The qpopper list archive ending on 12 Jun 2000
Topics covered in this issue include:
1. Re: How can I make qpopper create dot-lock file in different
Butch Kemper <kemper at tstar dot net>
Wed, 07 Jun 2000 10:03:59 -0500
2. Qpopper-3.0.2 on DEC-Unix-4.0e segmentation violation
Sylvain Robitaille <syl at alcor.concordia dot ca>
Wed, 7 Jun 2000 11:25:22 -0400 (EDT)
3. test
"Jack Barnett" <jbarnett at axil.netmate dot com>
Wed, 7 Jun 2000 10:25:47 -0500
4.
"jmcintyr" <jmcintyr at i-plus dot net>
Wed, 7 Jun 2000 14:33:51 -0400
5. RE:
"=?iso-8859-1?Q?Philipp_Gaschütz?=" <philipp at gng dot de>
Wed, 7 Jun 2000 20:57:00 +0200
6. Re:
Stephen Samuel <samuel at bcgreen dot com>
Wed, 07 Jun 2000 15:55:43 -0700
7. Re: How can I make qpopper create dot-lock file in different
Nurah Muhammad <nurah at mimos dot my>
Thu, 8 Jun 2000 08:56:24 +0800 (MYT)
8. Re:
"jmcintyr" <jmcintyr at i-plus dot net>
Thu, 8 Jun 2000 08:40:51 -0400
9. Corrupt Mailboxes
"Paul Trippett" <pjt at gmxuk dot net>
Thu, 8 Jun 2000 16:19:49 +0100
10. Re: Impact of -R option
James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Thu, 8 Jun 2000 14:32:03 -0400 (EDT)
11. Re: Corrupt Mailboxes
Joel Laing <joel at scripps dot edu>
Thu, 08 Jun 2000 11:53:21 -0700
12. Re: Corrupt Mailboxes
"Paul Trippett" <pjt at gmxuk dot net>
Thu, 8 Jun 2000 20:03:06 +0100
13. Re: Corrupt Mailboxes
Joel Laing <joel at scripps dot edu>
Thu, 08 Jun 2000 12:36:40 -0700
14. Re: Impact of -R option
John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Thu, 8 Jun 2000 14:14:02 -0600 (MDT)
15. Popper logging?
Chris Cook <ccook at tcworks dot net>
Thu, 08 Jun 2000 17:32:35 +0000
16. Mail security alert: Linux kernels prior to 2.2.16
"Kenneth Porter" <shiva at well dot com>
Thu, 08 Jun 2000 17:26:21 -0700
17. Re: Corrupt Mailboxes
John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Thu, 8 Jun 2000 20:32:12 -0600 (MDT)
18. qpopper or just popper in inetd.conf
PM WONG <pmwong at power25t.hkbu.edu dot hk>
Fri, 9 Jun 2000 12:10:54 +0800 (HKT)
19. .lock files in versions 2 & 3
PM WONG <pmwong at power25t.hkbu.edu dot hk>
Fri, 9 Jun 2000 12:45:16 +0800 (HKT)
20. Re: 3.0.2 distribution problem
Qpopper Support <qpopper at qualcomm dot com>
Fri, 9 Jun 2000 00:08:53 -0700
21. Unable to get canonical name of client
Serkan AKCIN <serkan at baskent.edu dot tr>
Fri, 9 Jun 2000 14:51:35 +0300 (EET DST)
22. Please remove dward at process-technology dot com
"Don Ward" <dward at process-technology dot com>
Fri, 09 Jun 2000 08:40:10 -0400
23. Re: Unable to get canonical name of client
"pestilence" <pestilence at netplan dot gr>
Fri, 9 Jun 2000 15:53:35 +0300
24. Re: Popper logging?
Jim Sander <jim at federation.addy dot com>
Fri, 9 Jun 2000 08:52:34 -0400 (EDT)
25. Re: Unable to get canonical name of client
"pestilence" <pestilence at netplan dot gr>
Fri, 9 Jun 2000 16:15:08 +0300
26. Re: 3.0.2 distribution problem
"jmcintyr" <jmcintyr at i-plus dot net>
Fri, 9 Jun 2000 10:30:00 -0400
27. Re: Unable to get canonical name of client
Ken Hornstein <kenh at cmf.nrl.navy dot mil>
Fri, 09 Jun 2000 10:48:41 -0400
28. Please remove poopie at jeeves.poopie dot net
P O O P I E <poopie at jeeves.poopie dot net>
Fri, 9 Jun 2000 12:11:19 +0000 (GMT)
29. Re: 3.0.2 distribution problem
Qpopper Support <qpopper at qualcomm dot com>
Fri, 9 Jun 2000 09:01:46 -0700
30. Please remove gwong at ascend dot com
Gary Wong <gwong at ascend dot com>
Fri, 9 Jun 2000 09:34:30 -0700 (PDT)
31. How To Unsubscribe from Qpopper
Daniel Norton <danorton at chsw dot com>
Fri, 09 Jun 2000 13:01:17 -0400
32. Re: Corrupt Mailboxes
Qpopper Support <qpopper at qualcomm dot com>
Fri, 9 Jun 2000 10:16:46 -0700
33. qpopper - I can't send any message!
"Eduardo M. A. M. Mendes" <emmendes at funrei dot br>
Thu, 08 Jun 2000 22:42:56 -0300
34. Re: 3.0.2 distribution problem
"jmcintyr" <jmcintyr at i-plus dot net>
Fri, 9 Jun 2000 13:30:54 -0400
35. Re: Please remove poopie at jeeves.poopie dot net
"Keith Warno" <keith at HaggleWare dot com>
Fri, 9 Jun 2000 14:01:29 -0400
36. Re: Popper logging?
James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Fri, 9 Jun 2000 14:01:02 -0400 (EDT)
37. qpoppper 3.0.2 inconsistency: LOG_LOCAL0 versus LOG_LOCAL1
James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Fri, 9 Jun 2000 14:11:40 -0400 (EDT)
38. "common" errors
"Keith Warno" <keith at HaggleWare dot com>
Fri, 9 Jun 2000 15:03:17 -0400
39. Re: Impact of -R option
James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Fri, 9 Jun 2000 15:33:43 -0400 (EDT)
40. Re: Please remove poopie at jeeves.poopie dot net
Leonard Hermens <Leonard.Hermens at rcity dot com>
Fri, 9 Jun 2000 13:07:19 -0700
41. RE: qpopper - I can't send any message!
"=?iso-8859-1?Q?Philipp_Gaschütz?=" <philipp at gng dot de>
Fri, 9 Jun 2000 22:31:25 +0200
42. Re: Impact of -R option
John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Fri, 9 Jun 2000 15:55:07 -0600 (MDT)
43. Re: Please remove poopie at jeeves.poopie dot net
John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Fri, 9 Jun 2000 15:59:24 -0600 (MDT)
44. Re: Please remove poopie at jeeves.poopie dot net
"Lisa Casey" <lisa at jellico dot com>
Fri, 9 Jun 2000 17:50:57 -0400
45. Re: How To Unsubscribe from Qpopper
Fergal Daly <fergal at esatclear dot ie>
Fri, 09 Jun 2000 23:12:14 +0100
46. qpopper -T timeout, How to use?
PM WONG <pmwong at power25t.hkbu.edu dot hk>
Sat, 10 Jun 2000 12:45:54 +0800 (HKT)
47. Problems with qpopper on mandrake 7.0
root <everett at gloryroad dot net>
Sat, 10 Jun 2000 23:09:33 -0400
48.
"Vladimir Nikolic" <vladimirn at ptt dot yu>
Mon, 12 Jun 2000 08:15:35 +0200
49. Re:
Kris Virtue <kvirtue at qualcomm dot com>
Mon, 12 Jun 2000 15:06:02 -0700
50. Compiling Qpopper 3.1b1
"Geoff Wilson" <gwilson at cbnet.ns dot ca>
Mon, 12 Jun 2000 20:30:02 -0300
Date: Wed, 07 Jun 2000 10:03:59 -0500
From: Butch Kemper <kemper at tstar dot net>
Subject: Re: How can I make qpopper create dot-lock file in different
Read the INSTALL file and you will find the answer to your question with
some other interesting options. But anyway, here is how you do it:
./configure --enable-temp-drop-dir=path
Butch
At 04:49 AM 06/07/2000, you wrote:
>Hi,
>I've set the user quota and enable the server mode. My problem is if my
>user exceeded their quota they will encountered below message:
>
>-ERR [SYS/TEMP] maillock error 3: '/var/mail/username'
>
>Any of you know how to create username.lock file in different directory
>other than mail directory.
>
>Any help is appreciated. Thanks.
>-Nurah-
TSTAR Internet, Inc | Making the Network Work
Marble Falls, TX | Serving Blanco, Burnet,
830-693-6967 | Llano, and Mason Counties
Date: Wed, 7 Jun 2000 11:25:22 -0400 (EDT)
From: Sylvain Robitaille <syl at alcor.concordia dot ca>
Subject: Qpopper-3.0.2 on DEC-Unix-4.0e segmentation violation
Yesterday I sent the appended message to qpopper at qualcomm dot com. Now
I know it's too soon for me to get impatient, but I only received an
automatic reply indicating that my message had been received.
The problem is that this is happening quite regularly on our machine at
this point, so I'm hoping that someone on the list has perhaps seen this
problem and can propose a solution.
I fully expect that Qualcomm will come up with a patch for this, but I'd
like to reduce the frequency with which core files are produced in the
meantime. Anyone have any suggestions?
--
----------------------------------------------------------------------
Sylvain Robitaille syl at alcor.concordia dot ca
Systems analyst Concordia University
Instructional & Information Technology Montreal, Quebec, Canada
----------------------------------------------------------------------
> Date: Tue, 6 Jun 2000 13:02:06 -0400 (EDT)
> From: Sylvain Robitaille <syl at alcor.concordia dot ca>
> To: qpopper at qualcomm dot com
> Subject: Qpopper-3.0.2 on DEC-Unix-4.0e segmentation violation
>
> Recently, in response to security announcements, we upgraded to
> Qpopper-3.0.2 on one of our systems (a Compaq DS-20 running Tru64 Unix
> 4.0e with patchkit 3 installed).
>
> I noticed yesterday a core file produced by the new Qpopper. Curious to
> see if it was the result of someone trying an exploit of the problems
> recently reported, I took a quick look, (stack trace appended below my
> signature).
>
> It seems to not be the result of an exploit attempt, but rather a
> segmentation violation resulting from an attempt to log a message
> regarding empty input.
>
> The code leading up to the segmentation violation is from the file
> qpopper-3.0.2/common/snprintf.c:
>
> 211 /*
> 212 * Get the string pointer. If NULL, or if the maximum
> l
> ength
> 213 * is zero, we're done with this conversion.
> 214 */
> 215 sval = va_arg ( ap, char * );
> 216 if ( sval == NULL || limit == 0 ) {
> 217 nState = IN_FORM;
> 218 break;
> 219 }
> 220
> 221 /*
> 222 * Make sure we have room if limit is unspecified.
> 223 */
> 224 if ( limit == -1 && strlen(sval) > nSize )
> 225 return -1;
>
> It appears that something is happening with sval, causing strlen to
> choke on it. If you note at the end of the stack trace below, it looks
> as though ap contains an argument consisting of an empty string.
>
> Unfortunately I'm not proficient enough with a debugger to be able
> to get much more information than that. I hope the information I've
> provided will be helpful to you and that you'll be able to work out
> precisely what went wrong...
>
> (dbx) print sval
> 0x40010800
> (dbx) print *sval
>
> can't read from process (address 0x40010800)
>
> If you need any additional information, please do not hesitate to ask,
> and I'll do my best to provide it...
>
> --
> ----------------------------------------------------------------------
> Sylvain Robitaille syl at alcor.concordia dot ca
>
> Systems analyst Concordia University
> Instructional & Information Technology Montreal, Quebec, Canada
> ----------------------------------------------------------------------
> dbx /local/pkg/qpopper/root-qpopper-3.0.2/bin/popper ./core
> dbx version 3.11.10
> Type 'help' for help.
> Core file created by program "popper"
>
> signal Segmentation fault at >*[NLstrdlen, 0x3ff800d5280] ldq_u r1,
> 0(r16)
> (dbx) where
> > 0 NLstrdlen(0xffffffffffffffff, 0x11fffd258, 0x11fffd2d8, 0x1, 0x1)
> [0x3ff800d5280]
> 1 Qvsnprintf(s = 0x1400063f8 = "", n = 2045, format = 0x1400019e0 = "EOF
> from %s at %s (%s): [%i] %i (%s); %d (%s)", ap = struct {
> _a0 = 0x11fffd770
> _offset = 88
> }) ["snprintf.c":224, 0x12001b1a4]
> 2 pop_log(p = 0x11fffdcd8, stat = POP_PRIORITY, fn = (nil), ln = 0, format =
> 0x1400019e0 = "EOF from %s at %s (%s): [%i] %i (%s); %d (%s)") ["pop_log.c":71,
> 0x1200098bc]
> 3 myfgets(p = 0x11fffdcd8, str = 0x11fffd8d8 = "", size = 1023, fp =
> 0x3ffc0080348) ["popper.c":295, 0x12000da5c]
> 4 tgets(p = 0x11fffdcd8, str = 0x11fffd8d8 = "", size = 1024, fp =
> 0x3ffc0080348, timeout = 120) ["popper.c":341, 0x12000dbfc]
> 5 main(argc = 2, argv = 0x11fffff08) ["popper.c":193, 0x12000d694]
> (dbx)
> (dbx) print ap
> struct {
> _a0 = 0x11fffd770
> _offset = 88
> }
> (dbx) print *ap._a0
> 0x11fffdcd8 = ""
From: "Jack Barnett" <jbarnett at axil.netmate dot com>
Subject: test
Date: Wed, 7 Jun 2000 10:25:47 -0500
test
Date: Wed, 7 Jun 2000 14:33:51 -0400
From: "jmcintyr" <jmcintyr at i-plus dot net>
i am running hpux 10.20 on a d250. have ver 3.0.1b2 installed. the most wonderful thing happened to a user, using the outlook client he connected fine to qpopper and it started building the .user.pop file. after it built this file it,qpopper, re-read the
whole dam mail file and loaded into the .user.pop file. so now the user has duplicate everything. a really neat option that i would like to have turned off. anybody else experience this wonderful option and how do i turn it off ?
jim mc....
below is a ll of the 2 files every 2 seconds:
-rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
-rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/issneerin.pop
-rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
-rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/.ssneerin.pop
-rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
-rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
-rw-rw---- 1 ssneerin mail 0 Jun 7 14:04 /var/mail/ssneerin
-rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
-rw-rw---- 1 ssneerin mail 425984 Jun 7 14:04 /var/mail/ssneerin
-rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
-rw-rw---- 1 ssneerin mail 2400256 Jun 7 14:04 /var/mail/ssneerin
-rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
-rw-rw---- 1 ssneerin mail 2613248 Jun 7 14:04 /var/mail/ssneerin
-rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
-rw-rw---- 1 ssneerin mail 3620864 Jun 7 14:04 /var/mail/ssneerin
-rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
-rw-rw---- 1 ssneerin mail 5288440 Jun 7 14:04 /var/mail/ssneerin
/tempmail/.ss* not found
-rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
/tempmail/.ss* not found
-rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
/tempmail/.ss* not found
From: "=?iso-8859-1?Q?Philipp_Gaschütz?=" <philipp at gng dot de>
Subject: RE:
Date: Wed, 7 Jun 2000 20:57:00 +0200
???
this is the way it has to go!
qpopper moves the whole mail file to .user.pop so that it can serve the
client with his mail, but on the same time your MTA can add incoming
messages to the users mail file...
-p
> -----Original Message-----
> From: jmcintyr [mailto:jmcintyr at i-plus dot net]
> Sent: Mittwoch, 7. Juni 2000 20:34
> To: Subscribers of Qpopper
> Subject:
>
>
> i am running hpux 10.20 on a d250. have ver 3.0.1b2 installed.
> the most wonderful thing happened to a user, using the outlook
> client he connected fine to qpopper and it started building
> the .user.pop file. after it built this file it,qpopper, re-read the
> whole dam mail file and loaded into the .user.pop file. so now
> the user has duplicate everything. a really neat option that i
> would like to have turned off. anybody else experience this
> wonderful option and how do i turn it off ?
>
> jim mc....
>
> below is a ll of the 2 files every 2 seconds:
>
> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04
> /tempmail/issneerin.pop
> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04
> /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04
> /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 0 Jun 7 14:04
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04
> /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 425984 Jun 7 14:04
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04
> /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 2400256 Jun 7 14:04
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04
> /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 2613248 Jun 7 14:04
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04
> /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 3620864 Jun 7 14:04
> /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04
> /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 5288440 Jun 7 14:04
> /var/mail/ssneerin
> /tempmail/.ss* not found
> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04
> /var/mail/ssneerin
> /tempmail/.ss* not found
> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04
> /var/mail/ssneerin
> /tempmail/.ss* not found
>
>
>
Date: Wed, 07 Jun 2000 15:55:43 -0700
From: Stephen Samuel <samuel at bcgreen dot com>
Subject: Re:
From this, it looks like your client is checking for mail WAAAY
to often. (I hope that it was only for testing purposes).
Make sure to set the check interval to something approaching sane
(probably 2 minutes, or more) I've seen people with really large
mail files (>100MB) that take over 10 minutes to complete he
copy/look/return cycle with 'leave messages on server' turned on.
jmcintyr wrote:
>
> i am running hpux 10.20 on a d250. have ver 3.0.1b2 installed. the most wonderful thing happened to a user, using the outlook client he connected fine to qpopper and it started building the .user.pop file. after it built this file it,qpopper, re-read
e
> whole dam mail file and loaded into the .user.pop file. so now the user has duplicate everything. a really neat option that i would like to have turned off. anybody else experience this wonderful option and how do i turn it off ?
>
> jim mc....
>
> below is a ll of the 2 files every 2 seconds:
>
> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/issneerin.pop
> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 0 Jun 7 14:04 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 425984 Jun 7 14:04 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 2400256 Jun 7 14:04 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 2613248 Jun 7 14:04 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 3620864 Jun 7 14:04 /var/mail/ssneerin
> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> -rw-rw---- 1 ssneerin mail 5288440 Jun 7 14:04 /var/mail/ssneerin
> /tempmail/.ss* not found
> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
> /tempmail/.ss* not found
> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
> /tempmail/.ss* not found
--
Stephen Samuel +1(604)876-0426 samuel at bcgreen dot com
(formerly samuel at result dot com)
The question, for this world, is not "will I die?".
It is, rather, "how will I live?".
Date: Thu, 8 Jun 2000 08:56:24 +0800 (MYT)
From: Nurah Muhammad <nurah at mimos dot my>
Subject: Re: How can I make qpopper create dot-lock file in different
With this we only create .username.pop file in temporary directory but not
.username.lock file
-Nurah-
On Wed, 7 Jun 2000, Butch Kemper wrote:
> Read the INSTALL file and you will find the answer to your question with
> some other interesting options. But anyway, here is how you do it:
>
> ./configure --enable-temp-drop-dir=path
>
> Butch
>
> At 04:49 AM 06/07/2000, you wrote:
> >Hi,
> >I've set the user quota and enable the server mode. My problem is if my
> >user exceeded their quota they will encountered below message:
> >
> >-ERR [SYS/TEMP] maillock error 3: '/var/mail/username'
> >
> >Any of you know how to create username.lock file in different directory
> >other than mail directory.
> >
> >Any help is appreciated. Thanks.
> >-Nurah-
>
>
> TSTAR Internet, Inc | Making the Network Work
> Marble Falls, TX | Serving Blanco, Burnet,
> 830-693-6967 | Llano, and Mason Counties
>
>
Date: Thu, 8 Jun 2000 08:40:51 -0400
From: "jmcintyr" <jmcintyr at i-plus dot net>
Subject: Re:
stephen:
the user in question originally had a 4 meg mail file. too large for my preference but he's my boss. his outlook client checks the mail every 10 minutes. it was taking almost that long to build the .pop file. this is on a little used d250 with 512meg memo
ry. the temp mail dir is on a different controller and drive. so i have 2 problems, why did qpopper duplicate the mail file and why does it run so slow building the .pop ? it took less than a minute for qpopper to move the mail from .pop back to /var/mail
.
thanks.....jim mc....
---------- Original Message ----------------------------------
From: Stephen Samuel <samuel at bcgreen dot com>
Date: Wed, 07 Jun 2000 15:55:43 -0700
>>From this, it looks like your client is checking for mail WAAAY
>to often. (I hope that it was only for testing purposes).
>Make sure to set the check interval to something approaching sane
>(probably 2 minutes, or more) I've seen people with really large
>mail files (>100MB) that take over 10 minutes to complete he
>copy/look/return cycle with 'leave messages on server' turned on.
>
>jmcintyr wrote:
>>
>> i am running hpux 10.20 on a d250. have ver 3.0.1b2 installed. the most wonderful thing happened to a user, using the outlook client he connected fine to qpopper and it started building the .user.pop file. after it built this file it,qpopper, re-
h
>e
>> whole dam mail file and loaded into the .user.pop file. so now the user has duplicate everything. a really neat option that i would like to have turned off. anybody else experience this wonderful option and how do i turn it off ?
>>
>> jim mc....
>>
>> below is a ll of the 2 files every 2 seconds:
>>
>> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/issneerin.pop
>> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/.ssneerin.pop
>> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
>> -rw-rw---- 1 ssneerin mail 0 Jun 7 14:04 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
>> -rw-rw---- 1 ssneerin mail 425984 Jun 7 14:04 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
>> -rw-rw---- 1 ssneerin mail 2400256 Jun 7 14:04 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
>> -rw-rw---- 1 ssneerin mail 2613248 Jun 7 14:04 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
>> -rw-rw---- 1 ssneerin mail 3620864 Jun 7 14:04 /var/mail/ssneerin
>> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
>> -rw-rw---- 1 ssneerin mail 5288440 Jun 7 14:04 /var/mail/ssneerin
>> /tempmail/.ss* not found
>> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
>> /tempmail/.ss* not found
>> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
>> /tempmail/.ss* not found
>
>--
>Stephen Samuel +1(604)876-0426 samuel at bcgreen dot com
> (formerly samuel at result dot com)
>The question, for this world, is not "will I die?".
>It is, rather, "how will I live?".
>
From: "Paul Trippett" <pjt at gmxuk dot net>
Subject: Corrupt Mailboxes
Date: Thu, 8 Jun 2000 16:19:49 +0100
Hi,
QPopper has been running now for over 30 days and has been very stable. It is running on the same
machine as QMail.
When i checked my email today i found I had 30+ email with no sender specified, no reciepient and no
message body.
At first i thought "spam" but after looking at the messages i noticed the email only had 3 lines and
then thought it must be a corrupt mailbox....
----- START MESSAGES -------
Return-Path: <freebsd-users-admin at uk.freebsd dot org>
Delivered-To: pjt at gmxuk dot net
X-UIDL: -Kl!!%hA!!HEF!!D5E!!
--
Return-Path: <Qpopper-errors at lists.pensive dot org>
Delivered-To: pjt at gmxuk dot net
X-UIDL: 54h"!dC,"!Fg,!!V;%#!
--
Return-Path: <xdhm-return-2372-34250999 at listbot dot com>
Delivered-To: pjt at gmxuk dot net
X-UIDL: ipg!!ddk"!92~"!b,5"!
----- END MESSAGES -------
Any reason as to why this could have happened would be grately appreciated so i can resolve the
problem quickly.
Thanks
Paul Trippett
pjt at gmxuk dot net
Date: Thu, 8 Jun 2000 14:32:03 -0400 (EDT)
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Subject: Re: Impact of -R option
On Sat, 3 Jun 2000, John Gonzalez/netMDC admin wrote:
> You also shouldnt be relying on reverse lookups for any security
> method,
There's no harm in relying on reverse lookups, *IF* you verify the
symmetry of the A and PTR records. This is the technique that
tcpwrappers, INN, and a host of other programs use.
Date: Thu, 08 Jun 2000 11:53:21 -0700
From: Joel Laing <joel at scripps dot edu>
Subject: Re: Corrupt Mailboxes
You're not running pop in server mode are you?
Paul Trippett wrote:
>
> Hi,
>
> QPopper has been running now for over 30 days and has been very stable. It is running on the same
> machine as QMail.
>
> When i checked my email today i found I had 30+ email with no sender specified, no reciepient and no
> message body.
>
> At first i thought "spam" but after looking at the messages i noticed the email only had 3 lines and
> then thought it must be a corrupt mailbox....
>
> ----- START MESSAGES -------
> Return-Path: <freebsd-users-admin at uk.freebsd dot org>
> Delivered-To: pjt at gmxuk dot net
> --
> Return-Path: <Qpopper-errors at lists.pensive dot org>
> Delivered-To: pjt at gmxuk dot net
> --
> Return-Path: <xdhm-return-2372-34250999 at listbot dot com>
> Delivered-To: pjt at gmxuk dot net
> ----- END MESSAGES -------
>
> Any reason as to why this could have happened would be grately appreciated so i can resolve the
> problem quickly.
>
> Thanks
>
> Paul Trippett
> pjt at gmxuk dot net
From: "Paul Trippett" <pjt at gmxuk dot net>
Subject: Re: Corrupt Mailboxes
Date: Thu, 8 Jun 2000 20:03:06 +0100
> You're not running pop in server mode are you?
what do you mean?
I start QPopper from inetd with the line: -
pop3 stream tcp nowait root /usr/sbin/popper qpopper -s -T 30
/PT
> Paul Trippett wrote:
> >
> > Hi,
> >
> > QPopper has been running now for over 30 days and has been very stable. It is running on the
same
> > machine as QMail.
> >
> > When i checked my email today i found I had 30+ email with no sender specified, no reciepient
and no
> > message body.
> >
> > At first i thought "spam" but after looking at the messages i noticed the email only had 3 lines
and
> > then thought it must be a corrupt mailbox....
> >
> > ----- START MESSAGES -------
> > Return-Path: <freebsd-users-admin at uk.freebsd dot org>
> > Delivered-To: pjt at gmxuk dot net
> > --
> > Return-Path: <Qpopper-errors at lists.pensive dot org>
> > Delivered-To: pjt at gmxuk dot net
> > --
> > Return-Path: <xdhm-return-2372-34250999 at listbot dot com>
> > Delivered-To: pjt at gmxuk dot net
> > ----- END MESSAGES -------
> >
> > Any reason as to why this could have happened would be grately appreciated so i can resolve the
> > problem quickly.
> >
> > Thanks
> >
> > Paul Trippett
> > pjt at gmxuk dot net
>
Date: Thu, 08 Jun 2000 12:36:40 -0700
From: Joel Laing <joel at scripps dot edu>
Subject: Re: Corrupt Mailboxes
When you built the package, did you use the --enable-server-mode flag to
configure? If so, then you are running in server mode and it's very
likely that you are getting corrupt spool files because there is another
agent (QMail) acting on the spool file.
-Joel
Paul Trippett wrote:
>
> > You're not running pop in server mode are you?
> what do you mean?
>
> I start QPopper from inetd with the line: -
>
> pop3 stream tcp nowait root /usr/sbin/popper qpopper -s -T 30
>
> /PT
>
> > Paul Trippett wrote:
> > >
> > > Hi,
> > >
> > > QPopper has been running now for over 30 days and has been very stable. It is running on the
> same
> > > machine as QMail.
> > >
> > > When i checked my email today i found I had 30+ email with no sender specified, no reciepient
> and no
> > > message body.
> > >
> > > At first i thought "spam" but after looking at the messages i noticed the email only had 3 lines
> and
> > > then thought it must be a corrupt mailbox....
> > >
> > > ----- START MESSAGES -------
> > > Return-Path: <freebsd-users-admin at uk.freebsd dot org>
> > > Delivered-To: pjt at gmxuk dot net
> > > --
> > > Return-Path: <Qpopper-errors at lists.pensive dot org>
> > > Delivered-To: pjt at gmxuk dot net
> > > --
> > > Return-Path: <xdhm-return-2372-34250999 at listbot dot com>
> > > Delivered-To: pjt at gmxuk dot net
> > > ----- END MESSAGES -------
> > >
> > > Any reason as to why this could have happened would be grately appreciated so i can resolve the
> > > problem quickly.
> > >
> > > Thanks
> > >
> > > Paul Trippett
> > > pjt at gmxuk dot net
> >
Date: Thu, 8 Jun 2000 14:14:02 -0600 (MDT)
From: John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Subject: Re: Impact of -R option
As has been mentioned many times in the past, using reverse lookups as
your sole method of security is severely flawed. Among spoofs, DNS hacks
can take place. IMHO, it's just plain not a good idea. YMMV.
On Thu, 8 Jun 2000, James Ralston wrote:
>On Sat, 3 Jun 2000, John Gonzalez/netMDC admin wrote:
>
>> You also shouldnt be relying on reverse lookups for any security
>> method,
>
>There's no harm in relying on reverse lookups, *IF* you verify the
>symmetry of the A and PTR records. This is the technique that
>tcpwrappers, INN, and a host of other programs use.
>
>
--
_ __ _____ __ _________
______________ /_______ ___ ____ /______ John Gonzalez/Net.Tech
__ __ \ __ \ __/_ __ `__ \/ __ /_ ___/ MDC Computers/netMDC!
_ / / / `__/ /_ / / / / / / /_/ / / /__ (505)437-7600/fax-437-3052
/_/ /_/\___/\__/ /_/ /_/ /_/\__,_/ \___/ http://www.netmdc.com
[---------------------------------------------[system info]-----------]
2:10pm up 28 days, 19:36, 3 users, load average: 0.12, 0.18, 0.17
Date: Thu, 08 Jun 2000 17:32:35 +0000
From: Chris Cook <ccook at tcworks dot net>
Subject: Popper logging?
Hello,
I am running Qpopper 2.54 on FreeBSD 3.1 and am needing to figure out
how to get it to log sucessful authentications to a logfile. I am
trying to get pop authenticated relaying working and the perl script
parses a logfile for the popper authentication message. I have mail.*
set to log to /var/log/maillog in the syslog.conf file but I do not see
login messages from popper. If someone could help me, please cc a copy
of the message to ccook at tcworks dot net because I am not on this mailing
list. Thanks!
--
Chris
o----< ccook at tcworks dot net >-----------------------------------------o
|Chris Cook - Network Admin | TCWORKS.NET - http://www.tcworks.net |
|The Computer Works ISP | FreeBSD - http://www.freebsd.org |
o------------------------------------------------------------------o
From: "Kenneth Porter" <shiva at well dot com>
Date: Thu, 08 Jun 2000 17:26:21 -0700
Subject: Mail security alert: Linux kernels prior to 2.2.16
A security vulnerability affecting setuid programs (including sendmail)
has been identified in Linux kernels prior to 2.2.16. It may affect
qpopper. It allows local users to gain root access with a trivial
exploit.
Details may be found at:
http://www.securityfocus.com/vdb/bottom.html?vid=1322
http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=11974
http://www.sendmail.org/sendmail.8.10.1.LINUX-SECURITY.txt
Ken
mailto:shiva at well dot com
http://www.sewingwitch.com/ken/
http://www.harrybrowne2000.org/
Date: Thu, 8 Jun 2000 20:32:12 -0600 (MDT)
From: John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Subject: Re: Corrupt Mailboxes
On Thu, 8 Jun 2000, Joel Laing wrote:
>When you built the package, did you use the --enable-server-mode flag to
>configure? If so, then you are running in server mode and it's very
>likely that you are getting corrupt spool files because there is another
>agent (QMail) acting on the spool file.
I run qmail and qpopper in servermode, is there a known problem in this
environment?
--
_ __ _____ __ _________
______________ /_______ ___ ____ /______ John Gonzalez/Net.Tech
__ __ \ __ \ __/_ __ `__ \/ __ /_ ___/ MDC Computers/netMDC!
_ / / / `__/ /_ / / / / / / /_/ / / /__ (505)437-7600/fax-437-3052
/_/ /_/\___/\__/ /_/ /_/ /_/\__,_/ \___/ http://www.netmdc.com
[---------------------------------------------[system info]-----------]
8:30pm up 29 days, 1:56, 4 users, load average: 0.11, 0.13, 0.14
Date: Fri, 9 Jun 2000 12:10:54 +0800 (HKT)
From: PM WONG <pmwong at power25t.hkbu.edu dot hk>
Subject: qpopper or just popper in inetd.conf
We've been using qpopper 2.x.y ,
the line in /etc/inetd.conf
has been
pop3 stream tcp nowait root /usr/local/lib/popper popper -s
I've upgraded it to qpopper 3.0 recently, but still used
the line above.
But i re-read the doc that the line for 3.0 should be
pop3 stream tcp nowait root /usr/local/lib/popper qpopper -s
Does it really matter ? (My experience is that even with
the old line, it still works)
regards
pmwong
Date: Fri, 9 Jun 2000 12:45:16 +0800 (HKT)
From: PM WONG <pmwong at power25t.hkbu.edu dot hk>
Subject: .lock files in versions 2 & 3
We've been using qpopper ver. 2 and i noticed very
few username.lock files are usually left under
/var/spool/mail (they were cleared up quite soon, i presume)
But for version 3, there are many of these .lock files
staying during the period when the user's .userhome.pop were
there, and their popper processes are still running.
Now why the differences ? Is this phenomenon normal in
both ver.2 and 3 and they are not bugs?
pmwong
Date: Fri, 9 Jun 2000 00:08:53 -0700
From: Qpopper Support <qpopper at qualcomm dot com>
Subject: Re: 3.0.2 distribution problem
At 12:39 PM -0400 5/24/00, Tadeusz Wiewiorowski wrote:
> I have just downloaded and compiled qpopper 3.0.2 on HP-UX 10.20.
> In order to get "make" working I had to change permissions on the
> etags.c [in the mmangle subdir] from 444 to 644 .
>
> Any comments ?
>
> Tadeusz Wiewiorowski
> CBC Toronto
There seems to be something funny about HP make. It wants to rebuild
'etags.c' even though it is not older than 'etags.gperf', which is
its dependency. Doing a 'touch' on 'etags.c' seems to get around it.
I don't yet know why this affects HP.
Date: Fri, 9 Jun 2000 14:51:35 +0300 (EET DST)
From: Serkan AKCIN <serkan at baskent.edu dot tr>
Subject: Unable to get canonical name of client
Hi,
We set up qpooper v3.0.2 on a Sun 5.7 system.
We are using qpopper and sendmail 8.9.3.
All our pop3 clients are working fine.
But one of my pop3 client can't set message to outside world.
When I ran qpooper with "-s -t /var/log/qpooper" options it gave an error like
"Fri Jun 9 13:49:49 2000 [24771] (v3.0.2) Unable to get canonical name of
client 195.174.253.87:Non-Authoritive answer: Host not found, or SERVERFAIL (2)
Fri Jun 9 13:49:49 2000 "
And when I ran qpooper with "-R -s -t /var/log/qpooper" options it didn't
give an error. But my pop3 client was unable to send e-mail to outside world.
What is the problem? Is there any idea?
P.S.1: My client's ip adres is 192.168.1.64 and she is behaind a firewall.
But I don't think it is a problem. Because she can receive her e-mail's by
pop. But she can't sent.
P.S.2: There is no DNS name record for 195.174.253.87. (nslookup can't
resolv it.) But I think "-R" switch can solve this problem.
Date: Fri, 09 Jun 2000 08:40:10 -0400
From: "Don Ward" <dward at process-technology dot com>
Subject: Please remove dward at process-technology dot com
Please remove dward at process-technology dot com listserv qpopper
From: "pestilence" <pestilence at netplan dot gr>
Subject: Re: Unable to get canonical name of client
Date: Fri, 9 Jun 2000 15:53:35 +0300
Your DNS is not working, or you haven't setup correctly the reverse mapping.
You need to have reverse lookups working, so that pop3 can reverse lookup
the IP's
----- Original Message -----
From: "Serkan AKCIN" <serkan at baskent.edu dot tr>
To: "Subscribers of Qpopper" <qpopper at lists.pensive dot org>
Sent: ?????????, 9 ??????? 2000 2:51 ??
Subject: Unable to get canonical name of client
> Hi,
>
> We set up qpooper v3.0.2 on a Sun 5.7 system.
> We are using qpopper and sendmail 8.9.3.
> All our pop3 clients are working fine.
>
> But one of my pop3 client can't set message to outside world.
> When I ran qpooper with "-s -t /var/log/qpooper" options it gave an error
like
>
> "Fri Jun 9 13:49:49 2000 [24771] (v3.0.2) Unable to get canonical name of
> client 195.174.253.87:Non-Authoritive answer: Host not found, or
SERVERFAIL (2)
> Fri Jun 9 13:49:49 2000 "
>
> And when I ran qpooper with "-R -s -t /var/log/qpooper" options it didn't
> give an error. But my pop3 client was unable to send e-mail to outside
world.
>
> What is the problem? Is there any idea?
>
>
> P.S.1: My client's ip adres is 192.168.1.64 and she is behaind a firewall.
> But I don't think it is a problem. Because she can receive her e-mail's by
> pop. But she can't sent.
>
> P.S.2: There is no DNS name record for 195.174.253.87. (nslookup can't
> resolv it.) But I think "-R" switch can solve this problem.
>
>
>
>
>
Date: Fri, 9 Jun 2000 08:52:34 -0400 (EDT)
From: Jim Sander <jim at federation.addy dot com>
Subject: Re: Popper logging?
> I am running Qpopper 2.54 on FreeBSD 3.1 and am needing to figure out
> how to get it to log sucessful authentications to a logfile. I have
> mail.* set to log to /var/log/maillog in the syslog.conf
Popper doesn't log under the mail facility. Use -vv on your syslog to
see what it's logging as, and of course 'man syslogd' for the details.
(you'll probably have to set up a !popper entry)
-=Jim=-
From: "pestilence" <pestilence at netplan dot gr>
Subject: Re: Unable to get canonical name of client
Date: Fri, 9 Jun 2000 16:15:08 +0300
Oh i forgot also, qpopper has nothing to do with sending email...you should
be checking your SMTP daemon, something like sendmail...
----- Original Message -----
From: "Serkan AKCIN" <serkan at baskent.edu dot tr>
To: "Subscribers of Qpopper" <qpopper at lists.pensive dot org>
Sent: ?????????, 9 ??????? 2000 2:51 ??
Subject: Unable to get canonical name of client
> Hi,
>
> We set up qpooper v3.0.2 on a Sun 5.7 system.
> We are using qpopper and sendmail 8.9.3.
> All our pop3 clients are working fine.
>
> But one of my pop3 client can't set message to outside world.
> When I ran qpooper with "-s -t /var/log/qpooper" options it gave an error
like
>
> "Fri Jun 9 13:49:49 2000 [24771] (v3.0.2) Unable to get canonical name of
> client 195.174.253.87:Non-Authoritive answer: Host not found, or
SERVERFAIL (2)
> Fri Jun 9 13:49:49 2000 "
>
> And when I ran qpooper with "-R -s -t /var/log/qpooper" options it didn't
> give an error. But my pop3 client was unable to send e-mail to outside
world.
>
> What is the problem? Is there any idea?
>
>
> P.S.1: My client's ip adres is 192.168.1.64 and she is behaind a firewall.
> But I don't think it is a problem. Because she can receive her e-mail's by
> pop. But she can't sent.
>
> P.S.2: There is no DNS name record for 195.174.253.87. (nslookup can't
> resolv it.) But I think "-R" switch can solve this problem.
>
>
>
>
>
Date: Fri, 9 Jun 2000 10:30:00 -0400
From: "jmcintyr" <jmcintyr at i-plus dot net>
Subject: Re: 3.0.2 distribution problem
tadeusz:
i have a d250 running 10.20 also and having the same problem you are.
i restored the distribution copy of etags.c and changed the permissions from 444 to 644. it still fails with the same errors in my environment though. any other suggestions ?
thanks....jim mc...
---------- Original Message ----------------------------------
From: Qpopper Support <qpopper at qualcomm dot com>
Date: Fri, 9 Jun 2000 00:08:53 -0700
>At 12:39 PM -0400 5/24/00, Tadeusz Wiewiorowski wrote:
>
>> I have just downloaded and compiled qpopper 3.0.2 on HP-UX 10.20.
>> In order to get "make" working I had to change permissions on the
>> etags.c [in the mmangle subdir] from 444 to 644 .
>>
>> Any comments ?
>>
>> Tadeusz Wiewiorowski
>> CBC Toronto
>
>There seems to be something funny about HP make. It wants to rebuild
>'etags.c' even though it is not older than 'etags.gperf', which is
>its dependency. Doing a 'touch' on 'etags.c' seems to get around it.
>I don't yet know why this affects HP.
>
Subject: Re: Unable to get canonical name of client
Date: Fri, 09 Jun 2000 10:48:41 -0400
From: Ken Hornstein <kenh at cmf.nrl.navy dot mil>
>We set up qpooper v3.0.2 on a Sun 5.7 system.
^^^^^^^
I have no joke here, I just like typing "qpooper".
--Ken
Date: Fri, 9 Jun 2000 12:11:19 +0000 (GMT)
From: P O O P I E <poopie at jeeves.poopie dot net>
Subject: Please remove poopie at jeeves.poopie dot net
Please remove poopie at jeeves.poopie dot net listserv qpopper
Date: Fri, 9 Jun 2000 09:01:46 -0700
From: Qpopper Support <qpopper at qualcomm dot com>
Subject: Re: 3.0.2 distribution problem
At 10:30 AM -0400 6/9/00, jmcintyr wrote:
> i have a d250 running 10.20 also and having the same problem you are.
> i restored the distribution copy of etags.c and changed the
> permissions from 444 to 644. it still fails with the same errors in
> my environment though. any other suggestions ?
Please try the following command from within the qpopper3.0 directory:
touch mmangle/etags.c
Then try 'make' again and let me know.
Check the mmangle/etags.c file first though. If you don't have gperf
on your system, changing the permissions from 444 to 644 probably
caused 'make' to zero it, in which case you will need to get a fresh
copy from the distribution again. No need to change the permissions,
just touch() it.
By the way, is your HP in 64-bit mode?
>
> ---------- Original Message ----------------------------------
> From: Qpopper Support <qpopper at qualcomm dot com>
> Date: Fri, 9 Jun 2000 00:08:53 -0700
>
>>At 12:39 PM -0400 5/24/00, Tadeusz Wiewiorowski wrote:
>>
>>> I have just downloaded and compiled qpopper 3.0.2 on HP-UX 10.20.
>>> In order to get "make" working I had to change permissions on the
>>> etags.c [in the mmangle subdir] from 444 to 644 .
>>>
>>> Any comments ?
>>>
>>> Tadeusz Wiewiorowski
>>> CBC Toronto
>>
>>There seems to be something funny about HP make. It wants to rebuild
>>'etags.c' even though it is not older than 'etags.gperf', which is
>>its dependency. Doing a 'touch' on 'etags.c' seems to get around it.
>>I don't yet know why this affects HP.
>>
Date: Fri, 9 Jun 2000 09:34:30 -0700 (PDT)
From: Gary Wong <gwong at ascend dot com>
Subject: Please remove gwong at ascend dot com
Please remove gwong at ascend dot com listserv qpopper
From: Daniel Norton <danorton at chsw dot com>
Subject: How To Unsubscribe from Qpopper
Date: Fri, 09 Jun 2000 13:01:17 -0400
To unsubscribe from Qpopper, put "unsubscribe" in the first line of the
body (not the Subject) of a message and send it to
<qpopper-request at lists.pensive dot org>.
You can also use this web page:
http://www.pensive.org/Mailing_Lists/QPopper_sub_form.html
For information on other commands available by E-mail, see this page:
http://www.pensive.org/Mailing_Lists/help.html
--
Daniel Norton
Date: Fri, 9 Jun 2000 10:16:46 -0700
From: Qpopper Support <qpopper at qualcomm dot com>
Subject: Re: Corrupt Mailboxes
At 12:36 PM -0700 6/8/00, Joel Laing wrote:
> When you built the package, did you use the --enable-server-mode flag to
> configure? If so, then you are running in server mode and it's very
> likely that you are getting corrupt spool files because there is another
> agent (QMail) acting on the spool file.
It's OK to use server mode if there is a delivery agent which appends
messages to the spool file. In server mode, it's important that the
only process other than Qpopper is a delivery agent which (1) only
modifies the spool by appending messages to the end and (2) properly
locks the spool.
There is a problem if other processes also alter the spool, such as
mail readers which delete messages from the middle, or otherwise
change the spool in ways other than appending new messages to the
end, and/or which don't use proper locking.
Date: Thu, 08 Jun 2000 22:42:56 -0300
From: "Eduardo M. A. M. Mendes" <emmendes at funrei dot br>
Subject: qpopper - I can't send any message!
Hello
I have just installed qpopper on solaris 2.6 and linux. Although
the installation went fine I can't send emails when using the pop3
accounts.
I am not sure whether this is a pop3 problem or something else. Could
someone out there help me out, please?
Thanks a lot.
regards
Eduardo
Date: Fri, 9 Jun 2000 13:30:54 -0400
From: "jmcintyr" <jmcintyr at i-plus dot net>
Subject: Re: 3.0.2 distribution problem
restored etags.c from the distribution tar file. did the touch and all is well. put it into production and no problems so far.
no it is not 64bit, 11.0 only.
thanks for your assistance....jim mc....
---------- Original Message ----------------------------------
From: Qpopper Support <qpopper at qualcomm dot com>
Date: Fri, 9 Jun 2000 09:01:46 -0700
>At 10:30 AM -0400 6/9/00, jmcintyr wrote:
>
>> i have a d250 running 10.20 also and having the same problem you are.
>> i restored the distribution copy of etags.c and changed the
>> permissions from 444 to 644. it still fails with the same errors in
>> my environment though. any other suggestions ?
>
>
>Please try the following command from within the qpopper3.0 directory:
>
> touch mmangle/etags.c
>
>Then try 'make' again and let me know.
>
>Check the mmangle/etags.c file first though. If you don't have gperf
>on your system, changing the permissions from 444 to 644 probably
>caused 'make' to zero it, in which case you will need to get a fresh
>copy from the distribution again. No need to change the permissions,
>just touch() it.
>
>By the way, is your HP in 64-bit mode?
>
>
>>
>> ---------- Original Message ----------------------------------
>> From: Qpopper Support <qpopper at qualcomm dot com>
>> Date: Fri, 9 Jun 2000 00:08:53 -0700
>>
>>>At 12:39 PM -0400 5/24/00, Tadeusz Wiewiorowski wrote:
>>>
>>>> I have just downloaded and compiled qpopper 3.0.2 on HP-UX 10.20.
>>>> In order to get "make" working I had to change permissions on the
>>>> etags.c [in the mmangle subdir] from 444 to 644 .
>>>>
>>>> Any comments ?
>>>>
>>>> Tadeusz Wiewiorowski
>>>> CBC Toronto
>>>
>>>There seems to be something funny about HP make. It wants to rebuild
>>>'etags.c' even though it is not older than 'etags.gperf', which is
>>>its dependency. Doing a 'touch' on 'etags.c' seems to get around it.
>>>I don't yet know why this affects HP.
>>>
>
>
From: "Keith Warno" <keith at HaggleWare dot com>
Subject: Re: Please remove poopie at jeeves.poopie dot net
Date: Fri, 9 Jun 2000 14:01:29 -0400
If you check the headers of a mail from just about any mailing list you'll
see info on how to get help or unsubscribe.
For example, in the headers of mails from this list:
List-Subscribe: <mailto:qpopper-request at lists.pensive dot org?body=subscribe>
List-Unsubscribe:
<mailto:qpopper-request at lists.pensive dot org?body=unsubscribe>
List-Archive: <mailto:autoshare at lists.pensive dot org?body=index%20QPopper>
List-Post: <mailto:qpopper at lists.pensive dot org>
*sigh*
kw
/* Keith Warno
** Developer & Sys Admin
** http://www.HaggleWare.com/
*/
Date: Fri, 9 Jun 2000 14:01:02 -0400 (EDT)
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Subject: Re: Popper logging?
On Thu, 8 Jun 2000, Chris Cook wrote:
> I have mail.* set to log to /var/log/maillog in the syslog.conf file
> but I do not see login messages from popper.
From the INSTALL file:
f) POP_FACILITY -- Specifies the log facility.
Default is LOG_LOCAL1 or LOG_MAIL, depending on
the OS. Pass --with-log-facility=name to
./configure to change.
As of qpopper 3.0.2, unless overriden as described in the INSTALL
file, if the OS is Tru64 or Linux, the LOG_MAIL facility is selected;
otherwise, the LOG_LOCAL0 facility is selected.
--
James Ralston, Information Technology
Software Engineering Institute
Carnegie Mellon University, Pittsburgh, PA, USA
Date: Fri, 9 Jun 2000 14:11:40 -0400 (EDT)
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Subject: qpoppper 3.0.2 inconsistency: LOG_LOCAL0 versus LOG_LOCAL1
On Fri, 9 Jun 2000, I wrote:
> From the INSTALL file:
>
> f) POP_FACILITY -- Specifies the log facility.
> Default is LOG_LOCAL1 or LOG_MAIL
> ^^^^^^^^^^
> As of qpopper 3.0.2, unless overriden as described in the INSTALL
> file, if the OS is Tru64 or Linux, the LOG_MAIL facility is selected;
> otherwise, the LOG_LOCAL0 facility is selected.
> ^^^^^^^^^^
I didn't realize it until I re-read my message, but there's a
inconsistency here: the qpopper documentation (and even the comments
in configure.in and config.h!) claim that LOG_LOCAL1 is the default
facility, but LOG_LOCAL0 is what winds up being used in the actual
code.
Obviously, either the code needs to be updated, or the comments need
to be updated...
From: "Keith Warno" <keith at HaggleWare dot com>
Subject: "common" errors
Date: Fri, 9 Jun 2000 15:03:17 -0400
This is a multi-part message in MIME format.
------=_NextPart_000_0005_01BFD223.D774AF20
Content-Type: text/plain;
charset="Windows-1252"
Content-Transfer-Encoding: 7bit
Building qpopper 3.0.2 on a linux box w/ "gcc version egcs-2.91.66
19990314/Linux (egcs-1.1.2 release)" generates a lot of "multiple common"
warnings at the linking stage.
ie,
pop_auth.o: warning: multiple common of `s'
pop_dele.o: warning: previous common is here
pop_auth.o: warning: multiple common of `size'
pop_dele.o: warning: previous common is here
pop_auth.o: warning: multiple common of `stream'
pop_dele.o: warning: previous common is here
This isn't really a big deal but for any other paranoid ppl in the audiance
I've attached a small patch to get rid of these warnings. It just moves the
s,size, and stream variable declarations out of popper.h into common.c and
declares them as externs in popper.h
Ciao,
kw
/* Keith Warno
** Developer & Sys Admin
** http://www.HaggleWare.com/
*/
------=_NextPart_000_0005_01BFD223.D774AF20
Content-Type: text/plain;
name="patch.txt"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
filename="patch.txt"
diff -urN qpopper3.0.2/popper/Makefile.in =
qpopper3.0.2-common_fixed/popper/Makefile.in
--- qpopper3.0.2/popper/Makefile.in Tue May 9 18:35:19 2000
+++ qpopper3.0.2-common_fixed/popper/Makefile.in Fri Jun 9 14:50:18 =
2000
@@ -68,7 +68,7 @@
pop_bull.o xtnd_xlst.o pop_uidl.o mktemp.o \
pop_rpop.o pop_apop.o md5.o pop_auth.o pop_pope.o \
pop_extend.o scram.o hmac.o base64.o pop_util.o =
\
- get_sub_opt.o msg_ptr.o
+ get_sub_opt.o msg_ptr.o common.o
SRCS = pop_dele.c pop_dropcopy.c \
pop_get_command.c pop_get_subcommand.c pop_init.c \
@@ -79,7 +79,7 @@
pop_bull.c xtnd_xlst.c pop_uidl.c mktemp.c \
pop_rpop.c pop_apop.c md5.c pop_auth.c pop_pope.c \
pop_extend.c scram.c hmac.c base64.c pop_util.c =
\
- get_sub_opt.c msg_ptr.c
+ get_sub_opt.c msg_ptr.c common.c
POPAUTHOBJS = base64.o scram.o md5.o \
hmac.o popauth.o
diff -urN qpopper3.0.2/popper/common.c =
qpopper3.0.2-common_fixed/popper/common.c
--- qpopper3.0.2/popper/common.c Wed Dec 31 19:00:00 1969
+++ qpopper3.0.2-common_fixed/popper/common.c Fri Jun 9 14:49:11 2000
@@ -0,0 +1,5 @@
+#include<stdio.h>
+
+char *s;
+int size;
+FILE *stream;
diff -urN qpopper3.0.2/popper/popper.h =
qpopper3.0.2-common_fixed/popper/popper.h
--- qpopper3.0.2/popper/popper.h Tue May 9 18:35:22 2000
+++ qpopper3.0.2-common_fixed/popper/popper.h Fri Jun 9 14:50:40 2000
@@ -503,9 +503,9 @@
char *get_time();
extern char *ZONE;
-char *s;
-int size;
-FILE *stream;
+extern char *s;
+extern int size;
+extern FILE *stream;
------=_NextPart_000_0005_01BFD223.D774AF20--
Date: Fri, 9 Jun 2000 15:33:43 -0400 (EDT)
From: James Ralston <qralston+ml.qpopper at andrew.cmu dot edu>
Subject: Re: Impact of -R option
On Thu, 8 Jun 2000, John Gonzalez/netMDC admin wrote:
> As has been mentioned many times in the past, using reverse lookups
> as your sole method of security is severely flawed.
I didn't recommend making security policy decisions based solely on
the information gleaned from reverse lookups. I merely stated that it
might be acceptable to use information from reverse lookups in certain
situations.
> Among spoofs, DNS hacks can take place
Yes, those are possible risks. But virtually any security policy has
possible risks. The real question is: is the risk minimal enough to
be employed in a particular situation or configuration? And can the
risks be minimized?
For example, let's assume bob at company dot com uses tcpwrappers to protect
the qpopper daemon on pop.company.com so that only hosts within the
company.com domain are permitted to establish a POP connection.
If eve at cracker dot org wants to connect to pop.company.com and, say, run
some new exploit against qpopper that she's dreamed up, she needs to
get pass tcpwrappers' checks before she'll be able to speak to the
qpopper daemon.
Since Eve controls the DNS server that is authoritative for both the
cracker.org domain and the netblocks that the hosts within cracker.org
use, it's trivial for her to change her DNS server so that when she
connects to pop.company.com from evil.cracker.org, and pop.company.com
performs a reverse-lookup of the IP address of evil.cracker.org, Eve's
DNS server will simply return a name within the company.com domain;
say, innocent.company.com. (The reverse-lookup will ultimately refer
back to Eve's DNS server, because it is her DNS server that is
authoritative for the IP address that connected to pop.company.com.)
At this point, if pop.company.com performs no further checks, Eve is
in.
But pop.company.com can easily defeat Eve's spoofing by simply
resolving innocent.company.com back to an IP address. Because
innocent.company.com is within the company.com domain, pop.company.com
will refer to company.com's nameserver (e.g., dns.company.com), not
Eve's, in order to resolve innocent.company.com to an address. The IP
address (if any) returned by dns.company.com will not match the IP
address for evil.cracker.org. At this point, pop.company.com has
enough information to realize that it is being spoofed, and can
discard the connection.
This is exactly the technique that tcpwrappers uses when validating
incoming connections against a domain-name match.
Eve might be able to flood dns.company.com so that the response to
pop.company.com's query will be lost. But that won't cut it; if the A
record lookup fails (or if the initial reverse lookup fails),
tcpwrappers will drop the connection. Eve will not only have to flood
dns.company.com (to prevent it from responding to pop.company.com),
but she will have to forge the reply to pop.company.com to make it
look as if it came from dns.company.com. To do that, company.com's
border routers must not be configured to drop forged-source packets,
and Eve will have to engage in some more advanced techniques like TCP
sequence number prediction. It's possible Eve will take this route,
but it's more likely that she'll take the easier (and usually more
productive route) of simply trying to find other machines within
company.com that haven't been kept up-to-date with patches and
therefore have remotely-exploitable security vulnerabilities.
If Eve could somehow attack or poison dns.company.com and cause it to
return bogus information, then that would also do the trick. But if
Eve is able to do that, then whether or not she can connect to
pop.company.com should be the least of Bob's concern.
If the network path between pop.company.com and dns.company.com is
insecure, and Eve can execute her attack from a machine located on
some component of the path, then all she has to do is flood
dns.company.com and then impersonate it when pop.company.com performs
the forward-lookup. But again, if Eve can masquerade as
dns.company.com at will, then the issue of whether Eve can connect to
pop.company.com pales in comparison.
If the network path between company.com machines and pop.company.com
is insecure, and Eve can get onto some component of the insecure path,
then Eve's best bet would be to simply impersonate a valid company.com
client. But again, if this is the case, there are all sorts of other
evil things Eve can do that are far more worrisome.
> IMHO, it's just plain not a good idea.
IMHO, making universally quantified statements such as "you also
shouldnt be relying on reverse lookups for any security method" is
just plain not a good idea. Relying on reverse lookups in certain
circumstances CAN be useful if 1) you perform the checking properly,
and 2) you understand the ramifications of using reverse lookups as
part of your security policy decisions.
This isn't meant as a personal attack. (And I certainly hope I've
been reasonably polite. :) It just really bugs me when people make
statements like "you should never do [x]", or "you should always do
[y]". Life just ain't that simple. Doing [x] or [y] might or might
not be appropriate in certain situations. One should rely on an
analysis of one's situation, not mantras.
--
James Ralston, Information Technology
Software Engineering Institute
Carnegie Mellon University, Pittsburgh, PA, USA
Date: Fri, 9 Jun 2000 13:07:19 -0700
From: Leonard Hermens <Leonard.Hermens at rcity dot com>
Subject: Re: Please remove poopie at jeeves.poopie dot net
Except that some clients hide the headers to make the e-mail look
"pretty" and the user doesn't see them. :-)
-- Leonard
At 2:01 PM -0400 6/9/00, Keith Warno wrote:
>If you check the headers of a mail from just about any mailing list you'll
>see info on how to get help or unsubscribe.
>
>For example, in the headers of mails from this list:
>
>List-Subscribe: <mailto:qpopper-request at lists.pensive dot org?body=subscribe>
>List-Unsubscribe:
><mailto:qpopper-request at lists.pensive dot org?body=unsubscribe>
>List-Archive: <mailto:autoshare at lists.pensive dot org?body=index%20QPopper>
>List-Post: <mailto:qpopper at lists.pensive dot org>
>
>*sigh*
>
From: "=?iso-8859-1?Q?Philipp_Gaschütz?=" <philipp at gng dot de>
Subject: RE: qpopper - I can't send any message!
Date: Fri, 9 Jun 2000 22:31:25 +0200
Qpopper is only a POP server, what you are looking for is a MTA
(Sendmail, Exim, Qmail, Smail).
Check the archives!
-p
> -----Original Message-----
> From: eduardo at funrei dot br [mailto:eduardo at funrei dot br]On Behalf Of Eduardo
> M. A. M. Mendes
> Sent: Freitag, 9. Juni 2000 03:43
> To: Subscribers of Qpopper
> Subject: qpopper - I can't send any message!
>
>
> Hello
> I have just installed qpopper on solaris 2.6 and linux. Although
> the installation went fine I can't send emails when using the pop3
> accounts.
> I am not sure whether this is a pop3 problem or something else. Could
> someone out there help me out, please?
>
> Thanks a lot.
>
> regards
>
> Eduardo
>
>
>
>
Date: Fri, 9 Jun 2000 15:55:07 -0600 (MDT)
From: John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Subject: Re: Impact of -R option
On Fri, 9 Jun 2000, James Ralston wrote:
>IMHO, making universally quantified statements such as "you also
>shouldnt be relying on reverse lookups for any security method" is
>just plain not a good idea. Relying on reverse lookups in certain
>circumstances CAN be useful if 1) you perform the checking properly,
>and 2) you understand the ramifications of using reverse lookups as
>part of your security policy decisions.
I agree with you, but i'm not sure if you (or perhaps me) remember the
original post correctly.
I think the poster mentioned something to the effect of 'what downsides do
i have of not using -R' and i mentioned:
there arent really any downsides, the only thing it will effect is reverse
lookup checking, and that he/she should not be using that as their sole
method of security anyway.
--
_ __ _____ __ _________
______________ /_______ ___ ____ /______ John Gonzalez/Net.Tech
__ __ \ __ \ __/_ __ `__ \/ __ /_ ___/ MDC Computers/netMDC!
_ / / / `__/ /_ / / / / / / /_/ / / /__ (505)437-7600/fax-437-3052
/_/ /_/\___/\__/ /_/ /_/ /_/\__,_/ \___/ http://www.netmdc.com
[---------------------------------------------[system info]-----------]
3:50pm up 29 days, 21:16, 4 users, load average: 0.15, 0.20, 0.16
Date: Fri, 9 Jun 2000 15:59:24 -0600 (MDT)
From: John Gonzalez/netMDC admin <ekool at ns1.netmdc dot com>
Subject: Re: Please remove poopie at jeeves.poopie dot net
I have yet to see an MUA that didnt allow you to view the header contents
in some way or another.
However, the people that cant figure out how to unsubscribe from a mailing
list are unlikely to be able to turn on that function....
<sigh>...
On Fri, 9 Jun 2000, Leonard Hermens wrote:
>Except that some clients hide the headers to make the e-mail look
>"pretty" and the user doesn't see them. :-)
>
>-- Leonard
>
>At 2:01 PM -0400 6/9/00, Keith Warno wrote:
>>If you check the headers of a mail from just about any mailing list you'll
>>see info on how to get help or unsubscribe.
>>
>>For example, in the headers of mails from this list:
>>
>>List-Subscribe: <mailto:qpopper-request at lists.pensive dot org?body=subscribe>
>>List-Unsubscribe:
>><mailto:qpopper-request at lists.pensive dot org?body=unsubscribe>
>>List-Archive: <mailto:autoshare at lists.pensive dot org?body=index%20QPopper>
>>List-Post: <mailto:qpopper at lists.pensive dot org>
>>
>>*sigh*
>>
>
--
_ __ _____ __ _________
______________ /_______ ___ ____ /______ John Gonzalez/Net.Tech
__ __ \ __ \ __/_ __ `__ \/ __ /_ ___/ MDC Computers/netMDC!
_ / / / `__/ /_ / / / / / / /_/ / / /__ (505)437-7600/fax-437-3052
/_/ /_/\___/\__/ /_/ /_/ /_/\__,_/ \___/ http://www.netmdc.com
[---------------------------------------------[system info]-----------]
3:55pm up 29 days, 21:21, 4 users, load average: 0.05, 0.15, 0.16
From: "Lisa Casey" <lisa at jellico dot com>
Subject: Re: Please remove poopie at jeeves.poopie dot net
Date: Fri, 9 Jun 2000 17:50:57 -0400
Hi,
> I have yet to see an MUA that didnt allow you to view the header contents
> in some way or another.
>
> However, the people that cant figure out how to unsubscribe from a mailing
> list are unlikely to be able to turn on that function....
>
> <sigh>...
One would think that anyone this clueless would not be running a POP3 mail
server...
<Double sigh>
Lisa Casey, Webmaster
Interstate 2000, INc.
lisa at jellico dot com
Date: Fri, 09 Jun 2000 23:12:14 +0100
From: Fergal Daly <fergal at esatclear dot ie>
Subject: Re: How To Unsubscribe from Qpopper
Despite having fairly low traffic, the qpopper list is has more botched
attempts at unsubscribing than any other listI've ever been on. Quite
ironic given that these people are actually running mail systems!! Maybe we
need to make it a little to subscribe in the first place ;-)
Fergal
At 18:01 09/06/00, Daniel Norton wrote:
>To unsubscribe from Qpopper, put "unsubscribe" in the first line of the
>body (not the Subject) of a message and send it to
><qpopper-request at lists.pensive dot org>.
>
>You can also use this web page:
>
> http://www.pensive.org/Mailing_Lists/QPopper_sub_form.html
>
>For information on other commands available by E-mail, see this page:
>
> http://www.pensive.org/Mailing_Lists/help.html
>
>--
>Daniel Norton
Date: Sat, 10 Jun 2000 12:45:54 +0800 (HKT)
From: PM WONG <pmwong at power25t.hkbu.edu dot hk>
Subject: qpopper -T timeout, How to use?
Our qpopper is very busy (due to a lot of users connecting and
most of their mailboxes are large, somewhere into 50Mb).
So for our case, is it better to set the timeout to be a large
value ( > 10 mins) ?
I've read the Release notes, but was confused by the explanation
of this timeout value. It says something that setting the value
too large has its own adverse effects too.
Could someone give me some advice ?
Date: Sat, 10 Jun 2000 23:09:33 -0400
From: root <everett at gloryroad dot net>
Subject: Problems with qpopper on mandrake 7.0
Hello
I have been useing linux-mandrake 6.1 for a while now and I just
upgraded to linux-mandrake 7.0.
Just before I upgraded I installed qpopper and it worked fine but afther
I upgraded or installed mandrake 7.0
qpopper stoped working.
Well what it's doing is not sending the messages to the client software.
My mail client says no new mail available it logs onto qpopper like it's
should but the messages just want come though.
Now when I say I upgraded mandrake I mean I install the full 7.0
systemso what was there from 6.1 is gone.
So I went back and reinstalled 6.1 and qpopper and then did a upgrade to
7.0 and qpopper worked then but some of my other hardware would not work
so I had to start over and do the full install of 7.0 now qpopper want
work.
I think it's something with linux-mandrake 7.0 I just wanted to see if
any one here might know what the problem is.
Thanks
From: "Vladimir Nikolic" <vladimirn at ptt dot yu>
Subject:
Date: Mon, 12 Jun 2000 08:15:35 +0200
Can anyone tell me where I can find patches to make qpopper look for
password in NT password database?
Subject: Re:
Date: Mon, 12 Jun 2000 15:06:02 -0700
From: Kris Virtue <kvirtue at qualcomm dot com>
Didn't see any response to this one...
Try moving up to 3.0.2. We ran into the same performance problems with
3.0.1 moving mail to the .user.pop too slowly, causing the client to
give up. Randy fixed this in 3.0.2. I'm not sure about the
duplication issues, but 3.0.2 should fix the performance problems while
building the .user.pop files.
-Kris
> stephen:
>
> the user in question originally had a 4 meg mail file. too large for my preference but he's my boss. his outlook client checks the mail every 10 minutes. it was taking almost that long to build the .pop file. this is on a little used d250 with 512meg
mo
> ry. the temp mail dir is on a different controller and drive. so i have 2 problems, why did qpopper duplicate the mail file and why does it run so slow building the .pop ? it took less than a minute for qpopper to move the mail from .pop back to /var
il
> .
>
> thanks.....jim mc....
>
>
> ---------- Original Message ----------------------------------
> From: Stephen Samuel <samuel at bcgreen dot com>
> Date: Wed, 07 Jun 2000 15:55:43 -0700
>
> >>From this, it looks like your client is checking for mail WAAAY
> >to often. (I hope that it was only for testing purposes).
> >Make sure to set the check interval to something approaching sane
> >(probably 2 minutes, or more) I've seen people with really large
> >mail files (>100MB) that take over 10 minutes to complete he
> >copy/look/return cycle with 'leave messages on server' turned on.
> >
> >jmcintyr wrote:
> >>
> >> i am running hpux 10.20 on a d250. have ver 3.0.1b2 installed. the most wonderful thing happened to a user, using the outlook client he connected fine to qpopper and it started building the .user.pop file. after it built this file it,qpopper
t
> h
> >e
> >> whole dam mail file and loaded into the .user.pop file. so now the user has duplicate everything. a really neat option that i would like to have turned off. anybody else experience this wonderful option and how do i turn it off ?
> >>
> >> jim mc....
> >>
> >> below is a ll of the 2 files every 2 seconds:
> >>
> >> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/issneerin.pop
> >> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 4333838 Jun 7 14:04 /tempmail/.ssneerin.pop
> >> -rw-rw---- 1 ssneerin mail 4333967 Jun 7 13:53 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> >> -rw-rw---- 1 ssneerin mail 0 Jun 7 14:04 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> >> -rw-rw---- 1 ssneerin mail 425984 Jun 7 14:04 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> >> -rw-rw---- 1 ssneerin mail 2400256 Jun 7 14:04 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> >> -rw-rw---- 1 ssneerin mail 2613248 Jun 7 14:04 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> >> -rw-rw---- 1 ssneerin mail 3620864 Jun 7 14:04 /var/mail/ssneerin
> >> -rw------- 1 ssneerin mail 8667805 Jun 7 14:04 /tempmail/.ssneerin.pop
> >> -rw-rw---- 1 ssneerin mail 5288440 Jun 7 14:04 /var/mail/ssneerin
> >> /tempmail/.ss* not found
> >> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
> >> /tempmail/.ss* not found
> >> -rw-rw---- 1 ssneerin mail 8668039 Jun 7 14:04 /var/mail/ssneerin
> >> /tempmail/.ss* not found
> >
> >--
> >Stephen Samuel +1(604)876-0426 samuel at bcgreen dot com
> > (formerly samuel at result dot com)
> >The question, for this world, is not "will I die?".
> >It is, rather, "how will I live?".
> >
>
From: "Geoff Wilson" <gwilson at cbnet.ns dot ca>
Date: Mon, 12 Jun 2000 20:30:02 -0300
Subject: Compiling Qpopper 3.1b1
Folks:
I have been trying to update qpopper on one of my servers.
This machine runs Solaris 2.4 and I have been unable to get the
compilation to finish for any of the versions available including the
version in the subject line. I am using gcc and the compiler keeps
barfing with this error :
../mmangle/libmangle.a ../common/libcommon.a
Undefined first referenced
symbol in file
srandom popper.o
ld: fatal: Symbol referencing errors. No output written to popper
*** Error code 1
Everything seems to work fine up to this point, and then it dies.
Has anyone seen this behaviour before and does anyone know
how to fix it ???
Thanks
Geoff Wilson
CBNet
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/
That money talks I don't deny
I heard it once, it said "goodbye"