Mailman 3 April 1999 - Mailman-Developers

automatically generated password too complicated?
by Gerhard Gonter 07 Jul '99

07 Jul '99

Some of our users complained about the automatically generated passwords that are sent out when a list is imported or if an admin subscribes someone. Especially the ` and ^ characters are major problem because these may be treated as parts of composite characters in some enviroments (` followed by a might be displayed as the same character as à in HTML) and so on. Also, upper case characters impose an extra mental burden ;) Anyway, I modified our Mailman which now has a function (method?) Utils.GetRandomPassword(length) which generates passwords of the given length with a restricted alphabet, namely: a-x, 2-9, excluding characters o and l as well as digits 0 and 1 which may be confused and y, z (german keyboards swap these, in the past, this cause trouble too ;) I would like to offer this patch unless there are good reasons why this should be avoided. The main concern is certainly a higher risk to crack such passwords (only 30 possibilities instead of 64) but this could easly be matched by using 5 character passwords: possibilities strength --------------------------- 64^4 = 16777216 1 30^4 = 810000 0.05 30^5 = 24300000 1.45 As far as I have seen, this patch involves replacing certain calls to GetRandomSeed in a few places such as: bin/add_members, Mailman/Cgi/admin.py, Mailman/MailCommandHandler.py Any comment? +gg -- Gerhard.Gonter(a)wu-wien.ac.at Fax: +43/1/31336/702 g.gonter(a)ieee.org Zentrum fuer Informatikdienste, Wirtschaftsuniversitaet Wien, Austria

3 3

dots in list name -> archive not accessible?
by Gerhard Gonter 02 May '99

02 May '99

Once upon a time, one group of users decided that they need a list called Assistent/Inn/En(a)wu-wien.ac.at and it took me a while to convince them, that the / would not work so they accepted Assistent.Inn.En(a)wu-wien.ac.at which worked quite well on Listproc... Today I moved the list, everything went fine except that the archives were not accessible from the list overview page. I tracked this problem down to Mailman/Cgi/private.py where this function is defined, which strips an extension from the directory name: | def getListName(path): | component = string.split(path, os.sep)[1] | root, ext = os.path.splitext(component) | return root If component is returned instead of root, everything seems to be working fine for this list and the other lists without dots in their name. Why should an extension be stripped? Is this a bug, a feature, or something that was left over from VMS times? P.S.: Yesterday I mailed an message about the problems our users have with automatically generated passwords. Neither my collegue nor I seem to have received the article, also, no flames appeared either ;) The article has been archived as http://www.python.org/pipermail/mailman-developers/1999-April/001040.html Did something go wrong with that? +gg -- Gerhard.Gonter(a)wu-wien.ac.at Fax: +43/1/31336/702 g.gonter(a)ieee.org Zentrum fuer Informatikdienste, Wirtschaftsuniversitaet Wien, Austria

2 1

mailman does not work with exim and python 1.5.2
by Gergely Madarasz 01 May '99

01 May '99

Hello! Mailman does not work with exim and python 1.5.2, probably because of the esmtp support in python 1.5.2's smtplib. Following is a trace from exim. Note that exim gets an ehlo and later a helo which resets it into smtp from esmtp. Please forward this to the appropriate python forums :) set_process_info: 15805 2.11 handling incoming connection from localhost [127.0.0.1] ready for new message smtp_setup_msg entered SMTP<< ehlo darmol.elte.hu sender_fullhost = localhost (darmol.elte.hu) [127.0.0.1] sender_rcvhost = localhost ([127.0.0.1] helo=darmol.elte.hu) set_process_info: 15805 2.11 handling incoming connection from localhost (darmol.elte.hu) [127.0.0.1] 250-darmol.elte.hu Hello localhost [127.0.0.1] 250-SIZE 2000000 250-PIPELINING 250 HELP SMTP<< helo darmol.elte.hu sender_fullhost = localhost (darmol.elte.hu) [127.0.0.1] sender_rcvhost = localhost ([127.0.0.1] helo=darmol.elte.hu) set_process_info: 15805 2.11 handling incoming connection from localhost (darmol.elte.hu) [127.0.0.1] 250 darmol.elte.hu Hello localhost [127.0.0.1] SMTP<< mail FROM:<teszt-admin(a)darmol.elte.hu> size=1335 LOG: 4 MAIN SMTP syntax error in "mail FROM:<teszt-admin(a)darmol.elte.hu> size=1335" H=localhost (darmol.elte.hu) [127.0.0.1]: malformed address: size=1335 may not follow <teszt-admin(a)darmol.elte.hu> 501 <teszt-admin(a)darmol.elte.hu> size=1335: malformed address: size=1335 may not follow <teszt-admin(a)darmol.elte.hu> SMTP<< rset 250 Reset OK 421 darmol.elte.hu lost input connection LOG: 4 MAIN SMTP connection from localhost (darmol.elte.hu) [127.0.0.1] lost child 15805 ended -- Madarasz Gergely gorgo(a)caesar.elte.hu gorgo(a)linux.rulez.org It's practically impossible to look at a penguin and feel angry. Egy pingvinre gyakorlatilag lehetetlen haragosan nezni. HuLUG: http://mlf.linux.rulez.org/

3 5

Error/crash in admindb - Solved
by Brian Ryner 01 May '99

01 May '99

First of all, the bug I reported earlier happens if you access this URL: http://host/mailman/admindb/ <-- note trailing slash It tries to cause an "Invalid options to CGI script" error, but can't because of the typo in Mailman/Cgi/admindb.py, line 77: doc.AddItem(eader(2, "Invalid options to CGI script.")) where Header is misspelled. That fixed the crash. But I think that this URL (admindb/) should work just like admindb without the slash -- it should say that you must specify a list. -Brian Ryner bryner(a)uiuc.edu

2 1

Error/crash in admindb - 1.0b11
by Brian Ryner 01 May '99

01 May '99

Hi- This crash happened when going to http://host/mailman/admindb. I have not been able to reproduce it since then. The error log is attached. The system is running Slackware 3.5, with Python 1.5.1. Any suggestions or patches? Also, is there a way that I can turn off the extraneous debugging information sent to the WWW client, and have it just be put in the log? The fact that it dumps so much information about the system could be seen as a security risk. Thanks. -Brian Ryner bryner(a)uiuc.edu Apr 29 22:33:41 1999 admin: @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ admin: [----- Mailman Version: 1.0b11 -----] admin: [----- Traceback ------] admin: Traceback (innermost last): admin: File "/admin/mailman/scripts/driver", line 112, in run_main admin: main() admin: File "/admin/mailman/Mailman/Cgi/admindb.py", line 77, in main admin: doc.AddItem(eader(2, "Invalid options to CGI script.")) admin: NameError: eader admin: [----- Environment Variables -----] admin: DOCUMENT_ROOT: /files/admin/httpd/htdocs admin: HTTP_ACCEPT_ENCODING: gzip, deflate admin: SERVER_PORT: 80 admin: PATH_TRANSLATED: /files/admin/httpd/htdocs/ admin: URL_COUNT: 6 admin: GATEWAY_INTERFACE: CGI/1.1 admin: HTTP_ACCEPT_LANGUAGE: en-us admin: HTTP_PROXY_CONNECTION: Keep-Alive admin: REMOTE_ADDR: xxx.yyy.zzz.xxx admin: SERVER_NAME: www.xxxxxxx.org admin: URL_COUNT_DB: /files/admin/httpd/logs/counters/ALL admin: HTTP_USER_AGENT: Mozilla/4.0 (compatible; MSIE 4.01; Windows 95) admin: HTTP_ACCEPT: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/vnd.ms-excel, application/msword, application/vnd.ms-powerpoint, */* admin: REQUEST_URI: /mailman/admindb/ admin: PATH: /usr/local/sbin:/usr/local/bin:/sbin:/usr/sbin:/bin:/usr/bin admin: QUERY_STRING: admin: SCRIPT_FILENAME: /files/admin/mailman/cgi-bin/admindb admin: PATH_INFO: / admin: HTTP_HOST: www.xxxxxxx.org admin: REQUEST_METHOD: GET admin: SERVER_SIGNATURE: admin: URL_COUNT_RESET: Thursday, 29-Apr-99 00:19:13 CDT admin: SCRIPT_NAME: /mailman/admindb admin: SERVER_ADMIN: admin(a)xxxxxx.org admin: SERVER_SOFTWARE: Apache/1.3.4 (Unix) admin: PYTHONPATH: /admin/mailman admin: SERVER_PROTOCOL: HTTP/1.0 admin: REMOTE_PORT: 6208

2 1

Re: [Mailman-Users] Error/crash in admindb - 1.0b11
by Brian Ryner 30 Apr '99

30 Apr '99

Is that line (admindb.py, line 77) supposed to say Header instead of eader ? > Apr 29 22:33:41 1999 admin: @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ > admin: [----- Mailman Version: 1.0b11 -----] > admin: [----- Traceback ------] > admin: Traceback (innermost last): > admin: File "/admin/mailman/scripts/driver", line 112, in run_main > admin: main() > admin: File "/admin/mailman/Mailman/Cgi/admindb.py", line 77, in main > admin: doc.AddItem(eader(2, "Invalid options to CGI script.")) > admin: NameError: eader

1 0

Re: [Mailman-Developers] Having kittens with the code
by Barry A. Warsaw 27 Apr '99

27 Apr '99

>>>>> "NJD" == Nicholson James D <James.Nicholson(a)amedd.army.mil> writes: NJD> I have been trying to fix the mailman code for some time now, NJD> and I would really appreciate any help. The basic problem is NJD> that the anonymous_list is broken on my machine, I don't know NJD> why, but, I can't even debug. James, here's another situation that I can't reproduce. If I set "Hide the sender" to yes, I get the expected header. Note that sendmail gets partly in the way here (and this may be your problem?). For me, sendmail munges the From header to not include my host name. This sucks because our mailhost knows nothing about my list. Oh well, sendmail's evil anyway. :-) BTW, on your previous message on the subject, I think From: should still come from the GetAdminEmail(), however I think maybe Reply-To: for anonymous lists should be set to GetListEmail(). And note further that turning on anonymous_list is probably a false comfort. Lots of other headers get through that /could/ be used to identify the user (e.g. my X-Face header gets through). We'd probably want to do a lot more stripping of headers to have a really effective anonymizer. NJD> So, if you find the code: | del msg['reply-to'] | del msg['sender'] NJD> I then add the following code: | msg.SetHeader('From', 'listname@machinename') | msg.SetHeader('To', 'listname@machinename') NJD> and comment out the following code beginning with 'if NJD> self.anonymous_list:' And what happens is nothing. Nothing I NJD> do here seems to affect the header of reflected mail. Is NJD> this the wrong place to make these changes or is the mail NJD> being postprocessed somewhere else that undoes my changes? Okay, apologies for the following "did-you-plug-it-in" suggestion. You're sure that you're either making the change to the installed version of the files, or that you're doing a make install after making the changes. In other words, you're absolutely positive that Mailman's running your modified version? Try also blowing away the .pyc files and see if they get regenerated. Finally, you can just do something like: fp = open('/tmp/rawlog.txt', 'w') fp.write('I know you are in there!\n') fp.close() right at the point in question. If that file doesn't show up with the message in it, you've got deeper problems. If that works, but LogMsg() still doesn't, it /could/ be a permission problem, but I don't think so. You might just want to use your rawlog.txt file to do your logging directly. -Barry

1 0

Having kittens with the code
by Nicholson James D 27 Apr '99

27 Apr '99

I have been trying to fix the mailman code for some time now, and I would really appreciate any help. The basic problem is that the anonymous_list is broken on my machine, I don't know why, but, I can't even debug. My Problem: I turn on the anonymous_list option so that all mail appears to the recipient as if it emanates from the listname@machine. However, it doesn't work. Everything sent out to the test list still appears to come from an individual user. My Attempt at a solution: Someone told me that the anonymizing of the list is done in the Post() routine in MailList.py. So, I go in there and hardwire a change to the 'From' and 'To' fields, with no luck. So, if you find the code: del msg['reply-to'] del msg['sender'] I then add the following code: msg.SetHeader('From', 'listname@machinename') msg.SetHeader('To', 'listname@machinename') and comment out the following code beginning with 'if self.anonymous_list:' And what happens is nothing. Nothing I do here seems to affect the header of reflected mail. Is this the wrong place to make these changes or is the mail being postprocessed somewhere else that undoes my changes? So, now I really want to debug. So, I use the handy self.LogMsg() function. But, I get nothing in the logs. Here's the code I added right after the hardwired changes above: subj = msg.getheader('from') self.LogMsg("error", "initially from %s ", subj ) NADA for output. Can someone please send me a clue here? I've been working on it for weeks and can't get anywhere. If I can't solve the problem, I just won't have a mailing list. -Jim

1 0

Mailman on NT
by Gianni Sandrucci 22 Apr '99

22 Apr '99

I have read some messages about future Mailman NT release. Since I have appreciated Mailman on a IBM AIX system I tried a porting test on NT with following steps: - installed Cygwin 2.0 (www.cygnus.com) - installed Python for Windows with all Windows specific extension - installed Mailman 1.0 b11 - made some minor changes in Mailmal files configure and src/Makefile.in - after some fighting with permissions (I was forced to comment tests on these matters in configuration script and C source code) succeded to ./configure --with-cgi-ext=.exe make install (no errors reported) - configured IIS to create a virtual site on /home/mailman named mailman - called: http://localhost/mailman/cgi-bin/admin.exe Got an error (reported below). It seems Python is not able to locate Mailman modules in /home/mailman/Mailman. However modules are there (in py and pyc format). I tried to go up in erorrs chain but every time the 'driver' script try to import something from Mailman I got the same error. Since I know very little about Python I wonder if anybody can help me. Thanks. ------------------------------------------- Bug in Mailman version <undetermined> We're sorry, we hit a bug! If you would like to help us identify the problem, please email a copy of this page to the webmaster for this site with a description of what happened. Thanks! Traceback: Traceback (innermost last): File "/home/mailman/scripts/driver", line 135, in print_traceback from Mailman.mm_cfg import VERSION ImportError: No module named Mailman.mm_cfg -------------------------------------------------------------------------------- Environment variables: Variable Value USERPROFILE C:\WINNT\Profiles\Default User HTTP_ACCEPT_ENCODING gzip, deflate REMOTE_HOST 127.0.0.1 HTTPS off OS2LIBPATH C:\WINNT\system32\os2\dll; SERVER_PORT_SECURE 0 SERVER_PORT 80 PATH_TRANSLATED C:\Inetpub\wwwroot PROCESSOR_LEVEL 6 GATEWAY_INTERFACE CGI/1.1 INSTANCE_ID 1 NUMBER_OF_PROCESSORS 1 HTTP_ACCEPT_LANGUAGE en-us PATH c:\agora\servletmanager\jrun\bin;c:\agora\servletmanager\jrun\jre\115\bin;C:\WINNT\system32;C:\WINNT;C:\Program Files\Mts SYSTEMDRIVE C: PATHEXT .COM;.EXE;.BAT;.CMD;.VBS;.JS SERVER_NAME localhost HTTP_CONNECTION Keep-Alive PROCESSOR_ARCHITECTURE x86 TERM cygwin WINDIR C:\WINNT HTTP_ACCEPT image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/vnd.ms-excel, application/msword, application/vnd.ms-powerpoint, */* COMPUTERNAME SVIL4 INCLUDE C:\Program Files\Mts\Include SYSTEMROOT C:\WINNT SERVER_PROTOCOL HTTP/1.1 PROCESSOR_REVISION 0502 CONTENT_LENGTH 0 HTTP_HOST localhost REQUEST_METHOD GET PROCESSOR_IDENTIFIER x86 Family 6 Model 5 Stepping 2, GenuineIntel PYTHONPATH /home/mailman LOCAL_ADDR 127.0.0.1 SCRIPT_NAME /mailman/cgi-bin/admin.exe SERVER_SOFTWARE Microsoft-IIS/4.0 HTTP_USER_AGENT Mozilla/4.0 (compatible; MSIE 4.01; Windows NT) OS Windows_NT LIB C:\Program Files\Mts\Lib COMSPEC C:\WINNT\system32\cmd.exe REMOTE_ADDR 127.0.0.1 @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ [----- Mailman Version: <undetermined> -----] [----- Traceback ------] Traceback (innermost last): File "/home/mailman/scripts/driver", line 135, in print_traceback from Mailman.mm_cfg import VERSION ImportError: No module named Mailman.mm_cfg [----- Environment Variables -----] USERPROFILE: C:\WINNT\Profiles\Default User HTTP_ACCEPT_ENCODING: gzip, deflate REMOTE_HOST: 127.0.0.1 HTTPS: off OS2LIBPATH: C:\WINNT\system32\os2\dll; SERVER_PORT_SECURE: 0 SERVER_PORT: 80 PATH_TRANSLATED: C:\Inetpub\wwwroot PROCESSOR_LEVEL: 6 GATEWAY_INTERFACE: CGI/1.1 INSTANCE_ID: 1 NUMBER_OF_PROCESSORS: 1 HTTP_ACCEPT_LANGUAGE: en-us PATH: c:\agora\servletmanager\jrun\bin;c:\agora\servletmanager\jrun\jre\115\bin;C:\WINNT\system32;C:\WINNT;C:\Program Files\Mts SYSTEMDRIVE: C: PATHEXT: .COM;.EXE;.BAT;.CMD;.VBS;.JS SERVER_NAME: localhost HTTP_CONNECTION: Keep-Alive PROCESSOR_ARCHITECTURE: x86 TERM: cygwin WINDIR: C:\WINNT HTTP_ACCEPT: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/vnd.ms-excel, application/msword, application/vnd.ms-powerpoint, */* COMPUTERNAME: SVIL4 INCLUDE: C:\Program Files\Mts\Include SYSTEMROOT: C:\WINNT SERVER_PROTOCOL: HTTP/1.1 PROCESSOR_REVISION: 0502 CONTENT_LENGTH: 0 HTTP_HOST: localhost REQUEST_METHOD: GET PROCESSOR_IDENTIFIER: x86 Family 6 Model 5 Stepping 2, GenuineIntel PYTHONPATH: /home/mailman LOCAL_ADDR: 127.0.0.1 SCRIPT_NAME: /mailman/cgi-bin/admin.exe SERVER_SOFTWARE: Microsoft-IIS/4.0 HTTP_USER_AGENT: Mozilla/4.0 (compatible; MSIE 4.01; Windows NT) OS: Windows_NT LIB: C:\Program Files\Mts\Lib COMSPEC: C:\WINNT\system32\cmd.exe REMOTE_ADDR: 127.0.0.1

1 0

FW: [Mailman-Users] Still need code info for privacy option
by Ken Manheimer 21 Apr '99

21 Apr '99

I'm sorry i can't help more with this, no time - but mailman-developers is a more appropriate venue for the question, maybe you'll get some other attention there. Context for mailman-developers - setting the "hide the message sender" privacy option doesn't seem to be working, and the code (in MailList.MailList.Post()) doesn't seem to be right - the reply-to is deleted, but the code is setting the address to the list admin, not the list proper. And it's not working anyway - as james d mentions below, wiring it to set the from to something else is not working... Ken Manheimer klm(a)digicool.com -----Original Message----- From: Nicholson James D [mailto:James.Nicholson@amedd.army.mil] Sent: Tuesday, April 20, 1999 5:17 PM To: Ken Manheimer Cc: mailman-users(a)python.org Subject: RE: [Mailman-Users] Still need code info for privacy option [Nicholson James D] I found the problem with the list not remailing anonymously. In routine Post(), it deletes the sender and reply-to fields and then sets the from field. However, it set the From field to self.GetAdminEmail. What I needed it to do (and what I believe it should be doing) is setting the From field to self.GetListEmail, which will return the list address. However, I can't get the damned thing to change the From field for anything. I've even done a msg.setheader( 'From', 'listname@machine' ). That's hardwired in the Post() routine just for testing, and it isn't bloody changing the header... at all. I altered the MailList.py file, moved the MailList.pyc to oldMailList.pyc, and it still does the same damned thing. It sets the From field to whoever sent the message. Was I supposed to recompile the package or something to get my changes to take effect? I just don't know python or mailman well enough to know even this. Sorry. -Jim

1 0

Re: [Mailman-Users] Moving Listproc lists
by Gerhard Gonter 17 Apr '99

17 Apr '99

According to Greg Connor: > I did it all by hand on my system... here are the commands I used to > transfer users, as closely as I can remember... Thanks a lot, in the meantime I moved the subscribers of the first few lists (100 more to go :) using a modified add_members: ./add_members --lp6 lists/LIST1/.subscribers list1 Only one file, add_members, had to be modified. This maps ACK, NOACK -> normal subscriptions DIGEST -> digest subscriptions and keeps the original password. Addresses are converted to lower case. I will offer this version to the developers, maybe it will show up in the next version. Right now, POSTPONE'd entries are dropped and the CONCEAL flag is ignored, keeping this information requires modifications in other parts of Mailman. I'm not sure if I'll try that, maybe not... > Also, I had problems subbing more than 20 people at a time; all would get > subbed, but only 20 or so would get welcome messages (containing their new > random password). If you think you might have this problem too, use > something like this instead.. Hmm... Thanks for the hint. The sendmail logs tell me that everyone in a list with 167 members should have received their welcome messages by now, so this might be a bug in other versions. The bug archive didn't mention anything about that. (I'm using 1.0b8, the original add_members was pulled out of CVS yesterday.) +gg -- Gerhard.Gonter(a)wu-wien.ac.at Fax: +43/1/31336/702 g.gonter(a)ieee.org Zentrum fuer Informatikdienste, Wirtschaftsuniversitaet Wien, Austria

1 0

jitterbug and #12
by Peter Gervai 16 Apr '99

16 Apr '99

Hello, Out of my curiousity I submitted a bug into the bug database. It got resolved really fast. I only have two problems. First: Maybe it's just me but I don't really see WHO did close the bug. Second: The bug (actually rather a cosmetic problem than a real bug) still exists, and the resolving was unsatisfactory (huh :)) for me, but there is no way to reply to the bug resolver, either because I do not know who he was or because I'm "logged in as a guest user". And about the #12: as the original text told the install overwrites the $BASE/templates directory without warning. Since I localized the templates to be able to create localized lists with newlist I wasn't REALLY happy to learn that all of them got overwritten. Of course lists/<listname>/templates don't get overwritten, but that's another matter I did not complain about... regards, grin

2 2

Re: [Mailman-Developers] Hypermail
by J C Lawrence 12 Apr '99

12 Apr '99

On Mon, 12 Apr 1999 18:43:37 +0200 (CEST) Dan Ohnesorg <dan(a)feld.cvut.cz> wrote: > On Mon, 12 Apr 1999 claw(a)kanga.nu wrote: >> I strongly suggest you look at MHonArc >> (http://www.oac.uci.edu/indiv/ehood/mhonarc.doc.html). It has >> far better MIME support, is significantly more configurable, etc >> etc etc yada yada. > Yes, I know, but it is nice on hypermail that it can be trigered > directly from mailman. When I will use another archiving tool, it > must be runed from crontab. MHonArc can archive a single message piped to sdtin. See the docs for the -add (IIRC) command line option. -- J C Lawrence Internet: claw(a)kanga.nu ---------(*) Internet: claw(a)varesearch.com ...Honorary Member of Clan McFud -- Teamer's Avenging Monolith...

1 0

Re: [Mailman-Developers] Hypermail
by claw＠kanga.nu 12 Apr '99

12 Apr '99

On Mon, 12 Apr 1999 17:45:19 +0200 (CEST) Dan Ohnesorg <dan(a)feld.cvut.cz> wrote: > I have troubles with hypermail archiver. I strongly suggest you look at MHonArc (http://www.oac.uci.edu/indiv/ehood/mhonarc.doc.html). It has far better MIME support, is significantly more configurable, etc etc etc yada yada. -- J C Lawrence Internet: claw(a)kanga.nu ----------(*) Internet: coder(a)kanga.nu ...Honorary Member of Clan McFud -- Teamer's Avenging Monolith...

4 4

Hypermail
by Dan Ohnesorg 12 Apr '99

12 Apr '99

I have troubles with hypermail archiver. On some ocasions it makes bad references between index page and articles, as you can see on http://poli.feld.cvut.cz/archives/public/kecy/1999-March/date.html 3-9 artiles from end are under one link. It is the most bad fall, but on other archives ot makes also too. It is version 1.0b10. BTW I will made templates for archives instead of static strings. Has it someone made alredy? cheers dan

1 0

Re: [Mailman-Developers] exim - Failure to exec script- gid question
by Harald Meland 06 Apr '99

06 Apr '99

[nbecker(a)fred.net] > I'm using mailman-1.0b10 + exim-2.10. > > Mar 30 13:10:08 nbeckerpc Mailman mail-wrapper: Failure to exec script. WANTED gid 12, GOT gid 99. (Reconfigure to take 99?) > > I can reconfigure mailman or exim to set the gid, but why is it needed > at all? Since wrapper is sgid, why does it care what gid was used to > invoke it? Just additional security check? Yes, it's "merely" a sanity check :) -- Harald

1 0

Envelope sender and list-admin address
by Brian Ryner 05 Apr '99

05 Apr '99

Hi- I reported a bug on this but I thought I'd also post here, becuause I might try to correct this myself (and any advice would be appreciated). Basically, when a message is sent to the admin address (i.e. mailman-developers-admin(a)python.org), the envelope sender is left untouched when the message is resent. I don't think this is a good idea. If an address on the admin list is not on the local host, and the email is not sent from the local host, then you basically have a message from a non-local address to a non-local address, which is a no-no in many anti-spam-relaying setups. It would make much more sense to set the envelope sender to the list-admin address. -- -Brian Ryner bryner(a)uiuc.edu

1 0

changes to mailpasswd to support multiple domains
by Jeff Hahn 04 Apr '99

04 Apr '99

Attached is a patch to cron/mailpasswds and templates/cronpass.txt to support multiple domains in the monthly password reminder. Patch is against latest CVS sources (1.0b11) WARNING: Apply with care! I'm just getting familiar with Python and Mailman so I could have made some really idiotic mistakes, but it works here... To Apply: cd /usr/src/mailman ("main" source directory) patch -p1 -b <password.dif Primary reason the patch file is so large is that I "de-tabbed" the source file (removed leading tabs and replaced with spaces). The original source was about half tabs and half spaces for indents. I'm not sure what the "official" position is on indenting for mailman. Let me know if this isn't necessary in the future. -Jeff diff -c -r MM.orig/cron/mailpasswds MM/cron/mailpasswds *** MM.orig/cron/mailpasswds Sat Apr 3 09:18:19 1999 --- MM/cron/mailpasswds Sun Apr 4 10:00:24 1999 *************** *** 51,79 **** The list can be any random one - it is only used for the message delivery mechanism.""" ! subj = '%s maillist memberships reminder\n' % list.host_name count = PAUSE_FREQUENCY for user, data in users.items(): ! table = [] ! for l, p, u in data: ! if len(l) > 9: ! table.append("%s\n %-10s\n%s\n" % (l, p, u)) ! else: ! table.append("%-10s %-10s\n%s\n" % (l, p, u)) ! header = ("%-10s %-10s\n%-10s %-10s" ! % ("List", "Password // URL", "----", "--------")) text = Utils.maketext( 'cronpass.txt', ! {'hostname': list.host_name, 'user' : user, ! 'one_list': l, }) # add this to the end so it doesn't get wrapped/filled text = text + header + '\n' + string.join(table, '\n') ! list.SendTextToUser(subject = subj, ! recipient = user, ! text = text, ! sender = mm_cfg.MAILMAN_OWNER, add_headers = ["X-No-Archive: yes", "Precedence: bulk"]) count = count - 1 --- 51,82 ---- The list can be any random one - it is only used for the message delivery mechanism.""" ! default_host = mm_cfg.DEFAULT_HOST_NAME ! mailman_owner = mm_cfg.MAILMAN_OWNER ! subj = '%s maillist memberships reminder\n' % default_host count = PAUSE_FREQUENCY for user, data in users.items(): ! table = [] ! for l, r, p, u in data: ! if len(l) > 39: ! table.append("%s\n %-10s\n%s\n" % (l, p, u)) ! else: ! table.append("%-40s %-10s\n%s\n" % (l, p, u)) ! header = ("%-40s %-10s\n%-40s %-10s" ! % ("List", "Password // URL", "----", "--------")) text = Utils.maketext( 'cronpass.txt', ! {'def_host': default_host, 'user' : user, ! 'ex_req' : r, ! 'mm_owner': mailman_owner, }) # add this to the end so it doesn't get wrapped/filled text = text + header + '\n' + string.join(table, '\n') ! list.SendTextToUser(subject = subj, ! recipient = user, ! text = text, ! sender = mailman_owner, add_headers = ["X-No-Archive: yes", "Precedence: bulk"]) count = count - 1 *************** *** 87,119 **** the user a single message with the info for all their lists on this site.""" list = None ! users = {} # user: (listname, password, url) # Constrain to specified users, if any. confined_to = args[1:] a_public_list = None for name in Utils.list_names(): ! list = MailList.MailList(name, lock = 0) if not a_public_list and list.advertised: a_public_list = list ! list_name = list.real_name ! umbrella_list = list.umbrella_list ! if not list.send_reminders: ! continue ! for user, password in list.passwords.items(): if confined_to: if user not in confined_to: continue else: # We're a bit verbose when operating "confined_to": print "%s in %s" % (user, list.real_name) ! url = list.GetAbsoluteOptionsURL(user) recipient = list.GetMemberAdminEmail(user) ! if users.has_key(recipient): ! users[recipient].append(list_name, password, url) ! else: ! users[recipient] = [(list_name, password, url)] if list: ! MailAllPasswords(a_public_list or list, users) def waitall(): """Return only when there are no forked subprocesses running.""" --- 90,125 ---- the user a single message with the info for all their lists on this site.""" list = None ! users = {} # user: (list address, request address, password, url) # Constrain to specified users, if any. confined_to = args[1:] a_public_list = None for name in Utils.list_names(): ! list = MailList.MailList(name, lock = 0) if not a_public_list and list.advertised: a_public_list = list ! list_address = list.GetListEmail() ! list_request = list.GetRequestEmail() ! umbrella_list = list.umbrella_list ! if not list.send_reminders: ! continue ! for user, password in list.passwords.items(): if confined_to: if user not in confined_to: continue else: # We're a bit verbose when operating "confined_to": print "%s in %s" % (user, list.real_name) ! url = list.GetAbsoluteOptionsURL(user) recipient = list.GetMemberAdminEmail(user) ! if users.has_key(recipient): ! users[recipient].append(list_address, list_request, ! password, url) ! else: ! users[recipient] = [(list_address, list_request, ! password, url)] if list: ! MailAllPasswords(a_public_list or list, users) def waitall(): """Return only when there are no forked subprocesses running.""" diff -c -r MM.orig/templates/cronpass.txt MM/templates/cronpass.txt *** MM.orig/templates/cronpass.txt Sat Apr 3 09:18:28 1999 --- MM/templates/cronpass.txt Sun Apr 4 08:07:38 1999 *************** *** 1,4 **** ! This is a reminder, sent out once a month, about your %(hostname)s mailing list memberships. It includes your subscription info and how to use it to change it or unsubscribe from a list. --- 1,4 ---- ! This is a reminder, sent out once a month, about your %(def_host)s mailing list memberships. It includes your subscription info and how to use it to change it or unsubscribe from a list. *************** *** 8,18 **** In addition to the URL interfaces, you can also use email to make such changes. For more info, send a message to the '-request' address of ! the list (for example, %(one_list)s-request@%(hostname)s) containing just the word 'help' in the message body, and an email message will be sent to you with instructions. If you have questions, problems, comments, etc, send them to ! mailman-owner@%(hostname)s. Thanks! Passwords for %(user)s: --- 8,19 ---- In addition to the URL interfaces, you can also use email to make such changes. For more info, send a message to the '-request' address of ! the list (for example, %(ex_req)s) containing just the word 'help' in the message body, and an email message will be sent to you with instructions. If you have questions, problems, comments, etc, send them to ! %(mm_owner)s. Thanks! Passwords for %(user)s: +

1 0

oops - hit a bug - 1.0b11
by Jeff Hahn 04 Apr '99

04 Apr '99

just noticed this in my logs from this afternoon - don't know if it's b11 specific... I am able to reproduce the problem by: 1. go to listinfo page 2. hit link for archive (archive restricted to members) 3. enter aaa(a)bbb.com for user ( any non member ) 4. enter 12345 (or anything else for password) then I get the "oops" page... I don't recall trying to access a restricted archive as a non-member before this, so it may not be b11 specific. I assume private.py should also be trapping the MMNotAMemberError error... -Jeff Apr 03 18:04:13 1999 admin: @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ admin: [----- Mailman Version: 1.0b11 -----] admin: [----- Traceback ------] admin: Traceback (innermost last): admin: File "/home/mailman/MM/scripts/driver", line 112, in run_main admin: main() admin: File "/home/mailman/MM/Mailman/Cgi/private.py", line 163, in main admin: if not list_name or not isAuthenticated(list_name): admin: File "/home/mailman/MM/Mailman/Cgi/private.py", line 120, in isAuthenticated admin: listobj.ConfirmUserPassword( username, password) admin: File "/home/mailman/MM/Mailman/SecurityManager.py", line 76, in ConfirmUserPassword admin: raise Errors.MMNotAMemberError admin: MMNotAMemberError

2 1

Re: [Mailman-Developers] mailpasswds issues with multiple domains
by bruce＠perens.com 03 Apr '99

03 Apr '99

From: Jeff Hahn <jeffh(a)streek.com> > Is this an issue for other people? In other words, should I make up a set > of clean diff's and submit them for inclusion in the distribution? To > where? Yes. I'm running mine in multiple domains, it would be very handy. Thanks Bruce

1 0

Release 1.0b11
by Barry A. Warsaw 03 Apr '99

03 Apr '99

Okay folks, 1.0b11 is now available at www.list.org. This is basically a bug fix release addressing problems with case-preserving addresses and archiving of Usenet gated newsgroups. For you hardcore debugging types, there's a new script called bin/withlist which you can run like so: % cd ~mailman % python -i bin/withlist mylist Loading list: mylist (unlocked) >>> m.members {'bwarsaw(a)cnri.reston.va.us': 0} >>> withlist leaves you with a global variable `m' which is the MailList object you opened. Give withlist and optional second argument if you want the list to be locked. A note about upgrading. I had to bump the DATA_FILE_VERSION number so that the config.db got updated to deal with case-preserving addresses in some situation. I've been running this on python.org for a few days now and it all seems to work, but please keep an eye out for anything strange with your config.db file. Enjoy, -Barry

2 1

problems with who ?
by Gergely Madarasz 03 Apr '99

03 Apr '99

I'm running mailman 1.0b10 This is all I get from a who request. ---------- Forwarded message ---------- Date: Wed, 31 Mar 1999 10:48:22 +0200 From: linux-flame-request(a)mlf.linux.rulez.org To: gorgo(a)caesar.elte.hu Subject: Mailman results for linux-flame ***** who Digest Members: anovak(a)nap-szam.hu atom(a)sophia.jpte.hu b-b0mb(a)freemail.c3.hu ba... Non-Digest Members: a7803268(a)unet.univie.ac.at a_linux(a)freemail.c3.hu acsosz(a)mail.da...

3 3

Re: [Mailman-Developers] Usenet gating and archiving
by Benjamin B. Thomas 03 Apr '99

03 Apr '99

> Barry A. Warsaw bwarsaw(a)cnri.reston.va.us writes: > Is anybody currently gating Usenet to a mailing list, /and/ doing > archiving with the default Pipermail? I am not yet using Mailman to do this, but I am doing the same thing with a hairy custom setup (majordomo + MHonArc + INN + gateway scripts). In about a month or so, I plan to switch our setup over to Mailman since it is *so* much nicer. > I would rather not add this hack to Mailman unless a lot of people > would benefit from it. We have a number of people who choose to interact via our local Usenet groups, and it is important that their contributions are archived. Without the ability to do so, I will need to stay with our current setup (icky as it is from an administrative standpoint, it does what I need) instead of moving to Mailman. Let me just say that Mailman is a wonderful piece of software. I have poked at it a little bit on a test installation, and it is beautifully conceived and implemented. Cheers, benjy 1975 Cahaba Valley Road (205) 988-4268 Indian Springs, AL 35124 benjy(a)alum.mit.edu

2 1

Usenet gating and archiving
by Barry A. Warsaw 03 Apr '99

03 Apr '99

Is anybody currently gating Usenet to a mailing list, /and/ doing archiving with the default Pipermail? If so, you may not have noticed that the archives are pretty broken; nothing that originates on Usenet will show up in the archive. I believe I understand why this is happening, and although I don't have a fix yet, it shouldn't be too difficult. I should be able to get it into 1.0b11. The problem occurs because Pipermail uses Python's mailbox.UnixMailbox objects which expect "From " separators. Message that originate on Usenet don't have these envelopes. The patch should be to simply synthesize these on messages gated off the newsgroup. However, there's a more severe breakage. You've actually lost information that is hard to reproduce because the flat listname.mbox files don't have the separators in them either. Running bin/arch over the file will give you even more corrupt archives (the incremental archiver just throws the Usenet messages away; bin/arch will tack them onto the last email-originated message). What you could do is to trek through the listname.mbox file looking for Newsgroup: headers (which appear, but I don't think are guaranteed, to be the first header in the message). When you find one, you jam in a synthesized "From " envelope. I would rather not add this hack to Mailman unless a lot of people would benefit from it. For the test lists that I'm running, we're willing to just trash the archives and start over; the lists aren't live anyway. I'd like to know who would be really adversely affected by this problem. Thanks, -Barry

2 2

Re: [Mailman-checkins] CVS: mailman/Mailman Deliverer.py
by Edgard Castro 01 Apr '99

01 Apr '99

On Wed, Mar 31, 1999 at 02:23:14PM -0500, Barry A. Warsaw wrote: > Modified Files: > Deliverer.py > Log Message: > DeliverToList(): Strip off any duplicate Sender: headers, leaving just > the one pointing to the list-admin. The same problems occurs with Errors-To, haven't checked agains other headers. e. -- Edgard Castro <castro(a)usmatrix.net> | 305-273-5003 -o) Chief Network Engineer - U.S. Matrix Internet, Inc. /\\ Linux Counter User #29078 -- http://counter.li.org _\_V

1 0