qmail Digest 2 Mar 2000 11:00:00 -0000 Issue 928

Topics (messages 38012 through 38094):

/etc/tcpcontrol/smtp.cdb not working (Re: can't stop qmail (supervise-scripts))
        38012 by: Smoerk

Re: Forwarding emails
        38013 by: Ricardo Cerqueira
        38014 by: Roger O. Svenning
        38015 by: Roger O. Svenning
        38036 by: iv0

Re: How big is a big queue?
        38016 by: brianb-qmail.technet.evoserve.com

RCPT <[EMAIL PROTECTED]> confuses qmail
        38017 by: Uncle George
        38077 by: Uncle George
        38078 by: Uncle George

Re: I spoke too soon.
        38018 by: Russell Nelson

Re: A complete log rolling & reporting system?
        38019 by: Russell Nelson
        38033 by: Dave Sill
        38065 by: Pavel Kankovsky
        38068 by: Dave Sill

Should this have been rejected?
        38020 by: Mark E. Drummond
        38059 by: Dave Sill

Re: qmail-smtpd and multilog
        38021 by: Bruno Wolff III

Making an smtproute to nowhere
        38022 by: torben fjerdingstad
        38024 by: Petr Novotny
        38025 by: Chris Johnson
        38037 by: torben fjerdingstad

ORBS database
        38023 by: Dan Ammellinn
        38028 by: Frank Tegtmeyer
        38030 by: Frank Tegtmeyer

Greetings and a dopey question:
        38026 by: Bennett
        38029 by: Chris Johnson
        38032 by: Frank Tegtmeyer
        38039 by: Bennett

Re: qmqp and local delivery
        38027 by: Fred Lindberg

Re: qmail-pop3d not conforming to RFC1939?!
        38031 by: iv0
        38034 by: iv0
        38035 by: Russell Nelson
        38038 by: iv0
        38042 by: Len Budney

Problem with tcpserver and pop3
        38040 by: Webmaster
        38058 by: Dave Sill

Re: Lost Mail
        38041 by: Uwe Ohse
        38043 by: Tom Reinertson

list server and rcpthosts file
        38044 by: clifford thurber
        38048 by: Chris Johnson
        38052 by: clifford thurber
        38054 by: Frank Tegtmeyer
        38055 by: Chris Johnson
        38061 by: clifford thurber
        38062 by: Frank Tegtmeyer
        38064 by: Dave Sill
        38066 by: Frank Tegtmeyer
        38069 by: Dave Sill
        38070 by: Chris Johnson

fastforward virtual domains [emergency]
        38045 by: Dan Laffin

Message 252 when VRFYing
        38046 by: Shera
        38049 by: Dave Sill
        38050 by: Anand Buddhdev
        38051 by: Timothy L. Mayo
        38060 by: Russell Nelson

daemontools v6.1 and _Life with qmail_
        38047 by: Grier Ellis
        38053 by: Dave Sill

receiving mail
        38056 by: Lee Trotter
        38057 by: Dave Sill

Re: Multiple Mails...........
        38063 by: Dave Sill

Re: ETRN
        38067 by: Robert Sanderson

Re: POP3 Slowdown solved (I think)
        38071 by: Faried Nawaz

Re: migrating virtuals from sendmail
        38072 by: Faried Nawaz

Unix as it should be
        38073 by: Russell Nelson

Re: Forward Messages to a secondary Mail Server
        38074 by: Jeff Russell, AIT

Qmail installation
        38075 by: Lee Trotter

Re: [qmail] Unix as it should be
        38076 by: ari
        38079 by: Bruno Wolff III

Re: Effective anti spamming
        38080 by: Aaron L. Meehan
        38081 by: Adam McKenna
        38082 by: Sascha Schumann
        38089 by: Sam

Re: Ineffective anti spamming
        38083 by: Ruben van der Leij

Cannot get Qmail to compile
        38084 by: BaimoonInc.aol.com
        38086 by: Stephen Mills

Re: Mailing list bandwidth
        38085 by: Steve Wolfe
        38088 by: net.ncal.verio.com
        38090 by: andy huhn

Qmail on FreeBSD 3.3
        38087 by: David Uzzell

Forward/Duplicate messages to hosts behind
        38091 by: Andy WONG
        38093 by: Stephen Mills

Mail doesn't go to Maildir
        38092 by: Stein Ma

vpopmail and Netscape
        38094 by: Erich Zigler

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]


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


>> >> >> 2. qmail-smtpd does not care about /etc/tcprules/smtp.cdb. I cannot
>> >> >> send mails localy via smtp, because the smtp does not relay. I put
>[...]
>> exec tcpserver -u "$uid" -g "$gid" -c "$concurrency" -v \
>>         -x /etc/tcpcontrol/smtp.cdb 0 smtp $rbl \
>>         qmail-pipe fixcr -- qmail-smtpd

>
>This uses /etc/tcpcontrol/smtp.cdb as opposed to /etc/tcprules/smtp.cdb, which
>                  ^^^^^^^                                ^^^^^
>is what you said you use.

Okay, the directory is /etc/tcpcontrol, not /etc/tcprules. My mail was
wrong not my configuration.

I still have the problem, that /etc/tcpcontrol/smtp.cdb is ignored. I
put

:deny

in smtp.rules (and created a new smtp.cdb) and connections were not
refused.

>As far a the supervise bit is concerned... try 'svc -dx <qmail_supervise_dir>'.
>If the process is still running after that, kill (-15) tcpserver.

Thank you, I will try it. But I found out, that stopping/starting
svscan starts/stops all the daemons (qmail, smtpd, log). So this
problem is solved.





On Wed, Mar 01, 2000 at 11:30:42AM +0100, Roger O. Svenning wrote:
> Hi
> 
> I using the vchkpw package to handle virtual domains on our server.
> How do I forward an existing users email ?
> I tried setting up a .qmail file but the mail still ends up in the users
> Maildir.

If you want to forward it, why do you have a Maildir for that user? :)
Anyway, that shouldn't happen... Are you setting up .qmail-<user> within the right 
directory? 

                                                Regards;
                                                        Ricardo

-- 
+-------------------
| Ricardo Cerqueira  
| PGP Key fingerprint  -  B7 05 13 CE 48 0A BF 1E  87 21 83 DB 28 DE 03 42 
| Novis  -  Rede T�cnica 
| P�. Duque Saldanha, 1, 7� E / 1050-094 Lisboa / Portugal




> On Wed, Mar 01, 2000 at 11:30:42AM +0100, Roger O. Svenning wrote:
> > Hi
> >
> > I using the vchkpw package to handle virtual domains on our server.
> > How do I forward an existing users email ?
> > I tried setting up a .qmail file but the mail still ends up in the users
> > Maildir.
>
> If you want to forward it, why do you have a Maildir for that user? :)
> Anyway, that shouldn't happen... Are you setting up .qmail-<user> within
the right directory?

Because this user will be away this year and we need to forward her email to
her replacement while she's away.
And yes I made a .qmail-<user> in the domains directory and inserted a
&<address> in that file just like the other forward files in that dir, the
only exception is that this user already has a Maildir.

I'll give it another try ...

Roger O. Svenning





> > On Wed, Mar 01, 2000 at 11:30:42AM +0100, Roger O. Svenning wrote:
> > > Hi
> > >
> > > I using the vchkpw package to handle virtual domains on our server.
> > > How do I forward an existing users email ?
> > > I tried setting up a .qmail file but the mail still ends up in the
users
> > > Maildir.
> >
> > If you want to forward it, why do you have a Maildir for that user? :)
> > Anyway, that shouldn't happen... Are you setting up .qmail-<user> within
> the right directory?
>
> Because this user will be away this year and we need to forward her email
to
> her replacement while she's away.
> And yes I made a .qmail-<user> in the domains directory and inserted a
> &<address> in that file just like the other forward files in that dir, the
> only exception is that this user already has a Maildir.
>
> I'll give it another try ...

My fault

This user had a dot in the address and I forgot to replace the dot with a
colon.

Everything works fine now

-Roger





"Roger O. Svenning" wrote:
> 
> Hi
> 
> I using the vchkpw package to handle virtual domains on our server.
> How do I forward an existing users email ?
> I tried setting up a .qmail file but the mail still ends up in the users
> Maildir.
> 
> Roger O. Svenning

Is the user in a virtual domain or in /etc/passwd?

both can use normal .qmail file processing since qmail-local
is delivering it to either the virtual domain directory or
the users home directory.

If the user is in a virtual domain do this:

echo "[EMAIL PROTECTED]" > ~vpopmail/domains/<virtual domain
name>/.qmail-<user>

Ken Jones
www.inter7.com




On Wed, 1 Mar 2000 [EMAIL PROTECTED] wrote:

> Hi, I'm currently managing a mailserver for a service with around 5500
> virtual users as of this writing, and we get over 1000 new users daily.
> I've been tracking our queue size since we launched, and it's currently
> just shy of 1000 messages, with ~650 messages unprocessed. Is this going
> to be a problem? How big are queues supposed to be?
> 
> Also, I've lowered my queuelifetime to 2 days and increased
> concurrencyremote to 120. Otherwise, it's a stock LWQ installation.

Oops. Actually, it WAS a completely standard LWQ installation. I had
forgotten to restart qmail-send, so it was using the concurrencyremote
default of 20. I restarted qmail-send, and watched qmail-qstat report a
rapidly dwindling queue. It's back to normal now.

I also watched my load average shoot up to 2.5 as all the qmail-remote
processes fired up. 

Brian
--
[EMAIL PROTECTED]
http://www.baquiran.com
AIM: bbaquiran








Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

I use fetchmail to retrieve messages from my ISP to my firewall.
fetchmail appears to append these lines at the top of the mail message

FROM <[EMAIL PROTECTED]>
RCPT <[EMAIL PROTECTED]>
BODY

But when i try to qmail-inject ( via the qmail sendmail program ), my
(own) DNS server gets this request ( PLEASE not that the ".com>" portion
appears to be bogus!

Mar  1 07:35:23 bigchill named[397]:
XX+/192.168.0.40/bigchill.gatworks.com>/ANY

If I remove the "<" and the ">" everything appears to be happy.

so is fetchmail confused, or qmail not learned enough?
gat










It seems that /usr/qmail/bin/sendmail -f<add@sender>  <gat@localhost>
where the "<>" in the <gat@localhost>  is not being taken take of ( is
removed ) before it is sent off to the dns server

Is there a bug list fot this stuff ?
gat

Uncle George wrote:

>   ------------------------------------------------------------------------
>
> Subject: RCPT <[EMAIL PROTECTED]> confuses qmail
> Date: Wed, 01 Mar 2000 07:51:56 -0500
> From: Uncle George <[EMAIL PROTECTED]>
> Organization: GatWorks
> To: [EMAIL PROTECTED]
>
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
>
> I use fetchmail to retrieve messages from my ISP to my firewall.
> fetchmail appears to append these lines at the top of the mail message
>
> FROM <[EMAIL PROTECTED]>
> RCPT <[EMAIL PROTECTED]>
> BODY
>
> But when i try to qmail-inject ( via the qmail sendmail program ), my
> (own) DNS server gets this request ( PLEASE not that the ".com>" portion
> appears to be bogus!
>
> Mar  1 07:35:23 bigchill named[397]:
> XX+/192.168.0.40/bigchill.gatworks.com>/ANY
>
> If I remove the "<" and the ">" everything appears to be happy.
>
> so is fetchmail confused, or qmail not learned enough?
> gat





Uncle George wrote:

>   ------------------------------------------------------------------------
>
> Subject: RCPT <[EMAIL PROTECTED]> confuses qmail
> Date: Wed, 01 Mar 2000 07:51:56 -0500
> From: Uncle George <[EMAIL PROTECTED]>
> Organization: GatWorks
> To: [EMAIL PROTECTED]
>
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
>
> I use fetchmail to retrieve messages from my ISP to my firewall.
> fetchmail appears to append these lines at the top of the mail message
>
> FROM <[EMAIL PROTECTED]>
> RCPT <[EMAIL PROTECTED]>
> BODY
>
> But when i try to qmail-inject ( via the qmail sendmail program ), my
> (own) DNS server gets this request ( PLEASE not that the ".com>" portion
> appears to be bogus!
>
> Mar  1 07:35:23 bigchill named[397]:
> XX+/192.168.0.40/bigchill.gatworks.com>/ANY
>
> If I remove the "<" and the ">" everything appears to be happy.
>
> so is fetchmail confused, or qmail not learned enough?
> gat





Stephen Bosch writes:
 > The server will only relay if the RELAYCLIENT variable is set,
 > correct? But I need the allow all at the end to ensure that I can receive
 > mail from the outside...

Right, but allow is the default condition, so all you need are
RELAYCLIENT lines matching the networks you want to allow to relay.

-- 
-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.




Pavel Kankovsky writes:
 > Ever heard of disk quotas? It might be a bit of overkill to create a
 > special user for every logfile (or group of thereof) but it works.

You've tried it?  But syslog runs as root.  Oops.

-- 
-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.




Pavel Kankovsky <[EMAIL PROTECTED]> wrote:

>Unfortunately, a message written to disk and erased from it before
>anyone had a chance to look at it is as good (read useless) as a message
>discarded immediately. :)

True, but a message written to disk and cycled through N log files
stands a better chance of being seen than one that never makes it to
disk. :-) And if the logs are monitored by a log watching process,
"seeing" them is guaranteed.

>The question is: do you prefer to LOSE old messages or new messages when
>you run out of space?

No, the question is: do you prefer to run out of disk space or keep
your logs under predetermined limits?

>Syslog says "new messages", cyclog says "old
>messages". I have to admit I do not understand why some people think one
>of the strategies is inherently better than the other...explanation?

Firstly, the choice is not that simple, as I explained
above. Secondly, there are a host of other reasons not to use syslog,
including performance, security, and reliability problems.

-Dave




On Wed, 1 Mar 2000, Russell Nelson wrote:

> Pavel Kankovsky writes:
>  > Ever heard of disk quotas? It might be a bit of overkill to create a
>  > special user for every logfile (or group of thereof) but it works.
> 
> You've tried it?  But syslog runs as root.  Oops.

Oh, you got me. I have mis-extrapolated the results of some experiments
done under very special conditions. Damned omnipotent root. I hate unix.


On Wed, 1 Mar 2000, Dave Sill wrote:

> Pavel Kankovsky <[EMAIL PROTECTED]> wrote:
> 
> >Unfortunately, a message written to disk and erased from it before
> >anyone had a chance to look at it is as good (read useless) as a message
> >discarded immediately. :)
> 
> True, but a message written to disk and cycled through N log files
> stands a better chance of being seen than one that never makes it to
> disk. :-) And if the logs are monitored by a log watching process,
> "seeing" them is guaranteed.

Unless the log is fed to the program directly (via a pipe of something) or
you have got a special scheduler guaranteeing cyclog can't save and rotate
logs faster than the program can grok them, the program can always miss a
file. Of course, the probability of missing a file is rather low.

> >The question is: do you prefer to LOSE old messages or new messages when
> >you run out of space?
> 
> No, the question is: do you prefer to run out of disk space or keep
> your logs under predetermined limits?

The size of a log is always kept under a predetermined limit: the
total capacity of a filesystem where the log is located. :)

--Pavel Kankovsky aka Peak  [ Boycott Microsoft--http://www.vcnet.com/bms ]
"Resistance is futile. Open your source code and prepare for assimilation."





Pavel Kankovsky <[EMAIL PROTECTED]> wrote:

>I hate unix.

Yeah, it's the *worst* operating system...except for all the others.

>> True, but a message written to disk and cycled through N log files
>> stands a better chance of being seen than one that never makes it to
>> disk. :-) And if the logs are monitored by a log watching process,
>> "seeing" them is guaranteed.
>
>Unless the log is fed to the program directly (via a pipe of something)

Like a "!processor" multilog action?

>> No, the question is: do you prefer to run out of disk space or keep
>> your logs under predetermined limits?
>
>The size of a log is always kept under a predetermined limit: the
>total capacity of a filesystem where the log is located. :)

You have each log file in a separate partition? Or do you allow a mail 
flood to kill ftp logging?

-Dave




Last night, someone somewhere attempted to send one of my users an email with
the W32/Fix trojan attached. Thanks to Jason's Scan4Virus script, it was picked
up and nixed at my mail gateway. Excellant (spoken with a Mr. Burns'esque
inflection) ... the hounds have been released.

However, The envelop sender was [EMAIL PROTECTED] rmc.ca is my domain, and of
course admin__ does not exist. Should this have been rejected? How can I tell
qmail to say that mail from rmc.ca should be from inside our network? I think I
am thinking about this the wrong way. I know qmail is supposed to verify the
sender domain, but is there some way I could prevetn this type of thing from
getting in? Had there been no virus, had it been just plain old spam, it would
have made it through because the domain, rmc.ca, is legit, but the sender was
not.

-- 
______________________________________________________
Mark Drummond|ICQ#19153754|mailto:[EMAIL PROTECTED]
         Gang Warily|http://signals.rmc.ca/
Kingston Linux Users Group|http://signals.rmc.ca/klug/




"Mark E. Drummond" <[EMAIL PROTECTED]> wrote:

>However, The envelop sender was [EMAIL PROTECTED] rmc.ca is my domain, and of
>course admin__ does not exist. Should this have been rejected?

No, qmail doesn't verify the ERP.

> How can I tell
>qmail to say that mail from rmc.ca should be from inside our network?

That would require a patch.

>I think I
>am thinking about this the wrong way. I know qmail is supposed to verify the
>sender domain,

No, it's not.

>but is there some way I could prevetn this type of thing from
>getting in? Had there been no virus, had it been just plain old spam, it would
>have made it through because the domain, rmc.ca, is legit, but the sender was
>not.

Again, the behavior you desire requires a patch.

-Dave




The tcpserver options I copied over included -q. I should know better than
to ask questions late at night.

On Wed, Mar 01, 2000 at 12:43:46AM -0600,
  Bruno Wolff III <[EMAIL PROTECTED]> wrote:
> I am setting up qmail on a new box and am trying to use multilog (which I
> wasn't on my old box). It seems to work OK for qmail, but I also was going
> to try to use it for qmail-smtpd. I think the multilog stuff is set up
> correctly, but I am not getting anything logged. Is this normal? Is there
> a condition I can test that should cause tcpserver or qmail-smtpd to write
> something that multilog will see?




Will the following line in smtproutes silently throw away all
outgoing mail to portal.mdr.net?

portal.mdr.net:

If not, how do I do it?

The problem is that I have thousands of error messages queued
for that host/domain. And it does does not exist. I want to
delete those messages from the queue, without further
retransmission attempts.

-- 
Med venlig hilsen / Regards 
Netdriftgruppen / Network Management Group
UNI-C          

Tlf./Phone   +45 35 87 89 41        Mail:  UNI-C                                
Fax.         +45 35 87 89 90               Bygning 304
E-mail: [EMAIL PROTECTED]       DK-2800 Lyngby





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

On 1 Mar 00, at 15:12, torben fjerdingstad wrote:

> Will the following line in smtproutes silently throw away all
> outgoing mail to portal.mdr.net?
> 
> portal.mdr.net:

No. It says "route portal.mdr.net according to MX record". It is used 
in this context:
portal.mdr.net:
:smart.host.at.isp
(it delivers to portal... directly, and the rest relays though the 
provider)

> If not, how do I do it?
> 
> The problem is that I have thousands of error messages queued
> for that host/domain. And it does does not exist. I want to
> delete those messages from the queue, without further
> retransmission attempts.

1. (more work) stop qmail, identify all those messages, delete 
them, start qmail
2. (simpler) Put
portal.mdr.net:alias-nowhere
into virtualdomains. Put
#
into ~alias/.qmail-nowhere-default. Put
portal.mdr.net:127.0.0.1
into smtproutes; if localhost is not allowed to relay, put 
portal.mdr.net also into rcpthosts. HUP qmail-send.
ALRM qmail-send. Watch the mails go to bit-heaven.


In other words, you don't make smtp to nowhere; you make virtual 
domain to nowhere.

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

iQA/AwUBOL0Y/FMwP8g7qbw/EQJyngCdE7sc1lK7OExb2WDJJsmuQZ5IRMIAoO3X
if99mVhZe+1OHAoLhMBGQ6j0
=fCqk
-----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 Wed, Mar 01, 2000 at 03:12:42PM +0100, torben fjerdingstad wrote:
> Will the following line in smtproutes silently throw away all
> outgoing mail to portal.mdr.net?
> 
> portal.mdr.net:
> 
> If not, how do I do it?
> 
> The problem is that I have thousands of error messages queued
> for that host/domain. And it does does not exist. I want to
> delete those messages from the queue, without further
> retransmission attempts.

Try this:

# echo '#' > ~alias/.qmail-baddomain-default
# echo portal.mdr.net:alias-baddomain >> /var/qmail/control/virtualdomains
# echo portal.mdr.net:127.0.0.1 >> /var/qmail/control/smtproutes

Then run /var/qmail/bin/qmail-tcpok, and send qmail-send a HUP and an ALRM. All
your queued mail for portal.mdr.net will get delivered by SMTP to 127.0.0.1 and
will be handled by ~alias/.qmail-baddomain-default. The single # in that file
will cause all the mail to be discarded.

Chris




On Wed, Mar 01, 2000 at 09:29:03AM -0500, Chris Johnson wrote:
> On Wed, Mar 01, 2000 at 03:12:42PM +0100, torben fjerdingstad wrote:
> > Will the following line in smtproutes silently throw away all
> > outgoing mail to portal.mdr.net?
> > 
> > portal.mdr.net:
> > 
> > If not, how do I do it?
> > 
> > The problem is that I have thousands of error messages queued
> > for that host/domain. And it does does not exist. I want to
> > delete those messages from the queue, without further
> > retransmission attempts.
> 
> Try this:
> 
> # echo '#' > ~alias/.qmail-baddomain-default
> # echo portal.mdr.net:alias-baddomain >> /var/qmail/control/virtualdomains
> # echo portal.mdr.net:127.0.0.1 >> /var/qmail/control/smtproutes
> 
> Then run /var/qmail/bin/qmail-tcpok, and send qmail-send a HUP and an ALRM. All
> your queued mail for portal.mdr.net will get delivered by SMTP to 127.0.0.1 and
> will be handled by ~alias/.qmail-baddomain-default. The single # in that file
> will cause all the mail to be discarded.

It works (except for 127.0.0.1 was not in my rcpthosts).

Honestly, at first I screewed it up, and generated a lot
of bounces.

Thank you very much

-- 
Med venlig hilsen / Regards 
Netdriftgruppen / Network Management Group
UNI-C          

Tlf./Phone   +45 35 87 89 41        Mail:  UNI-C                                
Fax.         +45 35 87 89 90               Bygning 304
E-mail: [EMAIL PROTECTED]       DK-2800 Lyngby





Help please! I's got to ORBS. I'd tried to solve it by the month. Read the
every letter from mailing list. But haven't the result. Can you help me?
Thes're my scripts

rc*:
tcpcontrol /etc/tcp.smtp.cdb tcpserver -x /etc/tcp.smtp.cdb -u 82 -g 81 0
smtp
rblsmtpd /var/qmail/bin/qmail-smtpd &
tcpserver -u 82 -g 81 0 qmtp /var/qmail/bin/qmail-qmtpd &
tcpserver 0 pop3 /var/qmail/bin/qmail-popup my.mail.server \
/bin/checkpassword /var/qmail/bin/qmail-pop3d Maildir &
qmail-start '|dot-forward .forward
./Mailbox' splogger qmail&

tcp.smtp

1.2.3.4:allow,RELAYCLIENT=""
127.0.0.1:allow,RELAYCLIENT=""
10.:allow,RELAYCLIENT=""
:allow

I have tcpserver of course. What's wrong here? Thanks.






> I have tcpserver of course. What's wrong here? Thanks.

Do you have a rcpthosts file?  Is ORBS possibly testing from a
10.x.x.x address?

Regards, Frank




> Is ORBS possibly testing from a
> 10.x.x.x address?

:) was missing :)




Hi all, i just joined the list and wanted to say Holla.    Also I had a
quick (if somewhat dopey) question.  Is there currently available a book
on Qmail?  I've just recently completed a standard install of qmail, and
now that it's been running more or less smoothly, I wanted to make some
changes, such as # of remotes, max lifetime in queue, etc.  


Anyways, any gentle nudges (or kicks) in the right direction would be
appreciated.

Thanks!

Bennett Thede






On Wed, Mar 01, 2000 at 09:34:04AM -0500, Bennett wrote:
> Hi all, i just joined the list and wanted to say Holla.    Also I had a
> quick (if somewhat dopey) question.  Is there currently available a book
> on Qmail?  I've just recently completed a standard install of qmail, and
> now that it's been running more or less smoothly, I wanted to make some
> changes, such as # of remotes, max lifetime in queue, etc.  

There's no book, yet. A couple of guys on the list are working on one though.

In the meantime, you might look at Dave Sill's excellent "Life with qmail."
It's at http://Web.InfoAve.Net/~dsill/lwq.html.

Chris




> quick (if somewhat dopey) question.  Is there currently available a book
> on Qmail?

There is work on one but the book is out of schedule.

The best is to look at www.qmail.org. Look there for "Life with qmail". 
There are also other useful resources.

Regards, Frank




Great, thank you very much.  I'm just a beginer in the world of Unix, long
time (l)user, new admin, and I definately look forward to learning more
about Qmail.


Thanks again,

Ben

On Wed, 1 Mar 2000, Chris Johnson wrote:

> On Wed, Mar 01, 2000 at 09:34:04AM -0500, Bennett wrote:
> > Hi all, i just joined the list and wanted to say Holla.    Also I had a
> > quick (if somewhat dopey) question.  Is there currently available a book
> > on Qmail?  I've just recently completed a standard install of qmail, and
> > now that it's been running more or less smoothly, I wanted to make some
> > changes, such as # of remotes, max lifetime in queue, etc.  
> 
> There's no book, yet. A couple of guys on the list are working on one though.
> 
> In the meantime, you might look at Dave Sill's excellent "Life with qmail."
> It's at http://Web.InfoAve.Net/~dsill/lwq.html.
> 
> Chris
> 





On Tue, 29 Feb 2000 21:18:12 -0800, Tracy R Reed wrote:

>Also, I don't see where in my ezmlm installation I can tell it to use another
>qmail installation. Do I have to recompile ezmlm? If so, then I may as well
>just bite the bullet one weekend soon and do the upgrade.

You can take your existing ezmlm and make it call qmail-newqueue
instead of qmail-queue. qmail-newqueue would be qmail-queue compiled
with a new conf-home.

You can do the same, but use "qmail-qmqpc" and set it up use QMQP to
transfer mail to a different qmail installation (all mail from the
lists). The advantage is that you can do it to a different host.

none of these require and change to ezmlm other than recompiling.
Remember to apply qmail-verh to the new installation, if you rely on
it.

THe advantage to the ezmlm-idx-0.40 setup is that admin messages and
posts are separated. This is useful to allow sub/unsub to function
immediately even while you are sending a post to 500K subscribers.


-Sincerely, Fred

(Frederik Lindberg, Infectious Diseases, WashU, St. Louis, MO, USA)






Russell Nelson wrote:
> 
> Markus Wuebben writes:
>  > Is this known?
>  > A complete description of the problem can be found
>  > at http://www.inter7.com/vpopmail/exploit.html
> 
> Yes, it's known.  The patch is still given using strlen(), though,
> which drags in the C library and makes qmail-pop3d gratuitiously
> bigger.

The proposed patch was not used to fix the exploit. A regular
bounds based read was used. However, good point about the strlen
usage in the package. All the str functions have been removed
from the 3.4.12 development verson. I'm not sure how using
string functions in the authentication program would effect
the size of qmail-pop3d.

Ken Jones
www.inter7.com




[EMAIL PROTECTED] wrote:
> 
> On Tue, Feb 29, 2000 at 10:24:39AM -0500, Russell Nelson wrote:
> > Markus Wuebben writes:
> >  > Is this known?
> >  > A complete description of the problem can be found
> >  > at http://www.inter7.com/vpopmail/exploit.html
> >
> > Yes, it's known.  The patch is still given using strlen(), though,
> > which drags in the C library and makes qmail-pop3d gratuitiously
> > bigger.
> 
> Patching qmail-pop3d is just plain wrong. qmail-pop3d is completely ok,
> it's vpopmail that should be fixed.
> 

Ah. I understand Russell's comment from before about the patch
using strlen and dragging in the string library. The patch was for
qmail-pop3d. Ahah. no no. The problem was in vpopmail and was fixed
January 7th, a few hours after we heard about it. Don't patch
qmail-pop3d. 

Ken Jones
www.inter7.com




iv0 writes:
 > Ah. I understand Russell's comment from before about the patch
 > using strlen and dragging in the string library. The patch was for
 > qmail-pop3d. Ahah. no no. The problem was in vpopmail and was fixed
 > January 7th, a few hours after we heard about it. Don't patch
 > qmail-pop3d. 

Yeah, but you're still serving up the Bugtraq posting with the
qmail-pop3d patch in it.

-- 
-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.




Russell Nelson wrote:
> 
> iv0 writes:
>  > Ah. I understand Russell's comment from before about the patch
>  > using strlen and dragging in the string library. The patch was for
>  > qmail-pop3d. Ahah. no no. The problem was in vpopmail and was fixed
>  > January 7th, a few hours after we heard about it. Don't patch
>  > qmail-pop3d.
> 
> Yeah, but you're still serving up the Bugtraq posting with the
> qmail-pop3d patch in it.

Funny thing. After reading that bugtraq posting again, I don't
agree with the contents. So I've taken it down. We orignally 
put it up to be complete.

Ken




iv0 <[EMAIL PROTECTED]> wrote:
> Russell Nelson wrote:
> > 
> > Yeah, but you're still serving up the Bugtraq posting with the
> > qmail-pop3d patch in it.
> 
> Funny thing. After reading that bugtraq posting again, I don't agree
> with the contents. So I've taken it down. We orignally put it up to
> be complete.

That was very honorable, since those comments tended to exhonorate
your software at the expense of Dan's. I congratulate you!

Len.


--
Peterson's basic problem is that he wasn't aware that ``radix sort''
could refer to anything other than LSD radix sort. Now he's going on an
ad hominem rampage, attacking me to try to cover up his own ignorance.
                                -- Dan Bernstein




Hi everybody,

I just installed tcpserver with pop3.

/usr/local/bin/tcpserver 0 110 /var/qmail/bin/qmail-popup hostname.com
 /bin/checkpassword /var/qmail/bin/qmail-pop3d Maildir &


i can't read mail with pop3 client,but if i do

telnet hostname.com 110
Connected to 
Escape character is '^]'.
+OK <>
user xxxxx
+OK
pass xxxxx
+OK
list

then i can read mail with pop3 client

Any idea?

thanks in advance


Juan Carlos Rodr�guez
Webmaster SEKER BBS, S.A.
[EMAIL PROTECTED]
Comte Borrell 209-211, Entlo. 1D
08029 BARCELONA
Tel. 93 444 76 00 - 902 338 338 - FAX 93 410 10 08








Webmaster <[EMAIL PROTECTED]> wrote:

>i can't read mail with pop3 client,but if i do
>
>telnet hostname.com 110
>Connected to 
>Escape character is '^]'.
>+OK <>
>user xxxxx
>+OK
>pass xxxxx
>+OK
>list
>
>then i can read mail with pop3 client
>
>Any idea?

Try another pop3 client, verify the configuration of the one you're
using, or, if you're really desparate, use recordio to record the POP3 
session and see what's going wrong.

-Dave




On Tue, Feb 29, 2000 at 05:47:25PM -0700, Tom Reinertson wrote:

> The mail log shows the sendmail on my workstation processing the messages just
> fine, I think, but I really don't know how to read output from sendmail.

indeed. Something like:

Feb 29 16:42:33 thecanyons sendmail[7437]: QAA07436: to=<tom@localhost>,
  delay=00:00:01, xdelay=00:00:00, mailer=local, relay=local, stat=Sent

means that the message QAA07436 has been successfully delivered locally.
(Whatever this means: the "local" mailer may have been "cat >/dev/null" 
or so). 
Note that my knowledge of sendmail internals is quite limited. I wish
it to stay that way.

btw: did you have netrape or emacs or any other mail reading program
running? If yes that one might have "stored" the missing mails somewhere.


> Would someone mind just looking at the sendmail log to see if anything obvious
> jumps out at them?  Sure would appreciate it.

Yes. You have a problem on a sendmail only system. This is a qmail forum.

And this:

> Feb 29 16:43:17 thecanyons sendmail[7461]: QAA07457: forward 
>/home/guest/.forward.thecanyons: World writable directory
> Feb 29 16:43:17 thecanyons sendmail[7461]: QAA07457: forward /home/guest/.forward: 
>World writable directory

This is obviously caused by the delivery of messages fetchmail retrieved
from the socialites account at pop.slkc.uswest.net. These two messages
cannot be delivered until the permission problem on ~guest has been 
fixed (hey, look, the sendmail people do something _right_).
"chmod 700 ~guest ; chown guest ~guest" is a first fix, but if it was
a multi user machine you might be in trouble now.

Regards, Uwe




Uwe,

> Note that my knowledge of sendmail internals is quite limited. I wish
> it to stay that way.

My sentiments exactly.

> btw: did you have netrape or emacs or any other mail reading program
> running? If yes that one might have "stored" the missing mails somewhere.

Nope.

> > Feb 29 16:43:17 thecanyons sendmail[7461]: QAA07457: forward 
>/home/guest/.forward.thecanyons: World writable directory
> > Feb 29 16:43:17 thecanyons sendmail[7461]: QAA07457: forward /home/guest/.forward: 
>World writable directory
> 
> This is obviously caused by the delivery of messages fetchmail retrieved
> from the socialites account at pop.slkc.uswest.net. These two messages
> cannot be delivered until the permission problem on ~guest has been 
> fixed (hey, look, the sendmail people do something _right_).
> "chmod 700 ~guest ; chown guest ~guest" is a first fix, but if it was
> a multi user machine you might be in trouble now.

The odd thing is that these files don't even exist.  Sendmail must presume
their existence and then do so badly.

I am grateful for your response and I really do realize this is a qmail forum. 
I'll just keep my fingers crossed and hope it doesn't happen again.

Tom




Hello,
We are running a lyris list server on a web machine. We run qmail(under tcp
server) on this same box as well. In the /etc/smtp.tcp file I listed the IP
of the machine as well as 127.0.0.1. When I look at  the SMTP converstation
I see the list server get the following error message:

"553 sorry, that domain isn't in my list of allowed rcpthosts (#5.7.1)" 

This domain isn't in the rcpthosts file of course but localhost as well as
the boxe's IP are listed in /etc/smtp.tcp.
Lyris the list server has a cname which points to the boxes IP. How do I
enable the list server to send mail out to the internet and yet have qmail
only accept incomming mail for the domains listed in the rcpthosts file ?
Thanks in advance as allways.
Clifford Thurber
Web Systems Administrator
LiveUniverse.com
[EMAIL PROTECTED]
565 5th Ave. 29th Fl.
New York, NY 10017
Ph:212 883 6940  (131)
Fax:212 856 9134




On Wed, Mar 01, 2000 at 11:32:46AM -0500, clifford thurber wrote:
> Hello,
> We are running a lyris list server on a web machine. We run qmail(under tcp
> server) on this same box as well. In the /etc/smtp.tcp file I listed the IP
> of the machine as well as 127.0.0.1. When I look at  the SMTP converstation
> I see the list server get the following error message:
> 
> "553 sorry, that domain isn't in my list of allowed rcpthosts (#5.7.1)" 
> 
> This domain isn't in the rcpthosts file of course but localhost as well as
> the boxe's IP are listed in /etc/smtp.tcp.

What are the exact contents of your smtp.tcp file, and how do you start your
SMTP invocation of tcpserver? When your lyris thing makes its SMTP connection,
from what IP address does it appear to be connecting? (Check the logs.)

If things were set up correctly, it would work. If you don't tell us how things
are set up, nobody will be able to discover your mistake. (I think I'll tag
this paragraph onto every message I send to the list from now on.)

Chris




The only entries listed in the /etc/tcp.smtp file are the loopback address
and the boxe's IP address. 
The invocation of qmail is done with: 

tcpserver -x/etc/tcp.smtp.cdb -u1120 -g1142 -c200 0 smtp
/path/to/qmail/qmail-smtpd

When lyris makes its SMTP connection the IP address is the boxes IP address
which is listed in the /etc/tcp.smtp file.


At 11:49 AM 3/1/00 -0500, Chris Johnson wrote:
>On Wed, Mar 01, 2000 at 11:32:46AM -0500, clifford thurber wrote:
>> Hello,
>> We are running a lyris list server on a web machine. We run qmail(under tcp
>> server) on this same box as well. In the /etc/smtp.tcp file I listed the IP
>> of the machine as well as 127.0.0.1. When I look at  the SMTP converstation
>> I see the list server get the following error message:
>> 
>> "553 sorry, that domain isn't in my list of allowed rcpthosts (#5.7.1)" 
>> 
>> This domain isn't in the rcpthosts file of course but localhost as well as
>> the boxe's IP are listed in /etc/smtp.tcp.
>
>What are the exact contents of your smtp.tcp file, and how do you start your
>SMTP invocation of tcpserver? When your lyris thing makes its SMTP
connection,
>from what IP address does it appear to be connecting? (Check the logs.)
>
>If things were set up correctly, it would work. If you don't tell us how
things
>are set up, nobody will be able to discover your mistake. (I think I'll tag
>this paragraph onto every message I send to the list from now on.)
>
>Chris
> 
Clifford Thurber
Web Systems Administrator
LiveUniverse.com
[EMAIL PROTECTED]
565 5th Ave. 29th Fl.
New York, NY 10017
Ph:212 883 6940  (131)
Fax:212 856 9134




> The only entries listed in the /etc/tcp.smtp file are the loopback address
> and the boxe's IP address. 

What is the exact content of /etc/tcp.smtp?

It should be something like that

127.0.0.1:allow,RELAYCLIENT=""
<box-ip-addr>:allow,RELAYCLIENT=""

Regards, Frank




On Wed, Mar 01, 2000 at 12:06:26PM -0500, clifford thurber wrote:
> The only entries listed in the /etc/tcp.smtp file are the loopback address
> and the boxe's IP address. 
> The invocation of qmail is done with: 
> 
> tcpserver -x/etc/tcp.smtp.cdb -u1120 -g1142 -c200 0 smtp
> /path/to/qmail/qmail-smtpd
> 
> When lyris makes its SMTP connection the IP address is the boxes IP address
> which is listed in the /etc/tcp.smtp file.

Right. And if you didn't have a mistake in your tcp.smtp file, it would work.
But you're resistant to telling anyone what exactly is in your tcp.smtp file,
so I'm afraid I can't be of any help.

Chris




Sorry I just didn't want to post the IPs etc. But it looks like this:

127.0.0.1:allow,RELAYCLIENT=""
<box-ip-addr>:allow,RELAYCLIENT=""
:allow

I did try telnet localhost 25
Mail From: [EMAIL PROTECTED]
250 ok
Rcpt To: [EMAIL PROTECTED]
553 sorry, that domain isn't in my list of allowed rcpthosts (#5.7.1)

So it looks like it is not necessarily a Lyris problem but a configuration
problem. Any help would be aprreciated.
Thanks

At 12:21 PM 3/1/00 -0500, Chris Johnson wrote:
>On Wed, Mar 01, 2000 at 12:06:26PM -0500, clifford thurber wrote:
>> The only entries listed in the /etc/tcp.smtp file are the loopback address
>> and the boxe's IP address. 
>> The invocation of qmail is done with: 
>> 
>> tcpserver -x/etc/tcp.smtp.cdb -u1120 -g1142 -c200 0 smtp
>> /path/to/qmail/qmail-smtpd
>> 
>> When lyris makes its SMTP connection the IP address is the boxes IP address
>> which is listed in the /etc/tcp.smtp file.
>
>Right. And if you didn't have a mistake in your tcp.smtp file, it would work.
>But you're resistant to telling anyone what exactly is in your tcp.smtp file,
>so I'm afraid I can't be of any help.
>
>Chris
> 
Clifford Thurber
Web Systems Administrator
LiveUniverse.com
[EMAIL PROTECTED]
565 5th Ave. 29th Fl.
New York, NY 10017
Ph:212 883 6940  (131)
Fax:212 856 9134




> I did try telnet localhost 25
> Mail From: [EMAIL PROTECTED]
> 250 ok
> Rcpt To: [EMAIL PROTECTED]
> 553 sorry, that domain isn't in my list of allowed rcpthosts (#5.7.1)
> 
> So it looks like it is not necessarily a Lyris problem but a configuration

What does the log say about tcpserver activity? 

Frank




clifford thurber <[EMAIL PROTECTED]> wrote:

>127.0.0.1:allow,RELAYCLIENT=""
><box-ip-addr>:allow,RELAYCLIENT=""
>:allow

Did you build/rebuild /etc/tcp.smtp.cdb?

-Dave





> I am unsure of how to direct tcpserver's output so that is gets logged to
> syslog. Currently I just get qmail's messages sent to /var/log./maillog.
> What is the syntax to start qmail and have both qmail's and tcpservers
> STDOUT go to the syslog facility?

My old configuration:
tcpserver -g 101 -u 1001 -R -v -x/etc/tcp.smtp.cdb 0 smtp qmail-smtpd 2>&1 | splogger 
smtpd &

The new conf. uses daemontools:

/service/smtp/run:
exec env - PATH="/var/qmail/bin:$PATH" /usr/local/bin/tcpserver \
-g 101 -u 1001 -Hl dns.tegtmeyer.com -R -v -c 40 -x /etc/tcp.smtp.cdb \
0 smtp qmail-smtpd 2>&1

/service/smtp/log/run:
exec /usr/local/bin/multilog t s4194304 n20 /var/log/smtp

Regards, Frank





clifford thurber <[EMAIL PROTECTED]> wrote:

>127.0.0.1:allow,RELAYCLIENT=""
><box-ip-addr>:allow,RELAYCLIENT=""
>:allow

Try replacing qmail-smtpd with a script that logs its environment.

-Dave




On Wed, Mar 01, 2000 at 12:37:53PM -0500, clifford thurber wrote:
> Sorry I just didn't want to post the IPs etc. But it looks like this:
> 
> 127.0.0.1:allow,RELAYCLIENT=""
> <box-ip-addr>:allow,RELAYCLIENT=""
> :allow
> 
> I did try telnet localhost 25
> Mail From: [EMAIL PROTECTED]
> 250 ok
> Rcpt To: [EMAIL PROTECTED]
> 553 sorry, that domain isn't in my list of allowed rcpthosts (#5.7.1)
> 
> So it looks like it is not necessarily a Lyris problem but a configuration
> problem. Any help would be aprreciated.

What's the output of:

tcprulescheck /etc/tcp.smtp.cdb 127.0.0.1

and:

tcprulescheck /etc/tcp.smtp.cdb <box-ip-addr>

By the way, your mail server at 208.48.247.73 is listed in both ORBS and RSS.
If your relay problem is fixed, you might want to do something about that.

Chris





I've got a broken ISP here.  Please take the time to read this and help if
you can.  I have read the archives and found similar problems, but no
solutions.

fastforward 0.51
qmail 1.03
 patches: big-todo, netscape-progress, popbull, tarpit

I can't seem to get fastforward to work with virtual domain forwardings
that aren't defined as the catch-all @acmegraphic.com

I have th following setup:

  acmegraphic.com IN MX 0 mail.mhonline.net

  /var/qmail/control/rcpthosts:acmegraphic.com
  /var/qmail/control/virtualdomains:acmegraphic.com:alias

  /var/qmail/alias/.qmail-default:| fastforward -d /etc/aliases.cdb

  /etc/aliases:[EMAIL PROTECTED]:    helen
  /etc/aliases:[EMAIL PROTECTED]:     helen
  /etc/aliases:[EMAIL PROTECTED]:    helen
  /etc/aliases:[EMAIL PROTECTED]:  helen

Understand that if I put @acmegraphic.com: helen into the file, the
forward would work, but the others would not.

The local user "helen" is able to be sent mail directly, but not though
the forwarding.  Also, all aliases in the fastforward database that
address local alias (ie games@localhost) seem to work.  The log shows the
following when an email is sent to a virtual domain:

  qmail: 951927935.773054 starting delivery 1319:
    msg 174181 to local [EMAIL PROTECTED]
  qmail: 951927935.820853 delivery 1319: failure:
    Sorry,_no_mailbox_here_by_that_name._(#5.1.1)/

When I test the fastforward database, everything seems to check out:

  # env DEFAULT=helen HOST=acmegraphic.com fastforward -nd /etc/aliases.cdb
  from <original envelope sender>
  to <[EMAIL PROTECTED]>


I'm at a total loss at this point.  Help.

Dan                              <[EMAIL PROTECTED]>
Network Systems Engineer          518.943.6600x3350







Hello,

I am very new to qmail, I have used sendmail for over 2 years for my
mini-lan.  I am having a slight problem, mail is delivered fine, but when I
telnet to port 25 and vrfy user I get the message "252 send some mail, i'll
try my best", Now if I check the RFC this just means that my users arent
found locally but the mail server will try to send the email (I am a self
taught sysadmin so excuse me for any term used improperly)  

In /var/qmail/control/locals   I have my qualified domain
also in /var/qmail/control/rcpthosts  I have my qualified domain

My users have there mail going to /$user/Maildir/
          ** note the mail does go there fine

I have everything running from one computer so directories and user db are
all in the same computer

Since the mail works fine except for this Message 252 there is no big
problem execpt that I have a program to verify users and it checks the mail
server with vrfy user and right now my program wont work.  So how do I fix
it so qmail sees my users as local?

Thank You,
Shera - Sysadmin Technology Unlimited PR







Shera <[EMAIL PROTECTED]> wrote:

>I am very new to qmail, I have used sendmail for over 2 years for my
>mini-lan.  I am having a slight problem, mail is delivered fine, but when I
>telnet to port 25 and vrfy user I get the message "252 send some mail, i'll
>try my best", Now if I check the RFC this just means that my users arent
>found locally but the mail server will try to send the email (I am a self
>taught sysadmin so excuse me for any term used improperly)

qmail doesn't implement VRFY because (1) qmail's modular design makes
it impractical, and (2) VRFY makes it easy to validate e-mail
addresses and local accounts--information that crackers and spammers
like.

>Since the mail works fine except for this Message 252 there is no big
>problem execpt that I have a program to verify users and it checks the mail
>server with vrfy user and right now my program wont work.  So how do I fix
>it so qmail sees my users as local?

I suggest you use another method to verify your users, such as looking 
in the password file.

-Dave




On Wed, Mar 01, 2000 at 12:45:19PM -0400, Shera wrote:

> Since the mail works fine except for this Message 252 there is no big
> problem execpt that I have a program to verify users and it checks the mail
> server with vrfy user and right now my program wont work.  So how do I fix
> it so qmail sees my users as local?

qmail-smtpd does not verify local users. This is how it is designed. It
always return 252 to clients who attempt to use VRFY, whether the user
exists or not.

-- 
See complete headers for more info




qmail does not and will not support VRFY.  Find another way to do what you
want.  VRFY will NOT work.

On Wed, 1 Mar 2000, Shera wrote:

> Hello,
> 
> I am very new to qmail, I have used sendmail for over 2 years for my
> mini-lan.  I am having a slight problem, mail is delivered fine, but when I
> telnet to port 25 and vrfy user I get the message "252 send some mail, i'll
> try my best", Now if I check the RFC this just means that my users arent
> found locally but the mail server will try to send the email (I am a self
> taught sysadmin so excuse me for any term used improperly)  
> 
> In /var/qmail/control/locals   I have my qualified domain
> also in /var/qmail/control/rcpthosts  I have my qualified domain
> 
> My users have there mail going to /$user/Maildir/
>           ** note the mail does go there fine
> 
> I have everything running from one computer so directories and user db are
> all in the same computer
> 
> Since the mail works fine except for this Message 252 there is no big
> problem execpt that I have a program to verify users and it checks the mail
> server with vrfy user and right now my program wont work.  So how do I fix
> it so qmail sees my users as local?
> 
> Thank You,
> Shera - Sysadmin Technology Unlimited PR
> 
> 
> 
> 

---------------------------------
Timothy L. Mayo                         mailto:[EMAIL PROTECTED]
Senior Systems Administrator
localconnect(sm)
http://www.localconnect.net/

The National Business Network Inc.      http://www.nb.net/
One Monroeville Center, Suite 850
Monroeville, PA  15146
(412) 810-8888 Phone
(412) 810-8886 Fax





Timothy L. Mayo writes:
 > qmail does not and will not support VRFY.

Well, actually it *could*, under certain conditions.  If you don't
have any .qmail-.*default's, you could create a CDB containing all the
valid addresses, which qmail-smtpd could consult after sufficient
patching.  However, that still has the problem of giving away your
valid addresses to spammers.  And, .qmail-.*default files are *so*
useful.

-- 
-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.





First, let me say thanks to Dave for _Life with qmail_!

But (oh, here comes the good part :-):, though your
startup script works, I have a hard time imagining that
DJB who said "svscan is designed to run forever" intended
for it to have to be killed to stop qmail.  Of course,
I have less experience than others on this list, so maybe 
I'm just missing something.

It occurs to me that another problem is that one might want 
cycle qmail-smtpd and qmail-pop3d (in my case, at least)
independently, while leaving qmail proper up.  

Of course, I can write my own scripts, but I thought that,
since he writes such great code, DJB might have had some
particular scheme in mind for using daemontools and qmail
together.  

I know I could forgo the use of svscan, and just use supervise, 
but having the pipe automagically created between a service and 
its logger is just *too* convenient!

grier




Grier Ellis <[EMAIL PROTECTED]> wrote:

>First, let me say thanks to Dave for _Life with qmail_!

You're welcome, as is everyone else who's taken the time to thank me
for it. (I don't always have the time to "welcome" everyone who thanks 
me.)

>But (oh, here comes the good part :-):,

There's always a "but"... :-)

>though your
>startup script works, I have a hard time imagining that
>DJB who said "svscan is designed to run forever" intended
>for it to have to be killed to stop qmail.  Of course,
>I have less experience than others on this list, so maybe 
>I'm just missing something.

The LWQ "qmail" script is designed to be plugged into the run-level
startup/shutdown script directories (.../rcN.d), and well-behaved
scripts, when run with the "stop" parameter, are supposed to kill off
*all* of their processes.

>It occurs to me that another problem is that one might want 
>cycle qmail-smtpd and qmail-pop3d (in my case, at least)
>independently, while leaving qmail proper up.  
>
>Of course, I can write my own scripts, but I thought that,
>since he writes such great code, DJB might have had some
>particular scheme in mind for using daemontools and qmail
>together.

Just add additional tags to the "case" statement, e.g.:

  stop-smtpd)
    echo -n "Stopping qmail-smtpd: "
    svc -d /var/supervise/qmail/smtpd
    echo "done."
    ;;
  start-smtpd)
    echo -n "Starting qmail-smtpd: "
    svc -u /var/supervise/qmail/smtpd
    echo "done."
    ;;

-Dave




Hello,
 
I installed qmail as per LFQ, now some questions,
1.    I followed the TEST.deliver instructions for testing, where is the syslog that is refered to in step one?
2.    I can send mail out (step 5) but mail is not being delivered locally or being recieved from an outside source (it get bounced back).
3.    What should I look for in the logs and I'll post the errors that appear there.
 
Thanks
 
Take care,
 
Lee Trotter




"Lee Trotter" <[EMAIL PROTECTED]> wrote:

>I installed qmail as per LFQ, now some questions,
>1.    I followed the TEST.deliver instructions for testing, where is
>      the syslog that is refered to in step one? 

As section 2.9 of LWQ points out, you're not using syslog for
logging. The log scripts in LWQ point to /var/log/qmail.

>2.    I can send mail out (step 5) but mail is not being delivered
>      locally or being recieved from an outside source (it get
>      bounced back).

Check your logs, obviously, but (1) where is mail supposed to be
delivered (/var/qmail/control/defaultdelivery), and where/how are you
looking for it?

>3.    What should I look for in the logs and I'll post the errors
>      that appear there.

Look for errors. :-) If you don't see any, post a snippet here.

-Dave




"kailash oswal" <[EMAIL PROTECTED]> wrote:

>Presently I am facing a problem on my client machine.It is using fetchmail 
>to recieve mails from my server which is running qmail.Now when someboby 
>sends a mail to that client and in 'To' field if more than one address's are 
>there(e.g 3 ) he recieves 3 same mails.What cn be the problem..please 
>suggest

qmail does what you tell it to do: no more, no less. Send a message
like:

  From: me
  To: you, you, you

  blah blah blah

and "you" will recieve three copies. Three is number of copies "you"
will receive. No more, no less. Three shall be the number of the
copies, and the number of the copies shall be three. Four shalt thou
not recieve, and neither receive thou two, excepting that thou then
goest on to receive three...

-Dave




Nick,

It sounds like you are changing the envelope recipient between the time you
receive the message and when you transfer it on to your customer.  How are you
delivering their queued messages?  Make sure that the tool you are using sends
the message with it's original envelope recipient.  The contents of the To:
field in the message play no part in this.

Robert Sanderson <[EMAIL PROTECTED]>


Nick Starai wrote:

>
>         Hello, we are having problems getting Exchange to work with our
> qmail etrn setup.  Seems that exchange reads the headers, and not from the
> TO field.  The customer logs in, sends the trigger, and DOES recieve all
> the mail, but the exchange server is denying it because of unknown
> reciepient. (the user where ALL the mail for that domain goes to be picked
> up must bebeing used.)  On another note, if their server is dialed
> into us at the time mail is sent, it goes to the proper users inbox!
> Anyone have ideas or answers?? Thank you.





"Jose de Leon" <[EMAIL PROTECTED]> writes:

  I soved the problem I was having earlier with POP3 authentication going very
  slow (30 secs to more than minute) or sometimes timing out.
  
  Server reboot.  Not just a stop and start of the qmail-pop3 or tcpserver
  daemon.  Just a plain cold restart.

I haven't encountered this problem.  What type of system do you have?  Was
anything else slow?




"Mark E. Drummond" <[EMAIL PROTECTED]> writes:

  I am moving a client from a sendmail box to a qmail box. I am not terribly
  familiar with sendmail, but they host over 100 domains, and email for many/most
  of those is handled by, I presume, sendmail's virtual domain capability.
  Whoever set this thing up originally has a script which parses a pile of text
  files (one per domain) which consist of virtual address:real address pairs and
  does whatever sendmail needs done.

What is the format of the text files and the file with address:real pairs?
Can you paste a few relevant lines?

  
  Now, I know I could set up virtual domains in qmail to handle this but the
  amount of handraulic work involved is staggering. I'd like to hear suggestions
  and experiences from the list. Is there an easier way? Any automation I can do?

I'm sure there's a perl script out there, waiting to be written, just to
handle your case...




Pavel Kankovsky writes:
 > Damned omnipotent root. I hate unix.

Well, my feeling is that Unix is well designed.  It's just the
programs that surround it that are not.  First, I'd start with most
programs written at BSD, and throw them out as "a nice try" by some
undergraduates.  I've already thrown out the programs *I* wrote as an
undergraduate.  I see no particular reason to worship anything in the
BSD.  BTW, Linus feels the same way, which is why you have to fsync "."
if you want information to be written into ".".

For example, the standard printing system (lpd/lpr/lpq) runs as root.
Bad idea.  No reason for it.  Run it under its own userid.  That's how 
CUPS does it.

That's how everything should be done under Unix -- as its own userid.
Other operating systems have Access Control Lists, which let you split
up permissions in a fine-grained manner.  Unix uses userids, group
membership, and file ownership, groups, and permissions to achieve the
same result.  If you don't do this because there's "no reason for all
those users", as Weitse Venema told me, then when you get a security
breach in one part of your system, it spreads out over the whole
system.

-- 
-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.




Yep
-----Original Message-----
From: Uwe Ohse <[EMAIL PROTECTED]>
To: Jeff Russell, AIT <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED] <[EMAIL PROTECTED]>
Date: Tuesday, February 29, 2000 6:03 PM
Subject: Re: Forward Messages to a secondary Mail Server


>On Tue, Feb 29, 2000 at 05:27:04PM -0500, Jeff Russell, AIT wrote:
>
>> This is a question regarding forwarding mail to a secondary server.  We
need to forward email on a per user email basis.  Is there a way that when
qmail receives email, and does not find that specific address for the user,
to direct it to another specified email server.
>
>See FAQ 4.1:
>
>Answer: Put
>
>   | forward "$LOCAL"@bigbang.af.mil
>
>   into ~alias/.qmail-default.
>
>(assuming bigbing is the secondary server)
>
>Regards, Uwe
>





I think something is messed up here, qmail appears to start ok on a reboot but if I stop manually and restart it here is what i get
 
[root@tester control]# /etc/rc.d/init.d/qmail stop
Stopping qmail: svscankill: No such pid at
kill: No such pid /var/run/svscan.pid
 qmail logging.
[root@tester control]# /etc/rc.d/init.d/qmail start
Starting qmail: svscan.
[root@tester control]# supervise: fatal: unable to acquire qmail-send/supervise/lock: temporary failure
supervise: fatal: unable to acquire log/supervise/lock: temporary failure
supervise: fatal: unable to acquire qmail-smtpd/supervise/lock: temporary failure
supervise: fatal: unable to acquire log/supervise/lock: temporary failure
Everything after the start then repeats itself everyfew minutes.
 
Take care,
 
Lee Trotter




Most of the programs which run as root don't need to even with their current
design.  All it would take is a few 'chown's, a few configuration file edits,
and you're set.  Hence, many times this problem is a configuration error
instilled by software vendors and distributors.  Other programs must run as
root to keep their current design, which is indeed unacceptable for the most
part.

Personally, i believe that running files as their own separate user is not good
enough; if at all possible, services should be in their own chroot()'ed
environment.  There is no excuse for named to run either as root or
system-wide.  There is no excuse for database servers to run either as root or
system-wide.  Most of these programs which are commonly used by crackers to
fully compromise systems can be reasonably secured as-is, without even doing a
major rewrite (though a major rewrite would obviously be the best COA for the
program authors if they wish to provide real security).

Why are so many distributors so oblivious to this?  Hell, i've replaced about
70% of the total software on our solaris machines (including all setuid files,
and virtually all daemons), and modified most of the replacements.  Redhat
distributions aren't really any better (which is only one reason i refuse to
work with a redhat distribution of linux)....  

Sigh... it really is so hard to find good software these days....

ari

[EMAIL PROTECTED] said this stuff:

> Pavel Kankovsky writes:
>  > Damned omnipotent root. I hate unix.
> 
> Well, my feeling is that Unix is well designed.  It's just the
> programs that surround it that are not.  First, I'd start with most
> programs written at BSD, and throw them out as "a nice try" by some
> undergraduates.  I've already thrown out the programs *I* wrote as an
> undergraduate.  I see no particular reason to worship anything in the
> BSD.  BTW, Linus feels the same way, which is why you have to fsync "."
> if you want information to be written into ".".
> 
> For example, the standard printing system (lpd/lpr/lpq) runs as root.
> Bad idea.  No reason for it.  Run it under its own userid.  That's how 
> CUPS does it.
> 
> That's how everything should be done under Unix -- as its own userid.
> Other operating systems have Access Control Lists, which let you split
> up permissions in a fine-grained manner.  Unix uses userids, group
> membership, and file ownership, groups, and permissions to achieve the
> same result.  If you don't do this because there's "no reason for all
> those users", as Weitse Venema told me, then when you get a security
> breach in one part of your system, it spreads out over the whole
> system.
> 
> -- 
> -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.




On Wed, Mar 01, 2000 at 03:44:44PM -0500,
  ari <[EMAIL PROTECTED]> wrote:
> 
> Personally, i believe that running files as their own separate user is not good
> enough; if at all possible, services should be in their own chroot()'ed
> environment.  There is no excuse for named to run either as root or
> system-wide.  There is no excuse for database servers to run either as root or
> system-wide.  Most of these programs which are commonly used by crackers to
> fully compromise systems can be reasonably secured as-is, without even doing a
> major rewrite (though a major rewrite would obviously be the best COA for the
> program authors if they wish to provide real security).

Another problem is that even normal users can't run untrusted programs
without giving the programs access to their files and the network. This
is getting to be a real problem as game and media programs are starting to
do some very unfriendly things.




Quoting Mark E. Drummond ([EMAIL PROTECTED]):
> I am currently using rblsmtpd to block spammers on the RBL. I may add ORBS as
> well. Think I'll wait, gather some stats on how much is being blocked by RBL,
> and then compare with RBL+ORBS.

My anti-spam mantra is "RSS+RBL+DUL"

I hardly ever get spam these days.  Perhaps an average of one a week.
Am I just lucky?  I dunno for sure, but I opened a hotmail account
just for grins, never used it, and a month later already have 30+
spams.  Wierd.

Since djb's quotes are hip right now, I'll invoke the "Profile, don't
speculate" rule (I think that was the quote...? ).  I really wanted to
see just how things were going and this thread is my excuse.

These stats are for Feb 24, 9:15pm local until now.

RSS has blocked 2294 smtp connections.
DUL has blocked 306 smtp connections.
RBL has blocked 3767 smtp connections.

note: RBL rejections output 421 error code, DUL and RSS 553.  That
does seem strange, don't ask my why I did that now, because I don't
know I have a particular reason.  It would seem more logical to have
RSS hosts receive a deferral, but there are a lot more hosts on RSS,
which would result in quite a few deferrals.  I *think* that was
my reasoning...

qmailanalog stats for the time period, and the actual hosts rejected,
are at http://defiant.coinet.com/rss.html for those who are interested
in looking.

Our local cable company got listed in RSS a couple weeks ago,
resulting in only the second time, I know of that is, that legit email
may have been affected (been using RSS since before it was affiliated
with MAPS).  The bozos got relay-raped by a viagra spammer.

Aaron




Right now I have DUL at 553, and RBL and RSS at 421.  It's annoying seeing
the RSS hosts retrying all the time, so I was considering changing to 553 for
RSS as well.

IMHO DUL should always be 553, because those hosts should never be allowed to
send mail to you..  So there's really no sense in deferring it.

--Adam

On Wed, Mar 01, 2000 at 01:21:43PM -0800, Aaron L. Meehan wrote:
> Quoting Mark E. Drummond ([EMAIL PROTECTED]):
> > I am currently using rblsmtpd to block spammers on the RBL. I may add ORBS as
> > well. Think I'll wait, gather some stats on how much is being blocked by RBL,
> > and then compare with RBL+ORBS.
> 
> My anti-spam mantra is "RSS+RBL+DUL"
> 
> I hardly ever get spam these days.  Perhaps an average of one a week.
> Am I just lucky?  I dunno for sure, but I opened a hotmail account
> just for grins, never used it, and a month later already have 30+
> spams.  Wierd.
> 
> Since djb's quotes are hip right now, I'll invoke the "Profile, don't
> speculate" rule (I think that was the quote...? ).  I really wanted to
> see just how things were going and this thread is my excuse.
> 
> These stats are for Feb 24, 9:15pm local until now.
> 
> RSS has blocked 2294 smtp connections.
> DUL has blocked 306 smtp connections.
> RBL has blocked 3767 smtp connections.
> 
> note: RBL rejections output 421 error code, DUL and RSS 553.  That
> does seem strange, don't ask my why I did that now, because I don't
> know I have a particular reason.  It would seem more logical to have
> RSS hosts receive a deferral, but there are a lot more hosts on RSS,
> which would result in quite a few deferrals.  I *think* that was
> my reasoning...
> 
> qmailanalog stats for the time period, and the actual hosts rejected,
> are at http://defiant.coinet.com/rss.html for those who are interested
> in looking.
> 
> Our local cable company got listed in RSS a couple weeks ago,
> resulting in only the second time, I know of that is, that legit email
> may have been affected (been using RSS since before it was affiliated
> with MAPS).  The bozos got relay-raped by a viagra spammer.
> 
> Aaron
> 




On Tue, Feb 29, 2000 at 11:20:06PM +0100, Ruben van der Leij wrote:
> On Tue, Feb 29, 2000 at 01:35:26PM -0800, Jon Rust wrote:
> 
> > Yes, ORBS catches a ton of spam. It also labels a lot of email that 
> > I'd like to see, as spam.
> 
> But that wasn't what ORBS is about. ORBS stands for Open Relay Blocking
> System, and it does exactly that. It blocks open relays. Most current
          ^^^^^^^^^^^^^^^^^^^^^^^

Wrong. It does more. It also fights against people/companies who
think that blocking ORBS' relay tester is correct. That effectively
means that email sent from the PHP project's email server will
not be received by anyone who subscribes to ORBS, because the server
happens to be hosted at Above.net.

> mailservers won't relay unless you specifically misconfigure them.

I prefer RSS. At least RSS doesn't block so much legatimate email
while it is quite good at killing spam.

- Sascha




On Wed, 1 Mar 2000, Adam McKenna wrote:

> Right now I have DUL at 553, and RBL and RSS at 421.  It's annoying seeing
> the RSS hosts retrying all the time, so I was considering changing to 553 for
> RSS as well.
> 
> IMHO DUL should always be 553, because those hosts should never be allowed to
> send mail to you..  So there's really no sense in deferring it.

It's not that it's 553 specifically, but that it's a 5xx code.  All 5xx
errors are permanent errors, and all 4xx errors are temporary errors.


--
Sam





On Tue, Feb 29, 2000 at 06:54:38PM -0500, John R. Levine wrote:

> >But that wasn't what ORBS is about. ORBS stands for Open Relay Blocking
> >System, and it does exactly that. It blocks open relays. 

> This isn't the place to have the ORBS arguement yet again. 

Which wasn't what I was planning on. I'm aware of the political issues.

ORBS is one way to keep away misconfigured mailservers. 

ip route bozo.mail.server/32 Null0 is another.

For many of this has become bigger than toys. To give you an example: One of
our customers decided to start playing with procmail. He put an
unconditional forward to his mailaccount at his ISP. No dupe-checks. Between
his ISP and us (we host his website) you'll find mostly STM-4. The
bottleneck between their and our mailserver is a local 100Mbps fast ethernet
link. The first looping bounce after he exceeded his quota there didn't
register. The second neither. By the time we noticed the mailloop, there
where tens of messages bouncing back and forth.

If this was transit-traffic, instead of peered traffic, this little blunder
would have costed him a five digit dollar-amount in traffic alone. Something
like 100 Gigabyte in bounced mail isn't cheap, and with current bandwidths
you'll need no more than a single night to burn away a years income in
traffic.

Granted, my example has nothing to do with ORBS, DUL or RBL, but does show
the risks you'll find on the 'modern' internet. People can do you a lot of
damage in a relative short time. Likewise, a single open relay in your
network can cost you a lot of money, if a spammer found it in the
christmass-weekend.

That alone might be a reason to at least have a look at ORBS, and check if
your own network is listed somewhere. Wether or not you'll use ORBS to block
open relays is a decision you should not take lightly, but some might find
it an option, for whatever reason. 

The fact that one can crash certain versions of Lotus Notes with a
ORBS-relaytest tells you more about the ability of Notes to handle
legitimate SMTP-transactions than wether or not relayscanning is a Good
Thing(tm).

-- 

Ruben





I cannot get mail to compile on my system. K6 350Mhz 128megs ram 4gig HD
Installed make and patch.

Using instructions from the following url:

http://dcfonline.sfu.ca/ying/linux/qmail/chap1.html

Using SPIRO Linux 7.0

Any help would be greatly appreciated!!

Greg


I execute rpm -bb /usr/src/redhat/SPECS/qmail.spec

I get the following:

<<clip>>

echo CC=\'`head -1 conf-cc`\'; \
echo LD=\'`head -1 conf-ld`\' \
) > auto-ccld.sh
cat auto-ccld.sh make-load.sh > make-load
chmod 755 make-load
cat auto-ccld.sh find-systype.sh > find-systype
chmod 755 find-systype
/find-systype > systype
( cat warn-auto.sh; ./make-load "`cat systype`" ) > load
chmod 755 load
cat auto-ccld.sh make-compile.sh > make-compile
chmod 755 make-compile
( cat warn-auto.sh; ./make-compile "`cat systype`" ) > \
compile
chmod 755 compile
( ( ./compile tryvfork.c && ./load tryvfork ) >/dev/null \
2>&1 \
&& cat fork.h2 || cat fork.h1 ) > fork.h
rm -f tryvfork.o tryvfork
/compile qmail-local.c
qmail-local.c:1: sys/types.h: No such file or directory
qmail-local.c:2: sys/stat.h: No such file or directory
make: *** [qmail-local.o] Error 1
Bad exit status from /var/tmp/rpm-tmp.53651 (%build

<<clip>>




>From the error message it seems you don't have your kernel headers installed
- install them and you should be right

Stephen


-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Thursday, March 02, 2000 10:13 AM
To: [EMAIL PROTECTED]
Subject: Cannot get Qmail to compile

<slip>

qmail-local.c:1: sys/types.h: No such file or directory
qmail-local.c:2: sys/stat.h: No such file or directory
make: *** [qmail-local.o] Error 1
Bad exit status from /var/tmp/rpm-tmp.53651 (%build

<<clip>>


<slip>





   One of the things that I am considering is hosting a number of mailing
lists, but I'm not sure of the bandwidth impact.  Now I can assume that an
average message is X bytes, that an average list has Y subscribers, and
that it will have an average of Z messages/day, and run the totals, but
that really doesn't give me a good "real-world" estimate of the bandwidth
necessities, especially of peak bandwidth.

   So, if I were to decide that I wanted to allocate 512k of my bandwidth
for mailing list purposes, what sort of load could I realistically expect
to handle?

   Some preliminary testing of mine gave fairly mediocre numbers.  A
program of mine sent 190 messages as fast as it could, to various
domains/users.  The program finished in 14 seconds, and watching the logs,
it took 30 to 45 seconds to deliver all messages that it could to remote
servers.  If I were to take that as indicative, I would say that it could
deliver (60 * 200) = 12,000 messages/hour without problems.  That means
that if an average list had 300 subscribers, and I have 40 lists, then I'm
looking at ten messages/hour to each list.  Averaged out, that would be 240
messages/day (not too bad), but realistically I know that most all traffic
will be concentrated during a part of the day...

  Any info would be appreciated.

steve





I'm not in a position to give an informed opinion, but in terms of
bandwidth, one significant parameter is the average size of messages.
Most lists with "well-behaved" users tend to have an average msg size of
2-4k, but some others - especially those that allow binary attachments,
be it pictures, documentc, etc. - may go up to 20-30k or even more
depending on the 'tolerance' level. 

I'd be interested to know what numbers you come up with!

Steve Wolfe wrote:
> 
> [...deleted...]
> 
>    So, if I were to decide that I wanted to allocate 512k of my bandwidth
> for mailing list purposes, what sort of load could I realistically expect
> to handle?
> 
> [...deleted...]




I'd be interested to know the answer to this, also.

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, March 01, 2000 8:07 PM
> To: Steve Wolfe
> Cc: [EMAIL PROTECTED]
> Subject: Re: OT: Mailing list bandwidth
> 
> 
> I'm not in a position to give an informed opinion, but in terms of
> bandwidth, one significant parameter is the average size of messages.
> Most lists with "well-behaved" users tend to have an average msg size of
> 2-4k, but some others - especially those that allow binary attachments,
> be it pictures, documentc, etc. - may go up to 20-30k or even more
> depending on the 'tolerance' level. 
> 
> I'd be interested to know what numbers you come up with!
> 
> Steve Wolfe wrote:
> > 
> > [...deleted...]
> > 
> >    So, if I were to decide that I wanted to allocate 512k of my 
> bandwidth
> > for mailing list purposes, what sort of load could I 
> realistically expect
> > to handle?
> > 
> > [...deleted...]
> 




I am having trouble setting up Qmail and supervise etc on FreeBSD.

I have in the past had qmail running as a local MTA on freebsd and all
worked fine.

I know need other things like time SMTP relay for external customers setup
from POP3 logins etc.

I am having trouble were by when I run the supervise command as in the HOWto
on the qmail.org site it comes back with setuser: not found

I can't seem to locate setuser or its man pages.

Thanks.
David Uzzell





Hello,

First please forgive my ignorance of qmail's world. I'm not familiar in qmail (just 
heard it's good and get installed) but I need to setup a mail hub in a hurry. Any 
advise are welcome.

I would like to have the qmail configurated to FORWARD/DUPLICATE EVERY incoming 
message (no matter whoever the receiver) to two different host behind. (Both hosts are 
not linux machines and one host is for production while the another is for 
development/migration)



                       +----------+                  + ----------+
                       |          |                  |           |
--incoming message --> |  qmail   |-------+----------|   Host A  |
                       |          |       |          |           |
                       +----------+       |          +-----------+
                                          |
                                          |          +-----------+
                                          |          |           |
                                          +----------|   Host B  |
                                                     |           |
                                                     +-----------+







man qmail-remote.

but basically, if you create a stmproutes files in /qmail/control with the
relevent info, it will do what you have requested

something like :

production.domain.com:hosta.domain.com
development.domain.com:hostb.domain.com

of course, hosta/b have to have the appropriate A records in the DNS, or you
can use IP addresses.

this will only forward, not copy the emails....

Cheers
--Stephen

-----Original Message-----
From: Andy WONG [mailto:[EMAIL PROTECTED]]
Sent: Thursday, March 02, 2000 3:00 PM
To: [EMAIL PROTECTED]
Subject: Forward/Duplicate messages to hosts behind


Hello,

First please forgive my ignorance of qmail's world. I'm not familiar in
qmail (just heard it's good and get installed) but I need to setup a mail
hub in a hurry. Any advise are welcome.

I would like to have the qmail configurated to FORWARD/DUPLICATE EVERY
incoming message (no matter whoever the receiver) to two different host
behind. (Both hosts are not linux machines and one host is for production
while the another is for development/migration)



                       +----------+                  + ----------+
                       |          |                  |           |
--incoming message --> |  qmail   |-------+----------|   Host A  |
                       |          |       |          |           |
                       +----------+       |          +-----------+
                                          |
                                          |          +-----------+
                                          |          |           |
                                          +----------|   Host B  |
                                                     |           |
                                                     +-----------+







Hi,
 
    As a newcomer I did read the docs .... but....hope somebody could help me.
 
    Case (1)::
    1. I logged in as root.
    2. Run fetchmail and I can retrive my e-mails([EMAIL PROTECTED]) from the ISP
    3. The emails were put in the ~alias/pppdir/new
    4. What did I miss out in order to direct the email to [EMAIL PROTECTED]
        or to put the e-mail into /home/stein/Maildir/new ??
 
    Case 2::
    1. I logged in as stein.
    2. Run fetchmail and I can retrive my e-mails([EMAIL PROTECTED]) from the ISP
    3. I saw the fecthmail output saying that the email has gone to stein@localhost  
    4. But I saw nothing in /home/stein/Maildir/new
    5. Then I tried:
            (i) echo To: stein@localhost | /var/qmail/bin/qmail-inject
            (ii) echo To: [EMAIL PROTECTED] | /var/qmail/bin/qmail-inject
    6. Only the (ii) procduced an e-amil in /home/stein/Maildir/new
    7. Why localhost != durian.fidamy.com ?
 
    I want to setup the qmail to serve multiple e-mail account, 
   [EMAIL PROTECTED]    (local name abc)
    [EMAIL PROTECTED]    (local name def)
    [EMAIL PROTECTED]    (local name ghi)
    How do I ask fetchmail to do so?
 
Thanks,
Stein Ma
 




Im having several complaints from people using Netscape 4.7's email client.
Seems when they are trying to log into the server that Netscape is only
sending user@ not [EMAIL PROTECTED] So the user then cannot get their mail.

-- 
Erich Zigler                    ----                  System Administrator
 Last night I played a blank tape at full blast. The mime next door went
                                nuts.


Reply via email to