The qpopper list archive ending on 18 May 2001
Topics covered in this issue include:
1. Newbie question
David Bengtsson <david.bengtsson at framfab dot se>
Fri, 11 May 2001 17:10:49 +0200
2. Re: Unable to process From lines (envelope), change recognitionmodes
Roy <garlic at garlic dot com>
Fri, 11 May 2001 09:09:22 -0700
3. Newbie questions two.
Mark on GCI Server <mweisman at gci dot net>
Fri, 11 May 2001 08:54:17 -0800
4. Re: Newbie questions two.
"Jeremy C. Reed" <reed at wcug.wwu dot edu>
Fri, 11 May 2001 10:32:55 -0700 (PDT)
5. Re: Newbie questions two.
"Kenneth Porter" <shiva at well dot com>
Fri, 11 May 2001 10:41:01 -0700
6. Re: Unable to process From lines (envelope), change
"Michael D. Sofka" <sofkam at rpi dot edu>
Fri, 11 May 2001 13:39:37 -0400
7. Re: Question about qpopper and domain name.
Mark on GCI Server <mweisman at gci dot net>
Fri, 11 May 2001 22:41:22 -0800
8. Re: Question about qpopper and domain name.
"Kenneth Porter" <shiva at well dot com>
Mon, 14 May 2001 11:00:54 -0700
9. Poor font in documentation
"Kenneth Porter" <shiva at well dot com>
Mon, 14 May 2001 11:05:23 -0700
10. TLS/SSL shutdown fails
"Kenneth Porter" <shiva at well dot com>
Mon, 14 May 2001 19:29:19 -0700
11. Re: TLS/SSL shutdown fails
"Kenneth Porter" <shiva at well dot com>
Mon, 14 May 2001 20:08:20 -0700
12. PRNGD and Qpopper?
Jacques Distler <distler at golem.ph.utexas dot edu>
Tue, 15 May 2001 00:21:29 -0700
13.
"Wilson Machasilla" <wilson at punto.net dot ec>
Tue, 15 May 2001 09:28:53 -0500
14. Problem with timeout option in qpopper
"Wilson Machasilla" <wilson at punto.net dot ec>
Tue, 15 May 2001 09:42:07 -0500
15. Re: Problem with timeout option in qpopper
James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Tue, 15 May 2001 16:09:43 -0400 (EDT)
16. Option: CHECK_SHELL
Butch Kemper <kemper at tstar dot net>
Tue, 15 May 2001 16:15:42 -0500
17. Problem with attachments?
"A. M. Salim" <salim at localweb dot com>
Tue, 15 May 2001 17:45:34 -0400 (EDT)
18. Re: Problem with attachments?
"Kenneth Porter" <shiva at well dot com>
Tue, 15 May 2001 15:16:42 -0700
19. Re: Qpopper 4.02/OpenSSL 0.9.6a - TLS/Alt port error
Randall Gellens <randy at qualcomm dot com>
Tue, 15 May 2001 16:59:50 -0700
20. Re: Newbie question
Randall Gellens <randy at qualcomm dot com>
Tue, 15 May 2001 17:05:05 -0700
21. hashed mail directory
Valter Nordh <valter at che.chalmers dot se>
Wed, 16 May 2001 13:58:12 +0200
22. Re: Problem with attachments?
"A. M. Salim" <salim at localweb dot com>
Wed, 16 May 2001 10:40:31 -0400 (EDT)
23. RE: Problem with timeout option in qpopper
"Wilson Machasilla" <wilson at punto.net dot ec>
Wed, 16 May 2001 10:21:16 -0500
24. RE: Problem with timeout option in qpopper
Randall Gellens <randy at qualcomm dot com>
Wed, 16 May 2001 11:24:45 -0700
25. Re: Problem with attachments?
Randall Gellens <randy at qualcomm dot com>
Wed, 16 May 2001 11:21:36 -0700
26. Re: Problem with attachments?
"A. M. Salim" <salim at localweb dot com>
Wed, 16 May 2001 14:45:26 -0400 (EDT)
27. correction for Qpopper FAQ (regarding xinetd on Red Hat)
James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Wed, 16 May 2001 17:02:56 -0400 (EDT)
28. Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
"Kenneth Porter" <shiva at well dot com>
Wed, 16 May 2001 14:19:45 -0700
29. Re: Problem with attachments?
Randall Gellens <randy at qualcomm dot com>
Wed, 16 May 2001 14:21:14 -0700
30. authorisation sources
simakin <root at simakin.spb dot ru>
Thu, 17 May 2001 10:23:19 +0400
31. How to make it with APOP
"Andrey V. Pevnev" <andrey at mgul.ac dot ru>
Thu, 17 May 2001 13:06:54 +0400
32. Qpopper with PAM on Solaris
simakin <root at simakin.spb dot ru>
Thu, 17 May 2001 12:17:24 +0400
33. Problem with Netscape Messenger, Qpopper with MySQL patch
"=?iso-8859-1?Q?Søren_P._Skou?=" <sps at arrowhead dot dk>
Thu, 17 May 2001 14:34:03 +0200
34. RE: Problem with Netscape Messenger, Qpopper with MySQL patch - solved
"=?iso-8859-1?Q?Søren_P._Skou?=" <sps at arrowhead dot dk>
Thu, 17 May 2001 15:04:20 +0200
35. Re: Problem with Netscape Messenger, Qpopper with MySQL patch
=?iso-8859-1?Q?Guðbjörn?= Hreinsson <gsh at skima dot is>
Thu, 17 May 2001 13:17:33 +0000
36. What is the .user.cahe file for ?
Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Thu, 17 May 2001 19:15:34 +0200
37. can't get NIS mail aliases to work
"Steven J. Backus" <backus at math.utah dot edu>
Thu, 17 May 2001 12:47:37 -0600 (MDT)
38. Re: can't get NIS mail aliases to work
"Steven J. Backus" <backus at math.utah dot edu>
Thu, 17 May 2001 13:39:38 -0600 (MDT)
39. Re: can't get NIS mail aliases to work
"Jeremy C. Reed" <reed at wcug.wwu dot edu>
Thu, 17 May 2001 13:23:25 -0700 (PDT)
40. Re: authorisation sources
Randall Gellens <randy at qualcomm dot com>
Thu, 17 May 2001 14:14:22 -0700
41. Re: hashed mail directory
Chuck Yerkes <chuck+qpopper at yerkes dot com>
Thu, 17 May 2001 14:28:46 -0700
42. Re: What is the .user.cahe file for ?
Randall Gellens <randy at qualcomm dot com>
Thu, 17 May 2001 14:26:34 -0700
43. Unable to process From lines (envelope)...
Michael Brennen <mbrennen at fni dot com>
Fri, 18 May 2001 02:41:25 -0500 (CDT)
44. Re: can't get NIS mail aliases to work
"James Raftery" <jrtest at spec.ch.man.ac dot uk>
Fri, 18 May 2001 09:25:49 +0100
45. BUG REPORT: module popper.c, function getline()
Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Fri, 18 May 2001 17:56:43 +0200
46. Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Fri, 18 May 2001 17:19:47 -0400 (EDT)
47. Re: Unable to process From lines (envelope)...
"Kenneth Porter" <shiva at well dot com>
Fri, 18 May 2001 14:25:20 -0700
48. Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
"Kenneth Porter" <shiva at well dot com>
Fri, 18 May 2001 14:44:10 -0700
49. kernel error
"Dan Trainor" <dan at concept-factory dot com>
Fri, 18 May 2001 16:06:40 -0700
50. Re: Unable to process From lines (envelope)...
Michael Brennen <mbrennen at fni dot com>
Fri, 18 May 2001 18:27:55 -0500 (CDT)
From: David Bengtsson <david.bengtsson at framfab dot se>
Subject: Newbie question
Date: Fri, 11 May 2001 17:10:49 +0200
Dear gurus,
A simple question. How do I change the domain name provided by Qpopper when
I telnet to port 110
i.e, I want my Qpopper to respond '+OK Qpopper (version 4.0.2) at
mail.mydomain.com starting.' Right now it reponds '+OK Qpopper (version
4.0.2) at mail.wrongdomain.com starting.'
Blue Skies,
/David
Date: Fri, 11 May 2001 09:09:22 -0700
From: Roy <garlic at garlic dot com>
Subject: Re: Unable to process From lines (envelope), change recognitionmodes
The better idea is to move to the sendmail that is available via sendmail.org.
The only IBM feature you lose is the ability to use "src" to control the
daemon.
Sendmail complies cleanly for me using the IBM C Compiler.
If you don't want to recompile each time, use the precompiled versions
available via Bull.
Steve Sullivan wrote:
> We're running qpopper 3.1.2 on an AIX 4.3.3 system. Recently, we have
> started getting reports of the From line message error from Eudora users (a
> ratio of about 1 out of every 2500 mails received). We found that a CTRL-@
> (unprintable) was the first character of the users mail file.
>
> Eventually, we found some one character mail files containing CTRL-@. By
> looking at the popper logs and mail logs, I found that the time of the file
> corresponded to the time sendmail was delivering mail to that user. The
> mail (from both internal and external sources) was not delivered to the
> user. popper logs don't show activity for that user at that time.
>
> Since we use the sendmail supplied by IBM, we decided to talk to
> them. After two days of explaining the symptoms, IBM still says the
> problem is popper, not bellmail (which actually writes the file). They
> suggest that popper and bellmail may handle file locking in different and
> incompatible ways. Their only suggestion is to run the IBM popper, but the
> performance hit makes me wary of this suggestion.
>
> Has this type of problem been reported before? I found the message in the
> FAQ, but it deals with procmail, which we're not using.
>
> Can you tell me how popper handles file locking, and do you know if it's
> compatible with the way bellmail handles file locking? Do you have any
> suggestions on solving this problem?
>
> Steve
Date: Fri, 11 May 2001 08:54:17 -0800
Subject: Newbie questions two.
From: Mark on GCI Server <mweisman at gci dot net>
Hey all,
Where exactly does the qpopper pull the domain name information. Does it
pull from active zones in the DNS, or from the setup of sendmail? Currently,
the domain name on my mail server includes the hostname, and I don't want
the hostname in the mail addresses that I set up for my users. Any help
would be greatly appreciated.
Thanks,
Mark
Date: Fri, 11 May 2001 10:32:55 -0700 (PDT)
From: "Jeremy C. Reed" <reed at wcug.wwu dot edu>
Subject: Re: Newbie questions two.
On Fri, 11 May 2001, Mark on GCI Server wrote:
> Where exactly does the qpopper pull the domain name information. Does it
What "domain name information"? (Do you have an example?)
> pull from active zones in the DNS, or from the setup of sendmail? Currently,
> the domain name on my mail server includes the hostname, and I don't want
> the hostname in the mail addresses that I set up for my users. Any help
> would be greatly appreciated.
That is an mail transfer agent issue. See the documentation for your mail
server. If you use sendmail, then look at the
http://www.sendmail.org/m4/masquerading.html and in your sendmail.cf look
for "DM" and define as needed.
Jeremy C. Reed
http://www.reedmedia.net/
http://bsd.reedmedia.net/ -- BSD news and resources
http://www.isp-faq.com/ -- find answers to your questions
From: "Kenneth Porter" <shiva at well dot com>
Date: Fri, 11 May 2001 10:41:01 -0700
Subject: Re: Newbie questions two.
On Fri, 11 May 2001 08:54:17 -0800, Mark on GCI Server wrote:
>Where exactly does the qpopper pull the domain name information. Does it
>pull from active zones in the DNS, or from the setup of sendmail? Currently,
>the domain name on my mail server includes the hostname, and I don't want
>the hostname in the mail addresses that I set up for my users.
Qpopper has no need to know its hostname. It just feeds from the local
mailspool.
The mail address of your users is only relevant to those who send mail
to your users and to the system used to route mail to them. If your MTA
is sendmail, you need to add the domain name you want to use to
/etc/mail/local-host-names. This tells sendmail that it should accept
mail for that domain. You'll need to create a DNS record like
"desired-domainname MX 0 mail-host-name". This tells others that mail
for user@desired-domainname should be delivered to mail-host-name.
BTW, please put something more informative in the subject line. Some of
us scan hundreds of messages a day and are likely to skip a message
with a boring subject line. You'll get a better response if you "sell"
your question a bit.
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: Fri, 11 May 2001 13:39:37 -0400
From: "Michael D. Sofka" <sofkam at rpi dot edu>
Subject: Re: Unable to process From lines (envelope), change
At 09:09 AM 5/11/2001 -0700, Roy wrote:
>The better idea is to move to the sendmail that is available via sendmail.org.
>The only IBM feature you lose is the ability to use "src" to control the
>daemon.
You don't even loose that.
chssys -s sendmail -f 9 -n 15 -S
changes src to use signals instead of sockets when stopping sendmail. Then
start sendmail with:
startsrc -s sendmail -a "-bD -q15m"
You'll have to change /etc/rc.tcpip to use -bD instead of -bd.
Mike
--
Michael D. Sofka sofkam at rpi dot edu
CIS/SSS Sr. Systems Programmer email, webmail, listproc, TeX, epistemology.
Rensselaer Polytechnic Institute, Troy, NY. http://www.rpi.edu/~sofkam/
Date: Fri, 11 May 2001 22:41:22 -0800
Subject: Re: Question about qpopper and domain name.
From: Mark on GCI Server <mweisman at gci dot net>
I got your answer, and it all made sense until I tried to implement it. I'm
running qpopper on a Redhat Linux box, and all I see in the /etc/mail folder
is "domaintable" and "domaintable.db", mailertable and mailertable.db,
access and access.db. I am unable to locate any other file with host info.
I've semi configured the sendmail MTA through the Linuxconf file. I'm not
sure what to do regarding the removal of the hostname from the email
address. Any help would be greatly appreciated.
Thanks,
Mark
From: "Kenneth Porter" <shiva at well dot com>
Date: Mon, 14 May 2001 11:00:54 -0700
Subject: Re: Question about qpopper and domain name.
On Fri, 11 May 2001 22:41:22 -0800, Mark on GCI Server wrote:
>I've semi configured the sendmail MTA through the Linuxconf file.
This is Very Bad. Linuxconf will wreck your sendmail.cf file. Its
sendmail module was written for sendmail-8.8, and that format of .cf
file is broken for a sendmail-8.9 setup.
The proper way to configure sendmail is to copy and edit redhat.mc (in
/usr/lib/sendmail-cf/cf), compile it with Make, and copy the resulting
.cf file to /etc/sendmail.cf.
You should ask your question on comp.mail.sendmail after recreating
your .cf file from a .mc file, and post your .mc file.
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: "Kenneth Porter" <shiva at well dot com>
Date: Mon, 14 May 2001 11:05:23 -0700
Subject: Poor font in documentation
I got burned this weekend by enabling the TLS option in qpopper 4 with
"-I 1". The font chosen for option characters in the PDF file is
sans-serif, so "-l" looks like "-I", and I ended up inadvertantly
leaving my server inoperable for the weekend. (Should've tested it, but
was rushing to get out of the office Friday.) Please use a serif font
(eg. Courier) for any text intended to be machine-read.
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: "Kenneth Porter" <shiva at well dot com>
Date: Mon, 14 May 2001 19:29:19 -0700
Subject: TLS/SSL shutdown fails
I'm trying to get TLS working. Using "-l 1", if I don't switch to TLS
mode, I get a message in the log saying that SSL_shutdown failed.
It appears from the OpenSSL documetation that SSL_shutdown should only
be called if an SSL connection is actually established, but not if only
SSL_init was called. You can check p->tls_started to see whether this
is necessary, as this is set if pop_tls_handshake was successful.
For those interested in trying this out, using the OpenSSL package
(http://www.openssl.org): I created a self-signed cert using "CA.pl
-newcert", moved the new file newcert.pem to /etc/mail/certs, and
removed the passphrase using "openssl rsa -in newcert.pem -out
qpopper.pem". I then cut the certificate section from newcert.pem and
pasted it into qpopper.pem. I created /etc/mail/qpopper.conf containing
"set tls-server-cert-file = /etc/mail/certs/qpopper.pem". I changed the
qpopper command to add "-l 1 -f /etc/mail/qpopper.conf".
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: "Kenneth Porter" <shiva at well dot com>
Date: Mon, 14 May 2001 20:08:20 -0700
Subject: Re: TLS/SSL shutdown fails
On Mon, 14 May 2001 19:29:19 -0700, Kenneth Porter wrote:
>I'm trying to get TLS working. Using "-l 1", if I don't switch to TLS
>mode, I get a message in the log saying that SSL_shutdown failed.
This message was "OpenSSL Error during shutdown", using qpopper-4.0.2.
I meanwhile set up a Eudora 5.1 client to force alternate port, added
the self-signed cert to its list of trusted servers (using the "last
TLS session" button next to the session type listbox), and now I get
"TLS shutdown error". The session otherwise seems to go ok. No details
are given for this error. The qpopper source suggests that this is a
failed system call inside OpenSSL.
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: Tue, 15 May 2001 00:21:29 -0700
From: Jacques Distler <distler at golem.ph.utexas dot edu>
Subject: PRNGD and Qpopper?
-----BEGIN PGP SIGNED MESSAGE-----
So, as far as I can tell, if configure does not find /dev/urandom,
qpopper 4.0.x reverts to its own home-grown random number generation.
Is there any way to get it to use PRNGD,
<http://www.aet.tu-cottbus.de/personen/jaenicke/postfix_tls/prngd.html>
instead?
If not, consider that a feature-request.
JD
-----BEGIN PGP SIGNATURE-----
Version: PGP
Comment: Public Key - http://golem.ph.utexas.edu/~distler/distler.asc
iQCVAwUBOwDZCqIBi34rsX+ZAQHtywQAjr3cci4bldFD0MyBm/vnhixdQ8rgSvLW
qFJI94xIZb9Zh4VgEQPOY2bONqqSWVXRkPefNHlj7JEPHya/yagX5NRxJLghuZxQ
GeZAaR17aa6X/ezazMP3wa6iEH4JJWExNXa1uMN7TaIuJaj8/dIiLpMbenYMo0nN
yuN4ky6hVX4
=qHzC
-----END PGP SIGNATURE-----
From: "Wilson Machasilla" <wilson at punto.net dot ec>
Subject:
Date: Tue, 15 May 2001 09:28:53 -0500
This is a multi-part message in MIME format.
------=_NextPart_000_0041_01C0DD21.74E38160
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hello friends.
I have a problem with qpopper-4.0.2 in the timeout option. It don´t
work!!!
I have a Linux Red Hat 7.0 System in a i686.
the /etc/xinetd.conf is:
service pop3
{
socket_type = stream
protocol = tcp
wait = no
user = root
server = /usr/local/sbin/popper
server_args = qpopper -T 60
port = 110
}
I have changed the -T option with 10, 60, 120, 180, 240 second and I get
the same result. I tried with server-mode, standalone-mode options in
the ./configure and nothing.
The lock file (.username.pop) when I crash a dial connection keeps for
10 to 20 minutes approximately (always).
I have many clients with poor dial-up connections.
Any ideas???
Thanks in advance
Wilson Machasilla
------=_NextPart_000_0041_01C0DD21.74E38160
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=iso-8859-1"
http-equiv=Content-Type>
<META content="MSHTML 5.00.2614.3500" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>Hello friends.</FONT></DIV>
<DIV><FONT face=Arial size=2>I have a problem with qpopper-4.0.2 in
the timeout
option. It don´t work!!!</FONT></DIV>
<DIV><FONT face=Arial size=2>I have a Linux Red Hat 7.0 System in a
i686.</FONT></DIV>
<DIV><FONT face=Arial size=2>the /etc/xinetd.conf is:</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>service pop3<BR>
{<BR>
socket_type =
stream<BR>
protocol =
tcp<BR>
wait
=
no<BR>
user
=
root<BR>
server =
/usr/local/sbin/popper<BR>
server_args = qpopper
-T 60<BR>
port
=
110</FONT></DIV>
<DIV><FONT face=Arial size=2> }</FONT></DIV>
<DIV><FONT face=Arial size=2>I have changed the -T option with 10,
60, 120, 180,
240 second and I get the same result. I tried with server-mode,
standalone-mode
options in the ./configure and nothing. </FONT></DIV>
<DIV><FONT face=Arial size=2>The lock file (.username.pop) when I
crash a dial
connection keeps for 10 to 20 minutes approximately
(always).</FONT></DIV>
<DIV><FONT face=Arial size=2>I have many clients with poor dial-up
connections.</FONT></DIV>
<DIV><FONT face=Arial size=2>Any ideas???</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Thanks in advance</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Wilson Machasilla</FONT></DIV>
<DIV> </DIV></BODY></HTML>
------=_NextPart_000_0041_01C0DD21.74E38160--
From: "Wilson Machasilla" <wilson at punto.net dot ec>
Subject: Problem with timeout option in qpopper
Date: Tue, 15 May 2001 09:42:07 -0500
This is a multi-part message in MIME format.
------=_NextPart_000_004E_01C0DD23.4E5A2BA0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hello friends.
I have a problem with qpopper-4.0.2 in the timeout option. It don´t
work!!!
I have a Linux Red Hat 7.0 System in a i686.
the /etc/xinetd.conf is:
service pop3
{
socket_type = stream
protocol = tcp
wait = no
user = root
server = /usr/local/sbin/popper
server_args = qpopper -T 60
port = 110
}
I have changed the -T option with 10, 60, 120, 180, 240 second and I get
the same result. I tried with server-mode, standalone-mode options in
the ./configure and nothing.
The lock file (.username.pop) when I crash a dial connection keeps for
10 to 20 minutes approximately (always).
I have many clients with poor dial-up connections.
Any ideas???
Thanks in advance
Wilson Machasilla
------=_NextPart_000_004E_01C0DD23.4E5A2BA0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=iso-8859-1"
http-equiv=Content-Type>
<META content="MSHTML 5.00.2614.3500" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2>Hello friends.</FONT></DIV>
<DIV><FONT face=Arial size=2>I have a problem with qpopper-4.0.2 in
the timeout
option. It don´t work!!!</FONT></DIV>
<DIV><FONT face=Arial size=2>I have a Linux Red Hat 7.0 System in a
i686.</FONT></DIV>
<DIV><FONT face=Arial size=2>the /etc/xinetd.conf is:</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>service pop3<BR>
{<BR>
socket_type =
stream<BR>
protocol =
tcp<BR>
wait
=
no<BR>
user
=
root<BR>
server =
/usr/local/sbin/popper<BR>
server_args = qpopper
-T 60<BR>
port
=
110</FONT></DIV>
<DIV><FONT face=Arial size=2> }</FONT></DIV>
<DIV><FONT face=Arial size=2>I have changed the -T option with 10,
60, 120, 180,
240 second and I get the same result. I tried with server-mode,
standalone-mode
options in the ./configure and nothing. </FONT></DIV>
<DIV><FONT face=Arial size=2>The lock file (.username.pop) when I
crash a dial
connection keeps for 10 to 20 minutes approximately
(always).</FONT></DIV>
<DIV><FONT face=Arial size=2>I have many clients with poor dial-up
connections.</FONT></DIV>
<DIV><FONT face=Arial size=2>Any ideas???</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Thanks in advance</FONT></DIV>
<DIV> </DIV>
<DIV><FONT face=Arial size=2>Wilson Machasilla</FONT></DIV>
<DIV> </DIV></BODY></HTML>
------=_NextPart_000_004E_01C0DD23.4E5A2BA0--
Date: Tue, 15 May 2001 16:09:43 -0400 (EDT)
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Subject: Re: Problem with timeout option in qpopper
On Tue, 15 May 2001, Wilson Machasilla wrote:
> I have a problem with qpopper-4.0.2 in the timeout option. It don´t
> work!!!
> service pop3
> {
> socket_type = stream
> protocol = tcp
> wait = no
> user = root
> server = /usr/local/sbin/popper
> server_args = qpopper -T 60
> port = 110
> }
Either remove "qpopper" from server_args or add a "flags = NAMEINARGS"
line to your pop3 service definition.
Suggestion: thoroughly read the xinetd.conf(5) man page. You'll save
yourself lots of headaches in the future.
--
James Ralston, Information Technology
Software Engineering Institute
Carnegie Mellon University, Pittsburgh, PA, USA
Date: Tue, 15 May 2001 16:15:42 -0500
From: Butch Kemper <kemper at tstar dot net>
Subject: Option: CHECK_SHELL
In the module pop_pass.c in Qpopper 3.1.2, there is code that checks on the
validity of the shell field in the passwd file and that code is surrounded
by a compile time option CHECK_SHELL.
This CHECK_SHELL option is documented in INSTALL file but is not supported
by CONFIGURE.
I have used this feature to disable accounts from checking mail but
recently, I found that it wasn't working. That is when I discovered the
compile time option CHECK_SHELL. When did this compile time option get
inserted?
The less in this is I should be more attentive to the changes.
Butch
TSTAR Internet, Inc | Making the Net Work
Marble Falls, TX | Serving Blanco, Burnet,
830-693-6967 | Llano, and Mason Counties
Date: Tue, 15 May 2001 17:45:34 -0400 (EDT)
From: "A. M. Salim" <salim at localweb dot com>
Subject: Problem with attachments?
Hi,
We are using qpopper v3.1.2 (last upgraded in March 2001). Seems to be
working fine but some clients (2 exactly) are complaining they cannot
download attachments, i.e. the server deletes the attachments.
There are about 2000-3000 active users on this mailserver and no other
users appear to be having this problem (at least, no complaints) and I
have tried and not been able to replicate any problem with attachments.
Any ideas what could be going on? Is there a client side or server side
setting that can cause problems with attachments for some users but not
others?
best regards
Mike
From: "Kenneth Porter" <shiva at well dot com>
Date: Tue, 15 May 2001 15:16:42 -0700
Subject: Re: Problem with attachments?
On Tue, 15 May 2001 17:45:34 -0400 (EDT), A. M. Salim wrote:
>We are using qpopper v3.1.2 (last upgraded in March 2001). Seems to be
>working fine but some clients (2 exactly) are complaining they cannot
>download attachments, i.e. the server deletes the attachments.
Qpopper, AFAIK, doesn't know about attachments. They're just part of
the message body. Might you have a procmail recipe that's stripping
attachments on delivery to the spool file? This is one way of
implementing antivirus policies.
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: Tue, 15 May 2001 16:59:50 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: Qpopper 4.02/OpenSSL 0.9.6a - TLS/Alt port error
At 3:56 PM +1000 5/11/01, Mathew Hudson wrote:
> The problem (from the server side):
>
> May 11 15:10:38 <host> -f [24170]: (null) at <address> (IP): -ERR
> POP EOF or I/O Error
> May 11 15:10:38 <host> -f[24170]: I/O error flushing output to
> client at <address> [IP]: Broken pipe
> (32)
>
>
> I am at a loss for getting STLS going with Qpopper working and
> would gladly accept others advice..
I'd suggest turning on debug tracing in Qpopper, and also enabling
logging in Eudora. You want to have Eudora log everything it sends
and receives from the server.
On Macs, drag the "esoteric settings" plug-in (which comes with
Eudora) from the "Extra Plugins" folder to the "Eudora Stuff"
folder. Then in your Eudora settings go to "Logging". Check "all
bytes transferred".
On Windows, move the "esoteric.epi" plug-in (which comes with
Eudora) from the "extrastuff" folder to the same folder as the
"eudora.exe" file. Then in your Eudora settings go to "Logging".
Check "all bytes sent" and "all bytes received".
To enable tracing in Qpopper:
1. Do a 'make clean'
2. Re-run ./configure, adding '--enable-debugging'.
3. Edit the inetd.conf line for Qpopper, adding '-d' or '-t tracefile'.
4. Send inetd a HUP signal.
This causes detailed tracing to be written to the syslog (if you
used '-d') or to the file specified as 'tracefile'.
Date: Tue, 15 May 2001 17:05:05 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: Newbie question
At 5:10 PM +0200 5/11/01, David Bengtsson wrote:
> A simple question. How do I change the domain name provided by Qpopper
You can always use 'sudo hostname mail.rightdomain.com' from a command prompt.
(Qpopper uses gethostname(3) to get the host name.)
Date: Wed, 16 May 2001 13:58:12 +0200
From: Valter Nordh <valter at che.chalmers dot se>
Subject: hashed mail directory
Hi!
I am running a server with approx 500 users.
The server is and "old" SGI R5000 machine, 200Mhz, 256 Mb.
From a performance angle should I use hashed mail-directory?
Or, my question really is:
When should I swith to a hashed mail-directory?
More then 100 users or?
regards, valter
Date: Wed, 16 May 2001 10:40:31 -0400 (EDT)
From: "A. M. Salim" <salim at localweb dot com>
Subject: Re: Problem with attachments?
Hi,
We are not using any procmailor any other filters etc. This user says he
can download attachments just fine from other hosts, just us he is having
a problem with, so that would lead me to believe it is something at the
server end (i.e. qpopper).
best regards
Mike
On Tue, 15 May 2001, Kenneth Porter wrote:
> >We are using qpopper v3.1.2 (last upgraded in March 2001). Seems to be
> >working fine but some clients (2 exactly) are complaining they cannot
> >download attachments, i.e. the server deletes the attachments.
> Qpopper, AFAIK, doesn't know about attachments. They're just part of
> the message body. Might you have a procmail recipe that's stripping
> attachments on delivery to the spool file? This is one way of
> implementing antivirus policies.
From: "Wilson Machasilla" <wilson at punto.net dot ec>
Subject: RE: Problem with timeout option in qpopper
Date: Wed, 16 May 2001 10:21:16 -0500
Thanks!!!
I already attempted this options but It don´t work.
I included this configuration in the mail because this is the config file
model in the FAQ of Eudora Site. The result is the same.
All options works fine: -r -S -f /etc/config.apopper, etc. But the
option -T don´t work!
I don´t undestand.
Wilson M.
----- Original Message -----
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
To: Subscribers of Qpopper <qpopper at lists.pensive dot org>
Sent: Tuesday, May 15, 2001 3:09 PM
Subject: Re: Problem with timeout option in qpopper
> On Tue, 15 May 2001, Wilson Machasilla wrote:
>
> > I have a problem with qpopper-4.0.2 in the timeout option. It don´t
> > work!!!
>
> > service pop3
> > {
> > socket_type = stream
> > protocol = tcp
> > wait = no
> > user = root
> > server = /usr/local/sbin/popper
> > server_args = qpopper -T 60
> > port = 110
> > }
>
> Either remove "qpopper" from server_args or add a "flags = NAMEINARGS"
> line to your pop3 service definition.
>
> Suggestion: thoroughly read the xinetd.conf(5) man page. You'll save
> yourself lots of headaches in the future.
>
> --
> James Ralston, Information Technology
> Software Engineering Institute
> Carnegie Mellon University, Pittsburgh, PA, USA
>
Date: Wed, 16 May 2001 11:24:45 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: RE: Problem with timeout option in qpopper
At 10:21 AM -0500 5/16/01, Wilson Machasilla wrote:
> All options works fine: -r -S -f /etc/config.apopper, etc. But the
> option -T don´t work!
I've not heard of this failing before. Try enabling debug tracing
and see what is going on. The debug trace records written by Qpopper
include the time, so you can see if the timeout is exceeded.
To enable tracing in Qpopper:
1. Do a 'make clean'
2. Re-run ./configure, adding '--enable-debugging'.
3. Edit the inetd.conf line for Qpopper, adding '-d' or '-t tracefile'.
4. Send inetd a HUP signal.
This causes detailed tracing to be written to the syslog (if you used
'-d') or to the file specified as 'tracefile'.
Qpopper simply sets an alarm for the timeout when it is ready for
client input. So you may want to use truss(1), strace(1), or
whatever the system call trace facility is on your platform. That
will show if the alarm is being set, and if the signal is being sent.
Date: Wed, 16 May 2001 11:21:36 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: Problem with attachments?
At 10:40 AM -0400 5/16/01, A. M. Salim wrote:
> This user says he
> can download attachments just fine from other hosts, just us he is having
> a problem with, so that would lead me to believe it is something at the
> server end (i.e. qpopper).
Try enabling debug tracing in Qpopper (at least for this user) and
see what the client sends and what Qpopper sends.
To enable tracing in Qpopper:
1. Do a 'make clean'
2. Re-run ./configure, adding '--enable-debugging'.
3. Edit the inetd.conf line for Qpopper, adding '-d' or '-t tracefile'.
4. Send inetd a HUP signal.
This causes detailed tracing to be written to the syslog (if you used
'-d') or to the file specified as 'tracefile'.
Note that in order to include the contents of messages in the trace,
you also need to add
'#define TRACE_MSG_BODY 1' (without the quotes) as the first line of
config.h after running ./configure but before running make.
Date: Wed, 16 May 2001 14:45:26 -0400 (EDT)
From: "A. M. Salim" <salim at localweb dot com>
Subject: Re: Problem with attachments?
Hi,
> Try enabling debug tracing in Qpopper (at least for this user) and see
> what the client sends and what Qpopper sends.
good suggestion but this being a live server with around 20-50,000 emails
per day going through it, it would fill up syslog pretty fast :-)
Maybe I will think about compiling a second copy of qpopper and running it
on a different port but that may take some doing (firewalls, etc. etc.)
best regards
Mike
Date: Wed, 16 May 2001 17:02:56 -0400 (EDT)
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Subject: correction for Qpopper FAQ (regarding xinetd on Red Hat)
On Wed, 16 May 2001, Wilson Machasilla wrote:
> I included this configuration in the mail because this is the config
> file model in the FAQ of Eudora Site.
So it is, and it's wrong. :(
This is what appears in the FAQ:
Create a file called pop3 in the /etc/xinetd.d directory that
contains the following lines (adjust the server line to contain
the path to the Qpopper executable, and the server_args line to
contain any Qpopper command-line flags you wish to use; this
example shows the executable located at /usr/local/lib/popper, and
the -s command-line flag):
service pop3
{
socket_type = stream
protocol = tcp
wait = no
user = root
server = /usr/local/lib/popper
server_args = qpopper -s
port = 110
}
Note that it may also be necessary to create an entry in the
/etc/hosts.allow file such as the following:
popper:ALL
Additionally, the standard distribution of RedHat Linux 7.0
installs the file /etc/xinetd.d/ipop3 in support of its default
POP3 service, which is likely to create a redundant POP3
references for the xinetd daemon. You may need to either delete
the /etc/xinetd.d/ipop3 file or comment out its service pop3 line.
I suggest replacing the above text with:
Create a file called pop3 in the /etc/xinetd.d directory that
contains the following lines (adjust the server line to contain
the path to the Qpopper executable, and the server_args line to
contain any Qpopper command-line flags you wish to use; this
example shows the executable located at /usr/local/lib/popper, and
the -s command-line flag):
service pop3
{
socket_type = stream
protocol = tcp
wait = no
user = root
server = /usr/local/lib/popper
server_args = -s
}
Then enable qpopper by using (as root) the chkconfig command:
$ chkconfig pop3 on
The standard distributions of RedHat Linux 7.0 and 7.1 install the
files /etc/xinetd.d/ipop2 and /etc/xinetd.d/ipop3 in support of
its default POP3 service, which is likely to create a redundant
POP3 references for the xinetd daemon. You should make sure the
ipop3 and ipop3 services are disabled:
$ chkconfig ipop2 off
$ chkconfig ipop3 off
Also note that it may also be necessary to create an entry in the
/etc/hosts.allow file such as the following:
popper:ALL
--
James Ralston, Information Technology
Software Engineering Institute
Carnegie Mellon University, Pittsburgh, PA, USA
From: "Kenneth Porter" <shiva at well dot com>
Date: Wed, 16 May 2001 14:19:45 -0700
Subject: Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
On Wed, 16 May 2001 17:02:56 -0400 (EDT), James Ralston wrote:
> service pop3
> {
Also make sure that the name specified for service is in the
/etc/services file for TCP port 110. On my old system (RH5.2 upgraded
to RH6.1), the service name is pop-3, so my /etc/xinetd.d/pop3 file
uses pop-3 for the service name.
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: Wed, 16 May 2001 14:21:14 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: Problem with attachments?
At 2:45 PM -0400 5/16/01, A. M. Salim wrote:
> Hi,
>
>> Try enabling debug tracing in Qpopper (at least for this user) and see
>> what the client sends and what Qpopper sends.
>
> good suggestion but this being a live server with around 20-50,000 emails
> per day going through it, it would fill up syslog pretty fast :-)
You can enable debug tracing on a per-user basis. Instead of using
the '-t' or '-d' command-line flags, use '-u' or '-U' and add 'set
debug' or 'set tracefile = path' to the configuration file for the
user. See the Administrator's Guide for more information.
>
> Maybe I will think about compiling a second copy of qpopper and running it
> on a different port but that may take some doing (firewalls, etc. etc.)
Just use a per-user configuration file.
Date: Thu, 17 May 2001 10:23:19 +0400
From: simakin <root at simakin.spb dot ru>
Subject: authorisation sources
Hello!
Does exists any another authorisation sources for qpopper
(plain text files user|passwd for example)?
Does exists any new ideas about subj except very old patches for Mysql
for very old versions of qpopper?
Does exists any another pop3 daemon that
NOT use /etc/passwd for authorisation pop3 users?
Al-X
Date: Thu, 17 May 2001 13:06:54 +0400
From: "Andrey V. Pevnev" <andrey at mgul.ac dot ru>
Subject: How to make it with APOP
Hello!
I'am trying to make qpopper4.0.2 with APOP support on my
FreeBSD 4.3-REL box and getting strange error.
I have compiled and installed gdbm-1.8.0
ls -l /usr/local/lib/libgdbm*
-rw-r--r-- 1 root wheel 50946 17 May 12:36 /usr/local/lib/libgdbm.a
-rwxr-xr-x 1 root wheel 432 17 May 12:36 /usr/local/lib/libgdbm.la
lrwxr-xr-x 1 root wheel 14 2 Mar 2000 /usr/local/lib/libgdbm.so -> libgdbm.so.2.0
-rwxr-xr-x 1 root wheel 25353 2 Mar 2000 /usr/local/lib/libgdbm.so.2.0
./configure --enable-apop
....
checking which database manager to use ... checking gdbm ... checking for gdbm_open in -lgdbm... no
found gdbm header file but not library
checking ndbm ... checking for dbm_open in -lndbm... no
found ndbm header file but not library
checking if BSD DBM... yes
checking dbm ... checking for dbm_open in -ldbm... no
didn't find dbm
....
(but configure process completes ok)
make
....
popauth.c: In function `check_db_err':
popauth.c:356: `DBM' undeclared (first use in this function)
popauth.c:356: (Each undeclared identifier is reported only once
popauth.c:356: for each function it appears in.)
popauth.c:356: syntax error before `)'
popauth.c: In function `main':
popauth.c:659: `DBM' undeclared (first use in this function)
popauth.c:659: `db' undeclared (first use in this function)
popauth.c:660: syntax error before `char'
popauth.c:693: `mode' undeclared (first use in this function)
popauth.c:879: `auth_dir' undeclared (first use in this function)
popauth.c:921: incompatible types in assignment
popauth.c:1034: `f' undeclared (first use in this function)
popauth.c:1056: incompatible types in assignment
popauth.c:1097: incompatible types in assignment
popauth.c:1097: incompatible types in assignment
popauth.c:1106: incompatible types in assignment
popauth.c:1183: incompatible types in assignment
popauth.c:1428: `DBM_REPLACE' undeclared (first use in this function)
*** Error code 1
....
--with-gdbm=/usr/local/lib - did not help :(
Can somebody explain where I'am wrong?
P.S. qpopper3.1.2 with APOP works perfectly.
--
Best regards,
MSFU LAN Admin
Andrey mailto:andrey at mgul.ac dot ru
http://www.mgul.ac.ru/~andrey
Date: Thu, 17 May 2001 12:17:24 +0400
From: simakin <root at simakin.spb dot ru>
Subject: Qpopper with PAM on Solaris
Does subj combination really allow to use another authorisation
sources then /etc/passwd?
Please point me that I have to do after i make steps written in
qpopper FAQ:
> PAM is an architecture which allows the use of various authentication modules with
> different applications. It is available on many platforms, including Linux and Solaris.
> To use PAM, add the --with-pam=service-name flag when running./configure. If you
> omit service-name it defaults to pop3.
> You must then create a file in /etc/pam.d with the same name as specified for
> service-name, for example, /etc/pam.d/pop3. This file contains the rules for authenti-cating
> using Qpopper. See your PAM documentation for more details.
> An example of such a file is:
> #%PAM-1.0
> auth required /lib/security/pam_pwdb.so shadow
> account required /lib/security/pam_pwdb.so
> password required /lib/security/pam_cracklib.so
> password required /lib/security/pam_pwdb.so nullok
> use_authtok md5 shadow
> session required /lib/security/pam_pwdb.so
Does changes must be like this:
in /etc/pam.d: auth required /lib/security/pam_pwdb.so /etc/pop3users
echo username:testpassword >/etc/pop3users
telnet 127.0.0.1 110
user username
etc.....
Or I'm not understand something very impotant?
Al-X
From: "=?iso-8859-1?Q?Søren_P._Skou?=" <sps at arrowhead dot dk>
Subject: Problem with Netscape Messenger, Qpopper with MySQL patch
Date: Thu, 17 May 2001 14:34:03 +0200
Hi there
Actually this has first come to my attention today despite the fact that
I've run Qpopper w/MySQL on top for quite some time. Here goes :)
In Netscape Messenger when you set your pop3 account up, the messenger
happily accepts a username like
sps at some dot domain, but when trying to log in to the pop3, Messenger (and
probably other mail programs as well) only ships off whatever you have in
front of the @. Now with a mailsetup where you're forced to login using the
whole email account + domain, this provides some problems, since those users
simply cannot get mail.
I cannot help but wonder that someone must have seen this before, and do
know the solution to this problem (and no, explaning X amount of users that
they unfortunately cannot use Netscape a.o. is not an option :).
Hope someone can help
Friendly Greetings
S. P. Skou
From: "=?iso-8859-1?Q?Søren_P._Skou?=" <sps at arrowhead dot dk>
Subject: RE: Problem with Netscape Messenger, Qpopper with MySQL patch - solved
Date: Thu, 17 May 2001 15:04:20 +0200
Oh.. Woe is me :) Nevermind, I think this summerheat must have gotten to my
brain or something :)
of course when browsing the nice homepage for the Patch I found that # is
the magical sign I've been looking for, sorry to disturb anyone :)
Friendly Greetings
S. P. Skou
Date: Thu, 17 May 2001 13:17:33 +0000
From: =?iso-8859-1?Q?Guðbjörn?= Hreinsson <gsh at skima dot is>
Subject: Re: Problem with Netscape Messenger, Qpopper with MySQL patch
Unfortunately this is true for Netscape (4.x and 6.x) clients,
but Mozilla does send the domain portion. If I remember this
correctly from testing this is also the case for SMTP AUTH.
We did solve this by having a unique ip address per each virtual
domain but not using qpopper but a commercial product.
Since there is no formal support in stndards for virtual hosting
for POP and IMAP access this is the result. It would be nice
to formalize this in standards either similarly to HTTP or
by formalizing the domain part during login.
Sadly, Netscape clients are lacking many of the features of
MS products, and those that are supported are often badly
implemented...
Rgds,
-GSH
"Søren P. Skou" wrote:
>
> Hi there
>
> Actually this has first come to my attention today despite the fact tha
t
> I've run Qpopper w/MySQL on top for quite some time. Here goes :)
>
> In Netscape Messenger when you set your pop3 account up, the messenger
> happily accepts a username like
> sps at some dot domain, but when trying to log in to the pop3, Messenger (and
> probably other mail programs as well) only ships off whatever you have
in
> front of the @. Now with a mailsetup where you're forced to login using
the
> whole email account + domain, this provides some problems, since those
users
> simply cannot get mail.
>
> I cannot help but wonder that someone must have seen this before, and d
o
> know the solution to this problem (and no, explaning X amount of users
that
> they unfortunately cannot use Netscape a.o. is not an option :).
>
> Hope someone can help
>
> Friendly Greetings
> S. P. Skou
From: Carles Xavier Munyoz =?iso-8859-1?q?Baldó?= <carles.munyoz at ctv-jet dot com>
Subject: What is the .user.cahe file for ?
Date: Thu, 17 May 2001 19:15:34 +0200
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi,
May someone tell me what is the .user.cache file for ?
It is used bye the Qpopper 4.0.2, but for what ?
I think that is for better perfomance, but how can this file improve the
perfomance ?
Many thanks in advance.
- ---
Carles Xavier Munyoz Baldó / cmunyoz at es.wanadoo dot com
- ---
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5.8
iQA/AwUBOwQHNhAGkoZz8//aEQJm2ACglsUacvME0uSNW3fkejG3EYUdTwMAn372
k/h4hbMiuqIHioCz/8hax82A
=2VPi
-----END PGP SIGNATURE-----
Date: Thu, 17 May 2001 12:47:37 -0600 (MDT)
From: "Steven J. Backus" <backus at math.utah dot edu>
Subject: can't get NIS mail aliases to work
I'm new to the list and apologize if this has been covered. I
recently changed from mail aliases in /etc/mail to the NIS aliases
map. Now qpopper doesn't recognize them, saying "user unknown"
when anyone mails to an alias. I'm on solaris 2.5.1 and I
downloaded the latest version of qpopper. Is there a way to fix
this?
Thanks,
Steve
Date: Thu, 17 May 2001 13:39:38 -0600 (MDT)
From: "Steven J. Backus" <backus at math.utah dot edu>
Subject: Re: can't get NIS mail aliases to work
rick pim <rick at post.queensu dot ca> writes:
> mail aliases are not a qpopper issue; they're an MTA issue:
> sendmail or postfix or whatever it is you use.
When I mail using a local mailer (elm, pine) it recognizes the
aliases just fine. When a remote user mails to an alias, the "user
unknown" error appears. When I put the /etc/mail/aliases file
back, everything worked normally. I'm not disagreeing with you,
but perhaps someone knows what's wrong? My /etc/nsswitch.conf has
nis in the aliases field and as far as I know everything is
configured correctly.
Thanks,
Steve
Date: Thu, 17 May 2001 13:23:25 -0700 (PDT)
From: "Jeremy C. Reed" <reed at wcug.wwu dot edu>
Subject: Re: can't get NIS mail aliases to work
Steve,
On Thu, 17 May 2001, Steven J. Backus wrote:
> rick pim <rick at post.queensu dot ca> writes:
>
> > mail aliases are not a qpopper issue; they're an MTA issue:
> > sendmail or postfix or whatever it is you use.
>
> When I mail using a local mailer (elm, pine) it recognizes the
> aliases just fine. When a remote user mails to an alias, the "user
> unknown" error appears. When I put the /etc/mail/aliases file
> back, everything worked normally. I'm not disagreeing with you,
> but perhaps someone knows what's wrong? My /etc/nsswitch.conf has
> nis in the aliases field and as far as I know everything is
> configured correctly.
Like stated above: this is not a qpopper issue. Consider explaining your
situation to a mailing list that is related to your MTA. Anyways, I can
only guess from your details; maybe your local mail client is using a
sendmail (or other) that uses one aliases file and your mail server (for
remote mails coming in) is using another. Don't continue this discussion
here on this list -- maybe discuss this on the list for your operating
system.
Jeremy C. Reed
http://www.reedmedia.net/
http://bsd.reedmedia.net/ -- BSD news and resources
http://www.isp-faq.com/ -- find answers to your questions
Date: Thu, 17 May 2001 14:14:22 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: authorisation sources
At 10:23 AM +0400 5/17/01, simakin wrote:
> Hello!
>
> Does exists any another authorisation sources for qpopper
> (plain text files user|passwd for example)?
>
> Does exists any new ideas about subj except very old patches for Mysql
> for very old versions of qpopper?
>
> Does exists any another pop3 daemon that
> NOT use /etc/passwd for authorisation pop3 users?
>
> Al-X
The FAQ has this to say: <http://www.eudora.com/qpopper/faq.html#which_auth>.
Date: Thu, 17 May 2001 14:28:46 -0700
From: Chuck Yerkes <chuck+qpopper at yerkes dot com>
Subject: Re: hashed mail directory
If you are using XFS as your file system, no,
you won't see a gain.
The issue really comes up under UFS/FFS when you
have more than 3000-5000 entries in a directory.
When you have 10,000 users on FFS, you need hashed dirs.
XFS - well, unlikely that you'd ever NEED to before
a few 100,000.
The times you might WANT to is if you split the
spool over a couple spindles, but then you'd
want your own hash structure.
500 users? Not even close to an issue, IMHO.
Quoting Valter Nordh (valter at che.chalmers dot se):
> Hi!
>
> I am running a server with approx 500 users.
> The server is and "old" SGI R5000 machine, 200Mhz, 256 Mb.
>
> From a performance angle should I use hashed mail-directory?
>
> Or, my question really is:
>
> When should I swith to a hashed mail-directory?
> More than 100 users or?
Date: Thu, 17 May 2001 14:26:34 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: What is the .user.cahe file for ?
At 7:15 PM +0200 5/17/01, Carles Xavier Munyoz Baldó wrote:
> May someone tell me what is the .user.cache file for ?
It holds the index into the mail spool. If no new mail has arrived
(bulletins are OK) then it is used at session start instead of
reading the spool, thus making a very big speed-up and drop in I/O.
Date: Fri, 18 May 2001 02:41:25 -0500 (CDT)
From: Michael Brennen <mbrennen at fni dot com>
Subject: Unable to process From lines (envelope)...
Two days ago a user received a 142 MB publisher file; I thought I
had a 10MB limit on inbound mail, but it must have dropped out
somewhere building sendmail.cf from m4; it is now back in. :)
Anyway, that file caused the mail server to choke rather badly until
I got rid of the offending message. That server stays busy, and it
has run without flaw for its 298 up days until now.
Since then I've had a few users that are getting the above message
from qpopper, so I figured that some corruption had occurred during
that time. Today I upgraded qpopper from 3.1 to 4.0.2, with no
change. I upgraded procmail from 3.13.1 to 3.15.1, the latest. I
checked the FAQ and docs thoroughly; nothing there fixed it. I've
been working with one spool in particular, and went so far as to run
the following, with no change.
# formail -I 'From ' -a 'From ' -s <oldspool >newspool
This completely rewrote the From_ headers, as evidenced by diff;
FWIW no other lines show changed except the From_ lines. However,
qpopper still chokes on it. The first line of the file is below;
the address is changed, but the spaces are exact.
>From user at there dot org Thu May 17 19:35:24 2001
I've turned on debug (not low level) and worked through the logs
with the source. The newline and overflow flags are normal, so I
can only figure that it is failing somewhere in the VALID macro. I
worked through parts of it, but there has to be a better solution to
than decipering that block of obfustication. The excerpt from the
trace log below shows the failure lines; these are after a long set
of traces validly parsing the messages with the UIDL lines. The
buffer count is wrong as I changed the email address, but everything
else is right.
May 17 20:15:16.685 2001 [7179] DROPCOPY: Reading the mail drop (p->msg_count = 20). [pop_dropcopy.c:795]
May 17 20:15:16.685 2001 [7179] newline=1; isOflow=0; wasOflow=0; buffer(56): From user at there dot org Thu May 17 19:35:24 2001 [pop_dropcopy.c:836]
May 17 20:15:16.685 2001 [7179] xxxxxxxx at 127.0.0.1 (127.0.0.1): -ERR [SYS/PERM] Unable to process From lines (envelopes), change recognition modes or check for corrupted mail drop. [pop_dropcopy.c:837]
Can anyone think of anything I might have missed? As near as I can
tell this error only comes from checking the first line in the file,
no others, so I'm baffled. TIA...
-- Michael
From: "James Raftery" <jrtest at spec.ch.man.ac dot uk>
Date: Fri, 18 May 2001 09:25:49 +0100
Subject: Re: can't get NIS mail aliases to work
I agree that this is not a qpopper question but because I had a not dissimilar
problem with qpopper maybe a reply is justified. I think that when compiling
your local delivery agent, perhaps procmail, you had conflicts in definitions
and the NIS stuff was dropped. Since you mention no problem with non-aliased
users,as a story this only hangs together if you did not have your users in the
NIS system.
Jim
On May 17, 1:23pm, Jeremy C. Reed wrote:
> Subject: Re: can't get NIS mail aliases to work
> Steve,
>
> On Thu, 17 May 2001, Steven J. Backus wrote:
>
> > rick pim <rick at post.queensu dot ca> writes:
> >
> > > mail aliases are not a qpopper issue; they're an MTA issue:
> > > sendmail or postfix or whatever it is you use.
> >
> > When I mail using a local mailer (elm, pine) it recognizes the
> > aliases just fine. When a remote user mails to an alias, the "user
> > unknown" error appears. When I put the /etc/mail/aliases file
> > back, everything worked normally. I'm not disagreeing with you,
> > but perhaps someone knows what's wrong? My /etc/nsswitch.conf has
> > nis in the aliases field and as far as I know everything is
> > configured correctly.
>
> Like stated above: this is not a qpopper issue. Consider explaining your
> situation to a mailing list that is related to your MTA. Anyways, I can
> only guess from your details; maybe your local mail client is using a
> sendmail (or other) that uses one aliases file and your mail server (for
> remote mails coming in) is using another. Don't continue this discussion
> here on this list -- maybe discuss this on the list for your operating
> system.
>
> Jeremy C. Reed
> http://www.reedmedia.net/
> http://bsd.reedmedia.net/ -- BSD news and resources
> http://www.isp-faq.com/ -- find answers to your questions
>-- End of excerpt from Jeremy C. Reed
--
James Raftery
Structural Chemistry Department|E-Mail:jrtest at spec.ch.man.ac dot uk
Manchester University | FAX:0161-275-4734
Manchester M13 9PL | Tel.:0161-275-4700
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: Fri, 18 May 2001 17:56:43 +0200
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
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/AwUBOwVGPRAGkoZz8//aEQLqogCfUaLHMHGA5cAIkApWPSUGEGUqpgMAnR9M
ZkgTgFEYrspPYhOzkykdMmQs
=2Pyg
-----END PGP SIGNATURE-----
Date: Fri, 18 May 2001 17:19:47 -0400 (EDT)
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Subject: Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
On Wed, 16 May 2001, Kenneth Porter wrote:
> Also make sure that the name specified for service is in the
> /etc/services file for TCP port 110. On my old system (RH5.2
> upgraded to RH6.1), the service name is pop-3, so my
> /etc/xinetd.d/pop3 file uses pop-3 for the service name.
That's true. However, only Red Hat 7.0 and 7.1 have xinetd, and both
of them have "pop3" listed as the service name in their default
/etc/services files.
Does anyone know of a Linux distribution that ships with xinetd but
*doesn't* have the "pop3" service name in /etc/services?
--
James Ralston, Information Technology
Software Engineering Institute
Carnegie Mellon University, Pittsburgh, PA, USA
From: "Kenneth Porter" <shiva at well dot com>
Date: Fri, 18 May 2001 14:25:20 -0700
Subject: Re: Unable to process From lines (envelope)...
On Fri, 18 May 2001 02:41:25 -0500 (CDT), Michael Brennen wrote:
>The first line of the file is below;
>the address is changed, but the spaces are exact.
>
>>From user at there dot org Thu May 17 19:35:24 2001
Is the ">" in the original file? Or did the "From" get quoted by my
MTA?
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: "Kenneth Porter" <shiva at well dot com>
Date: Fri, 18 May 2001 14:44:10 -0700
Subject: Re: correction for Qpopper FAQ (regarding xinetd on Red Hat)
On Fri, 18 May 2001 17:19:47 -0400 (EDT), James Ralston wrote:
>That's true. However, only Red Hat 7.0 and 7.1 have xinetd, and both
>of them have "pop3" listed as the service name in their default
>/etc/services files.
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.
>Does anyone know of a Linux distribution that ships with xinetd but
>*doesn't* have the "pop3" service name in /etc/services?
I've been upgrading my 6.1 system piece-meal, so I've now got xinetd.
/etc/services is from the "setup" RPM, so I guess I need to look into
updating that. The setup package includes a bunch of config files for
/etc, including hosts.{allow,deny}. If you upgrade yours, be sure to
back up any customizations. "rpm -ql setup" will list the files to look
at, and "rpm -V setup" will tell you which ones you've changed.
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: "Dan Trainor" <dan at concept-factory dot com>
Subject: kernel error
Date: Fri, 18 May 2001 16:06:40 -0700
Hello all.
I have a couple questions tonight, if that's alright.
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.
Thanks for the time, I appreciate it.
- Dan Trainor
- Systems Administrator
- Concept Factory, LLC.
- www.concept-factory.com
- dan at concept-factory dot com
Date: Fri, 18 May 2001 18:27:55 -0500 (CDT)
From: Michael Brennen <mbrennen at fni dot com>
Subject: Re: Unable to process From lines (envelope)...
That > was inserted by the MTA to escape the From_ in column 1.
Just to get service going again I fell back to an old cucipop
daemon. I still want to go back to qpopper, but I had to stop the
phone ringing.
-- Michael
On Fri, 18 May 2001, Kenneth Porter wrote:
> On Fri, 18 May 2001 02:41:25 -0500 (CDT), Michael Brennen wrote:
>
> >The first line of the file is below;
> >the address is changed, but the spaces are exact.
> >
> >>From user at there dot org Thu May 17 19:35:24 2001
>
> Is the ">" in the original file? Or did the "From" get quoted by my
> MTA?
Last updated on 18 May 2001 by Pensive Mailing List Admin