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

[MM3-users] Re: Hyperkitty CPU usage

by Alain Kohli

I have run "python manage.py rebuild_index" before, doesn't that do clear_index as well? Apart from that, I run hyperkitty in a docker container and didn't know fulltext_index should be persistent, so that got deleted after every version update for sure. On 4/26/19 10:18 PM, Mark Sapiro wrote: > On 4/26/19 11:14 AM, Alain Kohli wrote: >> I see loads of "A string literal cannot contain NUL (0x00) characters." >> messages, but I haven't found missing messages in the archives yet. Not >> sure how that could be related, though. Apart from that I don't see >> anything unusual. The other jobs (quarter_hourly, hourly, etc.) seem to >> run and finish normally. > > Did you upgrade from a Python 2.7 version of HyperKitty to a Python 3 > version? The Haystack/Whoosh search engine databases are not compatible > between the two and "A string literal cannot contain NUL (0x00) > characters." is the symptom. > > You need to run 'python manage.py clear_index' or just remove all the > files from the directory defined as 'PATH' under HAYSTACK_CONNECTIONS in > your settings file (normally 'fulltext_index' in the same directory that > contains your settings.py. >

7 years, 5 months

[MM3-users] Re: Hipperkitty failing to archive messages in a list

by Guillermo Hernandez (Oldno7)

On 6/1/22 19:19, Mark Sapiro wrote: > On 1/6/22 4:25 AM, Guillermo Hernandez (Oldno7) via Mailman-users wrote: >> >> In the last part of your suggestions you wrote: >> >> "I suggest you do >> >> pip install --upgrade hyperkitty mailman-hyperkitty >> >> and ensure it installs hyperkitty 1.3.5 and mailman-hyperkitty 1.2.0" >> >> But this is exactly what made me fall in this mess: > > > Sorry, I was confused. I thought the error message you were seeing was > changed before HyperKitty 1.3.5 was released, but that isn't correct. > > The problem is probably in Apache. See the last 2 posts in the thread > at > https://lists.mailman3.org/archives/list/mailman-users@mailman3.org/thread/… > I'll try tomorrow, now I'm a bit slow-head after a hard day (and the collateral effects of the third vaccine). It looks the right way (I have apache running in that server) to solve it. What intrigues me is the fact that I didn't change anything in the apache config or update apache itself. Well, tomorrow will be antoher day. Thanks for the point.

4 years, 9 months

[MM3-users] Re: Held for notification - runner exception: 'ascii' codec can't encode character

by Seth Seeger

> On Jan 2, 2022, at 3:47 PM, Mark Sapiro <mark(a)msapiro.net> wrote: > > Thanks for sending me the shunted .pck. It turns out I didn't need it. Had I looked more closely at the Traceback, I would have realized that the issue is that the list's list:admin:action:post template contains non-ascii. In particular, the character in position 5 is ’ which is a unicode right_single_quote, not an ascii apostrophe. > > There may be other unicode fancy quotes ad well. > > You can fix the template to contain only ascii or you can set the character set for English to utf-8 by adding > ``` > [language.en] > charset: utf-8 > ``` > to mailman.cfg. The downside of the latter is that outgoing message bodies will always be base64 encoded which generally precludes grepping mailboxes to find things. Ah hah! Right you are. I removed the offending fancy apostrophe. What about that traceback pointed to the template? (Is it something I could have figured out?) Is it possible to filter out UTF-8 chars from the templates when they are saved in Postorius? Thanks again, Seth

4 years, 9 months

[MM3-users] Re: Python backtrace from `mailman -h`

by Peter Chubb

>>>>> "Mark" == Mark Sapiro <mark(a)msapiro.net> writes: Mark> On 8/24/25 17:03, Peter Chubb via Mailman-users wrote: >> Hi Folks, I'm running the Debian-packaged mailman3 (yes I know ...) >> and running: sudo -u list mailman -h gives a python backtrace >> (which I've appended the tail end of). Can someone who has the >> upstream installed version (using venv) please check that `mailman >> -h` works there before I waste too much time investigating in the >> wrong places? Mark> Yes, `mailman -h` works and produces the same help output and Mark> list of subcommands as `mailman` with no arguments or Mark> options. Does it work with no options? Does it work if you do Mark> `mailman -C path/to/mailman.cfg -h`? Thanks. Without options it works; and it works with an explicit config path. So I know where to look to fix it (and will report it on the Debian bug tracker) Interestingly: sudo -u list mailman -C /etc/mailman3/mailman.cfg -h works, but sudo -u list env MAILMAN_CONFIG_FILE=/etc/mailman3/mailman.cfg mailman -h does not. And sudo -u list mailman info reports /etc/mailman3/mailman.cfg as the current configuration file. Peter C

1 year, 1 month

[MM3-users] Re: Avoiding non-members to be saved.

by Guillermo Hernandez (Oldno7)

On 7/8/21 20:03, Mark Sapiro wrote: > This is the way MM 3 works. The only reason *_these_nonmembers exists > in MM 3 is to support the legacy regexps from MM 2.1. For non-regexp > addresses the address is a nonmember with the appropriate moderation > action. > > I actually find this convenient. When a nonmember posts spam to a list > and that spam is held, I discard the spam in Postorius and at the same > time set the nonmember's moderation action to discard so I never see > any more from that address. Also, when someone who I recognize as a > list member posts from an alternate nonmember address, I can accept > that post and set the nonmember address moderation action to default > processing. > I find it very convenient too, but it concerns me that spammers, really "profesional" spammers and senders of phising mail change the forged address and sender domain too rapidly to track them. As of it, the non-member database is growing more and more. It could be a "monster" in two or three years... Not a problem now. And I know now how to delete them in one go :) --

5 years, 2 months

[MM3-users] Re: Fresh installation mailman3-suite via "pipx" - any experiences yet?

by Stephen J. Turnbull

Mark Sapiro writes: > But postorius and hyperkitty are dependencies of mailmanweb so only Are they? I'm pretty sure that in my current project which has a pre-existing archiver they wish to keep using I only installed postorius, not hyperkitty. ... oops, HyperKitty *is* there. OK, need to uninstall that. That's actually a bit annoying, since mailmanweb is a friendlier interface for control of either of HyperKitty and Postorius than the underlying Django utilities even if you don't want the other and its unique dependencies (such as Whoosh). (Issue #32 created) OK, so your two-venv approach should actually work. I still don't think it's much of a win since to finish setup you still have to do a lot of work as root and a lot of work outside of pipx's control. Basically it saves one "python3 -m venv", at the cost of "pipx run" to run each program in its own venv. Note, it doesn't save me any venv activationss because I put that in the mailman user's shell .rc. Wikinaut asks: > Has anyone tried that already? Why not you? Steve

2 years, 8 months

[MM3-users] Re: Hiding an ML from a site

by Odhiambo Washington

On Wed, Feb 7, 2024 at 9:45 PM Mark Sapiro <mark(a)msapiro.net> wrote: > On February 6, 2024 7:45:14 AM PST, Odhiambo Washington > <odhiambo(a)gmail.com> wrote: > > > >I had this set to True in settings_local.py, but False in settings.py. > >So it seems to not make a difference in my setup. > > In a standard setup, anything set in settings_local.py overrides any > corresponding setting in settings.py. > > That said, FILTER_VHOSTS is not perfect. It only works for Postorius if > the separate web hosts also have different mail hosts. See lines 1051 et > seq at > < > https://gitlab.com/mailman/postorius/-/blob/master/src/postorius/views/list… > > > What should I do then? Edit some file? Apply some patch? It's pretty easy for me to create a DNS record for the same server (IP) for use with the web host that I'd like to hide. -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 In an Internet failure case, the #1 suspect is a constant: DNS. "Oh, the cruft.", egrep -v '^$|^.*#' ¯\_(ツ)_/¯ :-) [How to ask smart questions: http://www.catb.org/~esr/faqs/smart-questions.html]

2 years, 8 months

[MM3-users] Re: Handling of SPAM mails in mailman3 - Check for X-Spam Header

by Torge Riedel

Am 16.01.20 um 21:43 schrieb Mark Sapiro: > On 1/16/20 10:51 AM, Torge Riedel wrote: >> Such SPAM mails are currently hold for moderation with reason "not from >> a list member". Is there no check for the "X-Spam: Yes" header or is >> this check done later? > > If you set a Header Filter (in Postorius the Header filters tab) those > checks are done before non-member moderation. > > This won't work for posts from members because the header checks are > done between member moderation checks and non-member moderation checks, > but if you create a Header Filter for header 'x-spam' and pattern 'yes' > with action hold, a message from a non-member which has an X-Spam: Yes > header should be held for that reason. > > Hi Mark, thank you for your answer. Just to understand that right: In case a posting with an "X-Spam: Yes" header comes from a member, it will not be held? It will be directly forwarded to the list? If yes, is there any chance to change or extend this? Best regards Torge

6 years, 8 months

[MM3-users] Re: postfix malformed address with confirm replies

by Stephen J. Turnbull

Mark Sapiro writes: > On 2/27/21 4:36 AM, Stephen J. Turnbull wrote: > > is legal, and I think that most servers will accept it, but RFC > > 5321 sets a limit of 64 octets on mailboxes[1], and your > > confirmation mailbox is 67 octets. > > Since the -confirm+<token> part is fixed at 49 octets, this would imply > a limit in list names of 15 octets. Ouch! I'm pretty sure we would have run into this before, if there were a lot of servers out there imposing this limit. 15 octets is pretty darn short, there are probably quite a few list names out there longer than that. Not to mention that the RFC itself says "you SHOULD NOT be a jerk about this". :-) So I don't think this is a big problem in practice, and I hope Steve can reconfigure the obstructive server to not be so picky. For Mailman, I doubt we need a full SHA1 (?) for the purpose. We can truncate the thing, probably opportunistically (ie, so the whole confirmation mailbox fits in 64 octets, or we could even parse the bounce message and resend only if we run into a problem). Steve

5 years, 7 months

[MM3-users] Re: External MTA incoming mail: configuration

by Odhiambo Washington

On Fri, Aug 2, 2024 at 6:44 PM Mark Sapiro <mark(a)msapiro.net> wrote: > On 8/2/24 08:12, Roland Giesler via Mailman-users wrote: > > > > It seems though that since I'm not running a local MTA, when then mail > > arrives at the mailman3 server, there's nothing that can process it. So > > should I run postfix in it's most basic setup to allow mail to be > > received or what is the best way to deal with it? Mailman3 has to > > receive mail like signup confirmations, etc after all, not? > > If you use transport_maps on the remote MTA to relay the list mail to > port 8024 on the Mailman server, Mailman's LMTP runner will receive the > mail and you don't need a local MTA. > Nice one! I'd love to know how this config is done in Postfix. Heading to ask Uncle G. -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 In an Internet failure case, the #1 suspect is a constant: DNS. "Oh, the cruft.", egrep -v '^$|^.*#' ¯\_(ツ)_/¯ :-) [How to ask smart questions: http://www.catb.org/~esr/faqs/smart-questions.html]

2 years, 2 months

Search results for query "sapiro"