9:57 a.m.
>>>> "sjt" == Stephen J Turnbull
<stephen(a)xemacs.org> writes:
sjt> OK, I've verified that my debug handler can catch Mailman
sjt> discarding the contents of the CC.
I've identified what is probably the problem module, and have a fix in
place to be live-tested (it passes the bench tests, we'll see what
happens with real data). For details see bug #1079760 on SourceForge
in the Mailman project.
Currently the fix is only in place for XEmacs Beta; I'll discuss with
the Tux admins whether it's useful to do this for all Tux lists
later. We may as well confirm (as well as possible) that it fixes the
problem.
The fix will be implemented immediately following this message. (Ie,
this is the last message before I change the Mailman pipeline for
XEmacs Beta.) I expect there to be no changes except that Cc
stripping should no longer occur. If something should go wrong, there
may be a delay up to 18 hours before I fix it; I'll check as I can,
but it's time to go home, and eventually to sleep, now. :-)
Steve
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
2:57 a.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
sjt> OK, I've verified that my debug handler can catch
Mailman
sjt> discarding the contents of the CC.
I've identified what is probably the problem module, and have
a fix in place to be live-tested (it passes the bench tests,
we'll see what happens with real data). For details see bug
#1079760 on SourceForge in the Mailman project.
Thanks for all this work! I really appreciate it.
Perhaps I should just turn off the eliminate-duplicates feature; it doesn't
really work well anyway, since there's no facility for specifying alternate
addresses [ben(a)666.com, ben(a)xemacs.org].
7:24 p.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
"Ben Wing" <ben(a)666.com> writes:
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
> sjt> OK, I've verified that my debug handler can catch
Mailman
> sjt> discarding the contents of the CC.
>
> I've identified what is probably the problem module, and have
> a fix in place to be live-tested (it passes the bench tests,
> we'll see what happens with real data). For details see bug
> #1079760 on SourceForge in the Mailman project.
Thanks for all this work! I really appreciate it.
Perhaps I should just turn off the eliminate-duplicates feature; it doesn't
really work well anyway, since there's no facility for specifying alternate
addresses [ben(a)666.com, ben(a)xemacs.org].
Seems to work well for me, except when people use Lotus Notes drafts
which seem to be stupid enough to allow sending multiple messeges with
the same message-id.
I'd have to check if it's important enough.
Ben, you're refering to
`nnmail-treat-duplicates' (buffer: *Hyper Apropos*, mode: Hyper-Apropos)
right?
That should track duplication based on message-id header, not addresses.
Adrian
7:57 p.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
On Wed, 08 Dec 2004 20:24:13 +0100, Adrian Aichner said:
Seems to work well for me, except when people use Lotus Notes drafts
which seem to be stupid enough to allow sending multiple messeges with
the same message-id.
Gaak. It isn't often that "A cauldron of boiling oil" seems a suitable
response
for MUA brain damage, but I have this sudden urge to run out and buy about 300
bottles of canola oil... ;)
8:18 p.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
Valdis.Kletnieks(a)vt.edu writes:
Gaak. It isn't often that "A cauldron of boiling oil"
seems a suitable response
for MUA brain damage, but I have this sudden urge to run out and buy about 300
bottles of canola oil... ;)
Oh yes, pretty please. Something that gives a blue flame :-)
norbert - who has to cope with Notes in the office ...
--
http://www.dreckstool.de/
12:44 a.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
Norbert Koch <viteno(a)xemacs.org> writes:
Valdis.Kletnieks(a)vt.edu writes:
> Gaak. It isn't often that "A cauldron of boiling oil" seems a suitable
response
> for MUA brain damage, but I have this sudden urge to run out and buy about 300
> bottles of canola oil... ;)
Oh yes, pretty please. Something that gives a blue flame :-)
norbert - who has to cope with Notes in the office ...
Locus Notes is what I call it.
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
7:57 p.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
>>>> "Norbert" == Norbert Koch
<viteno(a)xemacs.org> writes:
Norbert> norbert - who has to cope with Notes in the office ... --
What a polite way of putting it! A while ago I was forced to move from Gnus
to Notes at work and every day I find new ways to be disgusted by it.
Malcolm - Forced to top post as work due to corporate policy.
--
Malcolm Purvis <malcolmp(a)xemacs.org>
8:26 p.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
>>>> "Malcolm" == Malcolm Purvis
<malcolmp(a)xemacs.org> writes:
>>>> "Norbert" == Norbert Koch
<viteno(a)xemacs.org> writes:
Norbert> norbert - who has to cope with Notes
in the office ... --
Malcolm> What a polite way of putting it! A while ago I was forced to move
Malcolm> from Gnus to Notes at work and every day I find new ways to be
Malcolm> disgusted by it.
Malcolm> Malcolm - Forced to top post as work due to corporate policy.
I almost had that same problem when I worked at IBM (who forced everyone to
use Notes after they bought Lotus).
To get around it and continue using XEmacs for my mail I wrote some
(*shudder*) LotusScript (I hate admiting that) to package up my incoming email
(w/ it's header info in a special parsable format) and send it to my AIX
workstation where I then had a perl script reconstruct the original email as
best as possible (there were a *lot* of oddities I had to work around).
Thankfully I could send email out from the workstation directly so I didn't
have to hack up anything for that side.
Worked out pretty good for me (except when they attached Lotus documents to
their emails) and I think that code is still in use by a bunch of people back
there.
Now that I think about it, I *might* even have a copy of it lying around
somewhere if any of you currently suffering might be willing to hack at it
further to make it of use to you...
Chuck
ps - my officemate back there absolutely hated inline citations because they
were all so ingraned to top posting...
--
Charles K. Hines <ckh(a)requesttech.com> <chuck.hines(a)baesystems.com>
Principal Scientist at ReQuest Technologies Inc. (http://www.ReQuestTech.com/)
"If you can't be a good example, then you'll just have to be a
horrible warning." - Catherine Aird
4:57 a.m.
New subject: debugging Mailman pipeline on XEmacs Beta (CC stripping)
>>>> "APA" == Adrian Aichner
<adrian(a)xemacs.org> writes:
APA> "Ben Wing" <ben(a)666.com> writes:
> Perhaps I should just turn off the eliminate-duplicates
> feature; it doesn't really work well anyway, since there's no
> facility for specifying alternate addresses [ben(a)666.com,
> ben(a)xemacs.org].
I thought about implementing this, but then realized it would be an
invitation to DoS: any subscriber could DoS any other subscriber
simply by setting his own account to no-dupes then adding the other
subscriber's address.
APA> Ben, you're refering to `nnmail-treat-duplicates' (buffer:
APA> *Hyper Apropos*, mode: Hyper-Apropos) right?
No, he's talking about the no-dupes setting in the Mailman user
interface.
APA> That should track duplication based on message-id header, not
APA> addresses.
Which has the DoS potential you noticed. It's documented in the
procmail manpage. ;-)
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
7287
days inactive
7291
days old
8 comments
7 participants
participants (7)
-
Adrian Aichner -
Ben Wing -
Charles Hines -
Malcolm Purvis -
Norbert Koch -
Stephen J. Turnbull -
Valdis Kl=?iso8859-4?Q?=BA?=tnieks