 325205 325198 84 16:00 ? 00:06:56 /opt/mailman/venv/bin/python3 /opt/mailman/venv/bin/django-admin runjobs hourly --pythonpath /opt/mailman/mailman-suite/mailman-suite_project --settings settings /opt/mailman/venv/bin/django-admin runjobs --list will show you what the jobs are. The hourly jobs are new_lists_from_mailman, thread_starting_email and update_index. The one running long is almost certainly update_index. I would stop running the hourly jobs until you have successfully updated the search index for all lists. -- 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/

5 years, 8 months

[MM3-users] Re: Error Logging - what changed with the New Year releases?

by Odhiambo Washington

On Sat, Jan 14, 2023 at 7:13 PM Mark Sapiro <mark(a)msapiro.net> wrote: > On 1/14/23 01:19, Odhiambo Washington wrote: > > Apart from all the defaults in settings.py (derived from gitlab), I also > > have this line: > > LOGGING['handlers']['file']['filename'] = > > '/opt/mailman/mm/var/logs/mailmanweb.log' > > > > Before the New Year releases of MM3, I used to get errors written to this > > file. > > Not anymore after I updated to the New Year releases. > > Another thing that I have realized is that prior to the New Year > releases, > > I would get an error logged to that file and also mailed to me. > > I don't think that I have changed anything in settings,py related to > > logging. > > > Nothing in those releases per se would affect this. This is Django and > django-allauth. Perhaps when you updated the Mailman packages, you also > updated these dependencies. > Which dependencies would those be? When I upgraded, the command I ran was: pip install --upgrade --force mailman mailmanclient postorius hyperkitty mailman-hyperkitty django-mailman3 And how do I find out what is suppressing the logging for the social accounts? -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 "Oh, the cruft.", egrep -v '^$|^.*#' ¯\_(ツ)_/¯ :-)

3 years, 7 months

[MM3-users] Re: Confirm and Moderate Unsubscribe Requests

by Brian Carpenter

On 10/21/20 5:39 PM, Mark Sapiro wrote: > On 10/21/20 12:14 PM, Brian Carpenter wrote: >> >> Well I can't dictate that to my list owners and there is no confirm >> token sent with the moderation notice. Why include this option in >> Postorius when it can't be properly handled? > > Postorius is still a work in progress. The unsubscription policies are > in core and are exposed in Postorius. Certainly the ability to moderate > an unsubscription should be in Postorius, but it isn't yet. We are an > all volunteer project and there aren't that many of us, and perhaps I'm > overreacting, but the tone of some of your recent posts seems a bit > demanding. I apologize. I didn't think I was coming across demanding but I am somewhat exasperated at some of these issues. I have a list owner who is now going to get a moderated request notice everyday that he can't do anything about it. But my question still remains: why is this option in Postorius when it can't be handled properly? Surely the option can be easily removed, right? Is the ability to list (via the pending moderation notices tab) and handle (as in the List owner can decide a moderation response) an unsubscribe moderation request there in Mailman core so that I can properly address it via Affinity? -- Brian Carpenter Harmonylists.com Emwd.com

5 years, 10 months

[MM3-users] Re: Emergency moderation and clearing out a lot of held messages

by Abhilash Raj

On 12/14/22 03:31, Mark Sapiro wrote: > On 12/13/22 10:38, Dan Caballero wrote: >> Hi Mark, >> >> I ran the commands and it took about 35 seconds for the loop to run. >> >> Here's the result. >> >>>>> count >> 4168 > > > So that's at least part of the issue. How does 35 seconds compare to the > length of time to process one moderated message through Postorius? > It seems it has to do with in-efficient lookup of the actual entry in the pendings table which needs to be purged after handling a held message. It scales linearly with how many entries are there in pendedkeyvalue and pendings tables. I was actually browsing through the perf metrics on mail.python.org and found the slowest API response from Core is handling held messages and spent some time last night trying to improve it [1]. The linked MR should significantly reduce the time, although there is more that can be done to make it even better. I haven't had a chance to benchmark my solution yet, but I am hoping it will reduce the latency quite significantly. I'll report on the linked issue [2] after deploying [1] on m.p.o. how much it helped. [1]: https://gitlab.com/mailman/mailman/-/merge_requests/1068 [2]: https://gitlab.com/mailman/mailman/-/issues/1045 -- thanks, Abhilash Raj (maxking)

3 years, 8 months

[MM3-users] Re: New domain cannot set set host

by Alain D D Williams

On Thu, Nov 28, 2024 at 01:04:13PM -0800, Mark Sapiro wrote: > > I did not see this y/day and created the entry in table django_site using SQL. > > This might be the cause of the other problems that we have been talking about. > > > It's not likely that this is the cause. It is fixed - the mail list works. I had forgotten to put the reverse proxy into the Apache config -- this is what led to the many, rapid retries. I had a problem for a while in the nothing to one particular list was sent to users. This had several users who's delivery status was 'disabled' (correctly, as on the old instance). If was only after a lot of fiddling (trying to delete from the list) that it started to work for the others. There are still rough edges, eg trying to delete a list member gives a server error. In /var/log/mailman3/mailman.log I see a 404: "DELETE /3.1/lists/chat.bray.phcomp.co.uk/member/a.chentouf(a)hotmail.com HTTP/1.1" 404 26 "-" "GNU Mailman REST client v3.3.5" I am not to worried about that today, will look at it in the future. Thanks for your help -- Alain Williams Linux/GNU Consultant - Mail systems, Web sites, Networking, Programmer, IT Lecturer. +44 (0) 787 668 0256 https://www.phcomp.co.uk/ Parliament Hill Computers. Registration Information: https://www.phcomp.co.uk/Contact.html #include <std_disclaimer.h>

1 year, 9 months

Search results for query "sapiro"