Hello All,

I'm trying to get a qmail install up and running (first time), but came accross this problem...

By the way, I'm using this HOW_TO: http://lifewithqmail.org/lwq.html

I'm supposed to verify that the "qmailctl stop" script runs, but it doesn't kill the processes:

# netstat -a | grep smtp
tcp 0 0 localhost:smtp *:* LISTEN
unix 2 [ ACC ] STREAM LISTENING 2808 private/smtp
unix 2 [ ACC ] STREAM LISTENING 2844 private/bsmtp

# qmailctl stop
Stopping qmail...
qmail-smtpd
qmail-send

# netstat -a | grep smtp
tcp 0 0 localhost:smtp *:* LISTEN
unix 2 [ ACC ] STREAM LISTENING 2808 private/smtp
unix 2 [ ACC ] STREAM LISTENING 2844 private/bsmtp

So I went to the script itself:

<----BEGIN----->
.....

PATH=/var/qmail/bin:/bin:/usr/bin:/usr/local/bin:/usr/local/sbin
export PATH

QMAILDUID=`id -u qmaild`
NOFILESGID=`id -g qmaild`

case "$1" in
start)
.............
fi
;;
stop)
echo "Stopping qmail..."
echo " qmail-smtpd"
svc -d /service/qmail-smtpd /service/qmail-smtpd/log
echo " qmail-send"
svc -d /service/qmail-send /service/qmail-send/log
if [ -f /var/lock/subsys/qmail ]; then
rm /var/lock/subsys/qmail
fi
;;

<----END----->

I checked both /service/..../.../..../ locations and they are present.....

So, what is wrong with the script?? Why does't it stop qmail?

Are you sure you killed any running sendmail/postfix processes and their strartups?

What does 'qmailctl stat' return?

# ps -ef | grep sendmail
root 17981 17561 0 18:44 pts/5 00:00:00 grep sendmail
# ps -ef | grep postfix
root 1535 1 0 10:02 ? 00:00:00 /usr/lib/postfix/master
postfix 1553 1535 0 10:02 ? 00:00:00 qmgr -l -t fifo -u
postfix 16646 1535 0 18:22 ? 00:00:00 pickup -l -t fifo -u

How do I kill these gracefully?

# qmailctl stat
/service/qmail-send: down 23782 seconds, normally up
/service/qmail-send/log: down 23782 seconds, normally up
/service/qmail-smtpd: down 23781 seconds, normally up
/service/qmail-smtpd/log: down 23782 seconds, normally up
messages in queue: 0
messages in queue but not yet preprocessed: 0

It looks like you can successfully stop qmail, so your install appears to be working.

You will need to find out where your system is starting postfix and sendmail. Your first stop should be in /etc/inetd.conf and comment out any smtp references there with a # at the start of the line. Then you will need to re-read the inetd.conf - just do a 'kill -HUP <PID of inetd>' to reload it.

Check that there are no smtp lines (other than for qmail) in your /etc/inittab. If there are, comment them out and 'kill -HUP <PID of inittab>' (should be a PID of 1)

Since you're using SuSE you can use YaST to either remove sendmail and postfix or stop them from loading up at boot. Note that if you made the sendmail compatibily changes as outlined in the lifewithqmail docs that you may need to repeat the steps to create the symbolic links for sendmail compatibility.

# grep "smtp" *.*
inetd.conf:# smtp stream tcp nowait root /usr/sbin/sendmail sendmail -L sendmail -Am -bs

I couldn't find any enabled entry for smtp in the /etc dir......???? Strange huh?

Any other thoughts where it could be starting from?

Let me know if I am in the right track....

I tried this:

# find / -name "postfix"
/etc/sysconfig/postfix
/etc/init.d/postfix <----- went here
/etc/permissions.d/postfix
/etc/postfix
/lib/scpm/resource_types/service/get_deps/postfix
/var/adm/SuSEconfig/md5/etc/postfix
/var/spool/postfix
/usr/lib/postfix
/usr/sbin/postfix
/usr/share/doc/packages/postfix

This file /etc/init.d/postfix has all the initialization info.

Then I found this:

/etc/init.d/rc3.d/S14postfix

# ls -la S14postfix
lrwxrwxrwx 1 root root 10 Mar 13 17:03 S14postfix -> ../postfix

So as far as I understand...this is the one that's starting it....right?

What is the standard practice? Rename, edit....

So what should I do now?

Hmmmm, I now tried this:


<---------------BEGIN--------------------->

# find / -name "*postfix*"
/etc/sysconfig/postfix
/etc/init.d/rc3.d/K08postfix
/etc/init.d/rc3.d/S14postfix <---- 1
/etc/init.d/rc5.d/K08postfix
/etc/init.d/rc5.d/S14postfix <----- 2....???
/etc/init.d/postfix
/etc/permissions.d/postfix
/etc/postfix
/etc/postfix/postfix-script
/etc/postfix/postfix-files
/lib/scpm/resource_types/service/get_deps/postfix
/var/adm/fillup-templates/sysconfig.mail-postfix
/var/adm/fillup-templates/rc.config.postfix.del
/var/adm/fillup-templates/sysconfig.postfix
/var/adm/backup/postfix-20030408-0.tar.gz
/var/adm/SuSEconfig/md5/etc/postfix
/var/lib/YaST2/you/i386/update/8.1/patches/postfix-9660
/var/lib/YaST2/you-installed/postfix-9660
/var/spool/postfix
/usr/lib/postfix
/usr/sbin/postfix
/usr/share/doc/sdb/en/html/kngu_postfix_myorigin.html
/usr/share/doc/packages/postfix
/usr/share/doc/packages/postfix/html/postfix.1.html
/usr/share/doc/packages/postfix/pfixtls/contributed/make-postfix-cert.sh
/usr/share/man/man1/postfix.1.gz
/usr/share/susehelp/meta/Manuals/Productivity/postfix.desktop
/usr/share/susehelp/meta/Manuals/Productivity/amavis-postfix.desktop
/usr/share/YaST2/scrconf/cfg_postfix.scr
/sbin/rcpostfix
/sbin/conf.d/SuSEconfig.postfix


<---------------END--------------------->


Why are there 2 start scripts? One at level 3 and one at 5?

Are there more scripts that I should be hunting for? How can I know for sure that I have cancelled out all the scripts?

mcleodnine: YaST2 rocks!!!!!

I went with YaST2 to the runlevel editor and
found the entry for postfix.

Then Set/reset: disable
service;

Then Start/Stop/refresh stop service...done.
It saved the config....and that was it!


# netstat -a | grep smtp
#

Clean!

TESTING.....

I have tried this in two ways:

#1 echo to: this_local_user | /var/qmail/bin/qmail-inject

#2 echo to: <xuserx>@yahoo.com | /var/qmail/bin/qmail-inject


And then I check....

# qmailctl stat
/service/qmail-send: up (pid 13441) 110 seconds
/service/qmail-send/log: up (pid 13443) 110 seconds
/service/qmail-smtpd: up (pid 13446) 110 seconds
/service/qmail-smtpd/log: up (pid 13451) 110 seconds
messages in queue: 2
messages in queue but not yet preprocessed: 0


But they are stuck in that queue and I'm not getting the email.....?!

Then we need to look in the logs. Try 'tail -100f /var/log/qmail/current | tai64nlocal' and see if you can get any hints as to why your mail is stuck.

# pwd
/var/log/qmail


# ls -la
total 17
drwxr-xr-x 3 qmaill root 144 May 22 12:01 .
drwxr-xr-x 7 root root 816 May 24 10:04 ..
-rw-r--r-- 1 qmaill nofiles 14286 May 24 10:04 current
-rw------- 1 qmaill nofiles 0 May 22 12:01 lock
drwxr-xr-x 2 qmaill root 264 May 23 10:07 smtpd
-rw-r--r-- 1 qmaill nofiles 0 May 22 12:01 state

I believe the logs are not "enabled"....is this the case? How would I enable them?

[ looks like your logging is running fine. Did you try 'tail -100 /var/log/qmail/current' (the tai64nlocal just pretty-prints the datestamp format used by the ucspi-tcp logging)
looks like you have about 14k of logfiles you might want to invstigate.