Mailman 3 January 2003 - Mailman-coders

[ mailman-Bugs-621689 ] Subscribe requests dropped from admin db
by SourceForge.net 21 Jan '03

21 Jan '03

Bugs item #621689, was opened at 2002-10-10 20:10 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_i… Category: (un)subscribing Group: 2.1 (stable) Status: Open Resolution: None Priority: 7 Submitted By: Mark Dadgar (mdadgar) Assigned to: Nobody/Anonymous (nobody) Summary: Subscribe requests dropped from admin db Initial Comment: Occasionally, subscribe requests (via the web page) to a list that requires moderator approval for subscriptions never make it to the admin db for that list. They are listed in the subscribe log file, but never show up on the admin db page. Log file entries look like this: Oct 10 18:43:43 2002 (31168) eventmasters: pending XXXXXX.swfla.rr.com But the admin db page for the list shows no outstanding issues. I have verified that there was an actual subscription request behind this log entry. ---------------------------------------------------------------------- >Comment By: Mark Dadgar (mdadgar) Date: 2003-01-20 23:32 Message: Logged In: YES user_id=598228 I suppose it's possible that those two patches are causing it, but it seems unlikely. What information can I provide that will help you track this down? It's kind of a bummer. ---------------------------------------------------------------------- Comment By: Barry A. Warsaw (bwarsaw) Date: 2002-12-23 21:06 Message: Logged In: YES user_id=12800 I definitely need more information. I've neve been able to reproduce the problem. Do you think it could be related to the unofficial patches you've applied? ---------------------------------------------------------------------- Comment By: Mark Dadgar (mdadgar) Date: 2002-12-05 23:29 Message: Logged In: YES user_id=598228 Ok, I verified that it's still a problem under 2.1b5 with the htdig and indexing patches. It's a real drag, too. :( Please let me know if you need more debug info. ---------------------------------------------------------------------- Comment By: Mark Dadgar (mdadgar) Date: 2002-11-21 22:08 Message: Logged In: YES user_id=598228 Unfortunately, it's still occurring under 2.1b4. I'll check it under 2.1b5 as soon as the patches for htdig integration are released, as I need those to implement. ---------------------------------------------------------------------- Comment By: Barry A. Warsaw (bwarsaw) Date: 2002-10-22 07:11 Message: Logged In: YES user_id=12800 I'm going to defer this one until 2.1b4 goes out. I've fixed some related bugs in cvs, so hopefully this will be fixed in the new version. ---------------------------------------------------------------------- Comment By: Mark Dadgar (mdadgar) Date: 2002-10-20 10:14 Message: Logged In: YES user_id=598228 Sorry - I'm running 2.1b3. ---------------------------------------------------------------------- Comment By: Barry A. Warsaw (bwarsaw) Date: 2002-10-20 08:48 Message: Logged In: YES user_id=12800 What version of Mailman are you using. It's import to know the exact version, i.e. 2.1beta3, or if you're running from cvs. A lot of these problems have been fixed in cvs, which will soon be released as 2.1 beta 4. ---------------------------------------------------------------------- Comment By: Mark Dadgar (mdadgar) Date: 2002-10-10 21:29 Message: Logged In: YES user_id=598228 I just verified, by cat'ing pending.pck, that the appropriate changes seem to be in there. But the web interface never picks them up. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=621689&group_i…

1 0

[ mailman-Patches-668819 ] improve plain-ness and i18n-ness of plain-text digest
by SourceForge.net 21 Jan '03

21 Jan '03

Patches item #668819, was opened at 2003-01-16 00:17 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=668819&group_i… Category: internationalization Group: Mailman 2.1 Status: Open Resolution: None Priority: 5 Submitted By: Tokio Kikuchi (tkikuchi) Assigned to: Nobody/Anonymous (nobody) Summary: improve plain-ness and i18n-ness of plain-text digest Initial Comment: When the mail charset and list charset are different like Japanese, plain text digests become mixture of these charsets and result in mojibake. This patch converts the message into list specific charset and deliver the digest with single charset. It also takes care of the message headers to retain in the digest according to the setting of mm_cfg.DEFAULT_PLAIN_DIGEST_KEEP_HEADERS. ---------------------------------------------------------------------- >Comment By: Tokio Kikuchi (tkikuchi) Date: 2003-01-21 07:00 Message: Logged In: YES user_id=67709 Update of the patch: Use Scrubber.py to make plain text from multipart message. Problem: each attached file will be saved twice; when archiving and when digest. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=668819&group_i…

1 0

[ mailman-Bugs-671303 ] bin/arch --start and --end argument handling
by SourceForge.net 21 Jan '03

21 Jan '03

Bugs item #671303, was opened at 2003-01-20 13:52 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=671303&group_i… >Category: Pipermail >Group: 2.1 (stable) Status: Open Resolution: None Priority: 5 Submitted By: Bryan Fullerton (fehwalker) Assigned to: Nobody/Anonymous (nobody) Summary: bin/arch --start and --end argument handling Initial Comment: bin/arch says (among other things): -s N --start=N Start indexing at article N, where article 0 is the first in the mbox. Defaults to 0. -e M --end=M End indexing at article M. This script is not very efficient with respect to memory management, and for large archives, it may not be possible to index the mbox entirely. For that reason, you can specify the start and end article numbers. However, using --end=[some #] or --start=[some #] results in these errors: % bin/arch --wipe --end=25000 test [...] option --end must not have an argument % bin/arch --wipe --start=25001 test [...] option --start must not have an argument Using --end [some #] or --start [some #] (ie with no =) works fine. So either the help output of bin/arch needs to be amended to remove the ='s, or the argument handling needs to be tweaked to understand ='s. I'm unsure which is preferable. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=671303&group_i…

1 0

[ mailman-Bugs-671303 ] bin/arch --start and --end argument handling
by SourceForge.net 21 Jan '03

21 Jan '03

Bugs item #671303, was opened at 2003-01-20 13:52 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=671303&group_i… Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Bryan Fullerton (fehwalker) Assigned to: Nobody/Anonymous (nobody) Summary: bin/arch --start and --end argument handling Initial Comment: bin/arch says (among other things): -s N --start=N Start indexing at article N, where article 0 is the first in the mbox. Defaults to 0. -e M --end=M End indexing at article M. This script is not very efficient with respect to memory management, and for large archives, it may not be possible to index the mbox entirely. For that reason, you can specify the start and end article numbers. However, using --end=[some #] or --start=[some #] results in these errors: % bin/arch --wipe --end=25000 test [...] option --end must not have an argument % bin/arch --wipe --start=25001 test [...] option --start must not have an argument Using --end [some #] or --start [some #] (ie with no =) works fine. So either the help output of bin/arch needs to be amended to remove the ='s, or the argument handling needs to be tweaked to understand ='s. I'm unsure which is preferable. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=671303&group_i…

1 0

[ mailman-Patches-671300 ] allow more than one user group to be specified
by SourceForge.net 21 Jan '03

21 Jan '03

Patches item #671300, was opened at 2003-01-20 13:44 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671300&group_i… Category: cross platform Group: Mailman 2.1 Status: Open Resolution: None Priority: 5 Submitted By: John Dennis (johndennis) Assigned to: Nobody/Anonymous (nobody) Summary: allow more than one user group to be specified Initial Comment: Mailman security is in part enforced by requiring it execute SGID. When the mail process or the web server attempts to execute a mailman script a C program is invoked to verify the group permission. Mailman as it is shipped only allows one group to be specified at build time. For users who build and install on their own machine this is not a limitation. However, when making a binary package to be installed on an arbitrary machine it is hard to predict the correct group to use for that installation. Therefore this patch allows us to specify at build time a list of groups that will be iterated over, if the mailman process is executing as any of one of the group in the set of groups then the permission check passes. Since the groups we build with are limited to a small number of safe groups this does not lower the security much while at the same time provides a much more friendly way to package a binary installation that will run in a wider range of installations. It was necessary to add the macro MM_FIND_GROUP_LIST to the configure.in file replacing the original use of MM_FIND_GROUP_NAME, the former operates on a list of group names while the later on a single name. MM_FIND_GROUP_LIST includes a filter parameter that was added with the notion of supporting the with-permcheck option. If filter is true then only group names that exist on the build machine are permitted in the list, otherwise all names are permitted. However, note that whenever MM_FIND_GROUP_LIST is invoked it is currently hardcoded to disable filtering and is not tied to with-permcheck, this was done because of the observation that if one is passing a list of groups it is likely one is doing so to support installations that have a group not present on the build machine, but one might still want to take advantage of the other with-permcheck functionality. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671300&group_i…

1 0

[ mailman-Patches-671297 ] set directory permissions, partition install steps
by SourceForge.net 21 Jan '03

21 Jan '03

Patches item #671297, was opened at 2003-01-20 13:34 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671297&group_i… Category: configure/install Group: Mailman 2.1 Status: Open Resolution: None Priority: 5 Submitted By: John Dennis (johndennis) Assigned to: Nobody/Anonymous (nobody) Summary: set directory permissions, partition install steps Initial Comment: This patch accomplishes two things: 1) A number of the directories created during make installed had failed to set their permissions correctly, this caused the script check_perms to report errors, this adds the setting of correct permissions to those directories. 2) Most packaging/install tools divide the installation task into 2 distinct phases File Marshaling: This is where the files of the package are built and installed into a temporary root. The the tool collects all the files under the temporary root and adds them to some type of archive file. During installation on the target machine these files are extracted from the archive and placed in matching positions in the target file system with matching ownership and permissions. Target Update: After the files are placed on the target it is often necessary to run commands on the target system to do such things as register the files, register services, etc. The "make install" in the shipped version of mailman did not separate out these two steps. In particular it attempted to compile the python files during the first phase of file marshaling. During this phase the python files are in a temporary root directory, not their final install position on the target. The paths.py file which is included by many of the python files has hard coded in it the the target installation directory (from the configure step). Thus when attempting to compile the files in the marshaling area the compilations would fail because it could not located patch information. Also the make install attempt to run the bin/update, but once again this has to be deferred to target update time, not package build time. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671297&group_i…

1 0

[ mailman-Patches-671294 ] syntax error in tests/fblast.py
by SourceForge.net 21 Jan '03

21 Jan '03

Patches item #671294, was opened at 2003-01-20 13:30 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671294&group_i… Category: command line scripts Group: Mailman 2.1 Status: Open Resolution: None Priority: 5 Submitted By: John Dennis (johndennis) Assigned to: Nobody/Anonymous (nobody) Summary: syntax error in tests/fblast.py Initial Comment: The tests/fblast.py script in version 2.1 had a syntax error due to a missing close paren. diff -r -u mailman-2.1.orig/tests/fblast.py mailman-2.1/tests/fblast.py --- mailman-2.1.orig/tests/fblast.py 2002-03-13 00:59:40.000000000 -0500 +++ mailman-2.1/tests/fblast.py 2003-01-14 16:46:57.000000000 -0500 @@ -54,7 +54,7 @@ """ % {'num' : i, 'FROMADDR': FROMADDR, 'LISTADDR': LISTADDR, - } + }) time.sleep(snooze) finally: conn.quit() ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671294&group_i…

1 0

[ mailman-Patches-671283 ] make mailmanctl script compatible with RedHat chkconfig
by SourceForge.net 20 Jan '03

20 Jan '03

Patches item #671283, was opened at 2003-01-20 13:05 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671283&group_i… Category: configure/install Group: Mailman 2.1 >Status: Closed >Resolution: Rejected Priority: 5 Submitted By: John Dennis (johndennis) Assigned to: Nobody/Anonymous (nobody) Summary: make mailmanctl script compatible with RedHat chkconfig Initial Comment: For the mailman services to run across boots it is necessary to start the mailman queue daemon as part of init via the rc.d scripts. The tool to manage the rc.d directories is called chkconfig. The RedHat version demands that the script contain chkconfig: line listing the run level and start/stop priorities, plus a description: entry. If these two headers are not present chkconfig will refuse to manange the script. ---------------------------------------------------------------------- >Comment By: Barry A. Warsaw (bwarsaw) Date: 2003-01-20 13:09 Message: Logged In: YES user_id=12800 mailmanctl isn't the tool for this. Please see misc/mailman script, which already has the chkconfig lines. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671283&group_i…

1 0

[ mailman-Patches-671283 ] make mailmanctl script compatible with RedHat chkconfig
by SourceForge.net 20 Jan '03

20 Jan '03

Patches item #671283, was opened at 2003-01-20 13:05 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671283&group_i… Category: configure/install Group: Mailman 2.1 Status: Open Resolution: None Priority: 5 Submitted By: John Dennis (johndennis) Assigned to: Nobody/Anonymous (nobody) Summary: make mailmanctl script compatible with RedHat chkconfig Initial Comment: For the mailman services to run across boots it is necessary to start the mailman queue daemon as part of init via the rc.d scripts. The tool to manage the rc.d directories is called chkconfig. The RedHat version demands that the script contain chkconfig: line listing the run level and start/stop priorities, plus a description: entry. If these two headers are not present chkconfig will refuse to manange the script. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=300103&aid=671283&group_i…

1 0

[ mailman-Bugs-671239 ] no size shown by held messages
by SourceForge.net 20 Jan '03

20 Jan '03

Bugs item #671239, was opened at 2003-01-20 17:30 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=671239&group_i… Category: Web/CGI Group: 2.1 (stable) Status: Open Resolution: None Priority: 2 Submitted By: Peer Heinlein (pheinlein) Assigned to: Nobody/Anonymous (nobody) Summary: no size shown by held messages Initial Comment: When viewing details and body of a single held message mailman doesn`t list the size of the held message. -But mailman does that in the general overview of held messages. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=100103&aid=671239&group_i…

1 0