Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Alan McConnell
On Thu, Dec 07, 2006 at 06:23:40PM -0600, Brad Knowles wrote:
 At 4:57 PM -0500 12/7/06, Alan McConnell wrote:
 
  Meanwhile, I am adminning(sp?), through my ISP, a new but quite active
  E-list.  But their mailman install is incomplete; they haven't put in
  Pipermail(about which I know _nothing_).
 
 Uh, what version of Mailman is that?  I thought that Mailman had 
 fully integrated Pipermail along with the base code, for many years 
 now?  Are they running Mailman 1.x or something?
mm 2.1.5 .  But under Debian, so it has experienced/endured the
Debian security upgrade procedures.

  and massage these messages and then ship them off to the new very
  skilful tech staff that my ISP is allegedly hiring, and they will
  be able to slip this collection adroitly into place.  And it will
  be as if archiving was always in place . . .
 
 So long as you save the messages in mbox format, and you have access 
 to the command-line, you can always use the arch tool to import the 
 old messages into the new archives.  The instructions for doing this 
 kind of stuff are in the FAQ.
 
 But if you don't have direct command-line access to the server, then 
 you'll be dependant on the staff that do.  And from what you've 
 described, that's really the crux of the problem you're already faced 
 with.
You are exactly right.  It is, like so many other thingsg, a
political problem.  PatriotNet is a small ISP; its founder died,
the widow sold to another, totally incompetent, company . . . but
now it has passed to yet another company, which is known to have
Linux/Unix competent people, and which makes good noises in re
tech support.

Since you say mbox format and since my MUA is mutt, set up to use mbox,
I think/hope/believe I am OK . . .  but time will tell.

Many thanks to you and to the others who have chimed in.  This E-list is
wonderful; run the way a help list should be IMHO.

Best wishes,

Alan

-- 
Alan McConnell :  http://patriot.net/users/alan
There are many good Impeachment sites; one of the best is:
   www.waifllc.org
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


[Mailman-Users] virtual domains + postfix

2006-12-08 Thread Martin Marcher
Hello,

trying to setup mailman with virtual domains and postfix, no matter  
what I tried I get No such mailbox, I postfix I set up the  
alias_maps to hold the generated aliases that pipe to mailman and  
virtual_alias_maps points to the file that holds the fully qualified  
list addresses which in turn point to the aliases defined in the  
alias_maps.

I really have no Idea what to do to make it work, I tried several  
howtos from the web but none the approaches seem to work out

I'll provide any informations you need to help me analyze this but  
since I have no idea what's really going on here anymore I figured I  
wait what you say you need...

\martin

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Brad Knowles
At 7:52 AM -0500 12/8/06, Alan McConnell quoted me:

  Uh, what version of Mailman is that?  I thought that Mailman had
  fully integrated Pipermail along with the base code, for many years
  now?  Are they running Mailman 1.x or something?

   mm 2.1.5 .  But under Debian, so it has experienced/endured the
   Debian security upgrade procedures.

Okay, now that is one of the most bizarre things I've heard of in a 
very long time.  I cannot comprehend how they could possibly ship a 
version of Mailman 2.1.x that does not automatically include the 
bundled Pipermail component.

Congratulations -- I've been in this business long enough that I 
don't get too many surprises like this anymore.  Yet, Mailman has 
recently provided me these kinds of surprises twice in the last 
couple of weeks or so.

There should definitely be some sort of prize that you win as a 
result of your experience.

   You are exactly right.  It is, like so many other thingsg, a
   political problem.  PatriotNet is a small ISP; its founder died,
   the widow sold to another, totally incompetent, company . . . but
   now it has passed to yet another company, which is known to have
   Linux/Unix competent people, and which makes good noises in re
   tech support.

Ahh.  PatriotNet.  I had heard good things about them years ago, but 
it's been a while since I've seen that name.

Interesting to see what has since happened to that company.

  Since you say mbox format and since my MUA is mutt, set up to use mbox,
  I think/hope/believe I am OK . . .  but time will tell.

In terms of the mbox format itself, you should be fine.

I'm seriously thinking of switching back to mutt myself (after being 
away for so many years), because Qualcomm/Eudora have decided to kill 
the product as a separate program and the next major version is going 
to be based on the Thunderbird platform.  Of course, if I wanted 
Thunderbird, I'd be using Thunderbird.  I have over 6GB of archives 
in Eudora's slightly modified version of mbox archives, and I'm not 
looking forward to what I'm going to have to do in order to move to a 
different program.  I may take a look at what happens with Eudora 
after the change, I might go ahead and switch to real Thunderbird, 
or I might switch to a different program altogether -- which would 
almost certainly be mutt.  I haven't decided yet.

But regardless of what happens to me, you should definitely be fine with mutt.


I think your issue is going to be the level of support that you get 
from the people who have yet to be brought in to help manage the 
systems, and whether or not they have the level of clue and talent 
required to be able to properly import your mailboxes for the 
archives.

As far as that goes, about all I can do is to wish you luck, and then 
we all wait to see what happens.

  Many thanks to you and to the others who have chimed in.  This E-list is
  wonderful; run the way a help list should be IMHO.

I'm pretty much always glad to help, if I can -- as permitted by time 
and available resources.

-- 
Brad Knowles, [EMAIL PROTECTED]

Trend Micro has announced that they will cancel the stop.mail-abuse.org
mail forwarding service as of 15 November 2006.  If you have an old
e-mail account for me at this domain, please make sure you correct that
with the current address.
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] removing members with non-standard characters

2006-12-08 Thread Anne Ramey


Patrick Bogen wrote:
 On 12/6/06, Anne Ramey [EMAIL PROTECTED] wrote:
 I'm not sure how one of my list admins managed to do this, but they have
 four members on their list that look like this in the membership list:
 b
 http://lists.ncmail.net/mailman/options/nciin-network-ops/%00b%00r%00i%00a%00n%00v%00--at--%00n%00c%00c%00c%00s%00.%00c%00c%00.%00n%00c%00.%00u%00s%00%00%00%00%00%1F%00%00%00%01%00%00%00%00%00%00%00%03%00%00%00
  


 See FAQ 3.13. How do I remove a user name or email address with an
 illegal character in it?
 http://www.python.org/cgi-bin/faqw-mm.py?req=showfile=faq03.013.htp

I should have mentioned that will not work for this case.  You'll notice 
that these are not all nice ascii characters.  some are spaces, some 
deletes, some other hex values...I don't know what they all are because 
they will not copy and past nicely.  They don't appear at all when I do 
list_members.  Any other ideas
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Christopher X. Candreva
On Thu, 7 Dec 2006, Barry Warsaw wrote:

 Have I mentioned recently how long I've been looking for a volunteer  
 to help make all this not suck?  ;}  Pipermail is just one of those  
 things that people either live with or ditch.

I've used Hypermail for probably a decade to archive Majordomo lists.
It makes web archives from mbox files natively.


==
Chris Candreva  -- [EMAIL PROTECTED] -- (914) 967-7816
WestNet Internet Services of Westchester
http://www.westnet.com/
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Email arrives after a long delay

2006-12-08 Thread BERTHOLD Jean
Hello Brad,

First, many thanks for your long, useful and complete response (whaou...)
I'm not a email specialist but when I seen the message:

 [ID 801593 mail.warning] kB6BcXf1016562: collect: premature EOM:
  unexpected close )

I Think you are true, there is a problem on our Exchange server...

The second problem is now to convince our Exchange administrator... 
But I think your response will give them a good idea about the problem... :-)

Thanks again for your help and have a nice week-end

Jean


-Message d'origine-
De : Brad Knowles [mailto:[EMAIL PROTECTED] 
Envoyé : jeudi, 7. décembre 2006 17:52
À : BERTHOLD Jean; mailman-users@python.org
Cc : BERTHOLD Jean
Objet : Re: [Mailman-Users] Email arrives after a long delay

At 9:34 AM +0100 12/7/06, BERTHOLD Jean wrote:

  I have a configuration problem.
  All my lists works correctly excepted that the emails arrives with
  a delay of 10 minutes on the mailman server.

In the world of Internet e-mail, ten minutes is nothing.  In fact, 
that's pretty damn fast.

  Dec 06 12:49:49 2006 (4144) post to fpbg-dtr from
  [EMAIL PROTECTED], size=1880,
  message-id=[EMAIL PROTECTED],
  success

  Dec 06 13:08:18 2006 (4144) post to fpbg-dtr from
  [EMAIL PROTECTED], size=1875,
  message-id=[EMAIL PROTECTED],
  success

Okay, so here we have two different messages coming into Mailman from 
you.  So that we can make it easy to identify them, let's call them 
630 and 634, which comes from the last three digits of the 
Message-ID field before the at-symbol.

  Dec 06 12:49:49 2006 (4144)
  [EMAIL PROTECTED] smtp to
  fpbg-dtr for 1 recips, completed in 0.318 seconds

Okay, so we see that 630 came into Mailman from the MTA at 12:49:49 
and went back out to the MTA in less than a second.  That's good.

  Dec  6 12:48:46 bahamas sendmail[16562]: [ID 801593 mail.warning]
  kB6BcXf1016562: collect: premature EOM: unexpected close
  Dec  6 12:48:46 bahamas sendmail[16562]: [ID 801593 mail.notice]
  kB6BcXf1016562: collect: unexpected close on connection from
  swexch01lpro.sila.local, sender=[EMAIL PROTECTED]
  Dec  6 12:48:46 bahamas sendmail[16562]: [ID 801593 mail.info]
  kB6BcXf1016562: from=[EMAIL PROTECTED], size=1217,
  class=0, nrcpts=1, proto=ESMTP, daemon=MTA-v4,
  relay=swexch01lpro.sila.local [172.25.2.76]

These are signs that your Exchange server is screwing up the SMTP protocol.

  Dec  6 12:49:46 bahamas sendmail[16721]: [ID 801593 mail.info]
  kB6BnkRQ016721: from=[EMAIL PROTECTED], size=1315,
  class=0, nrcpts=1,
  msgid=[EMAIL PROTECTED],
  proto=ESMTP, daemon=MTA-v4, relay=swexch01lpro.sila.local [172.25.2.76]

Okay, so here is 630 coming into the system from Exchange.  Note that 
the sendmail queue-id in this case is kB6BnkRQ016721.  This is how we 
tie other log entries together back to this same message -- by the 
sendmail queue-id.

  Dec  6 12:49:47 bahamas sendmail[16722]: [ID 801593 mail.info]
  kB6BnkRQ016721: to=|/usr/local/mailman/mail/mailman post fpbg-dtr,
  ctladdr=[EMAIL PROTECTED] (1/0), delay=00:00:01,
  xdelay=00:00:01, mailer=prog, pri=31536, dsn=2.0.0, stat=Sent

Okay, so here is sendmail saying that it has now sent 630 to Mailman. 
We know that it's what we're calling message 630 because we see the 
same sendmail queue-id, namely kB6BnkRQ016721.

Note that Mailman says that it finished getting this message about 
two seconds later (at 12:49:49), and that Mailman then sends that 
message back to the MTA in less than a second (still 12:49:49).

  Dec  6 12:49:49 bahamas sendmail[16725]: [ID 801593 mail.info]
  kB6Bnnn6016725: from=[EMAIL PROTECTED],
  size=1880, class=-30, nrcpts=1,
  msgid=[EMAIL PROTECTED],
  proto=ESMTP, daemon=MTA-v4, relay=localhost [127.0.0.1]

Okay, so here is 630 having come back out of Mailman, and going into 
the MTA to be delivered to the recipient.  Note that the sendmail 
queue-id is now kB6Bnnn6016725, because as far as sendmail is 
concerned this is a totally different message.  Sendmail doesn't know 
that the content is exactly the same, and that what has happened is 
that the message has come through sendmail, into Mailman, and then 
back out again.  Sendmail sees the inbound and outbound legs as being 
two totally separate messages.

Any further delay is totally and completely out of the hands of 
Mailman.  There is absolutely nothing that we can do to help.

  Dec  6 13:08:20 bahamas sendmail[17179]: [ID 801593 mail.info]
  kB6C8Kge017177: to=[EMAIL PROTECTED], delay=00:00:00,
  xdelay=00:00:00, mailer=esmtp, pri=176820, relay=mta0.eosholding.ch.
  [193.8.222.23], dsn=2.0.0, stat=Sent (Ok: queued as 360A520C03B)

Dunno what message this is, but it doesn't look like it's 630.  Note 
that the sendmail queue-id is different -- Message 630 came in with 
queue-id kB6Bnnn6016725, and this has queue-id kB6C8Kge017177.  So, 
it looks like there are some log entries missing.

  Dec  6 13:08:21 bahamas sendmail[17180]: [ID 801593 mail.info]
  kB6C8LhF017180: from=[EMAIL PROTECTED], size=2777,
  

Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Todd Zullinger
Mark Sapiro wrote:
 Paul Tomblin wrote:
 
Quoting Barry Warsaw ([EMAIL PROTECTED]):
 
 It already does escape From lines in the body of the message.  It
 does this by way of the email package's Generator class, which is
 instantiated with mangle_from_=True.

Must be a newer version than the one in Debian stable.   I grepped
for mangle in /var/lib/mailman/Mailman/*, and didn't find it.  The
parameter does appear in /usr/lib/python2.3/email/Generator.py, but
since I don't know python I don't know how to pass it to it.  I'm
guessing it has something to do with changing the g =
Generator(fp) and g = Generator(outfp) lines in
Mailman/ListAdmin.py or more likely the g = Generator(self.fp)
line in Mailman/Mailbox.py?  Is it as simple as changing that last
one to g = Generator(self.fp,mangle_from_=True)?
 
 
 Yes it is.
 
 And I think Barry may have misspoken as I don't think that change is
 in the SVN trunk or Release_2_1-maint branch.

According to the docs for email.Generator, mangle_from defaults to
True and has since at least python 2.3.  So setting it shouldn't be
needed AFAICS.

In that case, shouldn't any message that reaches mailman with an
unescaped From_ line in the body already be handled properly?  It
seems like something else must be borked.  That or all of the messages
in a list mbox that contain unescaped From_ lines got there from
really old versions of Mailman/python.

-- 
ToddOpenPGP - KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp
==
You know an odd feeling?  Sitting on the toilet eating a chocolate
candy bar.
-- George Carlin, Napalm  Silly Putty



pgp0YagKhc7uk.pgp
Description: PGP signature
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp

Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Paul Tomblin
Quoting Todd Zullinger ([EMAIL PROTECTED]):
 Mark Sapiro wrote:
  Paul Tomblin wrote:
 Quoting Barry Warsaw ([EMAIL PROTECTED]):
  It already does escape From lines in the body of the message.  It
  does this by way of the email package's Generator class, which is
  instantiated with mangle_from_=True.
 
 According to the docs for email.Generator, mangle_from defaults to
 True and has since at least python 2.3.  So setting it shouldn't be
 needed AFAICS.

Oh right, now I get it.  I told you I didn't know Python, but I should have
been able to figure that 
def __init__(self, outfp, mangle_from_=True, maxheaderlen=78):
meant that it defaulted to True in Generator.py so didn't need to be
changed in Mailman.

 
 In that case, shouldn't any message that reaches mailman with an
 unescaped From_ line in the body already be handled properly?  It
 seems like something else must be borked.  That or all of the messages
 in a list mbox that contain unescaped From_ lines got there from
 really old versions of Mailman/python.

That is distinctly possible.  The archives in question go back to 1998.  I
didn't keep track of when the *last* unescaped From_ line was put in the
archives.



-- 
Paul Tomblin [EMAIL PROTECTED] http://blog.xcski.com/
It could have been raining flaming bulldozers, and those idiots would have
been standing out there smoking, going 'hey, look at that John Deere burn!'
  -- Texan AMD security guard
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread John A. Martin
 Brad == Brad Knowles
 Re: [Mailman-Users] Mailman archive messages(not rm, but install!)
  Fri, 8 Dec 2006 08:52:26 -0600

Brad At 7:52 AM -0500 12/8/06, Alan McConnell quoted me:
 Uh, what version of Mailman is that?  I thought that Mailman
 had fully integrated Pipermail along with the base code, for
 many years now?  Are they running Mailman 1.x or something?

 mm 2.1.5 .  But under Debian, so it has experienced/endured the
 Debian security upgrade procedures.

Brad Okay, now that is one of the most bizarre things I've heard
Brad of in a very long time.  I cannot comprehend how they could
Brad possibly ship a version of Mailman 2.1.x that does not
Brad automatically include the bundled Pipermail component.

Brad Congratulations -- I've been in this business long enough
Brad that I don't get too many surprises like this anymore.  Yet,
Brad Mailman has recently provided me these kinds of surprises
Brad twice in the last couple of weeks or so.

Brad There should definitely be some sort of prize that you win
Brad as a result of your experience.

See http://packages.debian.org/unstable/mail/mailman for a
description of the Debian Mailman package that integrates
... archiving   Further down that page under the heading
Download mailman click on one of the list of files buttons and see
among other things: usr/lib/mailman/Mailman/Archiver/pipermail.py

jam



pgpMEZkVN4gX6.pgp
Description: PGP signature
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp

Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Todd Zullinger
Paul Tomblin wrote:
 Oh right, now I get it.  I told you I didn't know Python, but I
 should have been able to figure that
 def __init__(self, outfp, mangle_from_=True, maxheaderlen=78):
 meant that it defaulted to True in Generator.py so didn't need to be
 changed in Mailman.

:-)

I actually trusted the docs (which I probably shouldn't do).  But
you're right, the __init__ line confirms the default setting for
mangle_from.

 In that case, shouldn't any message that reaches mailman with an
 unescaped From_ line in the body already be handled properly?  It
 seems like something else must be borked.  That or all of the
 messages in a list mbox that contain unescaped From_ lines got
 there from really old versions of Mailman/python.
 
 That is distinctly possible.  The archives in question go back to
 1998.  I didn't keep track of when the *last* unescaped From_ line
 was put in the archives.

It still sucks that you have to muck around with the mboxes to get
pipermail to archive them properly.  I feel your pain.  I am currently
trying to get another lists archives into better shape and have run
into the same issue.

I'm seriously lacking in the skill to fix pipermail or I'd take Barry
up on trying to fix it.

-- 
ToddOpenPGP - KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp
==
An idea is not responsible for the people who believe in it.
-- Anonymous



pgpKLGsxj1cPw.pgp
Description: PGP signature
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp

[Mailman-Users] A scrubber issue

2006-12-08 Thread Todd Zullinger
In honor of recent discussions on pipermail, I think I've found
another issue with archiving, though this seems to be in
Mailman.Scrubber.

In a few recent posts to the GnuPG lists, Werner Koch sent along some
signed patches fixing issues in the gpg code.  Unfortunately, the
archives ate his posts[1] so we can't point others to the patches in the
archives as nicely as one would like.

It seems that the problem is that both message parts lack a
Content-type: text/plain; charset=us-ascii header and the first part
also lacks a Content-disposition: inline header.  If I edit the raw
message to include a Content-type: text/plain; charset=us-ascii
header for each mime part, it passes the scrubber as is archived
properly.

From my limited reading of RFC 2045[2], it seems that a mime part
without a content-type header should be assumed to be text/plain;
charset=us-ascii.  Is the scrubber wrong to not assume this or are
there too many issues with making this (apparently quite standards
conformant) assumption?

[1] http://lists.gnupg.org/pipermail/gnupg-users/2006-December/029976.html
[2] http://www.ietf.org/rfc/rfc2045.txt

-- 
ToddOpenPGP - KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp
==
To know what is right and not to do it is the worst cowardice.
-- Confucius



pgpbumhTkm81V.pgp
Description: PGP signature
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp

Re: [Mailman-Users] removing members with non-standard characters

2006-12-08 Thread Mark Sapiro
Anne Ramey wrote:

I should have mentioned that will not work for this case.  You'll notice 
that these are not all nice ascii characters.  some are spaces, some 
deletes, some other hex values...I don't know what they all are because 
they will not copy and past nicely.  They don't appear at all when I do 
list_members.  Any other ideas


There are several options and I have lots of ideas.

Have you tried just checking the 'unsub' box next to the entry?

The 'address' you see looks a lot like the URL of the options page that
the address is a link to. Perhaps you are not seeing the address at
all, but rather, you are seeing the result of funny characters in the
address confusing the browser's rendering of the anchor tag. It looks
like
http://lists.ncmail.net/mailman/options/nciin-network-ops/%00b%00r%00i%00a%00n%00v%00--at--%00n%00c%00c%00c%00s%00.%00c%00c%00.%00n%00c%00.%00u%00s%00%00%00%00%00%1F%00%00%00%01%00%00%00%00%00%00%00%03%00%00%00
is a link to the options page of the user whose address is

%00b%00r%00i%00a%00n%00v%00--at--%00n%00c%00c%00c%00s%00.%00c%00c%00.%00n%00c%00.%00u%00s%00%00%00%00%00%1F%00%00%00%01%00%00%00%00%00%00%00%03%00%00%00

(if you replace --at-- with @, %00 with ^@, %1F with ^_, %01 with ^A
and %03 with ^C)

Do you see [EMAIL PROTECTED] on the list_members output? Does this
entry appear in the membership list on its own page at the beginning?

It also looks like someone mass subscribed a list pasted from or output
by a word processor

If the bad addresses don't appear in list_members (I don't know why
they wouldn't, but maybe they just appear with the control characters
that you don't see and thus look OK) you can do

  bin/list_members listname | bin/synch_members -f - -n listname

and if you like what that says, you can do

  bin/list_members listname | bin/synch_members -f - listname

You might also try

  bin/list_members -i listname

to see what that shows.

You could create a simple withlist script to validate member addresses
and delete invalid ones..

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Barry Warsaw
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Dec 8, 2006, at 11:10 AM, Paul Tomblin wrote:

 In that case, shouldn't any message that reaches mailman with an
 unescaped From_ line in the body already be handled properly?  It
 seems like something else must be borked.  That or all of the  
 messages
 in a list mbox that contain unescaped From_ lines got there from
 really old versions of Mailman/python.

 That is distinctly possible.  The archives in question go back to  
 1998.  I
 didn't keep track of when the *last* unescaped From_ line was put  
 in the
 archives.

Sorry, I should have been clearer that the /default/ behavior of the  
generator is to mangle From_ lines.  So it's true that nothing in  
Mailman should need to be changed.

However, it's also true that in the distant past, there were some  
bugs in the mbox implementation which would cause broken mbox files  
to be written.  A quick scan through the svn logs jogs my memory:  
r6341 on 2003-04-17 was added to fix a message separation bug.  I  
don't know how long that bug was lurking, but the fix puts it just  
before the 2.1.2 release according to the NEWS file.  I'll bet that  
it existed from 2.1 final (Dec 2002) until 2.1.2 (Apr 2003), the  
latter which was probably released specifically to fix this problem!

Note that this bug had no effect on the archiving of new messages on  
the fly.  Those always got archived correctly.  But the message was  
appended to the mbox file incorrectly which meant that if you  
regenerated your archives, you'd be screwed.  This was what bin/ 
cleanarch was intended to fix.

BTW, one less ambitious way to participate here to help fix things  
would be to improve bin/cleanarch.  At the very least, you should be  
able to run that script and get an mbox file that bin/arch can use to  
DTRT.  It would also be nice if bin/arch was able to compensate for  
running out of memory, possibly by changing it to fork a sub-process  
to do the actual archiving with the parent process pre-chunking the  
workload for the child.

Anyway, I'm cc'ing mailman-developers.  Further discussion of how to  
improve matters should be conducted on that list (and mailman-users  
should be removed).

- -Barry

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (Darwin)

iQCVAwUBRXmlMHEjvBPtnXfVAQIFYAP/W2LEOrKhqrB6sDniHKADAV5iMuLm19zu
nUkvrJpOumD78+tRDa1DCQG8RaCSAZ7bNkTA2VwIUgcX1I4+9d7ylklonQSiRJzB
xbg+OBD5+x5q+Cdo9qX1dhlGWTdmrSReN0CLRx6408JX8qtXhIh+3S0f3tG44bYE
lB76OX4HPXo=
=nhI8
-END PGP SIGNATURE-
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman stop delivering ... problem withApproval.py?

2006-12-08 Thread Mark Sapiro
[EMAIL PROTECTED] wrote:

The mailman lists on my server have suddenly stopped delivering mail. 


And what update to Mailman or Python was installed just before this
happened?


The error 
of one of the messages is cut and paste below ... All the messages are being 
shunted off and not delivered when it reaches the Approval.py and has trouble 
inporting ...? 

 File /usr/local/cpanel/3rdparty/mailman/Mailman/Handlers/Approve.py, line 
28, in ?
   from email.Iterators import typed_subpart_iterator
ImportError: cannot import name typed_subpart_iterator


The above import has been in Approve.py since 2.1.0. Something else in
your system changed.

See
http://www.python.org/cgi-bin/faqw-mm.py?req=showfile=faq06.011.htp
for the reason why I can't be more helpful.

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Mark Sapiro
Brad Knowles wrote:

At 7:52 AM -0500 12/8/06, Alan McConnell quoted me:

  Uh, what version of Mailman is that?  I thought that Mailman had
  fully integrated Pipermail along with the base code, for many years
  now?  Are they running Mailman 1.x or something?

  mm 2.1.5 .  But under Debian, so it has experienced/endured the
  Debian security upgrade procedures.

Okay, now that is one of the most bizarre things I've heard of in a 
very long time.  I cannot comprehend how they could possibly ship a 
version of Mailman 2.1.x that does not automatically include the 
bundled Pipermail component.


But the hosting service could set

  ARCHIVE_TO_MBOX = 1

in mm_cfg.py to disable pipermail archiving and archive only to the
.mbox file.

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman stop delivering ... problem withApproval.py?

2006-12-08 Thread SML
 The mailman lists on my server have suddenly stopped delivering
 mail. 
 
 And what update to Mailman or Python was installed just before this
 happened?

There was no update to either Mailman or Python ... but one of the
lists was hit with a *massive* amount of spam that caused too many
files open error, this same spam attack eventually caused the
server to run out of memory :(
 
 The above import has been in Approve.py since 2.1.0. Something else
 in your system changed.

When creating a temp list for a testing sandbox the Import errors were
replaced by AttributeError: 'str' object has no attribute
'get_sender' errors, and then all errors stopped and the lists on
the server started working again.

My assumption is that Mailman was using up server resources to deal
with spam and a (few) files got corrupted, which files
were then recompiled on the creation of a new list.

I'm now sending all mail to the list owner to dev/null and am
content filtering and discarding mail with images ... I'm hoping that
that will reduce the amount of resources that Mailman needs should
another load hit.

One interesting thing, the qfiles/shunt/*.pck files were owned:group
by mailman:mailman, but the most recent AttributeError .pck files
are owned:group root:mailman. How is it that Mailman started writing
files with root as the owner? 

 See
 http://www.python.org/cgi-bin/faqw-mm.py?req=showfile=faq06.011.htp
 for the reason why I can't be more helpful.

Yeah. I appreciate the time you've been able to give.  Thanks.

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] A scrubber issue

2006-12-08 Thread Mark Sapiro
Todd Zullinger wrote:

In honor of recent discussions on pipermail, I think I've found
another issue with archiving, though this seems to be in
Mailman.Scrubber.

In a few recent posts to the GnuPG lists, Werner Koch sent along some
signed patches fixing issues in the gpg code.  Unfortunately, the
archives ate his posts[1] so we can't point others to the patches in the
archives as nicely as one would like.


There seems to be another issue in that there is no 'link' to the
scrubbed part, just a relative URL which doesn't work.


It seems that the problem is that both message parts lack a
Content-type: text/plain; charset=us-ascii header and the first part
also lacks a Content-disposition: inline header.  If I edit the raw
message to include a Content-type: text/plain; charset=us-ascii
header for each mime part, it passes the scrubber as is archived
properly.

From my limited reading of RFC 2045[2], it seems that a mime part
without a content-type header should be assumed to be text/plain;
charset=us-ascii.  Is the scrubber wrong to not assume this or are
there too many issues with making this (apparently quite standards
conformant) assumption?

[1] http://lists.gnupg.org/pipermail/gnupg-users/2006-December/029976.html
[2] http://www.ietf.org/rfc/rfc2045.txt


You are correct in your assessment of why the part is scrubbed in the
first place. Also, it seems that according to sec 5.2 of RFC 2045 we
could assume 'us-ascii', but I expect this may cause other problems
with non-compliant MUAs. Tokio is responsible for a lot of scrubber
code, and he has a lot of experience with Japanese so I suspect there
is a reason we do it this way.

It seems that a good part of the problem in the above referenced
archive is that the scrubbed attachment is not given a clickable link,
and in fact the relative path given doesn't even work. I think at
least part of this must be specific to this site - perhaps a
(intentionally?) bad value for PUBLIC_ARCHIVE_URL in mm_cfg.py.

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman stop delivering ... problemwithApproval.py?

2006-12-08 Thread Mark Sapiro
SML wrote:

There was no update to either Mailman or Python ... but one of the
lists was hit with a *massive* amount of spam that caused too many
files open error, this same spam attack eventually caused the
server to run out of memory :(


And the reboot of the server could have caused things which were
previously downloaded and waiting for a reboot to install to finally
be installed.


 The above import has been in Approve.py since 2.1.0. Something else
 in your system changed.

When creating a temp list for a testing sandbox the Import errors were
replaced by AttributeError: 'str' object has no attribute
'get_sender' errors, and then all errors stopped and the lists on
the server started working again.


I've seen reports of this error before and it appears to me to be a
symptom of an underlying Python problem that causes incoming queue
entries to be missing critical metadata.


My assumption is that Mailman was using up server resources to deal
with spam and a (few) files got corrupted, which files
were then recompiled on the creation of a new list.


Creating a new list shouldn't cause anything to be recompiled or
reloaded.


One interesting thing, the qfiles/shunt/*.pck files were owned:group
by mailman:mailman, but the most recent AttributeError .pck files
are owned:group root:mailman. How is it that Mailman started writing
files with root as the owner? 


Because it was root that ran the 'bin/mailmanctl start' so root owns
the qrunner processes.

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] A scrubber issue

2006-12-08 Thread Todd Zullinger
Hi Mark,

Mark Sapiro wrote:
 There seems to be another issue in that there is no 'link' to the
 scrubbed part, just a relative URL which doesn't work.

Yes, there are other issues with the configuration there.  Click on
the list info link and then follow that to the archives.  You get into
the top level of the public archive dir, seeing a listing of all of
the lists and list mboxes.

However, I created a fresh list on my system to see whether this was a
list configuration issue or not and it's reproduceable using a default
list setup.  I initially thought it must be some over-agressive
content filtering, but after having it work on a virgin list I don't
think that's the case.  There may still be a way to configure the
content filtering to work around this, but I'm not familiar enough
with the various possibilities to know that.

 You are correct in your assessment of why the part is scrubbed in
 the first place. Also, it seems that according to sec 5.2 of RFC
 2045 we could assume 'us-ascii', but I expect this may cause other
 problems with non-compliant MUAs. Tokio is responsible for a lot of
 scrubber code, and he has a lot of experience with Japanese so I
 suspect there is a reason we do it this way.

I was guessing this might be the case as well.  Hopefully Tokio or
someone else who's intimate with the code can comment on whether
that's the case or not.  It always sucks when the choice is either not
conform to a standard or not work properly in the real world.  I don't
envy the choices you guys have to make when writing email parsing
code.  :)

If it came to it, perhaps there could be an option to be strict about
the parsing for those that would rather conform than be able to handle
all of the junk that people send?  I tend to go this direction when I
configure my MTA's, but I realize this isn't always a viable choice
for others.

 It seems that a good part of the problem in the above referenced
 archive is that the scrubbed attachment is not given a clickable
 link, and in fact the relative path given doesn't even work. I think
 at least part of this must be specific to this site - perhaps a
 (intentionally?) bad value for PUBLIC_ARCHIVE_URL in mm_cfg.py.

I don't think it's related.  My test list created a proper link to the
second message part, but it still scrubbed both mime parts.  If I
added a content-disposition: inline header to the first part, then it
was similarly scrubbed and a link inserted.  Without that header, the
part just disappeared completely.

I can send you config_list output for the test list if you like, but
there weren't any changes made to the config so it should be nothing
but Mailman default.  I don't know what OS the real gnupg-users list
runs on, but my test list was created on Fedora Core 6, using the
packaged mailman rpm there (version 2.1.9).  I don't think there are
significant deviations from Mailman's source other than the FHS patch
that they apply.  If you think it's relevant, I can install from
source and test as well.

Thanks,

-- 
ToddOpenPGP - KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp
==
It is not enough to be busy, so are the ants. The question is: what
are we busy about?
-- Henry David Thoreau



pgpn6IJzCLqa8.pgp
Description: PGP signature
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp

[Mailman-Users] problem with mailing lists general contact address

2006-12-08 Thread Henrik
All,

I am using mailman as installed and administered by a web host in 
vancouver, webserve.ca.

When I set up a mailing list, the general contact address on the 
mail.mail.domainname.ca/mailman/listinfo page is listed as 
[EMAIL PROTECTED], rather than [EMAIL PROTECTED]

When I send a test email to [EMAIL PROTECTED] I get a bounce 
back with this message:

This is an automatically generated Delivery Status Notification.

Delivery to the following recipients failed.

[EMAIL PROTECTED]


When I send a test email to [EMAIL PROTECTED] I get a bounce back 
with this message:

You are not allowed to post to this mailing list, and your message has
been automatically rejected.  If you think that your messages are
being rejected in error, contact the mailing list owner at [EMAIL PROTECTED]


I presume this is a configuration issue. Unfortunately I do not have 
access on the configuration server, and the webserve.ca support people 
seem to be at a loss as to what to do.

Judging from the fact that python's own mail list page uses 
[EMAIL PROTECTED], I presume that the issue is faulty configuration of 
the general list contact address in the configuration file.

Could someone explain to me how this can be fixed, so that I can pass it on?

Thanks much,

- Henrik

-- 

Henrik Bechmann
www.osscommons.ca
www.bechmannsoftware.com
Webmaster, www.dufferinpark.ca

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] A scrubber issue

2006-12-08 Thread Mark Sapiro
Todd Zullinger wrote:

However, I created a fresh list on my system to see whether this was a
list configuration issue or not and it's reproduceable using a default
list setup.  I initially thought it must be some over-agressive
content filtering, but after having it work on a virgin list I don't
think that's the case.  There may still be a way to configure the
content filtering to work around this, but I'm not familiar enough
with the various possibilities to know that.


It shouldn't be a content filtering issue. If a part is missing a
Content-Type: header, the message methods get_content_type() and
get_content_maintype() which are used by MimeDel.py (content
filtering) return the default types which are text/plain and text
except for subparts of multipart/digest when they are message/rfc822
and message.


 It seems that a good part of the problem in the above referenced
 archive is that the scrubbed attachment is not given a clickable
 link, and in fact the relative path given doesn't even work. I think
 at least part of this must be specific to this site - perhaps a
 (intentionally?) bad value for PUBLIC_ARCHIVE_URL in mm_cfg.py.

I don't think it's related.  My test list created a proper link to the
second message part, but it still scrubbed both mime parts.  If I
added a content-disposition: inline header to the first part, then it
was similarly scrubbed and a link inserted.  Without that header, the
part just disappeared completely.


I can't duplicate this. I am trying a multipart/mixed message with two
'no content-type' parts and a third 'Content-Type: text/plain;
charset=us-ascii' part. The first two parts are (individually)
either scrubbed and replaced with

 An embedded and charset-unspecified text was scrubbed...
 Name: not available
 Url: http://example.com/pipermail/list1/attachments/...

or left in the message body depending on whether or not they contain a
Content-Disposition: inline header.


I can send you config_list output for the test list if you like, but
there weren't any changes made to the config so it should be nothing
but Mailman default.  I don't know what OS the real gnupg-users list
runs on, but my test list was created on Fedora Core 6, using the
packaged mailman rpm there (version 2.1.9).  I don't think there are
significant deviations from Mailman's source other than the FHS patch
that they apply.  If you think it's relevant, I can install from
source and test as well.


Can you just send me the original message that has parts lost?

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] A scrubber issue

2006-12-08 Thread Todd Zullinger
Mark Sapiro wrote:
 It shouldn't be a content filtering issue. If a part is missing a
 Content-Type: header, the message methods get_content_type() and
 get_content_maintype() which are used by MimeDel.py (content
 filtering) return the default types which are text/plain and text
 except for subparts of multipart/digest when they are message/rfc822
 and message.

Okay, good to know.

 I can't duplicate this.

It's quite possible this is something that's tickled by the way Gnus
creates the message.  I know mutt doesn't create the sort of messages
that trigger this either.

 Can you just send me the original message that has parts lost?

Attached is the original message from the list mbox and one that I
munged up to included a content-type: text/plain; charset=us-ascii.

-- 
ToddOpenPGP - KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp
==
Women should be obscene and not heard.
-- Groucho Marx

From [EMAIL PROTECTED] Thu Dec 07 19:26:34 2006
Received: from kerckhoffs.g10code.com ([217.69.77.222])
by trithemius.gnupg.org with esmtp (Exim 4.50 #1 (Debian))
id 1GsNx4-0008G7-A7 for [EMAIL PROTECTED];
Thu, 07 Dec 2006 19:26:34 +0100
Received: from uucp by kerckhoffs.g10code.com with local-rmail (Exim 4.50 #1
(Debian)) id 1GsO7a-0001Qr-HL
for gnupg-users@gnupg.org; Thu, 07 Dec 2006 19:37:26 +0100
Received: from wk by localhost with local (Exim 4.62 #1 (Debian))
id 1GsNsD-0001gg-RX
for gnupg-users@gnupg.org; Thu, 07 Dec 2006 19:21:33 +0100
From: Werner Koch [EMAIL PROTECTED]
To: gnupg-users@gnupg.org
Subject: Signed patch against 2.0.1
Organisation: g10 Code GmbH
OpenPGP: id=5B0358A2; url=finger:[EMAIL PROTECTED]
Mail-Followup-To: gnupg-users@gnupg.org
Date: Thu, 07 Dec 2006 19:21:33 +0100
Message-ID: [EMAIL PROTECTED]
User-Agent: Gnus/5.110006 (No Gnus v0.6)
MIME-Version: 1.0
Content-Type: multipart/mixed;

boundary==KGB-Sundevil-gamma-Skipjack-government-Vince-Foster-Treasury-bce-S-B
X-Spam-Checker-Version: SpamAssassin 3.0.3 (2005-04-27) on 
trithemius.gnupg.org
X-Spam-Level: 
X-Spam-Status: No, score=0.0 required=5.0 tests=AWL autolearn=ham 
version=3.0.3
X-BeenThere: gnupg-users@gnupg.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: Help and discussion among users of GnuPG gnupg-users.gnupg.org
List-Unsubscribe: http://lists.gnupg.org/mailman/listinfo/gnupg-users,
mailto:[EMAIL PROTECTED]
List-Archive: /pipermail
List-Post: mailto:gnupg-users@gnupg.org
List-Help: mailto:[EMAIL PROTECTED]
List-Subscribe: http://lists.gnupg.org/mailman/listinfo/gnupg-users,
mailto:[EMAIL PROTECTED]
X-List-Received-Date: Thu, 07 Dec 2006 18:27:13 -
Content-Length: 8376
Lines: 294

--=KGB-Sundevil-gamma-Skipjack-government-Vince-Foster-Treasury-bce-S-B

Hi!

Here comes a signed patch against 2.0.1 for those who care to verify
signatures ;-).


Shalom-Salam,

   Werner


--=KGB-Sundevil-gamma-Skipjack-government-Vince-Foster-Treasury-bce-S-B
Content-Disposition: inline; filename=filter-context-20-small.diff
Content-Description: Patch against 2.0.1

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
NotDashEscaped: You need GnuPG to verify this message

This is a patch against GnuPG 2.0.1. Change the directory to g10/ and
apply this patch.

2006-12-02  Werner Koch  [EMAIL PROTECTED]

* encr-data.c: Allocate DFX context on the heap and not on the
stack.  Changes at several places.  Fixes CVE-2006-6235.


Index: encr-data.c
===
--- encr-data.c (revision 4352)
+++ encr-data.c (working copy)
@@ -39,16 +39,37 @@
 static int decode_filter ( void *opaque, int control, IOBUF a,
byte *buf, size_t *ret_len);
 
-typedef struct 
+typedef struct decode_filter_context_s
 {
   gcry_cipher_hd_t cipher_hd;
   gcry_md_hd_t mdc_hash;
   char defer[22];
   int  defer_filled;
   int  eof_seen;
-} decode_filter_ctx_t;
+  int  refcount;
+} *decode_filter_ctx_t;
 
 
+/* Helper to release the decode context.  */
+static void
+release_dfx_context (decode_filter_ctx_t dfx)
+{
+  if (!dfx)
+return;
+
+  assert (dfx-refcount);
+  if ( !--dfx-refcount )
+{
+  gcry_cipher_close (dfx-cipher_hd);
+  dfx-cipher_hd = NULL;
+  gcry_md_close (dfx-mdc_hash);
+  dfx-mdc_hash = NULL;
+  xfree (dfx);
+}
+}
+
+
+
 /
  * Decrypt the data, specified by ED with the key DEK.
  */
@@ -62,7 +83,11 @@
   unsigned blocksize;
   unsigned nprefix;
   
-  memset( dfx, 0, sizeof dfx );
+  dfx = xtrycalloc (1, sizeof *dfx);
+  if (!dfx)
+return gpg_error_from_syserror ();
+  dfx-refcount = 1;
+
   if ( opt.verbose  !dek-algo_info_printed )
 {
   const char *s = gcry_cipher_algo_name (dek-algo);
@@ -77,20 +102,20 @@
 goto leave;
   blocksize = gcry_cipher_get_algo_blklen (dek-algo);
   if ( !blocksize || blocksize  16 )
- 

Re: [Mailman-Users] problem with mailing lists general contact address

2006-12-08 Thread Mark Sapiro
Henrik wrote:

When I set up a mailing list, the general contact address on the 
mail.mail.domainname.ca/mailman/listinfo page is listed as 
[EMAIL PROTECTED], rather than [EMAIL PROTECTED]


The domain in this address depends on whether or not
VIRTUAL_HOST_OVERVIEW is set to Yes or No. The default is Yes. If you
visit the http://www.example.com/mailman/listinfo overview page and
you see only lists in the www.example.com domain, then it is set to
Yes. If you see all the advertised lists on the server, then it is set
to No (in mm_cfg.py).

So, if VIRTUAL_HOST_OVERVIEW is set to No, the domain in the site list
address is the result of looking up DEFAULT_URL_HOST in the
VIRTUAL_HOSTS dictionary - i.e. it is the second argument from

add_virtualhost(DEFAULT_URL_HOST, xxx)

from mm_cfg.py (normally this is DEFAULT_EMAIL_HOST).

If VIRTUAL_HOST_OVERVIEW is set to Yes, the domain is the result of
looking up the host domain from the URL used to access the page.

In your case, assuming no typo above even though it looks strange,
there is an entry in mm_cfg.py like

add_virtualhost('mail.mail.domainname.ca', 'mail.domainname.ca')

that should be

add_virtualhost('mail.mail.domainname.ca', 'domainname.ca')

instead.

There is another possibility. This add_virtualhost entry also affects
your own lists. Is the domain 'mail.domainname.ca' the correct domain
for mailing to your lists? If so, you don't want to change the
add_virtualhost entry; you need to add an alias or whatever to the MTA
for your domain for the mailman list.

On the third hand, I visited the listinfo page on the server at
ns38.servepower.com, and it appears to be a cPanel mailman so I relly
don't have a clue. See
http://www.python.org/cgi-bin/faqw-mm.py?req=showfile=faq06.011.htp.


When I send a test email to [EMAIL PROTECTED] I get a bounce 
back with this message:

This is an automatically generated Delivery Status Notification.

Delivery to the following recipients failed.

[EMAIL PROTECTED]


When I send a test email to [EMAIL PROTECTED] I get a bounce back 
with this message:

You are not allowed to post to this mailing list, and your message has
been automatically rejected.  If you think that your messages are
being rejected in error, contact the mailing list owner at [EMAIL PROTECTED]


I presume this is a configuration issue. Unfortunately I do not have 
access on the configuration server, and the webserve.ca support people 
seem to be at a loss as to what to do.


You get the above message because the installation has only one site
list (mailman) for all supported domains and it is configured to
reject posts from non-members

Yet, since this is presumably a cPanel Mailman, if it is the case that
the domain for mail to your lists IS mail.domainname.ca, you may be
able to solve the whole problem by just creating a 'mailman' list in
your own domain.


Judging from the fact that python's own mail list page uses 
[EMAIL PROTECTED], I presume that the issue is faulty configuration of 
the general list contact address in the configuration file.

Could someone explain to me how this can be fixed, so that I can pass it on?


If the above doesn't do it, let us know what else you need.

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] A scrubber issue

2006-12-08 Thread Mark Sapiro
Todd Zullinger wrote:

Attached is the original message from the list mbox and one that I
munged up to included a content-type: text/plain; charset=3Dus-ascii.


I see the symptom with the original message. I'll look further, but it
seems to have something to do with the fact that the first sub-part
has no headers at all. I have looked at RFC 2045 and I don't see that
any headers are required, but I don't yet know what the underlying
issue in Scrubber is.

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Brad Knowles
At 11:43 AM -0500 12/8/06, John A. Martin wrote:

  See http://packages.debian.org/unstable/mail/mailman for a
  description of the Debian Mailman package that integrates
   archiving   Further down that page under the heading
  Download mailman click on one of the list of files buttons and see
  among other things: usr/lib/mailman/Mailman/Archiver/pipermail.py

I've never claimed to be a Debian expert, and if they're mucking 
about with packages that include certain features by default in order 
to remove those features, then there's not much I can do to help the 
poor souls that are stuck with that kind of stuff.

However, no amount of your expecting me to do fact-checking with 
the way that Debian is building their highly modified packaged 
versions of our software is going to change that.  It's physically 
impossible to keep up with how every single vendor is choosing to 
ship our software.


As far as I'm concerned, Debian is now further in the doghouse with 
me with regards to Mailman than most any other vendor, with the 
possible exception of cPanel.  Even Apple ships a relatively 
plain-jane version of Mailman 2.1.5 with their MacOS X Server 
platform, even if they do have their own proprietary management 
system that they tack on.

Now, if you want to side with the Debian folks on this, you're 
welcome to do that.  But no one in that camp is going to be getting 
any sympathy from me.

-- 
Brad Knowles, [EMAIL PROTECTED]

Trend Micro has announced that they will cancel the stop.mail-abuse.org
mail forwarding service as of 15 November 2006.  If you have an old
e-mail account for me at this domain, please make sure you correct that
with the current address.
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Paul Tomblin
Quoting Brad Knowles ([EMAIL PROTECTED]):
 At 11:43 AM -0500 12/8/06, John A. Martin wrote:
   See http://packages.debian.org/unstable/mail/mailman for a
   description of the Debian Mailman package that integrates
    archiving   Further down that page under the heading
   Download mailman click on one of the list of files buttons and see
   among other things: usr/lib/mailman/Mailman/Archiver/pipermail.py
 
 I've never claimed to be a Debian expert, and if they're mucking 
 about with packages that include certain features by default in order 
 to remove those features, then there's not much I can do to help the 
 poor souls that are stuck with that kind of stuff.

I don't know what John is experiencing, but I'm using Mailman installed
from Debian Stable, and have been for a couple of years, and it's always
had pipermail.

-- 
Paul Tomblin [EMAIL PROTECTED] http://blog.xcski.com/
What we obtain too cheap we esteem too lightly; it is dearness only that
gives everything its value. - Thomas Paine.
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] problem with mailing lists general contact address

2006-12-08 Thread Henrik
Thanks Mark,

If VIRTUAL_HOST_OVERVIEW is set to Yes, the domain is the result of
looking up the host domain from the URL used to access the page.

This appears to be the case:

In fact if I access the page with mail.domainname.ca/mailman/, the 
contact address is [EMAIL PROTECTED], whereas if I access the 
page with domainname.ca/mailman/listinfo (ie without the mail. 
subdomain), the contact address shows as [EMAIL PROTECTED]

Reading the domain from the URL is desirable as I have a number of 
domain names associated with the account, and wish to have mailing lists 
set up for all or most of them. But is seems like what should be used is 
something like (in totally pseudo regex, I have no idea what the syntax 
is in python)

add_virtualhost('/(mail\.)?(*.)/', '$2')

Is this true? What would the entry look like? (I don't unfortunately 
have access to mm_cfg.py - is it in principle possible to have a 
mm_cfg.py for a user of a linux system -- ie for a web account). I would 
have to pass this on to the web host.

Then the other problem would appear to be that the default mail address 
([EMAIL PROTECTED]) is not set up properly on creation of the mail 
list.

Does all this seem like a reasonable theory?

BTW, mail list creation is done through cpanel, so I presume the 
installation was done through cpanel as well. But I don't know for sure.

Also, yes I did make a typo with mail.mail.domainname.ca. I should have 
typed mail.domainname.ca. Sorry about the confusion.

Is the domain 'mail.domainname.ca' the correct domain
for mailing to your lists? If so, you don't want to change the
add_virtualhost entry; you need to add an alias or whatever to the MTA
for your domain for the mailman list.


I have no control over this, and am not sure of the question. The pop3 
(and SMTP) mail server is mail.domainname.ca. Would this interfere?

Lists are sent to [EMAIL PROTECTED], ie no mail subdomain is used 
for mailing to the lists. Does this answer that question?

You get the above message because the installation has only one site
list (mailman) for all supported domains and it is configured to
reject posts from non-members

The ideal would be to have [EMAIL PROTECTED] go to the list owner 
associated with the domain name (ie multiple domain names on my web 
account), as opposed to my web account (in effect my web acount, read 
Linux user, is host to multiple domain names, but so, presumably, are 
other users on the same installation).

Yet, since this is presumably a cPanel Mailman, if it is the case that
the domain for mail to your lists IS mail.domainname.ca, you may be
able to solve the whole problem by just creating a 'mailman' list in
your own domain.


Sorry, don't quite get that. You mean create a list called 
[EMAIL PROTECTED], and this would pre-empt the central installation 
routing? And just make the list private, with only the owner as a 
recipient? This is an interesting possibility, though if I understand it 
right, the proper solution would be to have the mail list installation 
process check sendmail for the existence of a default address for the 
domain list administrator ([EMAIL PROTECTED]) and create one if it 
doesn't exist. But I may not be understanding this properly. What about 
creating a sendmail account called [EMAIL PROTECTED] Would this 
pre-empt the mail from going to the installation-wide address? If this 
were the case, then the remaining problem would be the aforementioned 
assignment by add_virtualhost. True?

Thanks again,

All the best,

- Henrik

Mark Sapiro wrote:
 Henrik wrote:
   
 When I set up a mailing list, the general contact address on the 
 mail.mail.domainname.ca/mailman/listinfo page is listed as 
 [EMAIL PROTECTED], rather than [EMAIL PROTECTED]
 


 The domain in this address depends on whether or not
 VIRTUAL_HOST_OVERVIEW is set to Yes or No. The default is Yes. If you
 visit the http://www.example.com/mailman/listinfo overview page and
 you see only lists in the www.example.com domain, then it is set to
 Yes. If you see all the advertised lists on the server, then it is set
 to No (in mm_cfg.py).

 So, if VIRTUAL_HOST_OVERVIEW is set to No, the domain in the site list
 address is the result of looking up DEFAULT_URL_HOST in the
 VIRTUAL_HOSTS dictionary - i.e. it is the second argument from

 add_virtualhost(DEFAULT_URL_HOST, xxx)

 from mm_cfg.py (normally this is DEFAULT_EMAIL_HOST).

 If VIRTUAL_HOST_OVERVIEW is set to Yes, the domain is the result of
 looking up the host domain from the URL used to access the page.

 In your case, assuming no typo above even though it looks strange,
 there is an entry in mm_cfg.py like

 add_virtualhost('mail.mail.domainname.ca', 'mail.domainname.ca')

 that should be

 add_virtualhost('mail.mail.domainname.ca', 'domainname.ca')

 instead.

 There is another possibility. This add_virtualhost entry also affects
 your own lists. Is the domain 'mail.domainname.ca' the correct domain
 for mailing to your lists? If so, you don't 

Re: [Mailman-Users] problem with mailing lists general contact address

2006-12-08 Thread Mark Sapiro
Henrik wrote:

If VIRTUAL_HOST_OVERVIEW is set to Yes, the domain is the result of
looking up the host domain from the URL used to access the page.

This appears to be the case:

In fact if I access the page with mail.domainname.ca/mailman/, the 
contact address is [EMAIL PROTECTED], whereas if I access the 
page with domainname.ca/mailman/listinfo (ie without the mail. 
subdomain), the contact address shows as [EMAIL PROTECTED]


Actually, my description in that reply was incomplete. I should have
added to the above

and if the host domain is not a key in VIRTUAL_HOSTS, the host domain
itself is used


Reading the domain from the URL is desirable as I have a number of 
domain names associated with the account, and wish to have mailing lists 
set up for all or most of them. But is seems like what should be used is 
something like (in totally pseudo regex, I have no idea what the syntax 
is in python)

add_virtualhost('/(mail\.)?(*.)/', '$2')


You need to do one of two things:

Either have one canonical host name per domain for web access, enforced
if necessary with rewrite rules in the web server, and then have one

add_virtualhost('web host name', 'corresponding email domain')

for each domain.

Or have two (assuming the email domain should be 'example.com')

add_virtualhost('example.com', 'example.com')
add_virtualhost('mail.example.com', 'example.com')

Note that it is OK to have two entries with the same email host if the
web hosts are different, but you can't have two entries with the same
web host as the second just replaces the first.


Is this true? What would the entry look like? (I don't unfortunately 
have access to mm_cfg.py - is it in principle possible to have a 
mm_cfg.py for a user of a linux system -- ie for a web account). I would 
have to pass this on to the web host.


There is one mm_cfg.py for the whole mailman installation. The only way
around this is to run separate instances of Mailman for separate
domains.


Then the other problem would appear to be that the default mail address 
([EMAIL PROTECTED]) is not set up properly on creation of the mail 
list.


It is not an attribute of the list at all. The local part is the name
of the site list and the domain is determined as above. This is all
determined on the fly when the listinfo page is sent.



Yet, since this is presumably a cPanel Mailman, if it is the case that
the domain for mail to your lists IS mail.domainname.ca, you may be
able to solve the whole problem by just creating a 'mailman' list in
your own domain.


Sorry, don't quite get that. You mean create a list called 
[EMAIL PROTECTED], and this would pre-empt the central installation 
routing? And just make the list private, with only the owner as a 
recipient?


Well, I didn't have enough information (and I still don't because I'm
shooting in the dark and don't know how cPanel does this), but I
thought your lists might be in the mail.domainname.ca domain and not
the domainname.ca domain. In that case, I was pretty sure you could
have created a [EMAIL PROTECTED] list. You may still be able
to create a [EMAIL PROTECTED] list which would override the site
list for your domain.

I don't know if it would work or not, but the worst that can happen is
it won't create the list, or it will but mail will still go to the
site list.


This is an interesting possibility, though if I understand it 
right, the proper solution would be to have the mail list installation 
process check sendmail for the existence of a default address for the 
domain list administrator ([EMAIL PROTECTED]) and create one if it 
doesn't exist. But I may not be understanding this properly. What about 
creating a sendmail account called [EMAIL PROTECTED] Would this 
pre-empt the mail from going to the installation-wide address?


It might. You could try that.


If this 
were the case, then the remaining problem would be the aforementioned 
assignment by add_virtualhost. True?


I think so.

-- 
Mark Sapiro [EMAIL PROTECTED]   The highway is for gamblers,
San Francisco Bay Area, Californiabetter use your sense - B. Dylan

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread John A. Martin
 Paul == Paul Tomblin
 Re: [Mailman-Users] Mailman archive messages(not rm, but install!)
  Fri, 8 Dec 2006 20:04:51 -0500

 Alan == Alan McConnell
 Re: [Mailman-Users] Mailman archive messages(not rm, but install!)
  Fri, 8 Dec 2006 07:52:33 -0500

Paul Quoting Brad Knowles ([EMAIL PROTECTED]):
 At 11:43 AM -0500 12/8/06, John A. Martin wrote:
   See http://packages.debian.org/unstable/mail/mailman for a
   description of the Debian Mailman package that integrates
    archiving   Further down that page under the
   heading Download mailman click on one of the list of
   files buttons and see among other things:
   usr/lib/mailman/Mailman/Archiver/pipermail.py

 I've never claimed to be a Debian expert, and if they're
 mucking about with packages that include certain features by
 default in order to remove those features,

What makes you, Brad, think that Debian removes pipermail when shown
where it can be seen by anybody that it is included!  What mucking
about or other removal of features are you, or someone else, referring
to?

 then there's not much I can do to help the poor souls that are
 stuck with that kind of stuff.

The first recourse when having trouble with a Debian package should
not be to the upstream but to the Debian maintainers, usually via a
Debian Bug report.

Paul I don't know what John is experiencing, 

I am experiencing dismay at the innuendo followed by disinformation
with respect to the Debian Mailman package.

Paul but I'm using Mailman installed from Debian Stable, and have
Paul been for a couple of years, and it's always had pipermail.

Yes.  AFICT the absence of pipermail from the Debian Mailman is a
fantasy held only by Brad Knowles as the explanation for difficulties
experienced by a user who remarked as follows:

Alan   mm 2.1.5 .  But under Debian, so it has
Alan   experienced/endured the Debian security upgrade
Alan   procedures.

jam



pgpETWwqkMUPs.pgp
Description: PGP signature
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp

Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Brad Knowles
At 10:17 PM -0500 12/8/06, John A. Martin wrote:

  I've never claimed to be a Debian expert, and if they're
  mucking about with packages that include certain features by
  default in order to remove those features,

  What makes you, Brad, think that Debian removes pipermail when shown
  where it can be seen by anybody that it is included!  What mucking
  about or other removal of features are you, or someone else, referring
  to?

I didn't say that Debian did.  Alan McConnell said that Mailman had 
been installed without pipermail:

Meanwhile, I am adminning(sp?), through my ISP, a new but
quite active E-list.  But their mailman install is
incomplete; they haven't put in Pipermail (about which
I know _nothing_).

When asked what kind of whacked-out version of Mailman they were 
running that didn't include the built-in version of pipermail, he 
said:

mm 2.1.5 .  But under Debian, so it has experienced/endured the
Debian security upgrade procedures.

To which my reply was:

Okay, now that is one of the most bizarre things I've heard of
in a very long time.  I cannot comprehend how they could
possibly ship a version of Mailman 2.1.x that does not
automatically include the bundled Pipermail component.

This lead to your mildly offensive reply, where you publicly said:

See http://packages.debian.org/unstable/mail/mailman
for a description of the Debian Mailman package that
integrates  archiving   Further down that page
under the heading Download mailman click on one of the
list of files buttons and see among other things:
usr/lib/mailman/Mailman/Archiver/pipermail.py

Yet nowhere on that page do I find any reference to pipermail.  If 
you had wanted to provide proof that Debian provides pipermail as 
part of the package, you should have been much less obtuse and 
offensive with your language, and much more explicit in the URL you 
provided.

Based on what I saw on that page, and the rude behaviour I was seeing 
from you, I concluded that Debian had actually done precisely what I 
had previously commented on to Alan, and led to my response:

I've never claimed to be a Debian expert, and if they're
mucking about with packages that include certain features
by default in order to remove those features, then there's
not much I can do to help the poor souls that are stuck
with that kind of stuff.

However, no amount of your expecting me to do fact-checking
with the way that Debian is building their highly modified
packaged versions of our software is going to change that.
It's physically impossible to keep up with how every single
vendor is choosing to ship our software.

Note that I do not, at any time, make an outright claim that Debian 
was stripping pipermail from the Mailman package that they were 
providing -- I said ... if they're mucking about with packages that 
include certain features by default to remove those features

Obviously the subtle difference in this statement was completely lost on you.

  The first recourse when having trouble with a Debian package should
  not be to the upstream but to the Debian maintainers, usually via a
  Debian Bug report.

I don't think it's appropriate for us to be filing bug reports on 
these sorts of things with package maintainers of a given platform. 
If the users of those packages wish to file bug reports, I would 
fully support that.  If the package maintainers wish to come back to 
us and file bug reports against our code in our bug tracking system, 
I welcome that.

But no one here has the time to go tracking down every single bloody 
bizarre behaviour that may or may not be a result of something 
strange that a package maintainer decided to do, and then to track 
them down and sit on them until they fix their bug.


That is, unless you're volunteering to do that, of course.  If so, 
then please just go ahead and do so, and quit making worse a 
situation that is already pretty bad to begin with.

 Paul I don't know what John is experiencing,

  I am experiencing dismay at the innuendo followed by disinformation
  with respect to the Debian Mailman package.

If you want clarity in a discussion, it would really help if you 
would actually provide some measure of clarity in your own postings.


If I've made a mistake, and that fact is pointed out to me in a 
reasonably neutral and constructive way, I generally accept and even 
welcome the correction and genuinely work towards a good resolution 
to the problem.

However, if your first reaction is obtuse and offensive bluster with 
baseball bats, then you damn well better be prepared for the kind of 
reaction you're going to receive.

 Paul but I'm using Mailman installed from Debian Stable, and have
 Paul been for a couple of years, and it's always had pipermail.

  Yes.  

Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread Stephen J. Turnbull
Brad Knowles writes:

  I've never claimed to be a Debian expert, and if they're mucking 
  about with packages that include certain features by default in order 
  to remove those features, then there's not much I can do to help the 
  poor souls that are stuck with that kind of stuff.

That's not what they do.  Debian splits packages into components for
ease of maintenance of the package infrastructure.  Users are expected
to select virtual packages that cause a collection of components to
be installed.  By design, the package named mailman is either a
complete virtual package, or a single package that contains all of
Mailman.  A package named after the upstream package is intended to
install all of it.  It works.  The mailman package did install a
complete Mailman on several Debian systems where I use Mailman.

This is not the same as cPanel, which *does* deliberately inhibit
capabilities that Mailman admins need.  This is a situation where the
user misjudges what the bug is, and reports to the wrong channel.
It's not terribly nice of Debian to risk your time on their ability to
create robust, non-buggy virtual packages, but it is the result of
providing a service that the Mailman project does not, and should not.

This policy will occasionally result in the kind of problem we see
here.  It's a tradeoff.  You don't like it as it manifests on
Mailman-Users, and you shouldn't---all we're ever going to see here is
the bugs.  But all you need to say is the Mailman project distributes
Mailman with the pipermail archiver included.  If you don't have it,
it is a packaging issue.  Your vendor Debian needs to know about it,
and you need to get support from them.

In sum, I think you are doing the Mailman project a disservice by
denigrating Debian.  If they're really doing their users such a
disservice, you (or somebody from Mailman who understands the issues)
should report it as a bug.  In my experience the various distros,
including Debian, are responsive to upstream maintainers.

Regards,

--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] A scrubber issue

2006-12-08 Thread Tokio Kikuchi
Hi all,

Mark Sapiro wrote:
 Todd Zullinger wrote:
 Attached is the original message from the list mbox and one that I
 munged up to included a content-type: text/plain; charset=3Dus-ascii.
 
 
 I see the symptom with the original message. I'll look further, but it
 seems to have something to do with the fact that the first sub-part
 has no headers at all. I have looked at RFC 2045 and I don't see that
 any headers are required, but I don't yet know what the underlying
 issue in Scrubber is.
 
It looks like the problem is something to do with email package 
behavior.  Here is a test code to reproduce the problem:


from email import message_from_string

s = '''MIME-Version: 1.0
Content-Type: multipart/mixed; boundary=ABCDEFG

Preamble

--ABCDEFG

A message without header

--ABCDEFG
Content-Disposition: inline

Another text with a header

--ABCDEFG--
'''

m = message_from_string(s)
for p in m.walk():
 print p.get_content_type()
 if p:
 print p.get_payload(decode=True)



This program prints out like this:

multipart/mixed
None
text/plain
text/plain
Another text with a header

Note that 'A message without header' is not printed out.

If I remove 'If p:' and print the payload unconditionally, I get:

multipart/mixed
None
text/plain
A message without header

text/plain
Another text with a header

Here, the no-header part is printed out.

The patch of revision 7281 may have been over-protective against the 
bug-id: 1099138 in message reconstruction part (line 327 in 
http://mailman.svn.sourceforge.net/viewvc/mailman/branches/Release_2_1-maint/mailman/Mailman/Handlers/Scrubber.py?r1=7207r2=7281
 
) but we may have to be paranoid against wired message like no-header 
text.  I believe well behaved MUAs won't generate no-header text parts. 
  (or, I believed ;-)

-- 
Tokio Kikuchi, [EMAIL PROTECTED]
http://weather.is.kochi-u.ac.jp/
--
Mailman-Users mailing list
Mailman-Users@python.org
http://mail.python.org/mailman/listinfo/mailman-users
Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py
Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/
Unsubscribe: 
http://mail.python.org/mailman/options/mailman-users/archive%40jab.org

Security Policy: 
http://www.python.org/cgi-bin/faqw-mm.py?req=showamp;file=faq01.027.htp


Re: [Mailman-Users] Mailman archive messages(not rm, but install!)

2006-12-08 Thread John A. Martin
 Brad == Brad Knowles
 Re: [Mailman-Users] Mailman archive messages(not rm, but install!)
  Fri, 8 Dec 2006 22:28:19 -0600

Brad At 10:17 PM -0500 12/8/06, John A. Martin wrote:
  I've never claimed to be a Debian expert, and if they're
  mucking about with packages that include certain features by
  default in order to remove those features,

 What makes you, Brad, think that Debian removes pipermail when
 shown where it can be seen by anybody that it is included!
 What mucking about or other removal of features are you, or
 someone else, referring to?

Oops.  The first sentence of mine above was intended to end with a
question mark rather than an exclamation mark.  Maybe that would have
sounded better.

Brad I didn't say that Debian did.  Alan McConnell said that
Brad Mailman had been installed without pipermail:

Does that mean that the Debian Package does not carry pipermail?

Brad   Meanwhile, I am adminning(sp?), through my ISP, a new
Brad   but quite active E-list.  But their mailman install is
Brad   incomplete; they haven't put in Pipermail (about which
Brad   I know _nothing_).

Brad When asked what kind of whacked-out version of Mailman they
Brad were running that didn't include the built-in version of
Brad pipermail, he said:

Brad   mm 2.1.5 .  But under Debian, so it has
Brad   experienced/endured the Debian security upgrade
Brad   procedures.

Does that mean that the Debian Package does not carry pipermail?

Brad To which my reply was:

Brad   Okay, now that is one of the most bizarre things I've
Brad   heard of in a very long time.  I cannot comprehend how
Brad   they could possibly ship a version of Mailman 2.1.x
Brad   that does not automatically include the bundled
Brad   Pipermail component.

Between you and Alan you suggest something that is not true.  You are
an authority on this list (and elsewhere).  Beginners will conclude
From your statement that they should avoid using the Debian package.
Whether that is good advice or not it is not justified by the line of
evidence above.

Excuse me if I have a tendency based upon the above to suspect a
readiness to assume that Debian does bizarre things when there is no
evidence supporting that assumption.

Brad This lead to your mildly offensive reply, where you publicly
Brad said:

Brad   See http://packages.debian.org/unstable/mail/mailman
Brad   for a description of the Debian Mailman package that
Brad   integrates  archiving   Further down that
Brad   page under the heading Download mailman click on one
Brad   of the list of files buttons and see among other
Brad   things:
Brad   usr/lib/mailman/Mailman/Archiver/pipermail.py

I do not know what there is offensive and no offense was intended.  I
generally try to be precise and succinct.  If one clicks one of the
buttons mentioned, one sees a URL something like
http://packages.debian.org/cgi-bin/search_contents.pl?searchmode=filelistword=mailmanversion=unstablearch=i386
(for the i386 architecture) which I thought was uglier than mentioning
the button to click at the URL I gave.

Brad Yet nowhere on that page do I find any reference to
Brad pipermail.  If you had wanted to provide proof that Debian
Brad provides pipermail as part of the package, you should have
Brad been much less obtuse and offensive with your language, and
Brad much more explicit in the URL you provided.

If you had bothered to click on one of the list of files buttons you
would have (for the i386 archetecture) seen the pipermail.py file as
the 19th of 3438 files.

 The first recourse when having trouble with a Debian package
 should not be to the upstream but to the Debian maintainers,
 usually via a Debian Bug report.

Brad I don't think it's appropriate for us to be filing bug
Brad reports on these sorts of things with package maintainers of
Brad a given platform.  If the users of those packages wish to
Brad file bug reports, I would fully support that.  If the
Brad package maintainers wish to come back to us and file bug
Brad reports against our code in our bug tracking system, I
Brad welcome that.

I agree with what you say in the paragraph above and would have
thought that went without saying.  However when packaging issues arise
I think it would be good to suggest that users of various
distributions should consult whatever support the distribution offers.
For the Debian Mailman package, which does not have a related Debian
mailing list, the Debian user should usually file a Debian Bug report.

Presumably the one having trouble with a Debian package is a Debian
user not the us which I take to mean you and the Mailman community
as a whole.

I've said enough.  My enthusiasm is well restrained when it is
necessary to craft each sentence