The qpopper list archive ending on 27 Jun 2001


Topics covered in this issue include:

  1. Re: Fw: log file and volume of data ...
       WSO Support <mailinglists at wso dot net>
       Thu, 21 Jun 2001 20:16:50 -0400
  2. Re: Fw: log file and volume of data ...
       Jesus Cea Avion <jcea at argo dot es>
       Fri, 22 Jun 2001 11:47:59 +0200
  3. question about qpopper v4.0.3
       "Damian Ramirez" <damian at highway.com dot py>
       Fri, 22 Jun 2001 08:30:43 -0400
  4. Re: question about qpopper v4.0.3
       Clifton Royston <cliftonr at lava dot net>
       Fri, 22 Jun 2001 08:31:12 -1000
  5. Re: question about qpopper v4.0.3
       Jeffrey Culverhouse <jafoc at rev dot net>
       Fri, 22 Jun 2001 14:36:46 -0400 (EDT)
  6. Re: mysql qpopper
       Clifton Royston <cliftonr at lava dot net>
       Fri, 22 Jun 2001 08:48:12 -1000
  7. Corrupted mail boxes
       Homer Wilson Smith <homer at lightlink dot com>
       Fri, 22 Jun 2001 15:39:43 -0400 (EDT)
  8. RE: question about qpopper v4.0.3
       "Damian Ramirez" <damian at highway.com dot py>
       Sat, 23 Jun 2001 20:16:45 -0400
  9. Re: question about qpopper v4.0.3
       Randall Gellens <randy at qualcomm dot com>
       Sun, 24 Jun 2001 19:15:03 -0700
 10. OpenSSL certficate signing
       "Ted Lufoma" <lufomat at coppernet dot zm>
       Mon, 25 Jun 2001 09:44:47 +0200
 11. RE: question about qpopper v4.0.3
       "Damian Ramirez" <damian at highway.com dot py>
       Mon, 25 Jun 2001 09:43:30 -0400
 12. Re: Qpopper+Kerberos in Solaris 8 for Intel
       Luis F V Gomes <lf at ele.puc-rio dot br>
       Mon, 25 Jun 2001 15:39:06 -0300
 13. QPopper 4.0.3 hanging? Or inetd? Help, please!
       Frazier Scott <scottf at oit dot edu>
       Mon, 25 Jun 2001 12:16:47 -0700
 14. Re: QPopper 4.0.3 hanging? Or inetd? Help, please!
       Admin Mailing Lists <mlist at intergrafix dot net>
       Mon, 25 Jun 2001 15:53:02 -0400 (EDT)
 15. RV: QPopper 4.0.3 hanging? Or inetd? Help, please!
       "Wilson Machasilla" <wilson at punto.net dot ec>
       Mon, 25 Jun 2001 18:17:02 -0500
 16. RE: question about qpopper v4.0.3
       Randall Gellens <randy at qualcomm dot com>
       Mon, 25 Jun 2001 16:48:58 -0700
 17. Re: OpenSSL certficate signing
       Randall Gellens <randy at qualcomm dot com>
       Mon, 25 Jun 2001 17:02:06 -0700
 18. Qpopper 4.0.3 doesn't support APOP?
       "Dapid Candra" <dapidc at cbn.net dot id>
       Tue, 26 Jun 2001 11:42:48 +0700
 19. Trouble Compiling Qpopper
       "Kent Morris" <gaunt at cophq dot org>
       Tue, 26 Jun 2001 01:34:52 -0400
 20. Re: Qpopper 4.0.3 doesn't support APOP?
       "Kenneth Porter" <shiva at well dot com>
       Mon, 25 Jun 2001 22:49:29 -0700
 21. 3.1 still safe ?
       hypnose at t-online dot de (guenter wessling)
       Tue, 26 Jun 2001 10:04:31 +0200
 22. Re: QPopper 4.0.3 hanging? Or inetd? Help, please!
       "Jeff Gallagher" <jeffg at nfld dot net>
       Tue, 26 Jun 2001 11:31:59 -0230
 23. Re: QPopper 4.0.3 hanging? Or inetd? Help, please!
       Admin Mailing Lists <mlist at intergrafix dot net>
       Tue, 26 Jun 2001 11:12:22 -0400 (EDT)
 24. web interface
       Wasim Bashir <wasim.bashir at broadband.co dot uk>
       Tue, 26 Jun 2001 16:40:30 +0100
 25. Re: web interface
       Clifton Royston <cliftonr at lava dot net>
       Tue, 26 Jun 2001 08:58:19 -1000
 26. Re: web interface
       "P S Graham" <patrick at gandatech dot com>
       Tue, 26 Jun 2001 15:40:10 -0400
 27. Re: QPopper 4.0.3 hanging? Or inetd? Help, please!
       Frazier Scott <scottf at oit dot edu>
       Tue, 26 Jun 2001 15:33:54 -0700
 28. Re: web interface
       "Kenneth Porter" <shiva at well dot com>
       Tue, 26 Jun 2001 18:23:56 -0700
 29. pop3/tcp server failing (looping), service terminated 
       Ilya Palagin <ipalagin at fioc dot kz>
       Wed, 27 Jun 2001 09:53:21 +0700
 30. Re: pop3/tcp server failing (looping), service terminated
       Jack Sasportas <jack at innovativeinternet dot com>
       Tue, 26 Jun 2001 23:45:02 -0400
 31. Re: pop3/tcp server failing (looping), service terminated
       peter at netlink.com dot au
       Wed, 27 Jun 2001 13:49:43 +1000 (EST)
 32. Re: pop3/tcp server failing (looping), service terminated
       Peter Evans <peter at gol dot com>
       Wed, 27 Jun 2001 12:59:13 +0900
 33. APOP Problem
       "Desmond Lim" <ddl76 at singnet.com dot sg>
       Wed, 27 Jun 2001 11:58:49 +0800
 34. APOP clients
       "Kenneth Porter" <shiva at well dot com>
       Wed, 27 Jun 2001 01:10:20 -0700
 35. How do clients keep status of maildrop
       =?iso-8859-1?Q?Guðbjörn?= Hreinsson <gsh at skima dot is>
       Wed, 27 Jun 2001 08:48:14 +0000
 36. Re: APOP clients
       Tim Jackson <tim at timj.co dot uk>
       Wed, 27 Jun 2001 12:20:47 +0100
 37. Re: APOP clients
       Frank Pineau <frank at pineaus dot com>
       Wed, 27 Jun 2001 08:05:00 -0400
 38. How do I set up Bulletins
       Jeff Donovan <jdonovan at beth.k12.pa dot us>
       Wed, 27 Jun 2001 09:39:06 -0400
 39. Re: How do I set up Bulletins
       Frank Pineau <frank at pineaus dot com>
       Wed, 27 Jun 2001 09:53:19 -0400
 40. Re: pop3/tcp server failing (looping), service terminated
       Admin Mailing Lists <mlist at intergrafix dot net>
       Wed, 27 Jun 2001 10:00:40 -0400 (EDT)
 41. Re: How do I set up Bulletins
       Jeff Donovan <jdonovan at beth.k12.pa dot us>
       Wed, 27 Jun 2001 10:26:39 -0400
 42. Re: pop3/tcp server failing (looping), service terminated
       Tim Olson <tolson at unionsemiconductor dot com>
       Wed, 27 Jun 2001 09:51:51 -0500
 43. SSL and XTND XMIT
       Steve Palm <stevep at sga dot org>
       Wed, 27 Jun 2001 10:15:20 -0500
 44. Bulletins * need clarification *
       Jeff Donovan <jdonovan at beth.k12.pa dot us>
       Wed, 27 Jun 2001 11:27:55 -0400
 45. Re: Bulletins * need clarification PLEASE *
       Jeff Donovan <jdonovan at beth.k12.pa dot us>
       Wed, 27 Jun 2001 14:31:06 -0400
 46. Re: Bulletins * need clarification PLEASE *
       Butch Kemper <kemper at tstar dot net>
       Wed, 27 Jun 2001 16:22:39 -0500
 47. Re: How do clients keep status of maildrop
       Clifton Royston <cliftonr at lava dot net>
       Wed, 27 Jun 2001 11:41:23 -1000
 48. I/O error flushing output to client ???
       Jack Sasportas <jack at innovativeinternet dot com>
       Wed, 27 Jun 2001 22:51:42 -0400
 49. .user.pop files in RAM disk, how faster?
       PM WONG <pmwong at power25t.hkbu.edu dot hk>
       Thu, 28 Jun 2001 11:20:53 +0800 (HKT)
 50. Tools/methods for deleting old email?
       Daniel Senie <dts at senie dot com>
       Thu, 28 Jun 2001 00:11:51 -0400

Date: Thu, 21 Jun 2001 20:16:50 -0400
From: WSO Support <mailinglists at wso dot net>
Subject: Re: Fw: log file and volume of data ...

Jesus,

Thanks for the code. So, your patch just adds an extra
field to the stats line that shows how much total data
was retrieved in that session?

Just looking for clarification.

Thanks!
-Chris
WSO


At 08:45 PM 6/21/2001 +0200, you wrote:
>> and my question can be reformulated in "why the statistics after each
>> session (enabled with option -t) don't still integrate one additionnal 
>> field (after the 4 others numbers), that give the amount of fetched
>> data (header + body)  ?"
>
>I've done trivial patch against current sources:
>
>>>>>>
>
>Index: popper/pop_updt.c
>==================================================================
>RCS file: /opt/src/cvsroot/qpopper/popper/pop_updt.c,v
>retrieving revision 1.1.1.2
>retrieving revision 1.1.1.2.14.1
>diff -c -u -r1.1.1.2 -r1.1.1.2.14.1
>cvs server: conflicting specifications of output style
>--- popper/pop_updt.c   2001/04/16 21:44:34     1.1.1.2
>+++ popper/pop_updt.c   2001/06/04 06:42:58     1.1.1.2.14.1
>@@ -293,10 +293,11 @@
>     DEBUG_LOG0 ( p, "Checking to see if all messages were deleted" );
> 
>     if ( p->bStats ) {
>-        pop_log ( p, POP_PRIORITY, HERE, "Stats: %s %d %ld %d %ld %s %s",
>+        pop_log ( p, POP_PRIORITY, HERE, "Stats: %s %d %ld %d %ld %ld %s %s",
>                   p->user, p->msgs_deleted, p->bytes_deleted,
>                   p->msg_count - p->msgs_deleted,
>                   p->drop_size - p->bytes_deleted,
>+                  ftell(p->output),
>                   p->client, p->ipaddr );
>     }
>   
><<<<<
>
>-- 
>Jesus Cea Avion                         _/_/      _/_/_/        _/_/_/
>jcea at argo.es http://www.argo dot es/~jcea/ _/_/    _/_/  _/_/    _/_/  _/_/
>                                      _/_/    _/_/          _/_/_/_/_/
>PGP Key Available at KeyServ   _/_/  _/_/    _/_/          _/_/  _/_/
>"Things are not so easy"      _/_/  _/_/    _/_/  _/_/    _/_/  _/_/
>"My name is Dump, Core Dump"   _/_/_/        _/_/_/      _/_/  _/_/
>"El amor es poner tu felicidad en la felicidad de otro" - Leibniz


Date: Fri, 22 Jun 2001 11:47:59 +0200
From: Jesus Cea Avion <jcea at argo dot es>
Subject: Re: Fw: log file and volume of data ...

> Thanks for the code. So, your patch just adds an extra
> field to the stats line that shows how much total data
> was retrieved in that session?

Right.

-- 
Jesus Cea Avion                         _/_/      _/_/_/        _/_/_/
jcea at argo.es http://www.argo dot es/~jcea/ _/_/    _/_/  _/_/    _/_/  _/_/
                                      _/_/    _/_/          _/_/_/_/_/
PGP Key Available at KeyServ   _/_/  _/_/    _/_/          _/_/  _/_/
"Things are not so easy"      _/_/  _/_/    _/_/  _/_/    _/_/  _/_/
"My name is Dump, Core Dump"   _/_/_/        _/_/_/      _/_/  _/_/
"El amor es poner tu felicidad en la felicidad de otro" - Leibniz

From: "Damian Ramirez" <damian at highway.com dot py>
Subject: question about qpopper v4.0.3
Date: Fri, 22 Jun 2001 08:30:43 -0400

Hello all

I running the following php script for connecting to my pop3 server runnig
qpopper v3.1.2 and display the number of messages stored in my mailbox :

<?php
$mbox = imap_open ("{correo.highway.com.py:110/pop3}", "damian",
"secret66");
$d = imap_num_msg ($mbox);
echo "$d";
?>

And work fine.

But running the same php script to my pop3 server with qpopper v4.0.3 , its
not work and the log of qpopper says the following:

Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
(208.209.45.1): -ERR POP EOF or I/O Error
Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
(208.209.45.1): -ERR POP EOF or I/O Error
Jun 22 08:19:22 hwns1 popper[88890]: I/O error flushing output to client  at
208.209.45.1 [208.209.45.1]: Operation not permitted(1)
Jun 22 08:19:22 hwns1 popper[88890]: I/O error flushing output to client  at
208.209.45.1 [208.209.45.1]: Operation not permitted(1)
Jun 22 08:19:22 hwns1 popper[88890]: (v4.0.3) Timing for @208.209.45.1
(error) auth=0init=0 clean=0

My pop3 server with qpopper v4.0.3 work fine for mail clients programs.
I install qpopper v4.0.3 with the following options:

su-2.03#
./configure --disable-check-pw-max --enable-log-login --enable-server-mode -
-
enable-shy --enable-timing --enable-chunky-writes=2

And running from inetd with the following options:

pop3    stream  tcp     nowait  root    /usr/local/bin/popper   popper -T
120 -R -s


Some ideas?

Thanks!

Damian Ramirez
damian at highway.com dot py


Date: Fri, 22 Jun 2001 08:31:12 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Re: question about qpopper v4.0.3

On Fri, Jun 22, 2001 at 08:30:43AM -0400, Damian Ramirez wrote:
> Hello all
> 
> I running the following php script for connecting to my pop3 server runnig
> qpopper v3.1.2 and display the number of messages stored in my mailbox :
> 
> <?php
> $mbox = imap_open ("{correo.highway.com.py:110/pop3}", "damian",
> "secret66");
> $d = imap_num_msg ($mbox);
> echo "$d";
> ?>
> 
> And work fine.

Not being terribly familiar with php, I assume that "imap_open" really
has some facility for connecting to POP ports as well?

My guess is that whatever is in that imap_open routine is broken. 
Maybe it is looking for some specific string in the banner.


> But running the same php script to my pop3 server with qpopper v4.0.3 , its
> not work and the log of qpopper says the following:
> 
> Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
> (208.209.45.1): -ERR POP EOF or I/O Error

The "(null)" in place of the username means it never got a login, and 
"-ERR POP EOF or I/O Error" means the remote side disconnected without
sending a "quit" command.

> Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
> (208.209.45.1): -ERR POP EOF or I/O Error
...

> My pop3 server with qpopper v4.0.3 work fine for mail clients programs.
> I install qpopper v4.0.3 with the following options:
> 
> su-2.03#
> ./configure --disable-check-pw-max --enable-log-login --enable-server-mode -
> -
> enable-shy --enable-timing --enable-chunky-writes=2

Try configuring it and making it again without the "shy" option; maybe
that PHP routine is expecting to see the string "pop" in the banner,
though it shouldn't. 
 
  -- Clifton

-- 
 Clifton Royston  --  LavaNet Systems Architect --  cliftonr at lava dot net
   WWJD?   "JWRTFM!" - Scott Dorsey (kludge)   "JWG" - Eddie Aikau

Date: Fri, 22 Jun 2001 14:36:46 -0400 (EDT)
From: Jeffrey Culverhouse <jafoc at rev dot net>
Subject: Re: question about qpopper v4.0.3

On Fri, 22 Jun 2001, Damian Ramirez wrote:

> And work fine.
>
> But running the same php script to my pop3 server with qpopper v4.0.3 ,
> its not work and the log of qpopper says the following:
>
> Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
> (208.209.45.1): -ERR POP EOF or I/O Error
> Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
> (208.209.45.1): -ERR POP EOF or I/O Error
> Jun 22 08:19:22 hwns1 popper[88890]: I/O error flushing output to client at
> 208.209.45.1 [208.209.45.1]: Operation not permitted(1)
> Jun 22 08:19:22 hwns1 popper[88890]: I/O error flushing output to client at
> 208.209.45.1 [208.209.45.1]: Operation not permitted(1)
> Jun 22 08:19:22 hwns1 popper[88890]: (v4.0.3) Timing for @208.209.45.1
> (error) auth=0init=0 clean=0


I think these messages relate to the POP connection closing without an
explicit QUIT command sent... someone please correct me if I am wrong!

jeff culverhouse
jafoc at rev dot net


Date: Fri, 22 Jun 2001 08:48:12 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Re: mysql qpopper

> <318685366298284412529 at lists.pensive.org> <474793172565112839593 at lists dot pensive dot org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 1.0us
In-Reply-To: <474793172565112839593 at lists.pensive dot org>; from nngodinh at ocic dot org on Sat, Jun 16, 2001 at 03:42:05PM +0200

On Sat, Jun 16, 2001 at 03:42:05PM +0200, Nhan NGO DINH (OCIC Missionary Service) wrote:
> I want to think about a cluster mail storage system, like an entry each 64k 
> of mail, to avoid big message bodies in a db field. Anyway this should 
> improve mailbox access speed and it could give mailbox availability for a 
> cluster of servers. It seems to me a good idea, but too difficult to 
> implement. May be there is already some stuff like that around here.

  Hmmm.  You want to improve mailbox speed by implementing a database,
spread across a cluster of servers, which efficiently handles records
varying in size from a few hundred bytes to many megabytes, in a
collection indexed on a single key, with multiple collections organized
for lookup by an ASCII string.

  Sounds kind of like a hierarchical UNIX directory structure over NFS
to me! (NFS+maildir format)

  Beyond that, I just don't see any benefit of storing the mail in a
database - are you ever going to try to do operations like "Look up the
3rd message in every user's mailbox" or "retrieve all messages with a
subject of 'Announcement' sent to every user whose name begins with
'z'"?  Probably not, in which case a database will provide no benefit
and only degrade your performance.  Unless of course you're doing it
for pure hack value (entertainment purposes) - in which case anything
goes.

  If availability and scalability of performance is really critical to
you (e.g. supporting >25K users with > 99.99% uptime is a requirement)
consider implementing a cluster of servers accessing a shared mailspool
on a redundantly configured NFS server, running something like the
Cyrus mail system or another system which supports a maildir type
format, and using the POP server that comes with that,if qpopper isn't
enhanced to support maildir by then.  I know this works because some
very large ISPs (Earthlink, PacBell) have used this general style of
architecture for their mail service; I've had some casual chats with
system designers at both of those.

  -- Clifton

-- 
 Clifton Royston  --  LavaNet Systems Architect --  cliftonr at lava dot net
   WWJD?   "JWRTFM!" - Scott Dorsey (kludge)   "JWG" - Eddie Aikau

Date: Fri, 22 Jun 2001 15:39:43 -0400 (EDT)
From: Homer Wilson Smith <homer at lightlink dot com>
Subject: Corrupted mail boxes


     Just FYI, we are running Linux 2.0.38 with qpopper 2.52,
upgraded to procmail 3.15.1 and are still getting corrupted
mail boxes on users that only use pop no shell.

     Will upgrade to 4.03

     Homer



From: "Damian Ramirez" <damian at highway.com dot py>
Subject: RE: question about qpopper v4.0.3
Date: Sat, 23 Jun 2001 20:16:45 -0400

> Try configuring it and making it again without the "shy" option; maybe
> that PHP routine is expecting to see the string "pop" in the banner,
> though it shouldn't.
>
>   -- Clifton
>

I get the same error.

Damian Ramirez
damian at highway.com dot py


----- Original Message -----
From: Clifton Royston <cliftonr at lava dot net>
To: Damian Ramirez <damian at highway.com dot py>
Cc: Subscribers of Qpopper <qpopper at lists.pensive dot org>
Sent: Friday, June 22, 2001 2:31 PM
Subject: Re: question about qpopper v4.0.3


> On Fri, Jun 22, 2001 at 08:30:43AM -0400, Damian Ramirez wrote:
> > Hello all
> >
> > I running the following php script for connecting to my pop3 server
runnig
> > qpopper v3.1.2 and display the number of messages stored in my mailbox :
> >
> > <?php
> > $mbox = imap_open ("{correo.highway.com.py:110/pop3}", "damian",
> > "secret66");
> > $d = imap_num_msg ($mbox);
> > echo "$d";
> > ?>
> >
> > And work fine.
>
> Not being terribly familiar with php, I assume that "imap_open" really
> has some facility for connecting to POP ports as well?
>
> My guess is that whatever is in that imap_open routine is broken.
> Maybe it is looking for some specific string in the banner.
>
>
> > But running the same php script to my pop3 server with qpopper v4.0.3 ,
its
> > not work and the log of qpopper says the following:
> >
> > Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
> > (208.209.45.1): -ERR POP EOF or I/O Error
>
> The "(null)" in place of the username means it never got a login, and
> "-ERR POP EOF or I/O Error" means the remote side disconnected without
> sending a "quit" command.
>
> > Jun 22 08:19:22 hwns1 popper[88890]: (null) at 208.209.45.1
> > (208.209.45.1): -ERR POP EOF or I/O Error
> ...
>
> > My pop3 server with qpopper v4.0.3 work fine for mail clients programs.
> > I install qpopper v4.0.3 with the following options:
> >
> > su-2.03#
> >
./configure --disable-check-pw-max --enable-log-login --enable-server-mode -
> > -
> > enable-shy --enable-timing --enable-chunky-writes=2
>
> Try configuring it and making it again without the "shy" option; maybe
> that PHP routine is expecting to see the string "pop" in the banner,
> though it shouldn't.
>
>   -- Clifton
>
> --
>  Clifton Royston  --  LavaNet Systems Architect --  cliftonr at lava dot net
>    WWJD?   "JWRTFM!" - Scott Dorsey (kludge)   "JWG" - Eddie Aikau
>


Date: Sun, 24 Jun 2001 19:15:03 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: question about qpopper v4.0.3

At 8:30 AM -0400 6/22/01, Damian Ramirez wrote:

>  But running the same php script to my pop3 server with qpopper v4.0.3 , its
>  not work and the log of qpopper says the following:

That log doesn't say much.  Try enabling debug tracing and see what 
Qpopper says is going on.

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'.

From: "Ted Lufoma" <lufomat at coppernet dot zm>
Subject: OpenSSL certficate signing
Date: Mon, 25 Jun 2001 09:44:47 +0200

Hello,
I have just downloded qpopper4.0.3 which requires TLS/SSL (OpenSSL)
OpenSSL have been installed. I have done create and install  certificates
for TLS/SSL and got to a point where it says "Send the certificate signing
request (file req.pem) to your Certificate Authority for signing. You will
receive
back a signed request."
Please advise the e-mail address of the Certificate Authority.

Kind regards

Ted
-------------------
Ted Lufoma                        e-mail: lufomat at coppernet dot zm
CopperNET Solutions       URL: www.coppernet.zm
P. O. Box 22149                Tel : 260-2-245087   Fax: 260-2-245734
Kitwe - ZAMBIA.



From: "Damian Ramirez" <damian at highway.com dot py>
Subject: RE: question about qpopper v4.0.3
Date: Mon, 25 Jun 2001 09:43:30 -0400

I running the following php script for connecting to my pop3 server runnig
qpopper v3.1.2 and display the number of messages stored in my mailbox :

<?php
$mbox = imap_open ("{correo.highway.com.py:110/pop3}", "damian",
"secret66");
$d = imap_num_msg ($mbox);
echo "$d";
?>

And work fine.

But running the same php script to my pop3 server with qpopper v4.0.3 , its
not work and the log of qpopper says the following:

----------------------------------------------------------------------------
-----
Ok, i reinstall with the debugging option, and the log says the following:

Jun 25 10:46:09.917 2001 [41860] Trace and Debug destination is file
"/var/log/pop.log
" [pop_init.c:855]
Jun 25 10:46:09.917 2001
Jun 25 10:46:09.917 2001 [41860] timeout = 120 (-T) [pop_init.c:860]
Jun 25 10:46:09.917 2001
Jun 25 10:46:09.917 2001 [41860] Avoiding reverse lookups (-R)
[pop_init.c:820]
Jun 25 10:46:09.917 2001
Jun 25 10:46:09.918 2001 [41860] Will generate stats records (-s)
[pop_init.c:825]
Jun 25 10:46:09.918 2001
Jun 25 10:46:09.918 2001 [41860] (v4.0.3) Servicing request from
"208.209.45.1" at 208
.209.45.1 [pop_init.c:1153]
Jun 25 10:46:09.918 2001
Jun 25 10:46:09.918 2001 [41860] before TLS; tls_support==0 [popper.c:172]
Jun 25 10:46:09.918 2001
Jun 25 10:46:09.918 2001 [41860] Skipped TLS Init [popper.c:196]
Jun 25 10:46:09.918 2001
Jun 25 10:46:09.918 2001 [41860] (v4.0.3) Intro [popper.c:238]
Jun 25 10:46:09.918 2001
Jun 25 10:46:09.918 2001 [41860] +OK ready   [popper.c:246]
Jun 25 10:46:09.918 2001
Jun 25 10:46:09.919 2001 [41860] Qpopper ready for input from (null) at
208.209.45.1 [
208.209.45.1] [popper.c:285]
Jun 25 10:46:09.919 2001
Jun 25 10:46:09.986 2001 [41860] Received (4): "CAPA"
[pop_get_command.c:105]
Jun 25 10:46:09.986 2001
Jun 25 10:46:09.986 2001 [41860] capa returned 1; CurrentState now auth1
[popper.c:329
]
Jun 25 10:46:09.986 2001
Jun 25 10:46:09.986 2001 [41860] Qpopper ready for input from (null) at
208.209.45.1 [
208.209.45.1] [popper.c:285]
Jun 25 10:46:09.986 2001
Jun 25 10:46:10.091 2001 [41860] (null) at 208.209.45.1 (208.209.45.1): -ERR
POP EOF o
r I/O Error [popper.c:794]
Jun 25 10:46:10.091 2001
Jun 25 10:46:10.091 2001 [41860] +OK Pop server at iserver.radar.com.py
signing off. [
popper.c:351]
Jun 25 10:46:10.091 2001
Jun 25 10:46:10.091 2001 [41860] (v4.0.3) Ending request from "" at
(208.209.45.1) 208
.209.45.1 [popper.c:369]
Jun 25 10:46:10.091 2001
Jun 25 10:46:10.091 2001 [41860] (v4.0.3) Timing for @208.209.45.1 (error)
auth=0 init
=0 clean=0 [popper.c:375]
Jun 25 10:46:10.091 2001
Damian Ramirez
damian at highway.com dot py
Departamento Tecnico
Highway I.S.P

----- Original Message -----
From: Randall Gellens <randy at qualcomm dot com>
To: Damian Ramirez <damian at highway.com dot py>; Subscribers of Qpopper
<qpopper at lists.pensive dot org>
Sent: Sunday, June 24, 2001 10:15 PM
Subject: Re: question about qpopper v4.0.3


> At 8:30 AM -0400 6/22/01, Damian Ramirez wrote:
>
> >  But running the same php script to my pop3 server with qpopper v4.0.3 ,
its
> >  not work and the log of qpopper says the following:
>
> That log doesn't say much.  Try enabling debug tracing and see what
> Qpopper says is going on.
>
> 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: Mon, 25 Jun 2001 15:39:06 -0300
From: Luis F V Gomes <lf at ele.puc-rio dot br>
Subject: Re: Qpopper+Kerberos in Solaris 8 for Intel

At 05:33 PM 6/21/2001, Kenneth Porter wrote:
>On Thu, 21 Jun 2001 15:56:27 -0300, Kenneth Porter wrote:
>
> >Undefined                       first referenced
> >  symbol                             in file
> >krb_mk_priv /usr/local/krb5-1.2.2/src/lib/libkrb5
> >util.a(compat_recv.o)
> >krb5_c_encrypt_length /usr/local/krb5-1.2.2/src/lib/libkrb5
> >.a(enc_helper.o)
>
>All of your missing symbols are referenced *from* the krb5 library.
>That suggests that krb5 might be built wrong. Which of the krb5
>libraries provide those symbols?
>
>Ken


Hi all

I don't know how to answer your question, Ken.
I also tried to compile using krb-5_1.1.2_beta2 and the undefined symbols 
were reduced to four.
I rebuilt krb-5_1.2.2 and I still have the same 16 errors:

-----
Undefined                       first referenced
  symbol                             in file
krb_mk_priv                         /usr/local/lib/libkrb5util.a(compat_recv.o)
krb5_c_encrypt_length               /usr/local/lib/libkrb5.a(enc_helper.o)
krb5_c_random_make_octets           /usr/local/lib/libkrb5.a(gen_seqnum.o)
krb5_c_decrypt                      /usr/local/lib/libkrb5.a(kdc_rep_dc.o)
krb5_c_block_size                   /usr/local/lib/libkrb5.a(auth_con.o)
krb5_c_string_to_key                /usr/local/lib/libkrb5.a(preauth2.o)
krb5_c_random_seed                  /usr/local/lib/libkrb5.a(init_ctx.o)
krb5_c_encrypt                      /usr/local/lib/libkrb5.a(enc_helper.o)
krb5_string_to_enctype              /usr/local/lib/libkrb5.a(init_ctx.o)
krb_net_write                       /usr/local/lib/libkrb5util.a(compat_recv.o)
krb_net_read                        /usr/local/lib/libkrb5util.a(compat_recv.o)
valid_enctype                       /usr/local/lib/libkrb5.a(init_ctx.o)
mit_des_key_sched                   /usr/local/lib/libdes425.a(key_sched.o)
krb_rd_req                          /usr/local/lib/libkrb5util.a(compat_recv.o)
krb5_encrypt_data                   /usr/local/lib/libkrb5.a(preauth.o)
krb5_c_enctype_compare              /usr/local/lib/libkrb5.a(ktf_g_ent.o)
ld: fatal: Symbol referencing errors. No output written to popper
collect2: ld returned 1 exit status
make: Fatal error: Command failed for target `popper'
Current working directory /usr/local/Hydrus/qpopper4.0.2/popper
-----

Maybe I will try with other older versions...
How do I know which libraries provide these symbols?

Thanks
Luis


!                           ,,,
!                          (@ @)
+---------------------oOO-/-(*)-\-OOo---+------------------------------------+
| Luis Fernando V. Gomes                |    Internet: lf at ele.puc-rio dot br
| Newtork Administrator                 |      postmaster at ele.puc-rio dot br
| Dept. Engenharia Eletrica             |            root at ele.puc-rio dot br
| PUC-Rio                               |   Voz: +(55) (21) 529-9417
| R. Marques de Sao Vicente 225/446L    |   Fax: +(55) (21) 511-5154
| 22453-900 - Rio de Janeiro/RJ, Brasil |   http://www.ele.puc-rio.br/~lf/
+---------------------------------------+------------------------------------+


Date: Mon, 25 Jun 2001 12:16:47 -0700
From: Frazier Scott <scottf at oit dot edu>
Subject: QPopper 4.0.3 hanging? Or inetd? Help, please!

Hi all,

I'm a newbie in charge of a production server.  I have Caldera eServer 
2.3.1 installed with a new 2.2.16 kernel.  Everything has been working like 
a charm.  Today is the first day of the switchover, and it is getting 
toward the afternoon.  Pop clients have begun to report "connection refused 
(10061)".  I have QPopper 4.0.3.  I restart inetd and all is well for 15 
minutes when it begins again.  Am I running out of threads?  I have 
debugging turned on, but nothing shows up.  Syslog just stops writing 
messages to the logs when the service, whichever it is, hangs.  ps aux 
shows the thread numbers climbing like crazy -- I'm up to 16,000 now, yet 
the number of processes is around 130.  Is this a clue?  Don't these thread 
numbers recycle themselves?  They seemed to when we were running Digital 
Unix (yeah, I was a newbie there, too.  Still am.).

Help me, please!
Frazier Scott - Internetwork Administrator
Oregon Institute of Technology
3201 Campus Drive
Klamath Falls, OR  97601
Ph (541) 885-1310
Fx (541) 885-1919
scottf at oit dot edu


Date: Mon, 25 Jun 2001 15:53:02 -0400 (EDT)
From: Admin Mailing Lists <mlist at intergrafix dot net>
Subject: Re: QPopper 4.0.3 hanging? Or inetd? Help, please!

I guess the first question would be, when you said switchover, what are
you switching from/to?? It's not really clear, from what i read anyway.
What does your inetd line for qpopper look like? are you using nowait.400
in it? Have you tried running in standalone mode? Running low/empty on
disk space?

-Tony
.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.
Anthony J. Biacco                       Network Administrator/Engineer
thelittleprince at asteroid-b612 dot org       Intergrafix Internet Services

    "Dream as if you'll live forever, live as if you'll die today"
http://www.asteroid-b612.org                http://www.intergrafix.net
.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.

On Mon, 25 Jun 2001, Frazier Scott wrote:

> Hi all,
> 
> I'm a newbie in charge of a production server.  I have Caldera eServer 
> 2.3.1 installed with a new 2.2.16 kernel.  Everything has been working like 
> a charm.  Today is the first day of the switchover, and it is getting 
> toward the afternoon.  Pop clients have begun to report "connection refused 
> (10061)".  I have QPopper 4.0.3.  I restart inetd and all is well for 15 
> minutes when it begins again.  Am I running out of threads?  I have 
> debugging turned on, but nothing shows up.  Syslog just stops writing 
> messages to the logs when the service, whichever it is, hangs.  ps aux 
> shows the thread numbers climbing like crazy -- I'm up to 16,000 now, yet 
> the number of processes is around 130.  Is this a clue?  Don't these thread 
> numbers recycle themselves?  They seemed to when we were running Digital 
> Unix (yeah, I was a newbie there, too.  Still am.).
> 
> Help me, please!
> Frazier Scott - Internetwork Administrator
> Oregon Institute of Technology
> 3201 Campus Drive
> Klamath Falls, OR  97601
> Ph (541) 885-1310
> Fx (541) 885-1919
> scottf at oit dot edu
> 
> 


From: "Wilson Machasilla" <wilson at punto.net dot ec>
Subject: RV: QPopper 4.0.3 hanging? Or inetd? Help, please!
Date: Mon, 25 Jun 2001 18:17:02 -0500

I have the same problem with Red Hat 7.0 and xinetd, I tried with almast all
options of xinetd but nothing.
The only solution that I found is qpopper in standalone mode, is very fast!
but the result is low securitty.

Best regards.
Wilson Machasilla
Network Administrator

> ----- Original Message -----
> From: Frazier Scott <scottf at oit dot edu>
> To: Subscribers of Qpopper <qpopper at lists.pensive dot org>
> Sent: Monday, June 25, 2001 2:16 PM
> Subject: QPopper 4.0.3 hanging? Or inetd? Help, please!
>
>
> > Hi all,
> >
> > I'm a newbie in charge of a production server.  I have Caldera eServer
> > 2.3.1 installed with a new 2.2.16 kernel.  Everything has been working
> like
> > a charm.  Today is the first day of the switchover, and it is getting
> > toward the afternoon.  Pop clients have begun to report "connection
> refused
> > (10061)".  I have QPopper 4.0.3.  I restart inetd and all is well for 15
> > minutes when it begins again.  Am I running out of threads?  I have
> > debugging turned on, but nothing shows up.  Syslog just stops writing
> > messages to the logs when the service, whichever it is, hangs.  ps aux
> > shows the thread numbers climbing like crazy -- I'm up to 16,000 now,
yet
> > the number of processes is around 130.  Is this a clue?  Don't these
> thread
> > numbers recycle themselves?  They seemed to when we were running Digital
> > Unix (yeah, I was a newbie there, too.  Still am.).
> >
> > Help me, please!
> > Frazier Scott - Internetwork Administrator
> > Oregon Institute of Technology
> > 3201 Campus Drive
> > Klamath Falls, OR  97601
> > Ph (541) 885-1310
> > Fx (541) 885-1919
> > scottf at oit dot edu
> >
>


Date: Mon, 25 Jun 2001 16:48:58 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: RE: question about qpopper v4.0.3

At 9:43 AM -0400 6/25/01, Damian Ramirez wrote:

>  Jun 25 10:46:09.918 2001 [41860] +OK ready   [popper.c:246]

It seems you're still in SHY mode, but it doesn't seem to matter, because:

>  Jun 25 10:46:09.986 2001 [41860] Received (4): "CAPA" 
> [pop_get_command.c:105]
>  Jun 25 10:46:09.986 2001 [41860] capa returned 1; CurrentState 
> now auth1 [popper.c:329]
>  Jun 25 10:46:10.091 2001 [41860] (null) at 208.209.45.1 
> (208.209.45.1): -ERR POP EOF or I/O Error [popper.c:794]

The client disconnected after seeing the results of CAPA.  Odd. 
Maybe it didn't like the TLS support?

Date: Mon, 25 Jun 2001 17:02:06 -0700
From: Randall Gellens <randy at qualcomm dot com>
Subject: Re: OpenSSL certficate signing

At 9:44 AM +0200 6/25/01, Ted Lufoma wrote:

>  Please advise the e-mail address of the Certificate Authority.

You can use any CA you choose.  You can also self-sign.  However, 
the point of the CA is that it is a third-party trusted by the 
client to certify that you are who you say you are.  So you may 
want to pick a CA that is by default trusted by the email clients 
that will access your system.  Otherwise your users will have to 
manually approve the certificate.

Netscape has a list of several server CAs at 
<https://certs.netscape.com/server.html>.



Date: Tue, 26 Jun 2001 11:42:48 +0700
From: "Dapid Candra" <dapidc at cbn.net dot id>
Subject: Qpopper 4.0.3 doesn't support APOP?

Hi all,

I am just installing Qpopper 4.0.3 on Linux RedHat 6.2 (kernel 2.2.16).
 When configuring I used --enable-APOP.
Then make and make install.

After all successfull, I tried to connect using APOP, but the server
 respond with: 
	Mail server does not support APOP.:

But if I use just POP, there were no problems.

What has gone wrong?

Thanks.

dapidC



From: "Kent Morris" <gaunt at cophq dot org>
Subject: Trouble Compiling Qpopper
Date: Tue, 26 Jun 2001 01:34:52 -0400

I'm having trouble compiling QPopper with Mandrake Linux 8.0.  I've checked
through the archive for the list and found another person having this
problem, but no solution.

The problem is that maillock.c uses tempnam, which gcc doesn't seem to like.
It says it's dangerous and would prefer the use of mkstemp instead.  I have
modified the code to use mkstemp, but I would prefer a simpler and possibly
more tested solution.  A way of making gcc ignore the tempnam "error" and
compile the code anyhow would be handy.

Any suggestions/solutions are quite welcome.

Thanks in advance,

Kent Morris


From: "Kenneth Porter" <shiva at well dot com>
Date: Mon, 25 Jun 2001 22:49:29 -0700
Subject: Re: Qpopper 4.0.3 doesn't support APOP?

On Tue, 26 Jun 2001 11:42:48 +0700, Dapid Candra wrote:

>I am just installing Qpopper 4.0.3 on Linux RedHat 6.2 (kernel 2.2.16). When configuring I used --enable-APOP.

The option is case-sensitive: --enable-apop. configure will probably
ignore unrecognized options.

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, 26 Jun 2001 10:04:31 +0200
From: hypnose at t-online dot de (guenter wessling)
Subject: 3.1 still safe ?

Hi,all.
Sorry to return to an "oldie" for a second:
I still use 3.1 (latest).
There is still no known security bugs / buffer overflows, is there ?

Thanks
gw
guenter wessling (hypnose at t-online dot de)

From: "Jeff Gallagher" <jeffg at nfld dot net>
Subject: Re: QPopper 4.0.3 hanging? Or inetd? Help, please!
Date: Tue, 26 Jun 2001 11:31:59 -0230

I am experiencing a similar (although perhaps unrelated) problem. On a HP-UX
10.20 box upgrading from 2.5.3 is causing grief. I compiled 4.0.3 with
server mode enabled, log-login,enable-temp-drop (nothing special so far) and
am running with inetd using  popper4.0.3_inetd -c -C -F -R  as the command.
Here's what happens:

Under normal connection load using top the box shows ~90 processes with ~88
sleeping and very low load averages. (1.2 - 1.7). When I swap to 4.0.3
things fall apart. The number of processes slowly climbs from 90 to around
300 with 29X sleeping. Load averages skyrocket into never, never land and
the box grinds to a halt. There appears to be an increase in zombie
processes. When I tried the same in standalone mode the box kept creating
new processes until it filled the proc table - then the parent died. This
cycle takes about 15 minutes under nominal load. Before attempting this
under any real traffic I soaked 4.0.3 overnight for about 9 hours under
fairly low connection volume without any problems at all.

I have tried the nowait. solution with no luck - I didn't think I was
getting THAT many connections.

Can anyone shed some light on this? BTW-> There was nothing useful in the
logs...

Jeff Gallagher
Advanced Internetworking Support
Aliant Telecom Inc.
709.739.2088
jeff.gallagher at aliant dot ca

----- Original Message -----
From: "Wilson Machasilla" <wilson at punto.net dot ec>
To: "Subscribers of Qpopper" <qpopper at lists.pensive dot org>
Sent: Monday, June 25, 2001 8:47 PM
Subject: RV: QPopper 4.0.3 hanging? Or inetd? Help, please!


> I have the same problem with Red Hat 7.0 and xinetd, I tried with almast
all
> options of xinetd but nothing.
> The only solution that I found is qpopper in standalone mode, is very
fast!
> but the result is low securitty.
>
> Best regards.
> Wilson Machasilla
> Network Administrator
>
> > ----- Original Message -----
> > From: Frazier Scott <scottf at oit dot edu>
> > To: Subscribers of Qpopper <qpopper at lists.pensive dot org>
> > Sent: Monday, June 25, 2001 2:16 PM
> > Subject: QPopper 4.0.3 hanging? Or inetd? Help, please!
> >
> >
> > > Hi all,
> > >
> > > I'm a newbie in charge of a production server.  I have Caldera eServer
> > > 2.3.1 installed with a new 2.2.16 kernel.  Everything has been working
> > like
> > > a charm.  Today is the first day of the switchover, and it is getting
> > > toward the afternoon.  Pop clients have begun to report "connection
> > refused
> > > (10061)".  I have QPopper 4.0.3.  I restart inetd and all is well for
15
> > > minutes when it begins again.  Am I running out of threads?  I have
> > > debugging turned on, but nothing shows up.  Syslog just stops writing
> > > messages to the logs when the service, whichever it is, hangs.  ps aux
> > > shows the thread numbers climbing like crazy -- I'm up to 16,000 now,
> yet
> > > the number of processes is around 130.  Is this a clue?  Don't these
> > thread
> > > numbers recycle themselves?  They seemed to when we were running
Digital
> > > Unix (yeah, I was a newbie there, too.  Still am.).
> > >
> > > Help me, please!
> > > Frazier Scott - Internetwork Administrator
> > > Oregon Institute of Technology
> > > 3201 Campus Drive
> > > Klamath Falls, OR  97601
> > > Ph (541) 885-1310
> > > Fx (541) 885-1919
> > > scottf at oit dot edu
> > >
> >
>
>


Date: Tue, 26 Jun 2001 11:12:22 -0400 (EDT)
From: Admin Mailing Lists <mlist at intergrafix dot net>
Subject: Re: QPopper 4.0.3 hanging? Or inetd? Help, please!

turn on --enable-debugging and run with -T /path/to/tracelog
see, if you can find any useful info in the log files for the children
getting spawned.
if you're C function savvy, you could also try stracing (truss?) the
qpopper parent, making sure to follow forks, etc.. 
strace -o strace.log -f -p qpopper_parent_pid

-Tony
.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.
Anthony J. Biacco                       Network Administrator/Engineer
thelittleprince at asteroid-b612 dot org       Intergrafix Internet Services

    "Dream as if you'll live forever, live as if you'll die today"
http://www.asteroid-b612.org                http://www.intergrafix.net
.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.

On Tue, 26 Jun 2001, Jeff Gallagher wrote:

> I am experiencing a similar (although perhaps unrelated) problem. On a HP-UX
> 10.20 box upgrading from 2.5.3 is causing grief. I compiled 4.0.3 with
> server mode enabled, log-login,enable-temp-drop (nothing special so far) and
> am running with inetd using  popper4.0.3_inetd -c -C -F -R  as the command.
> Here's what happens:
> 
> Under normal connection load using top the box shows ~90 processes with ~88
> sleeping and very low load averages. (1.2 - 1.7). When I swap to 4.0.3
> things fall apart. The number of processes slowly climbs from 90 to around
> 300 with 29X sleeping. Load averages skyrocket into never, never land and
> the box grinds to a halt. There appears to be an increase in zombie
> processes. When I tried the same in standalone mode the box kept creating
> new processes until it filled the proc table - then the parent died. This
> cycle takes about 15 minutes under nominal load. Before attempting this
> under any real traffic I soaked 4.0.3 overnight for about 9 hours under
> fairly low connection volume without any problems at all.
> 
> I have tried the nowait. solution with no luck - I didn't think I was
> getting THAT many connections.
> 
> Can anyone shed some light on this? BTW-> There was nothing useful in the
> logs...
> 
> Jeff Gallagher
> Advanced Internetworking Support
> Aliant Telecom Inc.
> 709.739.2088
> jeff.gallagher at aliant dot ca
> 
> ----- Original Message -----
> From: "Wilson Machasilla" <wilson at punto.net dot ec>
> To: "Subscribers of Qpopper" <qpopper at lists.pensive dot org>
> Sent: Monday, June 25, 2001 8:47 PM
> Subject: RV: QPopper 4.0.3 hanging? Or inetd? Help, please!
> 
> 
> > I have the same problem with Red Hat 7.0 and xinetd, I tried with almast
> all
> > options of xinetd but nothing.
> > The only solution that I found is qpopper in standalone mode, is very
> fast!
> > but the result is low securitty.
> >
> > Best regards.
> > Wilson Machasilla
> > Network Administrator
> >
> > > ----- Original Message -----
> > > From: Frazier Scott <scottf at oit dot edu>
> > > To: Subscribers of Qpopper <qpopper at lists.pensive dot org>
> > > Sent: Monday, June 25, 2001 2:16 PM
> > > Subject: QPopper 4.0.3 hanging? Or inetd? Help, please!
> > >
> > >
> > > > Hi all,
> > > >
> > > > I'm a newbie in charge of a production server.  I have Caldera eServer
> > > > 2.3.1 installed with a new 2.2.16 kernel.  Everything has been working
> > > like
> > > > a charm.  Today is the first day of the switchover, and it is getting
> > > > toward the afternoon.  Pop clients have begun to report "connection
> > > refused
> > > > (10061)".  I have QPopper 4.0.3.  I restart inetd and all is well for
> 15
> > > > minutes when it begins again.  Am I running out of threads?  I have
> > > > debugging turned on, but nothing shows up.  Syslog just stops writing
> > > > messages to the logs when the service, whichever it is, hangs.  ps aux
> > > > shows the thread numbers climbing like crazy -- I'm up to 16,000 now,
> > yet
> > > > the number of processes is around 130.  Is this a clue?  Don't these
> > > thread
> > > > numbers recycle themselves?  They seemed to when we were running
> Digital
> > > > Unix (yeah, I was a newbie there, too.  Still am.).
> > > >
> > > > Help me, please!
> > > > Frazier Scott - Internetwork Administrator
> > > > Oregon Institute of Technology
> > > > 3201 Campus Drive
> > > > Klamath Falls, OR  97601
> > > > Ph (541) 885-1310
> > > > Fx (541) 885-1919
> > > > scottf at oit dot edu
> > > >
> > >
> >
> >
> 
> 


Date: Tue, 26 Jun 2001 16:40:30 +0100
From: Wasim Bashir <wasim.bashir at broadband.co dot uk>
Subject: web interface

Hi,

I'm looking for a Web Interface to read a Mailbox that supports APOP, does
anyone know of one ?

Also what other mail clients support Apop, i've only found 3, these
are Eudora, The Bat and Pegasus.

Thanks

Wasim



Date: Tue, 26 Jun 2001 08:58:19 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Re: web interface

On Tue, Jun 26, 2001 at 04:40:30PM +0100, Wasim Bashir wrote:
> I'm looking for a Web Interface to read a Mailbox that supports APOP, does
> anyone know of one ?

Sorry, no. Most of the web interfaces I know of are based on IMAP.
  -- Clifton

-- 
 Clifton Royston  --  LavaNet Systems Architect --  cliftonr at lava dot net
   WWJD?   "JWRTFM!" - Scott Dorsey (kludge)   "JWG" - Eddie Aikau

From: "P S Graham" <patrick at gandatech dot com>
Subject: Re: web interface
Date: Tue, 26 Jun 2001 15:40:10 -0400

You may want to take a look at thinmailer
(http://www.oakbox.com/scripts/thinmailer.shtml)

I can't say how well it works with APOP, as I only use it
with POP. It is fairly customizable as well.

Patrick (returning to lurk mode)


----- Original Message -----
From: Wasim Bashir <wasim.bashir at broadband.co dot uk>
To: Subscribers of Qpopper <qpopper at lists.pensive dot org>
Sent: Tuesday, June 26, 2001 11:40 AM
Subject: web interface


> Hi,
>
> I'm looking for a Web Interface to read a Mailbox that supports APOP, does
> anyone know of one ?
>
> Also what other mail clients support Apop, i've only found 3, these
> are Eudora, The Bat and Pegasus.
>
> Thanks
>
> Wasim
>


Date: Tue, 26 Jun 2001 15:33:54 -0700
From: Frazier Scott <scottf at oit dot edu>
Subject: Re: QPopper 4.0.3 hanging? Or inetd? Help, please!

Many, many thanks to all who responded!!

It looks like this was the winner.  In this kernel, at least, inetd 
defaults to 40 if no number is specified with the nowait parameter.  I 
upped it and so far so good.  We'll see how it looks at the end of the week.

We were running Digital Unix 4.0G on a Dec Alpha 2100.  Way too much money 
in licensing and hardware annual fees, so linux on an Intel platform seemed 
a feasible and economical option.

Frazier.

At 03:53 PM 6/25/2001 -0400, Admin Mailing Lists wrote:

>I guess the first question would be, when you said switchover, what are
>you switching from/to?? It's not really clear, from what i read anyway.
>What does your inetd line for qpopper look like? are you using nowait.400
>in it? Have you tried running in standalone mode? Running low/empty on
>disk space?
>
>-Tony
>.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.
>Anthony J. Biacco                       Network Administrator/Engineer
>thelittleprince at asteroid-b612 dot org       Intergrafix Internet Services
>
>     "Dream as if you'll live forever, live as if you'll die today"
>http://www.asteroid-b612.org                http://www.intergrafix.net
>.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.
>
>On Mon, 25 Jun 2001, Frazier Scott wrote:
>
> > Hi all,
> >
> > I'm a newbie in charge of a production server.  I have Caldera eServer
> > 2.3.1 installed with a new 2.2.16 kernel.  Everything has been working 
> like
> > a charm.  Today is the first day of the switchover, and it is getting
> > toward the afternoon.  Pop clients have begun to report "connection 
> refused
> > (10061)".  I have QPopper 4.0.3.  I restart inetd and all is well for 15
> > minutes when it begins again.  Am I running out of threads?  I have
> > debugging turned on, but nothing shows up.  Syslog just stops writing
> > messages to the logs when the service, whichever it is, hangs.  ps aux
> > shows the thread numbers climbing like crazy -- I'm up to 16,000 now, yet
> > the number of processes is around 130.  Is this a clue?  Don't these 
> thread
> > numbers recycle themselves?  They seemed to when we were running Digital
> > Unix (yeah, I was a newbie there, too.  Still am.).
> >
> > Help me, please!
> > Frazier Scott - Internetwork Administrator
> > Oregon Institute of Technology
> > 3201 Campus Drive
> > Klamath Falls, OR  97601
> > Ph (541) 885-1310
> > Fx (541) 885-1919
> > scottf at oit dot edu
> >
> >


From: "Kenneth Porter" <shiva at well dot com>
Date: Tue, 26 Jun 2001 18:23:56 -0700
Subject: Re: web interface

On Tue, 26 Jun 2001 16:40:30 +0100, Wasim Bashir wrote:

>Also what other mail clients support Apop, i've only found 3, these
>are Eudora, The Bat and Pegasus.

PMMail, http://www.pmmail2000.com/, available for both Win32 and OS/2.
Handles multiple accounts simultaneously and has PGP support. Does NOT
do IMAP or TLS. Crude HTML display (but who wants HTML in email
anyway?). Pretty good filtering capability with a good filter pattern
and action language. If you have both the OS/2 and Win32 version in the
same directory, they'll use the same data.

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, 27 Jun 2001 09:53:21 +0700
From: Ilya Palagin <ipalagin at fioc dot kz>
Subject: pop3/tcp server failing (looping), service terminated

Hi,

Doing stupid experiments with my server (Mandrake 7.0, qpopper 4.0.3) I 
found an easy way to shut it down from any client, sending a bunch of 
requests by clicking (3 clicks per sec) on GetMsg button (in my case 
it's Mozilla). inetd terminates qpopper with the message:

inetd[379]: pop3/tcp server failing (looping), service terminated

How can I prevent a malicious user from clicking so fast? :-)

Thanks,
Ilya.



Date: Tue, 26 Jun 2001 23:45:02 -0400
From: Jack Sasportas <jack at innovativeinternet dot com>
Subject: Re: pop3/tcp server failing (looping), service terminated

After testing this on my setup,  I beleive there is something else wrong on
your side.
I'm running 4.03 along under RH 6.2, and I hit the get button like 20 times
in a row to see if I would get the same problem, and I it runs perfectly.

Jack

Ilya Palagin wrote:

> Hi,
>
> Doing stupid experiments with my server (Mandrake 7.0, qpopper 4.0.3) I
> found an easy way to shut it down from any client, sending a bunch of
> requests by clicking (3 clicks per sec) on GetMsg button (in my case
> it's Mozilla). inetd terminates qpopper with the message:
>
> inetd[379]: pop3/tcp server failing (looping), service terminated
>
> How can I prevent a malicious user from clicking so fast? :-)
>
> Thanks,
> Ilya.

--
___________________________________________________________
Jack Sasportas
Innovative Internet Solutions
Phone 305.665.2500
Fax 305.665.2551
www.innovativeinternet.com
www.web56.net



From: peter at netlink.com dot au
Subject: Re: pop3/tcp server failing (looping), service terminated
Date: Wed, 27 Jun 2001 13:49:43 +1000 (EST)

Ilya,

You can increase the 'tolerance' (number of requests it will handle)
for inetd with the 'nowait.(number)' parameter in the popper line of
inetd.conf. I think the default is 40 - you could try making it 200
or 400...
Apart from that, in my (limited) testing I can't cause this sort of
problem on our system - perhaps it has something to do with your
qpopper config since, in theory you shouldn't be able to make multiple
pop connections for the same user at the same time...
(Do you get multiple poppers for the same user in your process list?).

Regards, Peter

> 
> Hi,
> 
> Doing stupid experiments with my server (Mandrake 7.0, qpopper 4.0.3) I 
> found an easy way to shut it down from any client, sending a bunch of 
> requests by clicking (3 clicks per sec) on GetMsg button (in my case 
> it's Mozilla). inetd terminates qpopper with the message:
> 
> inetd[379]: pop3/tcp server failing (looping), service terminated
> 
> How can I prevent a malicious user from clicking so fast? :-)
> 
> Thanks,
> Ilya.
> 
> 
> 


Date: Wed, 27 Jun 2001 12:59:13 +0900
From: Peter Evans <peter at gol dot com>
Subject: Re: pop3/tcp server failing (looping), service terminated

Jack Sasportas (jack at innovativeinternet dot com) wrote:
> Ilya Palagin wrote:
> > inetd[379]: pop3/tcp server failing (looping), service terminated
> > How can I prevent a malicious user from clicking so fast? :-)

	Dont use inetd, or tune your inetd service to allow for a lot
	more pop sessions being started.

	xinetd is good, as is DJB's tcpserver.

	I used to use tcpserver, because it was quite happy spawning
	200+ / second which was close to our peak load, but the next
	version of the GOL.COM mail system uses exim/pop-proxy/qpopper
	and xinetd because its run by a linux drooly who refuses to
	use anything by DJB on principle.

	Whatever cranks your tractor ...

	P
	----:
	

From: "Desmond Lim" <ddl76 at singnet.com dot sg>
Subject: APOP Problem
Date: Wed, 27 Jun 2001 11:58:49 +0800


Hi all,

----------------------------------------------------------------------------
----
I'm using qpopper 4 on RH6.1. I'm also using APOP authentication. It seems
that something is wrong even if I follow the installation guide closely.
I get this error in my maillog
"-ERR [AUTH] You must use stronger authentication such as AUTH SCRAM-MD5 or
APOP to connect to this server".

Does anyone know how to resolve this?


From: "Kenneth Porter" <shiva at well dot com>
Date: Wed, 27 Jun 2001 01:10:20 -0700
Subject: APOP clients

On Tue, 26 Jun 2001 16:40:30 +0100, Wasim Bashir wrote:

>Also what other mail clients support Apop, i've only found 3, these
>are Eudora, The Bat and Pegasus.

Just found another, Ximian Evolution for Gnome
(http://www.ximian.com/). It uses the Camel mail library (modeled on
JavaMail), which includes support for APOP.

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, 27 Jun 2001 08:48:14 +0000
From: =?iso-8859-1?Q?Guðbjörn?= Hreinsson <gsh at skima dot is>
Subject: How do clients keep status of maildrop


Hi, this is perhaps slightly off topic, if so I apologize.

I am moving/migrating maildrops from NT Mail to Unix, I have all the 
details hammered out except that some clients will redownload all 
the messages in the maildrop if the user has selected to leave mail 
on the server... Outlook and Netscape at least but Eudora seems to 
retain the status between servers... Fetchmail also seems to realise 
that the messages have been read and should not be redownloaded.

So my question is how do clients determine the status of the maildrop?
Do they use TOP, UIDL, scan the Status header or X-UID[L] header or 
some esoteric thing?


Thanks for any pointers.
-GSH

Kenneth Porter wrote:
> 
> On Tue, 26 Jun 2001 16:40:30 +0100, Wasim Bashir wrote:
> 
> >Also what other mail clients support Apop, i've only found 3, these
> >are Eudora, The Bat and Pegasus.
> 
> PMMail, http://www.pmmail2000.com/, available for both Win32 and OS/2.
> Handles multiple accounts simultaneously and has PGP support. Does NOT
> do IMAP or TLS. Crude HTML display (but who wants HTML in email
> anyway?). Pretty good filtering capability with a good filter pattern
> and action language. If you have both the OS/2 and Win32 version in the
> same directory, they'll use the same data.
> 
> 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, 27 Jun 2001 12:20:47 +0100
From: Tim Jackson <tim at timj.co dot uk>
Subject: Re: APOP clients

Hi Kenneth, on Wed, 27 Jun 2001 01:10:20 -0700 you wrote:

> >Also what other mail clients support Apop, i've only found 3, these
> >are Eudora, The Bat and Pegasus.
> Just found another, Ximian Evolution for Gnome
> (http://www.ximian.com/). It uses the Camel mail library (modeled on
> JavaMail), which includes support for APOP.

Sylpheed (http://sylpheed.good-day.net/) is an excellent mail client that
apparently supports APOP, although I haven't personally tried using it in
APOP mode.

Tim

From: Frank Pineau <frank at pineaus dot com>
Subject: Re: APOP clients
Date: Wed, 27 Jun 2001 08:05:00 -0400

On Wed, 27 Jun 2001 12:20:47 +0100, you wrote:

>Hi Kenneth, on Wed, 27 Jun 2001 01:10:20 -0700 you wrote:
>
>> >Also what other mail clients support Apop, i've only found 3, these
>> >are Eudora, The Bat and Pegasus.
>> Just found another, Ximian Evolution for Gnome
>> (http://www.ximian.com/). It uses the Camel mail library (modeled on
>> JavaMail), which includes support for APOP.
>
>Sylpheed (http://sylpheed.good-day.net/) is an excellent mail client 
that
>apparently supports APOP, although I haven't personally tried using it 
in
>APOP mode.
>
>Tim

=46orte Agent does as well.  http://www.forteinc.com

=46P


-- 
<*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*>
=46rank Pineau  ------------>

        frank at pineaus.IHATESPAM dot com
          http://www.pineaus.com

It is pitch black. You are likely to be eaten by a grue.
<*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*>
PGP Public Key:
http://www.pineaus.com/frankpineau.asc

No mail from web-based mail systems such as Hotmail
or Yahoo will be accepted.

Date: Wed, 27 Jun 2001 09:39:06 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: How do I set up Bulletins

Greetings,

I want to be able to send notices out to my clients.
Im running qpopper 4.x on bsdi 4.2.

It has been a long time since i tried this, and the last time I had 
some problems. Im looking for a good tutorial on setting up bulletins.

Any info is helpful.

thanks

--jeff
-- 
------------------------------------------------------------------------
Jeff Donovan                    Network Analyst
Bethlehem Area School District	Network Operations
Bethlehem, PA  18020            (610) 807-5571  jdonovan at beth.k12.pa dot us

From: Frank Pineau <frank at pineaus dot com>
Subject: Re: How do I set up Bulletins
Date: Wed, 27 Jun 2001 09:53:19 -0400

On Wed, 27 Jun 2001 09:39:06 -0400, you wrote:

>Greetings,
>
>I want to be able to send notices out to my clients.
>Im running qpopper 4.x on bsdi 4.2.
>
>It has been a long time since i tried this, and the last time I had 
>some problems. Im looking for a good tutorial on setting up bulletins.

Have you read the Qpopper manual?  It's pretty well detailed in there.

=46P


-- 
<*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*>
=46rank Pineau  ------------>

        frank at pineaus.IHATESPAM dot com
          http://www.pineaus.com

It is pitch black. You are likely to be eaten by a grue.
<*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*><*>
PGP Public Key:
http://www.pineaus.com/frankpineau.asc

No mail from web-based mail systems such as Hotmail
or Yahoo will be accepted.

Date: Wed, 27 Jun 2001 10:00:40 -0400 (EDT)
From: Admin Mailing Lists <mlist at intergrafix dot net>
Subject: Re: pop3/tcp server failing (looping), service terminated

On Wed, 27 Jun 2001, Peter Evans wrote:

> 
> 	Dont use inetd, or tune your inetd service to allow for a lot
> 	more pop sessions being started.
> 
> 	xinetd is good, as is DJB's tcpserver.
> 

I used to use qpopper w/xinetd, but after a while (few weeks) xinetd RSS
would get up to like 50 meg. I guess it (xinetd) has a snowballing
memory leak somewhere.
So now i just run qpopper standalone and the system is happier memory
wise.


-Tony
.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.
Anthony J. Biacco                       Network Administrator/Engineer
thelittleprince at asteroid-b612 dot org       Intergrafix Internet Services

    "Dream as if you'll live forever, live as if you'll die today"
http://www.asteroid-b612.org                http://www.intergrafix.net
.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-._.-.

> 	I used to use tcpserver, because it was quite happy spawning
> 	200+ / second which was close to our peak load, but the next
> 	version of the GOL.COM mail system uses exim/pop-proxy/qpopper
> 	and xinetd because its run by a linux drooly who refuses to
> 	use anything by DJB on principle.
> 
> 	Whatever cranks your tractor ...
> 
> 	P
> 	----:
> 	
> 


Date: Wed, 27 Jun 2001 10:26:39 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: Re: How do I set up Bulletins

:)
I sent this message off before I got finished with the man page.

one question

how can i create a Document/ bulletin in "Mailbox Format", without 
going through sending a copy to myself then saving it back?

--jeff

>
>
>Have you read the Qpopper manual?  It's pretty well detailed in there.
>
>FP

-- 
------------------------------------------------------------------------
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, 27 Jun 2001 09:51:51 -0500
From: Tim Olson <tolson at unionsemiconductor dot com>
Subject: Re: pop3/tcp server failing (looping), service terminated

I don't know if this could help or not, but I did just notice that 
for RedHat 7.1 they have released a new Xinetd-2.1.8.9pre15-2.i386.rpm

Maybe this fixes something?

Tim


Admin Mailing Lists wrote:
> 
> On Wed, 27 Jun 2001, Peter Evans wrote:
> 
> >
> >       Dont use inetd, or tune your inetd service to allow for a lot
> >       more pop sessions being started.
> >
> >       xinetd is good, as is DJB's tcpserver.
> >
> 
> I used to use qpopper w/xinetd, but after a while (few weeks) xinetd RSS
> would get up to like 50 meg. I guess it (xinetd) has a snowballing
> memory leak somewhere.
> So now i just run qpopper standalone and the system is happier memory
> wise.
> 
> -Tony

Date: Wed, 27 Jun 2001 10:15:20 -0500
From: Steve Palm <stevep at sga dot org>
Subject: SSL and XTND XMIT

I was able to get SSL connections to work fine with Eudora clients and QPopper 4.0.3, but when trying to use XTND XMIT for sending over an SSL connection, it will not work.

So, SSL works by itself.

And, XTND XMIT works fine using with normal POP3 mode.

But SSL and XTND XMIT do not work.

Anyone out there ever try this combo?

-- 
-------------------------
Steven Palm
Slavic Gospel Association - Loves Park - IL - USA
stevep at sga dot org
-------------------------

Date: Wed, 27 Jun 2001 11:27:55 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: Bulletins * need clarification *

greetings,
I am having problems getting the right message format (in think).

In /etc/inetd.conf i have this:
pop     stream  tcp     nowait  root    /usr/libexec/tcpd 
popper -b /var/spool/bulls

I have created /var/spool/bulls
drwxr-xr-x   2 root  wheel  512 Jun 27 09:39 bulls


I then created a message using eudora and sent a copy to myself.
then I copied the message from /var/mail/myacct to /var/spool/bulls/1.test

then I killed and restarted popper.

the .popbull file is in my directory but the message did not go 
through to the users.

any hints.
what is the proper way of getting the bulletin into the bulls directory?

--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, 27 Jun 2001 14:31:06 -0400
From: Jeff Donovan <jdonovan at beth.k12.pa dot us>
Subject: Re: Bulletins * need clarification PLEASE *

So far the system is telling me this:
  Bulletin 1.test does not start with a valid "From " separator

Ok,..

i have tried editing this From line so many different ways, and Im 
still getting the same error.


What is this " From " separator supposed to be?
-- 
------------------------------------------------------------------------
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, 27 Jun 2001 16:22:39 -0500
From: Butch Kemper <kemper at tstar dot net>
Subject: Re: Bulletins * need clarification PLEASE *

It is the first line of every mail message in the mbox file format.  Use 
the vi editor and look at the file /var/mail/jdonovan and you will see a 
"From " line before each message.

Butch

At 01:31 PM 6/27/01, you wrote:
>So far the system is telling me this:
>  Bulletin 1.test does not start with a valid "From " separator
>
>Ok,..
>
>i have tried editing this From line so many different ways, and Im still 
>getting the same error.
>
>
>What is this " From " separator supposed to be?
>--
>------------------------------------------------------------------------
>Jeff Donovan                    Network Analyst
>Bethlehem Area School District  Network Operations
>Bethlehem, PA  18020            (610) 807-5571  jdonovan at beth.k12.pa dot us


TSTAR Internet, Inc         | Making the Net Work
Marble Falls, TX            | Serving Blanco, Burnet,
830-693-6967                | Llano, and Mason Counties


Date: Wed, 27 Jun 2001 11:41:23 -1000
From: Clifton Royston <cliftonr at lava dot net>
Subject: Re: How do clients keep status of maildrop

On Wed, Jun 27, 2001 at 08:48:14AM +0000, Guðbjörn Hreinsson wrote:
> Hi, this is perhaps slightly off topic, if so I apologize.
> 
> I am moving/migrating maildrops from NT Mail to Unix, I have all the 
> details hammered out except that some clients will redownload all 
> the messages in the maildrop if the user has selected to leave mail 
> on the server... Outlook and Netscape at least but Eudora seems to 
> retain the status between servers... Fetchmail also seems to realise 
> that the messages have been read and should not be redownloaded.
> 
> So my question is how do clients determine the status of the maildrop?
> Do they use TOP, UIDL, scan the Status header or X-UID[L] header or 
> some esoteric thing?

Generally speaking, I believe they *should* be using the output of the
UIDL command and comparing it to the UIDL headers on what they've
downloaded.  

I don't have any info on what the different clients actually are doing.

  -- Clifton

-- 
 Clifton Royston  --  LavaNet Systems Architect --  cliftonr at lava dot net
   WWJD?   "JWRTFM!" - Scott Dorsey (kludge)   "JWG" - Eddie Aikau

Date: Wed, 27 Jun 2001 22:51:42 -0400
From: Jack Sasportas <jack at innovativeinternet dot com>
Subject: I/O error flushing output to client ???

OK I have looked at past posts, and looked on the web with no answer to
this problem.
I noticed the few times I get this, there are literaly hundreds of
entries in the logfile.

Can someone from Qualcomm please help out and really define the
importance of this message?

Thanks !

Jack


___________________________________________________________
Jack Sasportas
Innovative Internet Solutions
Phone 305.665.2500
Fax 305.665.2551
www.innovativeinternet.com
www.web56.net



Date: Thu, 28 Jun 2001 11:20:53 +0800 (HKT)
From: PM WONG <pmwong at power25t.hkbu.edu dot hk>
Subject: .user.pop files in RAM disk, how faster?

We run qpopper 3.0.2 on an AIX unix box. As most users choose
to keep mails on server, hence it becomes quite slow when plenty
users are reading their mail at the same time (though i've
placed the .pop files directory separate from their /var/spool/mail
mailboxes). 
Now since the newest version of  AIX OS has this feature of
using part of the RAM as disk (using a command called mkramdisk)
, i wonder will it improve things much ( i did a simple test
on a test machine for 1 mailbox, but the difference is not significant)
for qpopper. Could it be that the reading of the mail from /var/spool/mail
to create the .pop files is also the bottleneck.
Any comments?


Date: Thu, 28 Jun 2001 00:11:51 -0400
From: Daniel Senie <dts at senie dot com>
Subject: Tools/methods for deleting old email?

I'd like to find some tools to go through mailboxes (mbox format, in 
/var/spool/mail) and delete anything that's been read, and which is more 
than 30 days old. I've got some users who just won't set the "delete from 
server" switch, and it's time to take action. I really don't want to do 
quotas enforcement at this point, though.

Anyone know of tools to do this? I'd love it if Qpopper had an option for 
this and would just perform the trim while it was doing its processing, but 
the only switch I find there is auto-delete.

An alternative would be to add auto-delete on a per-user basis, targetting 
the abusive users only.

Any thoughts or pointers to tools would be greatly appreciated.

Dan
-----------------------------------------------------------------
Daniel Senie                                        dts at senie dot com
Amaranth Networks Inc.                    http://www.amaranth.com


Last updated on 27 Jun 2001 by Pensive Mailing List Admin