CentOS Virtual Hosting Tutorial

Uptime

I have been testing the setup with my own domain for roughly three weeks now. Everything seems to be running smoothly, I haven't had to reboot it yet.

Spam Scanning

Four days ago I added a second domain which increased the amount of messages coming in to ~1000 per day. CPU usage remains around 3% to 5% and disk I/O averages 50.

Memory

free -m
             total       used       free     shared    buffers     cached
Mem:           360        263         97          0          3         76
-/+ buffers/cache:        183        177
Swap:          255         54        201

your tutorial is really good but

I prefer to use YUM when it's possible, and now it should be possible…

I'm doing something wrong?

I've followed the wiki how to (http://www.linode.com/wiki/index.php/CentOS) - selecting the Apache option.

When I do http://[Linode MyIP]/ I get:

Connection Failure

The requested URL could not be retrieved

Would it help if I post my config files?

Ok, fixed that error, now php. I've set up php in httpd.conf as follow:

LoadModule php5_module modules/libphp5.so
AddHandler php5-script php
DirectoryIndex index.html index.htm index.shtml index.cgi index.php index.php3 index.pl
#AddType application/x-httpd-php .php .php4 .php3
AddType text/html       php
AddType application/x-httpd-php-source phps

The removal of AddType application/x-httpd-php is recommended by http://dan.drydog.com/apache2php.html

I've tried it with or without the x-httpd-php. I've restarted apache with each change to httpd.conf.

When I surf to http://blahblah/setup.php it wants to download the file saying it is a PHP file (is this a case issue?)

if I surf to http://blahblah/ it wants to download the page because it is an application/x-httpd-php.

What am I missing here?

What does the following command show on your Linode?

yum list installed '*php*'

yum list installed '*php*' Loaded plugins: fastestmirror, priorities Installed Packages php.i386 5.1.6-23.2.el5_3 installed php-cli.i386 5.1.6-23.2.el5_3 installed php-common.i386 5.1.6-23.2.el5_3 installed php-devel.i386 5.1.6-23.2.el5_3 installed php-gd.i386 5.1.6-23.2.el5_3 installed php-imap.i386 5.1.6-23.2.el5_3 installed php-ldap.i386 5.1.6-23.2.el5_3 installed php-mbstring.i386 5.1.6-23.2.el5_3 installed php-mcrypt.i386 5.1.6-15.el5.centos.1 installed php-mysql.i386 5.1.6-23.2.el5_3 installed php-odbc.i386 5.1.6-23.2.el5_3 installed php-pdo.i386 5.1.6-23.2.el5_3 installed php-pear.noarch 1:1.4.9-4.el5.1 installed php-pear-Auth-SASL.noarch 1.0.2-4.el5.centos installed php-pear-DB.noarch 1.7.13-1.el5.centos installed php-pear-Date.noarch 1.4.7-2.el5.centos installed php-pear-File.noarch 1.2.2-1.el5.centos installed php-pear-HTTP-Request.noarch 1.4.2-1.el5.centos installed php-pear-Log.noarch 1.9.13-1.el5.centos installed php-pear-MDB2.noarch 2.4.1-2.el5.centos installed php-pear-Mail.noarch 1.1.14-1.el5.centos installed php-pear-Mail-Mime.noarch 1.4.0-1.el5.centos installed php-pear-Net-SMTP.noarch 1.2.10-1.el5.centos installed php-pear-Net-Sieve.noarch 1.1.5-2.el5.centos installed php-pear-Net-Socket.noarch 1.0.8-1.el5.centos installed php-pear-Net-URL.noarch 1.0.15-1.el5.centos installed php-pecl-Fileinfo.i386 1.0.4-3.el5.centos installed php-pecl-memcache.i386 2.2.3-1.el5_2 installed php-sqlite2.i386 5.1.6-200705230938 installed php-xml.i386 5.1.6-23.2.el5_3 installed php-xmlrpc.i386 5.1.6-23.2.el5_3 installed php5-ioncube-loader.i386 3.1-07051711 installed phpmyadmin.noarch 2.11.9.3-1.fc7.rf installed psa-php5-configurator.noarch 1.5.1-cos5.build92090422.13 installed

Well, you certainly don't seem to be missing the PHP package. Maybe the module isn't being loaded for some reason? Does php5_module show up when you run the following?

/usr/sbin/httpd -M

This is what I get:

# /usr/sbin/httpd -M
[Fri Jun 26 08:26:38 2009] [warn] module php5_module is already loaded, skipping
...
php5_module (shared)

Hmm, afraid I can't offer much help then. When I installed Apache and PHP on CentOS, it "just worked." If you don't mind abandoning your current configuration, maybe try deleting it, removing the httpd and php* packages, and re-installing them?

Paste your httpd.conf

I've uploaded it to my other server:

http://www.paranoidandroid.co.za/files/httpd.conf

http://www.paranoidandroid.co.za/files/ … hosts.conf">http://www.paranoidandroid.co.za/files/httpd_vhosts.conf

Most of my changes are marked with [ALE MOD]

I checked your httpd.conf after sent you a pm.

I think there is a mime type problem.

But your httpd.conf is seems ok.

LoadModule php5_module modules/libphp5.so

AddHandler php5-script php

AddType text/html php

Php module is loaded too.

:?

Ok, just to update all and to ask a new question(s):

I've re-installed Centos (twice now, last time following the wiki article to the letter)

Now the first question:

I get the following in mailllog file

Jul 17 16:59:15 li73-31 postfix/postfix-script[27010]: warning: not owned by root: /etc/postfix/sql
Jul 17 16:59:15 li73-31 postfix/postfix-script[27011]: warning: not owned by root: /etc/postfix/sql/mysql_virtual_domains_maps.cf
Jul 17 16:59:15 li73-31 postfix/postfix-script[27012]: warning: not owned by root: /etc/postfix/sql/mysql_virtual_alias_maps.cf
Jul 17 16:59:15 li73-31 postfix/postfix-script[27013]: warning: not owned by root: /etc/postfix/sql/mysql_virtual_mailbox_maps.cf
Jul 17 16:59:15 li73-31 postfix/postfix-script[27027]: starting the Postfix mail system
Jul 17 16:59:15 li73-31 postfix/master[27028]: daemon started -- version 2.6.2, configuration /etc/postfix

Do I need to worry about it?

If so what should I do to fix it?

Thank you :)

[Fixed] Also…

Postfix is having problems…

Jul 17 20:06:39 li73-31 postfix/smtpd[16986]: fatal: open database /etc/aliases.db: No such file or directory

And I don't have newaliases on my installation (I've un-installed sendmail as per the wiki.) What I can figure out is that newaliases is part of sendmail.

Should I re-install sendmail? [Yes]

Will it break my current postfix setup? [Doesn't seem to break postfix]

Newaliases is part of Postfix's Sendmail compatibility interface. If you had both installed simultaneously, I can see how that might cause Postfix's versions of newaliases / sendmail / mailq to not be present. You almost certainly want them, as other applications tend to rely on their presence.

You might want to first try running system-switch-mail (if not present, just do yum install system-switch-mail) to choose Postfix as your system's MTA. That might be enough to get all the pieces in place. Otherwise, I'd suggest removing both Sendmail and Postfix and then re-installing Postfix only, followed by system-switch-mail.

Hi

Thank you for the response.

It seems to be working if I have sendmail installed but not running.

Hi All

I'm posting here because I use this setup. I'm having trouble setting up the mail.

My mail seems to be eaten up…

In the maillog file I get:

Aug  7 09:49:18 li73-31 postfix/smtpd[9962]: connect from ey-out-2122.google.com[74.125.78.26]      
Aug  7 09:49:18 li73-31 postfix/smtpd[9962]: B9E752727A: client=ey-out-2122.google.com[74.125.78.26]
Aug  7 09:49:18 li73-31 postfix/cleanup[9968]: B9E752727A: hold: header Received: from ey-out-2122.google.com (ey-out-2122.google.com [74.125.78.26])??by mail.paranoidandroid.co.za (Postfix) with ESMT
P id B9E752727A??for <andre@paranoidandroid.co.za>; Fri,  7 Aug 2009 09: from ey-out-2122.google.com
[74.125.78.26]; from= <tungsten15@gmail.com>to= <andre@paranoidandroid.co.za>proto=ESMTP helo= <ey-ou t-2122.google.com="">Aug  7 09:49:18 li73-31 postfix/cleanup[9968]: B9E752727A: message-id=<4A7BDC6B.5080403@googlemail.c
om>                                                                                                 
Aug  7 09:49:23 li73-31 MailScanner[9833]: New Batch: Scanning 1 messages, 2821 bytes               
Aug  7 09:49:23 li73-31 MailScanner[9833]: Virus and Content Scanning: Starting                     
Aug  7 09:49:23 li73-31 MailScanner[9833]: Requeue: B9E752727A.A4428 to 65A0F27283                  
Aug  7 09:49:23 li73-31 MailScanner[9833]: Uninfected: Delivered 1 messages                         
Aug  7 09:49:23 li73-31 postfix/qmgr[7425]: 65A0F27283: from=<tungsten15@gmail.com>, size=2115, nrcp
t=1 (queue active)                                                                                  
Aug  7 09:49:23 li73-31 postfix/error[9972]: 65A0F27283: to=<andre@paranoidandroid.co.za>, relay=non
e, delay=4.9, delays=4.8/0/0/0.05, dsn=4.3.0, status=deferred (mail transport unavailable)          
Aug  7 09:49:23 li73-31 MailScanner[9833]: Deleted 1 messages from processing-database</andre@paranoidandroid.co.za></tungsten15@gmail.com></ey-ou></andre@paranoidandroid.co.za></tungsten15@gmail.com></andre@paranoidandroid.co.za>

And also:

Aug  7 09:56:20 li73-31 postfix/error[9998]: 37FE62727E: to=<andre@paranoidandroid.co.za>, relay=non
e, delay=940, delays=940/0.03/0/0.01, dsn=4.3.0, status=deferred (mail transport unavailable) 

Aug  7 09:56:20 li73-31 postfix/qmgr[7425]: warning: connect to transport private/dovecot: No such file or directory</andre@paranoidandroid.co.za> 

I guess the MailScanner doesn't copy the email to where it should be, just delete it…?

Your mail is in the postfix local delivery queue. postfix is reporting an error:````
Aug 7 09:49:23 li73-31 postfix/error[9972]: 65A0F27283: to=andre@paranoidandroid.co.za, relay=none, delay=4.9, delays=4.8/0/0/0.05, dsn=4.3.0, status=deferred (mail transport unavailable)

because the specified local delivery transport is unavailable:

Aug 7 09:56:20 li73-31 postfix/qmgr[7425]: warning: connect to transport private/dovecot: No such file or directory
````

Thank you for the response.

The week was long and I'm a bit slow.

Is private/dovecot a directory or the bin file? I've created a directory in /var/spool/postfix/private/dovecot (just after I posted the previous post). Now I get a Connection refused error

Did you install Dovecot yet?

Yes this is my setting:

# 1.2.1: /usr/etc/dovecot.conf
# OS: Linux 2.6.18.8-x86_64-linode1 x86_64 CentOS release 5.3 (Final)
protocols: imap imaps pop3 pop3s
ssl_cert_file: /etc/pki/paranoidandroidCA/server.crt
ssl_key_file: /etc/pki/paranoidandroidCA/server.key
login_dir: /usr/var/run/dovecot/login
login_executable(default): /usr/libexec/dovecot/imap-login
login_executable(imap): /usr/libexec/dovecot/imap-login
login_executable(pop3): /usr/libexec/dovecot/pop3-login
mail_executable(default): /usr/libexec/dovecot/imap
mail_executable(imap): /usr/libexec/dovecot/imap
mail_executable(pop3): /usr/libexec/dovecot/pop3
mail_plugin_dir(default): /usr/lib/dovecot/imap
mail_plugin_dir(imap): /usr/lib/dovecot/imap
mail_plugin_dir(pop3): /usr/lib/dovecot/pop3
auth default:
  mechanisms: plain login cram-md5 ntlm
  passdb:
    driver: sql
    args: /etc/dovecot/dovecot_sql_passdb.conf
  userdb:
    driver: sql
    args: /etc/dovecot/dovecot_sql_userdb.conf
  socket:
    type: listen
    client:
      path: /var/spool/postfix/private/auth
      mode: 432
      user: postfix
      group: postfix
    master:
      path: /usr/var/run/dovecot/auth-master
      mode: 384
      user: vmail