Exim love apache

S

sspt

Verified User
Joined
Oct 27, 2005
Messages
63
[root@server customapache]# service httpd stop
Stopping httpd: [FAILED]
[root@server customapache]# service httpd start
Starting httpd: (98)Address already in use: make_sock: could not bind to address 0.0.0.0:443
no listening sockets available, shutting down
Unable to open logs
[FAILED]
[root@server customapache]# netstat -lnp | grep 443
tcp 0 0 0.0.0.0:443 0.0.0.0:* LISTEN 15658/exim
[root@server customapache]# service exim stop
Shutting down exim:
[root@server customapache]# service httpd start
Starting httpd: (98)Address already in use: make_sock: could not bind to address 0.0.0.0:443
no listening sockets available, shutting down
Unable to open logs
[FAILED]
[root@server customapache]# netstat -lnp | grep 443
tcp 0 0 0.0.0.0:443 0.0.0.0:* LISTEN 15658/exim
[root@server customapache]# kill -9 15658
[root@server customapache]# netstat -lnp | grep 443
tcp 0 0 0.0.0.0:443 0.0.0.0:* LISTEN 15659/exim
[root@server customapache]# kill -9 15659
[root@server customapache]# netstat -lnp | grep 443
[root@server customapache]# service httpd start
Starting httpd: [ OK ]
[root@server customapache]# service exim start
Starting exim:

Why exim start using port 443 after an apache crash?
 
maybe apache was used to run exim or something like that?
like 5 seconds after "service httpd stop" i consider a good idea to do "killall httpd"
and then start
 
sspt,

You should probably grep as follows:
Code: 
$ grep 443 /etc/exim.conf
just to make sure your exim.conf file isn't incorrectly attempting to start exim on port 443.

Jeff
 
I didn't find nothing in exim conf about ports 80 or 443, i have no idea why it happens.
 
This exact error happened to me today when httpd was shut down for log processing and won't restart automatically until manual intervention. I'm running Apache 2.0.55 and Exim 4.60. Anyone with ideas to why it happened?
 
same problem here, happens randomly.
Looked @ google and there was a discussion about this issue on the exim-devel mailing list but no-one knew what it was.

My guess is that exim is being ran through the apache process (exploit? crash?) and because of that it uses the apache ports.
 
Exim run through any other system would still only use exim ports.

Is this happening only with Apache 2.x? Or also with 1.x?

Jeff
 
Happens here with Apache 2.x.

Also I couldn't find anything odd in the logs but I might have missed something.
 
I don't personally support Apache version 2, so I can't help you with anything specific.

I do remember reading somewhere about exim TLS which might be causing a problem.

I can't help further :( .

Jeff
 
*kick*

Got this problem last night on Fedora 3 + Apache 2.0.58 and Exim 4.61.

It wasn't the first time I had this error.

Very very ugly bug it is.

I Wonder who is the troublemaker?

Code: 
Jun 15 00:18:02 www httpd: httpd shutdown succeeded
Jun 15 00:18:04 www httpd: (98)Address already in use: make_sock: could not bind to address 0.0.0.0:443
Jun 15 00:18:04 www httpd: no listening sockets available, shutting down
Jun 15 00:18:04 www httpd: Unable to open logs
Jun 15 00:18:04 www httpd: httpd startup failed
Jun 15 00:18:04 www httpd: httpd shutdown failed

I need to kill Exim and then start Apache to get everything back to normal. After that starting Exim again.
 
Last edited:
well this happend to me on apache 1.X series

so its not apache 2.X

I know this an old topic.. just bringing it back to life since I had a 4 hour downtime ( while I was asleep) and come to find out the httpd was down.
 
I have this problem too and I've reported DirectAdmin staff about it. First answer was "reinstall exim", hah... It's not very great offer for sysadmin, who is administrating 1000+ servers. It's not a problem in one server software :) Then I did strace at ~9 servers and sent results to DirectAdmin support, waiting for answer now (still for ~24 hours). Also I sent all topics URL with this problem in this forum :)
 
Well... Here I thought it was just me going mental.

I too can confirm this on 2 CentOS 4.x boxes, though I must admit that I haven't seen it happen for some time now. *shrug*
 
The only thing that comes to mind might be the openssl version (this is just a guess). Our installers have pre-assembled service packages with various rpms for a specific OS. Newer OS's in the same family will often still use the same services packages, so there may be different/newer openssl and openssl-devel rpms avaiable for your system.

Try doing a yum upgrade for openssl and openssl-devel, do a recompile of exim and apache, and see if that makes any difference.

I'm not an exim programmer myself, so I'd would have a hard time at trying to figure out what's going on if it is code related (I'm guessing it's not)

John
 
hi, i've the same problem...

ive just tryed to reinstall exim from /usr/local/directadmi/scripts/exim.sh

i get this:

>./exim.sh
No matching processes were found
pkg_add: can't stat package file 'da_exim-4.62-1.tgz'
Substitution replacement not terminated at -e line 1.
`wrapper' is up to date.
Testing for perl (/usr/bin/perl)...
Configuring scripts...
./install.sh -m 751 -O 1006 -g 1 . /etc/virtual/majordomo
./install.sh -m 755 -O 1006 -g 1 . /etc/virtual/majordomo/bin
Copying tools to /etc/virtual/majordomo/bin
Copying Majordomo files to /etc/virtual/majordomo
Copying archiving and other tools to /etc/virtual/majordomo/Tools
./install.sh -m 755 -O 1006 -g 1 . /etc/virtual/majordomo/Tools
Using installed majordomo.cf
Installing manual pages in /etc/virtual/majordomo/man

To finish the installation, 'su' to root and type:

make install-wrapper

If not installing the wrapper, type

cd /etc/virtual/majordomo; ./wrapper config-test

(no 'su' necessary) to verify the installation.
./install.sh -o root -g 1 -m 4755 wrapper /etc/virtual/majordomo/wrapper

To verify that all the permissions and etc are correct,
run the command

cd /etc/virtual/majordomo; ./wrapper config-test
cp: /var/www/html/squirrelmail-1.4.9a/data and /var/www/html/squirrelmail/data are identical (not copied).


This is Ok? or there a problem while reinstlalin exim?

Im Using FreeBSD 5 with Apache 1.3.37 Php 4.4.4, dunno what exim version

Thank you.
 
I am experiencing the same on 9 of my Debian Sarge machine's.

They all run the latest version of Exim and Apache 2.

I think Apache segfaults when it has a pipe open to Exim (/usr/sbin/sendmail) wich leaves Exim on port 80.

I made a nasty script that checks for Exim on port 80:

Code: 
#!/bin/sh

NUM=`netstat -nap --inet|grep ":80"|grep LISTEN|grep exim|wc -l`

if test ! $NUM = "0"; then
        killall -9 exim
        echo "exim on port 80 on `hostname`"|mail -s "Exim on port 80 on `hostname`" [email protected]
        sleep 5
        /etc/init.d/exim start
        /etc/init.d/httpd start
fi

It is a little workaround to prevent myself from getting out of bed in the middle of the night for only restarting Exim.
 
You must log in or register to reply here.
Back
Top