Mailman 3 February 2019 - Mailman-Users

Installing Mailman on a Debian system with Apache 2.4.2, CGI error!
by Jayson Smith Dec. 5, 2019

Dec. 5, 2019

Hi, This is a weird one. I think of myself as knowing my way around Mailman pretty well for a user/admin/installer/upgrader. I'm upgrading to a Debian 9 system from CentOS 6.5. Debian's Apache configs took a bit of getting used to, but I actually like them better. It uses Apache 2.4.2. I wanted to install Mailman from source, since the latest version is 2.1.26, Debian has 2.1.23, and IIRC that's too old to have some screen reader optimizations I want. So I built, installed, all working well. I copied over the Mailman config from my CentOS system to use in Apache, making some changes so it'd work under 2.4.2. Here it is: -------------------- # # httpd configuration settings for use with mailman. # ScriptAlias /mailman/ /usr/lib/mailman/cgi-bin/ <Directory /usr/lib/mailman/cgi-bin/> AllowOverride None Options ExecCGI Require all granted </Directory> Alias /pipermail/ /var/lib/mailman/archives/public/ <Directory /var/lib/mailman/archives/public> Options Indexes MultiViews FollowSymLinks AllowOverride None Require all granted AddDefaultCharset Off </Directory> # Uncomment the following line, to redirect queries to /mailman to the # listinfo page (recommended). # RedirectMatch ^/mailman[/]*$ /mailman/listinfo -------------------- The problem is that the CGI isn't working. If I go to http://temphostname/mailman/listinfo/mailman for example, Apache says /usr/lib/mailman/cgi-bin/listinfo/mailman doesn't exist. If I just go to /mailman/listinfo I get an Elf binary thrown at me, rather than the page saying there are no advertised lists. If I do the same thing from the server using the Lynx web browser, I get the same binary thrown at me, with a page title, "Mailman CGI error!" It sounds to me like Apache 2.4.2 isn't seeing that /usr/lib/mailman/cgi-bin is, in fact, CGI scripts, and is trying to treat them like regular files. Has something else changed between Apache versions? Thanks, Jayson

3 4

Debugging mass invitations using web page interface...
by The Mailing List Server Admin Sept. 5, 2019

Sept. 5, 2019

Dear Mailman Cognoscenti, I'm helping one of my list owners send out 5K plus invitations to students to subscribe to his mailing list. Our current configuration: Mailman v2.1.20 RHEL v5.11 Semdmail v8.13.8 Apache v2.2.3 Since this was the first time doing this, I suggested breaking the batch input into 3 groups, 50, 500, and the rest. The 50 went fine, as did the 500, but the largest batch gave him a generic web server error: > Internal Server Error > > The server encountered an internal error or misconfiguration and was > unable to complete your request. > > Please contact the server administrator, root(a)conundrum.unh.edu and > inform them of the time the error occurred, and anything you might > have done that may have caused the error. > > More information about this error may be available in the server > error log. I looked at the logs and I couldn't find anything that hinted at what went wrong. So I asked the owner to send me the last back and I'd give it a try. I wrote a script that removed folks already subscribed to his list and split the remaining subscribers up into 6 files with a thousand records each. I just tried uploading the 1st batch of 1K, with the following options: Subscribe these users now... (*) Invite Send welcome message... (*) No Send notifications... (*) No And entered a 7 line paragraph explaining the invitation. I ended up having the same error happen. Looking at the Mailman logs, I can't see any difference before or after my submission. In the HTTPD logs, I see: >> [Fri Aug 26 19:59:23 2016] [warn] [client 132.177.215.132] Timeout >> waiting for output from CGI script >> /usr/local/mailman/cgi-bin/admin, referer: >> https://lists.unh.edu/mailman/admin/campus.connection/members/add >> [Fri Aug 26 19:59:23 2016] [error] [client 132.177.215.132] >> Premature end of script headers: admin, referer: >> https://lists.unh.edu/mailman/admin/campus.connection/members/add So is there an inherent limit to the number of invites that can be submitted via the web form? As a work around, how would I do large invites on behalf of the owner from the command line, including the 'extra text' that is allowed via the web interface? -- Cordially, the UNH Mailing List Server Admins Bill Costa, senior admin (603) 862-3056

2 7

Uncaught runner exception
by Lothar Schilling March 8, 2019

March 8, 2019

Hi everybody, a few weeks ago I upgraded from 2.1.16 (as far as I can remember...) to 2.1.29. Everything seemed to work fine at first. But then I found out that a lot of posts - actually far more than half of them - aren't archived any longer. What logging the errors tells me is this: Feb 28 12:29:02 2019 (3123) Uncaught runner exception: 'ascii' codec can't decode byte 0xb5 in position 26: ordinal not in range(128) Feb 28 12:29:02 2019 (3123) Traceback (most recent call last): File "/usr/lib/mailman/Mailman/Queue/Runner.py", line 119, in _oneloop self._onefile(msg, msgdata) File "/usr/lib/mailman/Mailman/Queue/Runner.py", line 190, in _onefile keepqueued = self._dispose(mlist, msg, msgdata) File "/usr/lib/mailman/Mailman/Queue/ArchRunner.py", line 77, in _dispose mlist.ArchiveMail(msg) File "/usr/lib/mailman/Mailman/Archiver/Archiver.py", line 216, in ArchiveMail h.processUnixMailbox(f) File "/usr/lib/mailman/Mailman/Archiver/pipermail.py", line 596, in processUnixMailbox self.add_article(a) File "/usr/lib/mailman/Mailman/Archiver/pipermail.py", line 640, in add_article author = fixAuthor(article.decoded['author']) File "/usr/lib/mailman/Mailman/Archiver/pipermail.py", line 63, in fixAuthor while i>0 and (L[i-1][0] in lowercase or [error message stops right here] As I read in a previous thread the reason for this may be non-ascii compliant characters in the post, especially the "from:"-line. But why would Python or Mailman now all of a sudden use ASCII instead of UTF-8 in the first place? And if so: How can I change that behaviour? I am using Python 2.6.6 on a Centos 6 system. The output of /usr/lib64/python >>> import string >>> string.lowercase is 'abcdefghijklmnopqrstuvwxyz' (as it is supposed to be). Any help would be appreciated. Thank you Lothar Schilling

2 8

Trying to secure the list server web page
by Jeffrey Westgate March 2, 2019

March 2, 2019

Help? We have been running the list server on port 80 for ... years. Yesterday we created an ssl certificate, and move the http to https. Seems to be going fine, except for a few thing -- when a list owner tends to the moderator requests, and hits the Submit All Data button... depending on the browser, it says it is going to send the information over an insecure connection.... and nothing happens results-wise - the request is not handled. Is there some config we need to change in mailman to secure it properly? -- Jeff Westgate DIS UNIX/Linux System Administrator Facebook<https://mail.arkansas.gov/owa/redir.aspx?C=5LO1-Ca8cEq-DdT7YSljUIRyFa2JitUI…>|Twitter<https://mail.arkansas.gov/owa/redir.aspx?C=5LO1-Ca8cEq-DdT7YSljUIRyFa2JitUI…>|Linkedin<https://mail.arkansas.gov/owa/redir.aspx?C=5LO1-Ca8cEq-DdT7YSljUIRyFa2JitUI…>

4 5

User doesn't get added to mailinglist
by Pascal Christen March 2, 2019

March 2, 2019

Hi We have a problem since we updated from Mailman 2.1.27 -> 2.1.29 (FreeBSD). If we try to add an User on the Webgui "members/add" the Gui says successfully added and in the logs we see this: > logs/subscribe:Sep 20 10:04:59 2018 (20772) ugh_panteghini.ch: new > shouldfail(a)pascalchristen.ch, admin mass sub The user even get his confirmation mail with the password and link to his account. But if he tries to log in it says he's not authorized: > logs/mischief:Sep 20 10:09:48 2018 (33059) Login failure with private > rosters: shouldfail(a)pascalchristen.ch from X.X.X.X > logs/security:Sep 20 10:09:48 2018 (33059) Authorization failed > (private): user=shouldfail(a)pascalchristen.ch: list=ugh_panteghini.ch: > remote=X.X.X.X And if I check it with the command line tool the user isn't added: > ./list_members ugh_panteghini.ch > ... > shouldbeok(a)pascalchristen.ch Sometimes we're able to add an user when we set the "Send notifications about new registrations to the list owner?" to No. But it's like 50/50 if it works then or not. Do you guys have any idea whats wrong? Greetings Pascal

3 16

email to sms?
by Dave Stevens Feb. 28, 2019

Feb. 28, 2019

Hi, I've been using mailman to send routine announcements for a long time and more and more what people want is a text message. I've been able to discover gateways for individual carriers so that I can send to <number>@<gateway>.<carrier>.com and the subscriber gets a text, so phone notification which is quick and handy. This has been well received. I've been looking at the list of carriers getting longer and have looked into email to sms services but what I've seen has been commercial and too expensive. Is there a collaborative or open source email to sms project? Can anyone refer me to better information? I've been pretty much just casting around so far. TIA. Dave -- In modern fantasy (literary or governmental), killing people is the usual solution to the so-called war between good and evil. My books are not conceived in terms of such a war, and offer no simple answers to simplistic questions. ----- Ursula Le Guin

4 3

Starting mailman on reboot
by Odhiambo Washington Feb. 28, 2019

Feb. 28, 2019

Server: Ubuntu 18.01 I have followed these: http://www.gnu.org/software/mailman/mailman-install/node42.html % update-rc.d mailman defaults But still, mailman doesn't start on reboot. Nothing in the logs suggest any reason. -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 "Oh, the cruft.", grep ^[^#] :-)

3 10

Re: [Mailman-Users] Postfix relay_domains and MySQL tables (uncertainties of the beginner)
by Davide Marchi Feb. 28, 2019

Feb. 28, 2019

> > On 2/25/19 8:51 AM, Davide Marchi wrote: >> Hi Friends, >> I've just installed Mailman 2.1.23 on Debian Stretch, Postfix+Dovecot >> with "virtual_domains" "virtual_users" "virtual_aliases" on MySQL, and >> all seems works fine, but not tested again. >> I write you for clarification before to start. >> >> I'm not shure about the Postfix option: >> >> postconf -e 'relay_domains = lists.example.com' > > This looks like Debian stuff and probably has to do with > postfix_to_mailman.py. Please see <https://wiki.list.org/x/12812344> > and > <https://wiki.list.org/x/15564817>. > > -- > Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, > San Francisco Bay Area, California better use your sense - B. Dylan > Well, I will investigate.. Many thanks for the tips Mark! Davide -- cosmogoniA n o p r o v a r e n o f a r e o n o n f a r e n o n c e p r o v a r e

1 0

Re: [Mailman-Users] Starting mailman on reboot
by John Feb. 27, 2019

Feb. 27, 2019

On 2/27/19 12:22 PM, Mark Sapiro wrote: > On 2/27/19 9:36 AM, John wrote: >> hmmm... I have mailman on 2 systems. One does not start on boot. Referencing >> the manual link, I have no misc/mailman on either system.(?) Both systems are >> ubuntu (different versions), with Virtualmin. > > misc/mailman is in the source distribution. If you installed the > Debian/Ubuntu package, you probably don't have that, but it appears you > do have /etc/init.d/mailman > >> From the failing system: >> root>systemctl status mailman >> ● mailman.service - LSB: Mailman Master Queue Runner >> Loaded: loaded (/etc/init.d/mailman; bad; vendor preset: enabled) > These are normal. > >> Active: inactive (dead) > This of course is not. > > Are the ownership and mode of /etc/init.d/mailman the same on both > systems and is there any difference in the contents? > > You might try 'systemctl enable mailman' on the non-starting system. > > -- Mark Sapiro <mark(a)msapiro.net> The highway is for gamblers, San Francisco Bay > Area, California better use your sense - B. Dylan I forgot I found the problem before, reminded myself by debugging again... ;-) The directory /var/run/mailman is going away on reboot, so mailman can't create the pid file. The work-around is to create /var/run/mailman with the proper permissions & ownership and start the mailman service. I'm going to compare my 2 init.d files, if there is no difference, I am going to patch the init.d file to check for /var/run/mailman... Thanks, John

1 1

Re: [Mailman-Users] Starting mailman on reboot
by John Feb. 27, 2019

Feb. 27, 2019

On 2/27/19 8:19 AM, Mark Sapiro wrote: > On 2/27/19 1:45 AM, Odhiambo Washington wrote: >> The script in misc/ is the only one named mailman though. There is none in >> scripts/. > That was a mistake in the manual. It is now fixed. See > <https://www.gnu.org/software/mailman/mailman-install/node42.html>. > > >> I copied that into etc/init.d/ and rebooted, but mailman did not come up >> still. > > You may need 'chmod +x etc/init.d/'. > > What does 'systemctl status mailman' show? > hmmm... I have mailman on 2 systems. One does not start on boot. Referencing the manual link, I have no misc/mailman on either system.(?) Both systems are ubuntu (different versions), with Virtualmin. >From the failing system: root>systemctl status mailman ● mailman.service - LSB: Mailman Master Queue Runner Loaded: loaded (/etc/init.d/mailman; bad; vendor preset: enabled) Active: inactive (dead) Docs: man:systemd-sysv-generator(8) John

2 1