The qpopper list archive ending on 29 May 2001
Topics covered in this issue include:
1. Re: kernel error
Michael Brennen <mbrennen at fni dot com>
Fri, 18 May 2001 18:39:45 -0500 (CDT)
2. Re: kernel error
"Kenneth Porter" <shiva at well dot com>
Fri, 18 May 2001 16:33:13 -0700
3. problem with outlook?
Luiz Felipe Ceglia <lceglia at terenet.com dot br>
Sat, 19 May 2001 20:35:11 -0300
4. -ERR POP EOF or I/O Error
"Nick Belnap" <nickb at lansrc dot com>
Mon, 21 May 2001 16:00:48 -0600
5. Newbie question about standalone mode.
"Nick Belnap" <nickb at lansrc dot com>
Mon, 21 May 2001 16:02:41 -0600
6. Re: -ERR POP EOF or I/O Error
Steven Champeon <schampeo at hesketh dot com>
Mon, 21 May 2001 17:56:26 -0400
7. Re: Newbie question about standalone mode.
Chuck Yerkes <chuck+qpopper at yerkes dot com>
Tue, 22 May 2001 00:05:16 -0700
8. Re: Newbie question about standalone mode.
Ken Hornstein <kenh at cmf.nrl.navy dot mil>
Tue, 22 May 2001 08:49:50 -0400
9. Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
Jason Englander <jason at fiber-city dot com>
Tue, 22 May 2001 09:56:27 -0400 (EDT)
10. Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
Gregory Hicks <ghicks at cadence dot com>
Tue, 22 May 2001 10:06:42 -0700 (PDT)
11. Re: problem with outlook?
Randall Gellens <randy at qualcomm dot com>
Tue, 22 May 2001 10:08:43 -0700
12. Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
Jason Englander <jason at fiber-city dot com>
Tue, 22 May 2001 16:15:10 -0400 (EDT)
13. Patch for qpopper4.0.2 to set server-mode based on user shell
Clifton Royston <cliftonr at lava dot net>
Tue, 22 May 2001 17:12:40 -1000
14. popper.core dumps with outlook express
Jeff Donovan <jdonovan at beth.k12.pa dot us>
Wed, 23 May 2001 11:21:44 -0400
15. Re: popper.core dumps with outlook express
Jeff Donovan <jdonovan at beth.k12.pa dot us>
Wed, 23 May 2001 11:40:14 -0400
16. Re: popper.core dumps with outlook express
Jeff Donovan <jdonovan at beth.k12.pa dot us>
Wed, 23 May 2001 11:38:46 -0400
17. Re: popper.core dumps with outlook express
Jack Sasportas <jack at innovativeinternet dot com>
Wed, 23 May 2001 12:16:20 -0400
18. Re: popper.core dumps with outlook express
Jack Sasportas <jack at innovativeinternet dot com>
Wed, 23 May 2001 11:58:01 -0400
19. Re: popper.core dumps with outlook express
Joseph S D Yao <jsdy at cospo.osis dot gov>
Wed, 23 May 2001 12:05:25 -0400
20. Re: popper.core dumps ::THANKS::
Jeff Donovan <jdonovan at beth.k12.pa dot us>
Wed, 23 May 2001 13:09:45 -0400
21. Re: popper.core dumps with outlook express
Clifton Royston <cliftonr at lava dot net>
Wed, 23 May 2001 08:09:53 -1000
22. Re: Patch for qpopper4.0.2 to set server-mode based on user
Randall Gellens <randy at qualcomm dot com>
Wed, 23 May 2001 14:39:19 -0700
23. Re: popper.core dumps with outlook express
Randall Gellens <randy at qualcomm dot com>
Wed, 23 May 2001 14:42:24 -0700
24. Re: Patch for qpopper4.0.2 to set server-mode based on user shell
Clifton Royston <cliftonr at lava dot net>
Wed, 23 May 2001 12:20:46 -1000
25. Re: Patch for qpopper4.0.2 to set server-mode based on user
Randall Gellens <randy at qualcomm dot com>
Wed, 23 May 2001 17:06:57 -0700
26. Re: Patch for qpopper4.0.2 to set server-mode based on user shell
"Kenneth Porter" <shiva at well dot com>
Thu, 24 May 2001 01:52:12 -0700
27. trimming username
James Eastwood <JamesEastwood at businessserve.co dot uk>
Thu, 24 May 2001 10:47:50 +0100
28. Re: popper.core dumps ::THANKS::
"Doryce E . Moore" <demoore at ccsalpha3.nrl.navy dot mil>
Thu, 24 May 2001 14:47:23 -0400
29. Feature request
Forrest Aldrich <forrie at forrie dot com>
Thu, 24 May 2001 15:12:13 -0400
30. Re: Patch for qpopper4.0.2 to set server-mode based on user shell
Robert Brewer <rbrewer at lava dot net>
Thu, 24 May 2001 13:45:24 -1000
31. Re: Patch for qpopper4.0.2 to s ...
Peter Evans <peter at gol dot com>
Fri, 25 May 2001 08:58:27 +0900
32. BUG REPORT: module popper.c, function getline()
Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Mon, 28 May 2001 11:56:00 +0200
33. user can not get messages
Joerg Hartmann <J.Hartmann at des.mediadesign dot de>
Mon, 28 May 2001 13:03:22 +0200
34. Re: BUG REPORT: module popper.c, function getline()
Gregory Hicks <ghicks at cadence dot com>
Mon, 28 May 2001 07:46:34 -0700 (PDT)
35. Re: BUG REPORT: module popper.c, function getline()
Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Mon, 28 May 2001 17:40:57 +0200
36. Re: BUG REPORT: module popper.c, function getline()
Homer Wilson Smith <homer at lightlink dot com>
Mon, 28 May 2001 11:52:35 -0400 (EDT)
37. Re: BUG REPORT: module popper.c, function getline()
Derek Balling <dredd at megacity dot org>
Mon, 28 May 2001 08:50:49 -0700
38. RE: user can not get messages
"PuntoNet" <wilson at punto.net dot ec>
Mon, 28 May 2001 10:30:57 -0500
39. Re: BUG REPORT: module popper.c, function getline()
Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Mon, 28 May 2001 18:10:11 +0200
40. Re: BUG REPORT: module popper.c, function getline()
Chuck Yerkes <chuck+qpopper at yerkes dot com>
Mon, 28 May 2001 12:41:02 -0700
41. compiling problems
"Eduardo A. Alcocer Rosado" <ealcocer at hotmail dot com>
Mon, 28 May 2001 16:06:08 -0500
42. qpopper.config
applein <applein at tutopia.com dot br>
Mon, 28 May 2001 18:28:39 -0300
43. Re: user can not get messages
Joerg Hartmann <J.Hartmann at des.mediadesign dot de>
Tue, 29 May 2001 09:05:29 +0200
44. Outlook Express troubles following Qpopper 3 -> 4 upgrade.
Eric Luyten <Eric.Luyten at vub.ac dot be>
Tue, 29 May 2001 15:36:10 +0200 (MET DST)
45. Outlook Express - - Qpopper - - Openssl
"Christopher Crowley" <ccrowley at tulane dot edu>
Tue, 29 May 2001 09:51:28 -0500
46. Re: Outlook Express troubles following Qpopper 3 -> 4 upgrade.
"Eduardo E. Silva" <esilva at silvex dot com>
Tue, 29 May 2001 08:48:55 -0700
47. Re: Outlook Express troubles following Qpopper 3 -> 4 upgrade.
Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Tue, 29 May 2001 18:19:28 +0200
48. Re: Outlook Express troubles following Qpopper 3 -> 4 upgrade.
Roy <garlic at garlic dot com>
Tue, 29 May 2001 10:25:47 -0700
49. Re: qpopper.config
Clifton Royston <cliftonr at lava dot net>
Tue, 29 May 2001 11:01:12 -1000
50. 3.0 to 4.02 upgrade went smoothly
Clifton Royston <cliftonr at lava dot net>
Tue, 29 May 2001 14:04:40 -1000
Date: Fri, 18 May 2001 18:39:45 -0500 (CDT)
From: Michael Brennen <mbrennen at fni dot com>
Subject: Re: kernel error
If you are running linux, you will want to get the kernel source,
then see /usr/src/linux/Documentation/proc.txt; look for 'file-max'.
You might well need to look at inode-max as well as compile time
kernel limits if you are out of open files.
-- Michael
On Fri, 18 May 2001, Dan Trainor wrote:
> First off, I do see this message quite often in /var/log/messages:
>
> May 18 15:37:54 mail inetd[672]: pid 4150: exit status 1
>
> Looks like inetd just.. well, dies. I have yet to find a reason
> why, but I think it may have something to do with qpopper.
> Would anyone care to explain why, if it is the cause, qpopper
> makes it do that?
>
> Also, when doing some little tests on qpopper, I did come out with this
> error in /var/log/messages:
>
> May 18 15:39:16 mail kernel: VFS: file-max limit 4096 reached
>
> ..and until some messages were sent out or that number was
> lowered, procmail could not locally deliver messages. I do
> believe it is a kernel option that must be set (right?), but
> even so, it is something that others might want to take note
> when handling a bunch of connections.
From: "Kenneth Porter" <shiva at well dot com>
Date: Fri, 18 May 2001 16:33:13 -0700
Subject: Re: kernel error
On Fri, 18 May 2001 16:06:40 -0700, Dan Trainor wrote:
>May 18 15:37:54 mail inetd[672]: pid 4150: exit status 1
>May 18 15:39:16 mail kernel: VFS: file-max limit 4096 reached
Can't tell much without knowing what kind of system this is, or what
version of programs are in use.
If this is Linux, you might want to check comp.os.linux.*, and check
Google's archives of the Linux newsgroups for similar reports.
Ken
mailto:shiva at well dot com
http://www.sewingwitch.com/ken/
[If answering a mailing list posting, please don't cc me your reply. I'll take my answer on the list.]
Date: Sat, 19 May 2001 20:35:11 -0300
From: Luiz Felipe Ceglia <lceglia at terenet.com dot br>
Subject: problem with outlook?
Hi Folks,
I am not sure wheter this is an inherent problem of qpopper or not.
I am now using Qpopper4.02, and was using last version (3.xx) before.
Many of my users who use Microsoft Outlook are having problems of timing out.
The problem persist after I upgraded to qpopper4.02
Was that noticed before?
Do you guys know of any possible cause?
Thank you for your attention,
Luiz Felipe Ceglia
+55-21-9135-3579
From: "Nick Belnap" <nickb at lansrc dot com>
Subject: -ERR POP EOF or I/O Error
Date: Mon, 21 May 2001 16:00:48 -0600
I'm running qpopper 4.01 in standalone mode on Redhat 7.0
Anybody have any idea what this error is? -ERR POP EOF or I/O Error. This
is all the info it gives me.
From: "Nick Belnap" <nickb at lansrc dot com>
Subject: Newbie question about standalone mode.
Date: Mon, 21 May 2001 16:02:41 -0600
Hi all.
I am running qpopper 4.01 in standalone mode on RedHat 7.0. When I launch
qpopper it launches only one daemon. Is there anyway to have it launch
say...10 daemons and would that increase my performance?
Thanks.
Date: Mon, 21 May 2001 17:56:26 -0400
From: Steven Champeon <schampeo at hesketh dot com>
Subject: Re: -ERR POP EOF or I/O Error
on Mon, May 21, 2001 at 04:00:48PM -0600, Nick Belnap wrote:
> I'm running qpopper 4.01 in standalone mode on Redhat 7.0
>
> Anybody have any idea what this error is? -ERR POP EOF or I/O Error. This
> is all the info it gives me.
http://www.eudora.com/qpopper/faq.html#EOF
Date: Tue, 22 May 2001 00:05:16 -0700
From: Chuck Yerkes <chuck+qpopper at yerkes dot com>
Subject: Re: Newbie question about standalone mode.
Quoting Nick Belnap (nickb at lansrc dot com):
> I am running qpopper 4.01 in standalone mode on RedHat 7.0. When I launch
> qpopper it launches only one daemon. Is there anyway to have it launch
> say...10 daemons and would that increase my performance?
It would if qpopper were built on this model.
The Apache group has some really nice libraries and routines
in development for Apache 2.0. Sort of a best of preforking
and threading and multiple processes.
Again, not really part of a non-threading qpopper.
I do use a commercial POP/IMAP in my life that has 1 daemon.
Ever. New connections are just threads. Scales quite well.
[and take a look at qpopper 4.02 and perhaps Redhat 7.1 - the
previous versions have some issues]
Subject: Re: Newbie question about standalone mode.
Date: Tue, 22 May 2001 08:49:50 -0400
From: Ken Hornstein <kenh at cmf.nrl.navy dot mil>
>The Apache group has some really nice libraries and routines
>in development for Apache 2.0. Sort of a best of preforking
>and threading and multiple processes.
>
>Again, not really part of a non-threading qpopper.
You don't really need threads for preforking, which I think would help
a fair amount (that would require other changes to qpopper, of course).
--Ken
Date: Tue, 22 May 2001 09:56:27 -0400 (EDT)
From: Jason Englander <jason at fiber-city dot com>
Subject: Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
On Fri, 18 May 2001, Kenneth Porter wrote:
> I believe "pop3" is the "correct" name, adopted by IANA. (I found an
> official service name list somewhere on the standards website.) "pop-3"
> predates this adoption.
http://www.iana.org/assignments/port-numbers/
Jason
--
Jason Englander <jason at fiber-city dot com>
Systems Administrator - FiberCity Networks
Date: Tue, 22 May 2001 10:06:42 -0700 (PDT)
From: Gregory Hicks <ghicks at cadence dot com>
Subject: Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
> Date: Tue, 22 May 2001 09:56:27 -0400 (EDT)
> From: Jason Englander <jason at fiber-city dot com>
>
> On Fri, 18 May 2001, Kenneth Porter wrote:
>
> > I believe "pop3" is the "correct" name, adopted by IANA. (I found an
> > official service name list somewhere on the standards website.) "pop-3"
> > predates this adoption.
>
> http://www.iana.org/assignments/port-numbers/
This actually ought to be:
http://www.iana.org/assignments/port-numbers
(Note, no trailing "/" ...)
Regards,
Gregory Hicks
---------------------------------------------------------------------
Gregory Hicks | Principal Systems Engineer
Cadence Design Systems | Direct: 408.576.3609
555 River Oaks Pkwy M/S 6B1 | Fax: 408.894.3479
San Jose, CA 95134
Date: Tue, 22 May 2001 10:08:43 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: problem with outlook?
At 8:35 PM -0300 5/19/01, Luiz Felipe Ceglia wrote:
> Many of my users who use Microsoft Outlook are having problems of
> timing out.
> The problem persist after I upgraded to qpopper4.02
This happens a lot with Outlook users on congested networks, because
th default timeouts in Outlook are very small. You can try having
the users increase the timeouts, try a different client (such as
Netscape or Eudora), or use packet traces to see what is going on and
fix the problem.
Date: Tue, 22 May 2001 16:15:10 -0400 (EDT)
From: Jason Englander <jason at fiber-city dot com>
Subject: Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
On Tue, 22 May 2001, Gregory Hicks wrote:
> This actually ought to be:
>
> http://www.iana.org/assignments/port-numbers
>
> (Note, no trailing "/" ...)
Oops! Thanks for the correction.
Jason
Date: Tue, 22 May 2001 17:12:40 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Patch for qpopper4.0.2 to set server-mode based on user shell
Greetings to the list. Would there be interest in my posting a new
patch that I have submitted for incorporation into qpopper 4.0?
This patch allows qpopper to automatically select or deselect server
mode based on whether the user has shell access - the main factor in
whether server mode is "safe" - by looking at their login shell.
Introduction:
This patch grew out of considering what had stood in the way of our
enabling server mode for Qpopper at our site. Here, there are a
substantial minority of users who have shell access on the server which
handles mail delivery (though the vast majority do not.)
Prudence kept us from simply enabling server mode across the board
and "hoping for the best." Changing the login group for thousands of
users, so that we could use the group-based server mode selection
seemed likely to cause problems. One of our group (Robert Brewer)
asked if we could base the server-mode on whether the user actually has
shell access, based on their shell setting - and that was enough of an
idea for me to dig in and code it. It seems to work nicely, and it
looks like it will solve all the issues with rolling server mode out
for 90% of our userbase.
Usage:
Server mode can be either enabled, or disabled, for a given class of
users, based on whether their actual login shell (from their password
entry) is one of those listed in a configuration variable, in the form
of a colon-separated string.
Thanks to the much cleaner design of the latest qpopper, this proved
very easy for me to integrate "across the board" in the style of the
other options: the option can be set either from the configure script,
to get it in as a default, or from a run-time configuration file.
An example of the former:
./configure --enable-shell-server-mode=/sbin/nologin
An example of the latter:
set shell-server-mode = "/usr/bin/false:/sbin/nologin:/usr/local/etc/no_shell"
You could also exclude specific shells via either mechanism:
set server-mode
set shell-no-server-mode = "/bin/sh:/bin/bash:/bin/ksh:/bin/csh:/bin/tcsh"
etc.
Integration:
I have tried to make this code as clean, elegant, portable, and
transparent as feasible in a small amount of time. I don't believe any
new portability issues should come up with the constructs I'm using,
since the tests for group membership already presume a successful
getpwent() call.
I have a unified recursive diff for all of the files which were
updated to make this change, and the C source for the one new source
file (popper/check_shell.c). I will happily post it to the list if
that meets the expectations for attachments and source to this
particular list. (The whole patch + new code totals about 20K.) I've
also submitted it to the qpopper-patches address.
Cheers,
-- Clifton
--
Clifton Royston -- LavaNet Systems Architect -- cliftonr at lava dot net
WWJD? "JWRTFM!" - Scott Dorsey (kludge) "JWG" - Eddie Aikau
Date: Wed, 23 May 2001 11:21:44 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: popper.core dumps with outlook express
Greetings
I have a problem with outlook express clients (like who doesn't)
running on a macintosh system.
My server is BSDi 4.2 patched up to M20-010.
qpopper version (i think) is 3.0 Im not sure how to check my version.
the clients are not able to pop into their maillboxes and receive
their mail. when the pop in, i get a popper.core file in the / dir.
I may have to update to the latest qPopper unless there is a better
remedy, I have never done this before so if anyone could point me to
a nice README file that will show me how to do this without killing
the thousands of other functioning accounts I have.
TIA
--jeff
--
------------------------------------------------------------------------
Jeff Donovan Network Analyst
Bethlehem Area School District Network Operations
Bethlehem, PA 18020 (610) 807-5571 jdonovan at beth.k12.pa dot us
Date: Wed, 23 May 2001 11:40:14 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: Re: popper.core dumps with outlook express
How much trouble would it be do upgrade?
would anything get lost. This machine hosts many many accounts.
--jeff
>
>Not sure about your specific problem, but it's probably best
>to upgrade to 4.0.2 anyway.
--
------------------------------------------------------------------------
Jeff Donovan Network Analyst
Bethlehem Area School District Network Operations
Bethlehem, PA 18020 (610) 807-5571 jdonovan at beth.k12.pa dot us
Date: Wed, 23 May 2001 11:38:46 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: Re: popper.core dumps with outlook express
> > qpopper version (i think) is 3.0 Im not sure how to check my version.
>
>
>Either:
>
>1) popper -v
>
>or
>
>
>2) telnet pop.host 110
> quit
ok it's
3.0.2-krb-IV
--
------------------------------------------------------------------------
Jeff Donovan Network Analyst
Bethlehem Area School District Network Operations
Bethlehem, PA 18020 (610) 807-5571 jdonovan at beth.k12.pa dot us
Date: Wed, 23 May 2001 12:16:20 -0400
From: Jack Sasportas <jack at innovativeinternet dot com>
Subject: Re: popper.core dumps with outlook express
I have read many issues on the list with 4 and so I would recommend 3.12,
until you get a chance to setup on a test box the 4.x version....
3.12 will resolve many small issues that may address your problem.
Jack
Jeff Donovan wrote:
> How much trouble would it be do upgrade?
> would anything get lost. This machine hosts many many accounts.
>
> --jeff
>
> >
> >Not sure about your specific problem, but it's probably best
> >to upgrade to 4.0.2 anyway.
>
> --
> ------------------------------------------------------------------------
> Jeff Donovan Network Analyst
> Bethlehem Area School District Network Operations
> Bethlehem, PA 18020 (610) 807-5571 jdonovan at beth.k12.pa dot us
--
___________________________________________________________
Jack Sasportas
Innovative Internet Solutions
Phone 305.665.2500
Fax 305.665.2551
www.innovativeinternet.com
www.web56.net
Date: Wed, 23 May 2001 11:58:01 -0400
From: Jack Sasportas <jack at innovativeinternet dot com>
Subject: Re: popper.core dumps with outlook express
I would recommend to upgrade at least to 3.12.....
Jeff Donovan wrote:
> Greetings
>
> I have a problem with outlook express clients (like who doesn't)
> running on a macintosh system.
> My server is BSDi 4.2 patched up to M20-010.
>
> qpopper version (i think) is 3.0 Im not sure how to check my version.
>
> the clients are not able to pop into their maillboxes and receive
> their mail. when the pop in, i get a popper.core file in the / dir.
>
> I may have to update to the latest qPopper unless there is a better
> remedy, I have never done this before so if anyone could point me to
> a nice README file that will show me how to do this without killing
> the thousands of other functioning accounts I have.
>
> TIA
>
> --jeff
> --
> ------------------------------------------------------------------------
> Jeff Donovan Network Analyst
> Bethlehem Area School District Network Operations
> Bethlehem, PA 18020 (610) 807-5571 jdonovan at beth.k12.pa dot us
--
___________________________________________________________
Jack Sasportas
Innovative Internet Solutions
Phone 305.665.2500
Fax 305.665.2551
www.innovativeinternet.com
www.web56.net
Date: Wed, 23 May 2001 12:05:25 -0400
From: Joseph S D Yao <jsdy at cospo.osis dot gov>
Subject: Re: popper.core dumps with outlook express
On Wed, May 23, 2001 at 11:40:14AM -0400, Jeff Donovan wrote:
> How much trouble would it be do upgrade?
> would anything get lost. This machine hosts many many accounts.
I just upgraded. I changed one file - the executable. I made sure
that I used all my 'configure' arguments from the previous build. The
account names are stored in the /etc/passwd file, so nothing is lost.
--
Joe Yao jsdy at cospo.osis dot gov - Joseph S. D. Yao
OSIS Center Computer Support EMT-B
-----------------------------------------------------------------------
This message is not an official statement of COSPO policies.
Date: Wed, 23 May 2001 13:09:45 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: Re: popper.core dumps ::THANKS::
Thanks to all who replied.
the upgrade sounds simple enough.
--jeff
--
------------------------------------------------------------------------
Jeff Donovan Network Analyst
Bethlehem Area School District Network Operations
Bethlehem, PA 18020 (610) 807-5571 jdonovan at beth.k12.pa dot us
Date: Wed, 23 May 2001 08:09:53 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Re: popper.core dumps with outlook express
On Wed, May 23, 2001 at 11:40:14AM -0400, Jeff Donovan wrote:
> How much trouble would it be do upgrade?
> would anything get lost. This machine hosts many many accounts.
Past qpopper upgrades (both minor versions and major versions) have
been pretty painless in my experience.
-- Clifton
--
Clifton Royston -- LavaNet Systems Architect -- cliftonr at lava dot net
WWJD? "JWRTFM!" - Scott Dorsey (kludge) "JWG" - Eddie Aikau
Date: Wed, 23 May 2001 14:39:19 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: Patch for qpopper4.0.2 to set server-mode based on user
At 5:12 PM -1000 5/22/01, Clifton Royston wrote:
> Changing the login group for thousands of
> users, so that we could use the group-based server mode selection
> seemed likely to cause problems.
There's no need to change the login group. Secondary group is
fine. The user only needs to be a member of the group.
Also, you can use per-user configuration files to set server mode
on or off for specific users. You can omit files for users with
nothing to set. So you could have server mode default to on, but
be turned off for specific users.
Date: Wed, 23 May 2001 14:42:24 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: popper.core dumps with outlook express
At 12:16 PM -0400 5/23/01, Jack Sasportas wrote:
> I have read many issues on the list with 4 and so I would recommend 3.12,
> until you get a chance to setup on a test box the 4.x version....
The only issue that comes up with 4 is increased timeouts on
congested networks, because of 4's default of grouping small writes
into large packets. This can be avoided by setting chuinky-writes
to never.
Date: Wed, 23 May 2001 12:20:46 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Re: Patch for qpopper4.0.2 to set server-mode based on user shell
On Wed, May 23, 2001 at 02:39:19PM -0700, Randall Gellens wrote:
> At 5:12 PM -1000 5/22/01, Clifton Royston wrote:
>
> > Changing the login group for thousands of
> > users, so that we could use the group-based server mode selection
> > seemed likely to cause problems.
>
> There's no need to change the login group. Secondary group is
> fine. The user only needs to be a member of the group.
At least on the BSD systems we use, I *believe* we have found problems
in the past with putting more than <mumble> users into a single given
secondary group. I forget what the exact number was, maybe several
hundred. (This may have been a bug resulting from some maximum buffer
length or line size in the login program parsing the /etc/groups file
at login time.) We could certainly have worked around this by using
different groups for different subsets of the shell users, but that was
starting to look a bit convoluted and hard to maintain if shell access
gets added or revoked
It just seemed logical to me that if the main contraindication for
server_mode is shell access, I should implement a way of making
server_mode hinge on shell access. At this site, at least, it looks
like it will make configuration very simple.
But of course it's your call as to whether it should be integrated into
the main line of code.
-- Clifton
--
Clifton Royston -- LavaNet Systems Architect -- cliftonr at lava dot net
WWJD? "JWRTFM!" - Scott Dorsey (kludge) "JWG" - Eddie Aikau
Date: Wed, 23 May 2001 17:06:57 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: Patch for qpopper4.0.2 to set server-mode based on user
At 12:20 PM -1000 5/23/01, Clifton Royston wrote:
> It just seemed logical to me that if the main contraindication for
> server_mode is shell access, I should implement a way of making
> server_mode hinge on shell access. At this site, at least, it looks
> like it will make configuration very simple.
>
> But of course it's your call as to whether it should be integrated into
> the main line of code.
If people find it useful, I'm all for including it. At this point
I don't have much of a feel either way for how useful it would be.
From: "Kenneth Porter" <shiva at well dot com>
Date: Thu, 24 May 2001 01:52:12 -0700
Subject: Re: Patch for qpopper4.0.2 to set server-mode based on user shell
On Wed, 23 May 2001 17:06:57 -0700, Randall Gellens wrote:
>If people find it useful, I'm all for including it. At this point
>I don't have much of a feel either way for how useful it would be.
I've got a small user base (about 75) with about 2/3 being mail-only,
no shell. Not enough load to justify enabling server mode, but if I had
lots of users, it would be nice to be able to trivially turn on server
mode just by filtering on a shell of "/POPPER/ANY/SHELL", which I use
for mail-only users.
Ken
mailto:shiva at well dot com
http://www.sewingwitch.com/ken/
[If answering a mailing list posting, please don't cc me your reply. I'll take my answer on the list.]
From: James Eastwood <JamesEastwood at businessserve.co dot uk>
Subject: trimming username
Date: Thu, 24 May 2001 10:47:50 +0100
qpopper 4.01 running on bsdi 4.1
I am having problems trimming the @domain name from the username
i have tried setting popper -C in inetd.conf
and i have also "set trim-cd domain = true" in a config file
does anybody have any ideas where im going wrong
Date: Thu, 24 May 2001 14:47:23 -0400
From: "Doryce E . Moore" <demoore at ccsalpha3.nrl.navy dot mil>
Subject: Re: popper.core dumps ::THANKS::
At 1:09 PM -0400 5/23/01, Jeff Donovan wrote:
>Thanks to all who replied.
>
>the upgrade sounds simple enough.
>
>--jeff
>--
>------------------------------------------------------------------------
>Jeff Donovan Network Analyst
>Bethlehem Area School District Network Operations
>Bethlehem, PA 18020 (610) 807-5571 jdonovan at beth.k12.pa dot us
--
I would rename the old popper.exe to something like popper.3.02.orig so that
if anything goes wrong you can move it back.
Doryce
Date: Thu, 24 May 2001 15:12:13 -0400
From: Forrest Aldrich <forrie at forrie dot com>
Subject: Feature request
What about a connection rate throttle.... this can be done through xinetd,
I know. But it might be useful in the daemon itself.
Date: Thu, 24 May 2001 13:45:24 -1000
From: Robert Brewer <rbrewer at lava dot net>
Subject: Re: Patch for qpopper4.0.2 to set server-mode based on user shell
--On Wednesday, May 23, 2001 5:06 PM -0700 Randall Gellens
<randy at qualcomm dot com> wrote:
> At 12:20 PM -1000 5/23/01, Clifton Royston wrote:
>
>> It just seemed logical to me that if the main contraindication for
>> server_mode is shell access, I should implement a way of making
>> server_mode hinge on shell access. At this site, at least, it looks
>> like it will make configuration very simple.
>>
>> But of course it's your call as to whether it should be integrated into
>> the main line of code.
>
> If people find it useful, I'm all for including it. At this point I don't
> have much of a feel either way for how useful it would be.
I think Clifton already covered the basic points, but it's worth mentioning
that this patch is about to change the deployment of server mode from
something that we have been thinking about doing for years into something
that we can drop into place in a week's time.
Date: Fri, 25 May 2001 08:58:27 +0900
From: Peter Evans <peter at gol dot com>
Subject: Re: Patch for qpopper4.0.2 to s ...
> > At 12:20 PM -1000 5/23/01, Clifton Royston wrote:
> >> It just seemed logical to me that if the main contraindication for
> >> server_mode is shell access, I should implement a way of making
> >> server_mode hinge on shell access. At this site, at least, it looks
> >> like it will make configuration very simple.
However,
Most Shell users (there are still places that let people use shell!)
are unlikely to be using pop, so I am afraid that I find this
patch to be a solution for a problem that doesnt exist.
pop users pop,
shell users mutt/pine/elm/cur/mew etc.
Conflict only occurs when the shell user is dumb enough to pop at
the same time he pines. Not that dangerous usually, since the mailspool
is still there for the shelly process. Simple warning to shell uses should
avoid even that. "dont pop while reading email, you might lose something"
should be sufficient.
In the bad old days when gol did shell, we never had problems running
server mode, most of our problems were caused by users doing shelly
things like irc bots, running muds diguised as bash and trying to crack
other accounts on the same box.
P
----+
--
Remember The 5 K's. Don't forget to make your feng go schwee~
From: Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Subject: BUG REPORT: module popper.c, function getline()
Date: Mon, 28 May 2001 11:56:00 +0200
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi,
Some days ago I send you this bug report.
Have you seen something about it ?
[...]
Hi,
I'm using Qpopper 4.0.2 and I have found a bug in the module 'popper.c',
function 'getline()'.
The bug is in the line:
425: if ( pPOP->pcInEnd != pPOP->pcInStart && *p == '\n' ) {
The solution is modify the line this way:
OLD 425: if ( pPOP->pcInEnd != pPOP->pcInStart && *p == '\n' ) {
NEW 425: if ( p != pPOP->pcInEnd && *p == '\n' ) {
I'm going to explain the bug.
Its something hard to explain, but I'll do my best :-)
The problem appears when I try to test one of my pop users accounts using the
telnet client from Windows.
This telnet client sends one TCP packet for every keystroke.
The function getline stores one line from the client TCP connection in one
buffer. If this BUFFER IS NOT INITIALIZED it will contain RANDOM BYTES.
Imagin that the buffer has this random bytes:
--------------------
| a | | g | 1 | | F | k |\n | X | ....
--------------------
and that I type the line 'pass pass00' in my Windows telnet client.
When Qpopper calls the function getline() the status of the buffer and the
pointers will be:
--------------------
| a | | g | 1 | | F | k |\n | X | ....
--------------------
^------pPOP->pcInStart
^------pPOP->pcInEnd
^------p
After the first read system call over the connected socket:
526: nRead = read ( pPOP->input_fd, pPOP->pcInEnd, nRoom );
the status of the buffer and the pointers will be:
--------------------
| p | | g | 1 | | F | k |\n | X | ....
--------------------
^--------pPOP->pcInStart
^------pPOP->pcInEnd
^--------p
Here the Qpopper process has readed the first letter of the pass keyword from
the connected socket. Due that my telnet client sends one TCP packet for
every keystroke, when Qpopper issues the read system call over the socket, it
will read only one byte.
Next the program flow goes to the begining of the main while loop of the
getline() function and this is the code executed:
418: /*
419: * Look for line in our buffer
420: */
421: p = pPOP->pcInStart;
422: for ( ; p < pPOP->pcInEnd; p++ )
423: if ( *p == '\n' )
424: break;
After this loop the status will be:
--------------------
| p | | g | 1 | | F | k |\n | X | ....
--------------------
^--------pPOP->pcInStart
^------pPOP->pcInEnd
^------p
And then the code line in which the bug is will be executed:
OLD 425: if ( pPOP->pcInEnd != pPOP->pcInStart && *p == '\n' ) {
At this point we we can see the bug.
If you see the status of the p pointer, its POINTING A BYTE WITH RANDOM DATA
and we are using it as part of the if conditions of the code line with the
bug !!!!!!
If we follow the program, we will reach the status:
--------------------
| p | a | s | s | | p | a |\n | X | ....
--------------------
^--- -----------------pPOP->pcInStart
^------pPOP->pcInEnd
^------p
In this status the conditios for the if stament in the line 425:
OLD 425: if ( pPOP->pcInEnd != pPOP->pcInStart && *p == '\n' ) {
become both true, making that the getline() function finish returning the
line 'pass pa', instead of the line I typed 'pass pass00', and my Qpopper
finish giving the error '-ERR [AUTH] Password supplied for "user" is
incorrect.'.
How I have said, the solution to the bug is very simple, modify the line 425
this way:
OLD 425: if ( pPOP->pcInEnd != pPOP->pcInStart && *p == '\n' ) {
NEW 425: if ( p != pPOP->pcInEnd && *p == '\n' ) {
With this modification, if the p pointer points to the same random byte that
pPOP->pcInEnd, the first condition of the if stament is not true and the
function getline() continues until a '\n' caracter is read from the socket.
I have observed that the bug only appears with users with the length of its
password string greater than the length of its user name. I believe that this
is because the same buffer is used to store the POP request 'user username',
and the next 'pass password', but the buffer is not cleaned before the
receive of the pass POP request.
This problem is difficult to see because the email clients like Netscape,
Outlook, ... sends all the request in one TCP packet and when Qpopper calls
the read system call in the getline() function, it obtains all the bytes from
the request, including the '\n' byte. This way the bug never appears because
the getline function reads the '\n' byte in the request and, then, the p
pointer will never point to random data.
UFffff, I hope that my explanation will be clear enougth :-)
Well, this is my second bug report for the Qpopper development team :-)
Do you have any award or prize for acumulation of bug reporting ????? ;-)
[...]
Greetings.
- ---
Carles Xavier Munyoz Baldó / cmunyoz at es.wanadoo dot com
VAS - Experto en Sistemas IP
Wanadoo España - http://www.wanadoo.es/
Tel: +34 96 5040046 - Fax: +34 96 5040047
- ---
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8
iQA/AwUBOxIgsBAGkoZz8//aEQITCgCdGFM6Tw2YXxtn2KjeNKnTYVTK+kQAn22D
a6JzUziAwuzuw+e18WkHA21y
¨06
-----END PGP SIGNATURE-----
From: Joerg Hartmann <J.Hartmann at des.mediadesign dot de>
Subject: user can not get messages
Date: Mon, 28 May 2001 13:03:22 +0200
Hallo ,
some of my users on different linux-mailers have problems with qpopper (Version 3.1.2 and 4.0.1) from time to time.
The Mailclient (Outlook or Netscape) stops downloading the mails from the spool and hangs until timeout. Not for all users, only one or two from some dozends.
Sometimes this happens if the outlook of the user crashes, sometimes the mails works one day and stops the other.
Most of the time i can put the user back to live with mail if i delete the very first message in his spoolfile. But the is NO good solution ...
Dont know if the spool was coruppted - i can not see any problems
in the spoolfile and mutt opens them without errors.
In the server logs are I/O-Errors at the end of this sessions, but i dont understand what
"operation not permitted" means to me :-( in this environment.
popper[9981]: Running in server mode; spool not changed; deleting temp drop [pop_updt.c:313]
popper[9981]: +OK Pop server at mail.dus.mediadesign.de signing off. [popper.c:347]
popper[9981]: I/O error flushing output to client golschew at dus-kiste [192.168.120.18]: Operation not permitted (1) [pop_send.c:685]
popper[9981]: (v4.0.2) Ending request from "golschew" at (dus-kiste) 192.168.120.18 [popper.c:365]
And a second question:
popper[9991]: HDR (len=54; hangup=0; errno=2) *** X-Mailer: Microsoft Internet E-Mail/MAPI - 8.0.0.4 [pop_send.c:432]
popper[9991]: HDR (len=18; hangup=0; errno=2) *** MIME-Version: 1.0 [pop_send.c:432]
popper[9991]: HDR (len=47; hangup=0; errno=2) *** Content-Type: text/plain; charset="iso-8859-1" [pop_send.c:432]
popper[9991]: HDR (len=32; hangup=0; errno=2) *** Content-Transfer-Encoding: 8bit [pop_send.c:432]
popper[9991]: HDR (len•; hangup=0; errno=2) *** X-MIME-Autoconverted: from quoted-printable to 8bi [pop_send.c:432]
There is alway errno=2 at the header lines. What kind of error ist this ?
Can someone please point me to a documentation where i can learn about this ?
Ù@ù
Thanks for any help ...
Joerg
--
Joerg Hartmann
mediadesign akademie
Phone +49 340 253 2023 Fax +49 340 253 2010
J.Hartmann at des.mediadesign dot de
Date: Mon, 28 May 2001 07:46:34 -0700 (PDT)
From: Gregory Hicks <ghicks at cadence dot com>
Subject: Re: BUG REPORT: module popper.c, function getline()
Um-m-m-m... I'll start of with "I'm not a qpopper developer, but...",
and I'm not picking on you, but...
I've always found that, when testing software, one should test software
in the mode it was designed for... Ie: if the software is designed to
receive lines, to send it lines. If it designed to recognize character
at a time inputs AND line at a time, to test both. If designed for only
'character input', send only character inputs...
Qpopper is designed for software that sends "line" at a time input. Why
not test it in that mode? Most telnet implementations that I have seen
have a switch that allows line transmissions instead of character
transmissions...
Regards,
Gregory Hicks
---------------------------------------------------------------------
Gregory Hicks | Principal Systems Engineer
Cadence Design Systems | Direct: 408.576.3609
555 River Oaks Pkwy M/S 6B1 | Fax: 408.894.3479
San Jose, CA 95134 | Internet: ghicks at cadence dot com
> From: Carles Xavier Munyoz Baldó <carles.munyoz at ctv-jet dot com>
> Date: Mon, 28 May 2001 11:56:00 +0200
>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi,
> Some days ago I send you this bug report.
> Have you seen something about it ?
>
> [...]
> Hi,
> I'm using Qpopper 4.0.2 and I have found a bug in the module
'popper.c',
> function 'getline()'.
>
> The bug is in the line:
> 425: if ( pPOP->pcInEnd != pPOP->pcInStart && *p == '\n' ) {
>
> The solution is modify the line this way:
> OLD 425: if ( pPOP->pcInEnd != pPOP->pcInStart && *p == '\n' )
{
> NEW 425: if ( p != pPOP->pcInEnd && *p == '\n' ) {
>
>
> I'm going to explain the bug.
> Its something hard to explain, but I'll do my best :-)
>
> The problem appears when I try to test one of my pop users accounts
> using the telnet client from Windows.
> This telnet client sends one TCP packet for every keystroke.
[...snip rest of explanation...]
From: Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Subject: Re: BUG REPORT: module popper.c, function getline()
Date: Mon, 28 May 2001 17:40:57 +0200
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Monday 28 May 2001 16:46, Gregory Hicks wrote:
> I've always found that, when testing software, one should test software
> in the mode it was designed for... Ie: if the software is designed to
> receive lines, to send it lines. If it designed to recognize character
> at a time inputs AND line at a time, to test both. If designed for only
> 'character input', send only character inputs...
Sorry, but I don't agree with your opinion.
I'm a software developer and when I develope my software I make it to be
prepared for any input it can receive.
You can not develope a software and wait that all users send you the input
in the format you have specified (not all the people is good user, in the net
exists hackers and other kinds of bad users :).
One of the most important tasks of a software developer is the syntax parse
of the inputs received from the users.
If this task is not done well, you will have problems like buffer overflows,
segmentation faults, security problems, etc...
Great part of the bugs reported to software developers like BIND, SENDMAIL,
QPOPPER, ..., are bugs about buffer overflows generated due to a bad input
parse.
Greetings.
- ---
Carles Xavier Munyoz Baldó / cmunyoz at es.wanadoo dot com
VAS - Experto en Sistemas IP
Wanadoo España - http://www.wanadoo.es/
Tel: +34 96 5040046 - Fax: +34 96 5040047
- ---
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8
iQA/AwUBOxJxihAGkoZz8//aEQKN6QCfQg3+MOMx+Rc0F0colYPO2Eea1aEAn1Ky
77SRusQeNScHKm4C9R6uWihC
=7/V6
-----END PGP SIGNATURE-----
Date: Mon, 28 May 2001 11:52:35 -0400 (EDT)
From: Homer Wilson Smith <homer at lightlink dot com>
Subject: Re: BUG REPORT: module popper.c, function getline()
> Qpopper is designed for software that sends "line" at a time input. Why
> not test it in that mode? Most telnet implementations that I have seen
> have a switch that allows line transmissions instead of character
> transmissions...
Since most telnet clients work in character at a time mode, and since
testing the popper with telnet clients is a common activity for us, as I
imagine it is for many ISP's, it would be nice to have a standard
character at a time input mode.
Homer
Date: Mon, 28 May 2001 08:50:49 -0700
From: Derek Balling <dredd at megacity dot org>
Subject: Re: BUG REPORT: module popper.c, function getline()
At 5:40 PM +0200 5/28/01, Carles Xavier Munyoz Baldó wrote:
>Sorry, but I don't agree with your opinion.
>I'm a software developer and when I develope my software I make it to be
>prepared for any input it can receive.
>You can not develope a software and wait that all users send you the input
>in the format you have specified (not all the people is good user, in the n
et
>exists hackers and other kinds of bad users :).
>
>One of the most important tasks of a software developer is the syntax parse
>of the inputs received from the users.
>If this task is not done well, you will have problems like buffer overflows
,
>segmentation faults, security problems, etc...
>Great part of the bugs reported to software developers like BIND, SENDMAIL,
>QPOPPER, ..., are bugs about buffer overflows generated due to a bad input
>parse.
The ultimate summation of the RFC's:
"Be Conservative in what you send, and liberal in what you accept."
If there's something in qpopper that breaks when its getting
1-char-per-packet type of thing, then it needs to get fixed, even if
that isn't the "normal mode of operation".
(caveat: I haven't been following this thread, but it sounds to me
like Carles has the right mindset)
D
--
+---------------------+-----------------------------------------+
| dredd at megacity dot org | "Conan! What is best in life?" |
| Derek J. Balling | "To crush your enemies, see them |
| | driven before you, and to hear the |
| | lamentation of their women!" |
+---------------------+-----------------------------------------+
From: "PuntoNet" <wilson at punto.net dot ec>
Subject: RE: user can not get messages
Date: Mon, 28 May 2001 10:30:57 -0500
I have the same problem with qpopper version 4.0.2 in a RedHat Linux 7.0
system
If I get mail (the same spool) from the LAN don´t exist this problem, but if
I try to get mail from a dial-up conexion the problem persist. --> I don´t
think that the spool is corrupted.
Is this a qpopper bug?
Thanks in advance.
Wilson Machasilla.
----- Original Message -----
From: Joerg Hartmann <J.Hartmann at des.mediadesign dot de>
To: Subscribers of Qpopper <qpopper at lists.pensive dot org>
Sent: Monday, May 28, 2001 6:03 AM
Subject: user can not get messages
> Hallo ,
>
> some of my users on different linux-mailers have problems with qpopper
(Version 3.1.2 and 4.0.1) from time to time.
> The Mailclient (Outlook or Netscape) stops downloading the mails from the
spool and hangs until timeout. Not for all users, only one or two from some
dozends.
>
> Sometimes this happens if the outlook of the user crashes, sometimes the
mails works one day and stops the other.
> Most of the time i can put the user back to live with mail if i delete the
very first message in his spoolfile. But the is NO good solution ...
> Dont know if the spool was coruppted - i can not see any problems
> in the spoolfile and mutt opens them without errors.
>
> In the server logs are I/O-Errors at the end of this sessions, but i dont
understand what
> "operation not permitted" means to me :-( in this environment.
>
> popper[9981]: Running in server mode; spool not changed; deleting temp
drop [pop_updt.c:313]
> popper[9981]: +OK Pop server at mail.dus.mediadesign.de signing off.
[popper.c:347]
> popper[9981]: I/O error flushing output to client golschew at dus-kiste
[192.168.120.18]: Operation not permitted (1) [pop_send.c:685]
> popper[9981]: (v4.0.2) Ending request from "golschew" at (dus-kiste)
192.168.120.18 [popper.c:365]
>
> And a second question:
>
> popper[9991]: HDR (len=54; hangup=0; errno=2) *** X-Mailer: Microsoft
Internet E-Mail/MAPI - 8.0.0.4 [pop_send.c:432]
> popper[9991]: HDR (len=18; hangup=0; errno=2) *** MIME-Version: 1.0
[pop_send.c:432]
> popper[9991]: HDR (len=47; hangup=0; errno=2) *** Content-Type:
text/plain; charset="iso-8859-1" [pop_send.c:432]
> popper[9991]: HDR (len=32; hangup=0; errno=2) ***
Content-Transfer-Encoding: 8bit [pop_send.c:432]
> popper[9991]: HDR (len•; hangup=0; errno=2) *** X-MIME-Autoconverted:
from quoted-printable to 8bi [pop_send.c:432]
>
> There is alway errno=2 at the header lines. What kind of error ist this ?
> Can someone please point me to a documentation where i can learn about
this ?
> Ù@ù
> Thanks for any help ...
> Joerg
> --
> Joerg Hartmann
> mediadesign akademie
> Phone +49 340 253 2023 Fax +49 340 253 2010
> J.Hartmann at des.mediadesign dot de
>
From: Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Subject: Re: BUG REPORT: module popper.c, function getline()
Date: Mon, 28 May 2001 18:10:11 +0200
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Monday 28 May 2001 17:50, Derek Balling wrote:
> If there's something in qpopper that breaks when its getting
> 1-char-per-packet type of thing, then it needs to get fixed, even if
> that isn't the "normal mode of operation".
This is the case.
When qpopper receives 1-char-per-packet it access a non initialized memory
array and may have no normal behavior because it uses the random bytes of the
array in a conditional stament.
This is not a very important bug because the email clients doesn't send
1-char-per-packet when stablish a POP connection and there is no buffer
overflow problems associated with this, but its a bug and must be corrected.
If this "non important" bug is not corrected, it may be converted in an
important bug in future versions of Qpopper.
Greetings.
- ---
Carles Xavier Munyoz Baldó / cmunyoz at es.wanadoo dot com
VAS - Experto en Sistemas IP
Wanadoo España - http://www.wanadoo.es/
Tel: +34 96 5040046 - Fax: +34 96 5040047
- ---
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8
iQA/AwUBOxJ4ZBAGkoZz8//aEQKXowCePi2yClSG8sbb2GrsWvwweo8t12AAoOc7
UaMErZg3BKmQV9KmjngtS4XC
=0ZbE
-----END PGP SIGNATURE-----
Date: Mon, 28 May 2001 12:41:02 -0700
From: Chuck Yerkes <chuck+qpopper at yerkes dot com>
Subject: Re: BUG REPORT: module popper.c, function getline()
Um, qpopper doesn't receive packets. The kernel, via the stack,
receive packets. While QPopper could receive and examine each
packet, it uses higher level system calls.
Should input from the sender be trusted? Hell, no. IMHO,
NO input should be used without examination. Program so that
you presume that inetd has been compromised.
But that packet -> input line conversion is done before the
application ever sees it. Whether your program sends 1
char/packet or a whole line, the Application has no awareness
of that.
Quoting Carles Xavier Munyoz Baldó (carles.munyoz at ctv-jet dot com):
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Monday 28 May 2001 16:46, Gregory Hicks wrote:
> > I've always found that, when testing software, one should test software
> > in the mode it was designed for... Ie: if the software is designed to
[...]
> One of the most important tasks of a software developer is the syntax parse
> of the inputs received from the users.
> If this task is not done well, you will have problems like buffer overflows,
> segmentation faults, security problems, etc...
From: "Eduardo A. Alcocer Rosado" <ealcocer at hotmail dot com>
Subject: compiling problems
Date: Mon, 28 May 2001 16:06:08 -0500
Sorry for my bad english..
I´m trying to install qpopper4.0.2 in Linux 6.2 but I´m having problems
with some libraries, I have 4 days ago trying to resolv it.
I executed the file "./configure", when I check the file config.cache I saw
that no found some libraries and the config.log tell me that "Failed program
was: " and show me some code lines, the numbers are:
3042, 3394, 3433, 3507, 3549, 3589, 3669, 3712, 3752, 3796, 4855
so I check that numbers in the "configure" file and saw the libraries wich
didn´t find. So I did a "find / -name <librarie.h> -print" and some
libraries
are in my server but not all of them. maybe are the enviroment variables?
when I use make, it generated the popper file and I have configured the
inetd.conf
file, the services file too but I never can did a telnet at the port 110.
I don´t know what to do.
the archive config.cache was :
ac_cv_func_pw_encrypt=${ac_cv_func_pw_encrypt='no'}
ac_cv_func_set_auth_parameters=${ac_cv_func_set_auth_parameters='no'}
ac_cv_func_setproctitle=${ac_cv_func_setproctitle='no'}
ac_cv_header_sys_netinet_in_h=${ac_cv_header_sys_netinet_in_h='no'}
ac_cv_header_sys_security_h=${ac_cv_header_sys_security_h='no'}
ac_cv_lib_dir_opendir=${ac_cv_lib_dir_opendir='no'}
ac_cv_lib_krb_krb_recvauth=${ac_cv_lib_krb_krb_recvauth='no'}
ac_cv_lib_mail_maillock=${ac_cv_lib_mail_maillock='no'}
ac_cv_lib_nsl_t_accept=${ac_cv_lib_nsl_t_accept='no'}
ac_cv_lib_posix4_clock_gettime=${ac_cv_lib_posix4_clock_gettime='no'}
ac_cv_lib_security_set_auth_parameters=${ac_cv_lib_security_set_auth_paramet
ers='no'}
ac_cv_lib_socket_socket=${ac_cv_lib_socket_socket='no'}
ac_cv_prog_cc_cross=${ac_cv_prog_cc_cross='no'}
Thanx on advance
Eduardo Alcocer
Date: Mon, 28 May 2001 18:28:39 -0300
From: applein <applein at tutopia.com dot br>
Subject: qpopper.config
I'm trying use qpopper.config file for all my users how to make it?
From: Joerg Hartmann <J.Hartmann at des.mediadesign dot de>
Subject: Re: user can not get messages
Date: Tue, 29 May 2001 09:05:29 +0200
Am Montag, 28. Mai 2001 17:30 schrieben Sie:
> I have the same problem with qpopper version 4.0.2 in a RedHat Linux
> 7.0 system
> If I get mail (the same spool) from the LAN don´t exist this problem,
> but if I try to get mail from a dial-up conexion the problem persist.
> I don´t think that the spool is corrupted.
I have this kind of problem on LAN traffic and on leased lines (T1).
So i dont think it's a speed/bandwith problem.
Someone pointed me to the size of the messages and wrote it could be
a M$ Outlook problem, but - i tried to get mail for such a problem user
on different machines with different mailclients (Outlook, OExpress,
Netscape, Eudora, Pegasus) without any luck.
I even copyed the problematic spool file to the end of my own mail
spool file and been hit by this sh**** :(
So i assume i could not only be the client, but i have no clue where to
search for the reason of the problem.
regards
Jörg
--
Joerg Hartmann
mediadesign akademie
Phone +49 340 253 2023 Fax +49 340 253 2010
J.Hartmann at des.mediadesign dot de
Subject: Outlook Express troubles following Qpopper 3 -> 4 upgrade.
Date: Tue, 29 May 2001 15:36:10 +0200 (MET DST)
From: Eric Luyten <Eric.Luyten at vub.ac dot be>
We upgraded Qualcomm popper version 3.1.2 to 4.0.2 last Friday
and had to help a number of Outlook Express users (other admins
have already posted the same on this mailing list).
In a number of cases (all ?) the situation could be normalised by
entering the user password in the appropriate OE configuration panel.
We found that a combination of empty password field and a checked
"remember password" box makes the client exhibit different behaviour
with Qpopper 4 server versus version 3.
Small illustration :
% telnet XXXX pop3
Trying [SNIP]...
Connected to XXXX.
Escape character is '^]'.
+OK QPOP (version 3.1.2) at XXXX starting.
user joeuser<CR>
+OK Password required for joeuser.
pass<CR> [several seconds wait]
-ERR [AUTH] Password supplied for "joeuser" is incorrect.
+OK Pop server at XXXX signing off.
Connection closed by foreign host.
% telnet YYYY pop3
Trying [SNIP]...
Connected to YYYY.
Escape character is '^]'.
+OK Qpopper (version 4.0.2) at YYYY starting.
user joeuser<CR>
+OK Password required for joeuser.
pass<CR> [several MINUTES wait]
-ERR POP timeout from YYYY
+OK Pop server at YYYY signing off.
Connection closed by foreign host.
Other data :
both servers are Solaris 7, shadow password files
Qpopper was compiled in server mode, no APOP or Kerberos, no TLS
(de)selecting "chunky-writes" makes no difference
Hope this helps,
Eric Luyten, Computing Centre VUB/ULB.
P.S. Moving from Qpopper 3 to 4 makes for half the CPU utilisation
and 60% less I/O on the mail spool partitions.
Great job, Randall et al.
From: "Christopher Crowley" <ccrowley at tulane dot edu>
Subject: Outlook Express - - Qpopper - - Openssl
Date: Tue, 29 May 2001 09:51:28 -0500
Anyone experienced this problem?
I have Qpopper built with OpenSSL support.
When using Qpopper configuration directive:
set tls-support = alternate-port
I am using a server cert and key (signed by Verisign) that was on the server
already for another use.
I can connect to the server using the openssl s_client (see below for
handshake transcript). I can not connect using Outlook Express.
- - - - - - - - - - -
When using Qpopper configuration directive:
set tls-support = stls (which is supposed to allow plain and encrypted
"negotiations on the standard (or any) port...")
I can access using Outlook Express on the port I started qpopper on:
/export/home/root/src/qpopper4*/popper/popper 8110 -f
/etc/mail/pop/qpopper.config -l 1
Date: Tue, 29 May 2001 08:48:55 -0700
From: "Eduardo E. Silva" <esilva at silvex dot com>
Subject: Re: Outlook Express troubles following Qpopper 3 -> 4 upgrade.
This is a cryptographically signed message in MIME format.
--------------msAEEA876A939F4292B16B42E5
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Try this:
make clean
./configure --enable-specialauth <-- This enables /etc/shadow usage.
make
Eric Luyten wrote:
> We upgraded Qualcomm popper version 3.1.2 to 4.0.2 last Friday
> and had to help a number of Outlook Express users (other admins
> have already posted the same on this mailing list).
>
> In a number of cases (all ?) the situation could be normalised by
> entering the user password in the appropriate OE configuration panel.
>
> We found that a combination of empty password field and a checked
> "remember password" box makes the client exhibit different behaviour
> with Qpopper 4 server versus version 3.
>
> Small illustration :
>
> % telnet XXXX pop3
> Trying [SNIP]...
> Connected to XXXX.
> Escape character is '^]'.
> +OK QPOP (version 3.1.2) at XXXX starting.
> user joeuser<CR>
> +OK Password required for joeuser.
> pass<CR> [several seconds wait]
> -ERR [AUTH] Password supplied for "joeuser" is incorrect.
> +OK Pop server at XXXX signing off.
> Connection closed by foreign host.
>
> % telnet YYYY pop3
> Trying [SNIP]...
> Connected to YYYY.
> Escape character is '^]'.
> +OK Qpopper (version 4.0.2) at YYYY starting.
> user joeuser<CR>
> +OK Password required for joeuser.
> pass<CR> [several MINUTES wait]
> -ERR POP timeout from YYYY
> +OK Pop server at YYYY signing off.
> Connection closed by foreign host.
>
> Other data :
> both servers are Solaris 7, shadow password files
> Qpopper was compiled in server mode, no APOP or Kerberos, no TLS
> (de)selecting "chunky-writes" makes no difference
>
> Hope this helps,
> Eric Luyten, Computing Centre VUB/ULB.
>
> P.S. Moving from Qpopper 3 to 4 makes for half the CPU utilisation
> and 60% less I/O on the mail spool partitions.
> Great job, Randall et al.
--------------msAEEA876A939F4292B16B42E5
Content-Type: application/x-pkcs7-signature; name="smime.p7s"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="smime.p7s"
Content-Description: S/MIME Cryptographic Signature
--------------msAEEA876A939F4292B16B42E5--
From: Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Subject: Re: Outlook Express troubles following Qpopper 3 -> 4 upgrade.
Date: Tue, 29 May 2001 18:19:28 +0200
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On Tuesday 29 May 2001 15:36, Eric Luyten wrote:
> % telnet YYYY pop3
> Trying [SNIP]...
> Connected to YYYY.
> Escape character is '^]'.
> +OK Qpopper (version 4.0.2) at YYYY starting.
> user joeuser<CR>
> +OK Password required for joeuser.
> pass<CR> [several MINUTES wait]
> -ERR POP timeout from YYYY
> +OK Pop server at YYYY signing off.
> Connection closed by foreign host.
Hi,
I have seen that the problem is that Qpopper 4.0.2 ignores the POP pass
command with empty string and waits for a new pass command, but Qpopper 4.0.2
doesn't shows any warning message about this.
I have discovered this using this test:
% telnet YYYY pop3
Trying [SNIP]...
Connected to YYYY.
Escape character is '^]'.
+OK Qpopper (version 4.0.2) at YYYY starting.
user joeuser<CR>
+OK Password required for joeuser.
pass<CR>
At this point if you wait several minutes, the Qpopper will show the timeout
error message because it is waiting for input and it has received nothing in
the timeout seconds.
If, at this point, you write this:
nocommand<CR>
-ERR Unknown command: "nocommand".
you get an answer from Qpopper POP3 server. This indicates us that the server
is waiting for input and that it has ignored the previous pass command.
Greetings.
- ---
Carles Xavier Munyoz Baldó / cmunyoz at es.wanadoo dot com
VAS - Experto en Sistemas IP
Wanadoo España - http://www.wanadoo.es/
Tel: +34 96 5040046 - Fax: +34 96 5040047
- ---
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8
iQA/AwUBOxPMEBAGkoZz8//aEQKDQgCfbYi4lGSLqWr7jtiEkuFk0RR/ibAAoPR1
rjT32IEI283FSyF6xYe7Ga2k
=56p9
-----END PGP SIGNATURE-----
Date: Tue, 29 May 2001 10:25:47 -0700
From: Roy <garlic at garlic dot com>
Subject: Re: Outlook Express troubles following Qpopper 3 -> 4 upgrade.
If you enter the pass command followed by four blanks, it works correctly. Its
some sort of bug in pop_parse.c in the special handling of the pass command for
the special case of nothing after the pass command.
Carles Xavier Munyoz Baldó wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Tuesday 29 May 2001 15:36, Eric Luyten wrote:
> > % telnet YYYY pop3
> > Trying [SNIP]...
> > Connected to YYYY.
> > Escape character is '^]'.
> > +OK Qpopper (version 4.0.2) at YYYY starting.
> > user joeuser<CR>
> > +OK Password required for joeuser.
> > pass<CR> [several MINUTES wait]
> > -ERR POP timeout from YYYY
> > +OK Pop server at YYYY signing off.
> > Connection closed by foreign host.
>
> Hi,
> I have seen that the problem is that Qpopper 4.0.2 ignores the POP pass
> command with empty string and waits for a new pass command, but Qpopper 4.0.2
> doesn't shows any warning message about this.
>
> I have discovered this using this test:
> % telnet YYYY pop3
> Trying [SNIP]...
> Connected to YYYY.
> Escape character is '^]'.
> +OK Qpopper (version 4.0.2) at YYYY starting.
> user joeuser<CR>
> +OK Password required for joeuser.
> pass<CR>
> At this point if you wait several minutes, the Qpopper will show the timeout
> error message because it is waiting for input and it has received nothing in
> the timeout seconds.
> If, at this point, you write this:
> nocommand<CR>
> -ERR Unknown command: "nocommand".
> you get an answer from Qpopper POP3 server. This indicates us that the server
> is waiting for input and that it has ignored the previous pass command.
>
> Greetings.
> - ---
> Carles Xavier Munyoz Baldó / cmunyoz at es.wanadoo dot com
> VAS - Experto en Sistemas IP
> Wanadoo España - http://www.wanadoo.es/
> Tel: +34 96 5040046 - Fax: +34 96 5040047
> - ---
> -----BEGIN PGP SIGNATURE-----
> Version: PGP 6.5.8
>
> iQA/AwUBOxPMEBAGkoZz8//aEQKDQgCfbYi4lGSLqWr7jtiEkuFk0RR/ibAAoPR1
> rjT32IEI283FSyF6xYe7Ga2k
> =56p9
> -----END PGP SIGNATURE-----
Date: Tue, 29 May 2001 11:01:12 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Re: qpopper.config
On Mon, May 28, 2001 at 06:28:39PM -0300, applein wrote:
> I'm trying use qpopper.config file for all my users how to make it?
Are you asking how to make popper use it, or what should go in it?
If you want one file to apply to all the users, put -f and the name of
the file on the qpopper command line (in inetd.conf) like this:
popper -f /usr/local/etc/qpopper.config
If you want each user to have their own separate options file in the
mail spool directory, put -U on the command line like this:
popper -U
If you want each user to have their own separate options file in their
own home directory, put -u on the command line like this:
popper -u
If you want to know what should go in the file, there is no shortcut
there, you need to read the administrators manual enough to understand
what the options do.
-- Clifton
--
Clifton Royston -- LavaNet Systems Architect -- cliftonr at lava dot net
WWJD? "JWRTFM!" - Scott Dorsey (kludge) "JWG" - Eddie Aikau
Date: Tue, 29 May 2001 14:04:40 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: 3.0 to 4.02 upgrade went smoothly
Another testimonial for 4.0.2:
I cut in the new version 4.0.2 seamlessly this morning. Load is
dramatically reduced over what we were seeing with a patched 3.0, I'm
sure mostly due to our being able to enable server mode + fast updates.
Late last week we regularly saw the load average go to 100 on this
server, and it looked like it was heading that way just before 9:00am
when I cut the new version in. Since then the load has not gone over
22, and that was in the early stages after the cutover (when the UIDL
.cache files were still getting built.) Now it's back down to 7.
Zero (0) customer calls or support problems so far.
I was thinking that I'm getting more odd messages syslogged for
various client problems. However, after comparing to last week's logs,
I think it is just that I am looking closer, not that version 4.0.2 is
more "talkative" than the old one. There are two exceptions:
May 29 13:40:50 malasada akiltd[10184]: popper[21213]: noop has null function
May 29 13:40:53 malasada akiltd[10184]: popper[21213]: noop has null function
...
May 29 13:41:28 malasada popper[21938]: (null) at 66.8.255.197 (66.8.255.197): -
ERR Unknown authentication mechanism: MSN
May 29 13:41:28 malasada popper[21938]: (null) at 66.8.255.197 (66.8.255.197): -
ERR POP EOF or I/O Error
I don't think any of these messages were getting logged before,
specifically that "noop has null function" which is showing up for a
couple of users, but a *lot* for one of them. Is this something which
really needs to be logged? I assume it's some mail client sending
"noop" commands to keep the session from timing out.
Looking at last weeks logs, I do see that "Unknown authentication
mechanism" message, but they were mostly showing up with "NTLM" and
"twinkie" as parameters instead of "MSN". Any idea what this comes
from? The IP address there is not one of ours, and doesn't seem to
have valid reverse DNS.
-- Clifton
--
Clifton Royston -- LavaNet Systems Architect -- cliftonr at lava dot net
WWJD? "JWRTFM!" - Scott Dorsey (kludge) "JWG" - Eddie Aikau
Last updated on 29 May 2001 by Pensive Mailing List Admin