Recently we have found that many of Plesk users are complaining about receiving Blank Email to the Plesk Admin address like below

As per Parallels this is the latest bug in Plesk which will be fixed in future releases. They have also provided a work around for the problem; as per them this issue can be temporary fixed by adjusting the permissions of mail handlers hooks. Execute following command from the Shell of your Server.

Recently we faced a problem where Plesk was not accessible over Internet i.e. using port 8443 (https) and 8880 (normal). The first thing we checked was the Firewall; it happens sometime that the problem might be with firewall which was blocking port 8443 and 8880 but there was no problem with Firewall. Then we tried to check the status of Plesk (PSA) Service and got following response

While checking for logs (Plesk Logs) on /usr/local/psa/admin/logs/error_log it was showing following Messages

Now we have used command “/usr/sbin/lsof -i | grep http” (without quotes) and then killed the process using the ID. After that we have removed the Plesk Lock

But starting Plesk service didn’t fixed the problem. Still we were getting “could not bind to port 80” Error.

Now we have opened Plesk Apache Configuration file which include Apache (Web Server) Configuration of Plesk Control Panel.

and found that the User has changed the Plesk Port and set it to listen on port 80 which is default port of HTTPD

So, we have made necessary changes httpsd.conf file i.e. disabled Port 80 and allowed 8443 and 8880 in Plesk Apache Configuration and restart PSA.

NOTE:- Here the problem was with Plesk Apache configuration but there might be some different problems too. I have posted this fix which will help to understand Plesk Error Log file and Plesk Apache Configuration files. If any of the USER face different problem and knows the FIX Let me know through Contact form I will be happy to post here.

After installation of Atmail Webmail Client, while restarting Apache it throws following Syntax Error:

First we will have to check where atmail is installed. We can do that by querying the RPM files using option “rpm -ql“

The above output shows that the files are created in “/etc/psa-webmail/” directory instead of “/etc/psa/webmail/“

So, copy the file to relevant directory as in the conf file of httpd service

After that we were able to start the httpd service successfully.

While changing Plesk admin password it throws following

Above error arises when the problem is when Plesk is unable to copy files in /var/backup/psa/tmp/ directory. As per http://kb.parallels.com/en/5201 the main cause of the problem is “Such error occurs in case if symbolic link is used as the path to the Plesk product directory. PHP does not follow symlinks for security reasons.”

When we checked the server we found a Symlink for tmp in /var/backup/psa/ directory

So We removed the symbolic link and remounted the tmp folder physically using following steps:

Now please try to reset the passwords.

**Note: This settings will save till next system reboot.

To make this setting permanent execute following command from the shell of the server which will modify the /etc/fstab file of the Server.

Introduction:

Whenever PHP runs as an Apache Module it get executes as “user/group” of the web server which is usually “nobody” or “apache”. SuExec is a mechanism supplied with Apache which allows executing CGI scripts as the user to which they belongs to, rather than Apache’s user. This improves security in possibilities where multiple mutually distrusting users can put CGI content on the server and these scripts are executed as the user that created them. If user “admin” uploaded a PHP OR CGI script, you would see it was “admin” running the script when looking at the running processes on your server? This also provides an additional layer of security where script permissions can’t be set to 777 (read/write/execute at user/group/world level).

Installation:

**Make sure yum is installed on the server as it will help to install the dependencies.

1) To get the latest version of mod_suphp RPM enable the Atomic Repository for Yum. This can be done using Following command

# wget -q -O - http://eitwebguru.com/scripts/atomic |sh 

2) After this mod_suPHP can be installed using command “yum install mod_php”

Configuration:

Once mod_suphp is installed, you will have to configure it with Apache.

1) First you will have to create a suphp.conf file in /etc/httpd/conf.d/ Directory and Paste the following code

LoadModule suphp_module modules/mod_suphp.so

php_admin_value engine off
suPHP_Engine On
AddHandler x-httpd-php .php .php3 .php4 .php5
suPHP_AddHandler x-httpd-php

2) Restart Apache Service using command “service httpd restart”

3) Now you will have to creae suphp configuration file. So, create a new file in /etc directory named “suphp.conf” (rename the existing file and create a new one) and add following lines in it:

[global]
logfile=/var/log/suphp.log
loglevel=info
webserver_user=apache
docroot=/var/www/vhosts
allow_file_group_writeable=false
allow_file_others_writeable=false
allow_directory_group_writeable=false
allow_directory_others_writeable=false
check_vhost_docroot=false
errors_to_browser=false
env_path=/bin:/usr/bin
umask=0022
min_uid=30
min_gid=30

[handlers]
x-httpd-php=php:/usr/bin/php-cgi
x-suphp-cgi=execute:!self

4) Restart Apache once more i.e. “service httpd restart”

In previous Article I have explain how to enable Firewall modules on Virtuozzo or OpenVZ based node. Today I am going to explain how to install and configure CSF Firewall on a VPS.

ConfigServer Security and Firewall (CSF) is a free product developed by “ConfigServer Services” and this Firewall is successfully tested on following Operating System

RedHat v7.3, v8.0, v9.0
RedHat Enterprise v3, v4, v5 (32/64 bit)
CentOS v3, v4, v5 (32/64 bit)
Fedora Core v1 to v12(32/64 bit)
*Ubuntu v6.06 LTS, v8.10, v9.10
*Debian v3.1, v4.0, v5.0
*Mandriva 2009
*Gentoo
*openSUSE v10, v11
*Slackware v12.2
* may require custom regex patterns for some functions

Installation of CSF

Before installation of CSF make sure you have Perl 5.8.8 and perl-libwww-perl (Perl Module) is installed on your Server. If they are not installed you can install it using YUM.

yum install perl perl-libwww-perl

Now copy and paste Following Commands After “#” to install CSF on a VPS

To Download CSF:- # wget http://www.configserver.com/free/csf.tgz
Extract zip:- # tar -xzvf csf.tgz
Go to CSF Directory:- # cd csf
Install CSF:- # ./install.sh

Configuration of CSF

To Configure CSF (Basic Configuration) open/Edit CSF Configuration File at /etc/csf/csf.conf file and make following changes

1) Disable Testing Mode

Change
TESTING = "1"
to
TESTING = "0"

2) Enable incoming and outgoing Ports.

Search for “TCP_IN” and add following ports for respective control panels

Cpanel = "20,21,22,25,26,53,80,110,143,443,465,993,995,2082,2083,2086,2087,2095,2096"
Plesk = "20,21,22,25,53,80,110,143,443,465,993,995,8443,8880"
DirectAdmin = "20,21,22,25,53,80,110,143,443,2222"

Search for “TCP_OUT” and add following Ports

Cpanel = "21,22,25,26,27,37,43,53,80,110,113,443,465,873,2089"
Plesk =  "20,21,22,25,53,37,43,80,113,443,465,873,5443"

Search “UDP_IN“

Cpanel = "20,21,53,953"
Plesk =  "37,53,873"
DirectAdmin = "21,53,873"

Search “UDP_OUT“

Cpanel = "20,21,53,113,123,873,953"
Plesk = "53,873"

**I have used Default port for CSF Configuration and if you have any customized OR additional Port, you can add it in CSF Configuration.

Save and Exit the file. Now you can restart the Firewall using command “csf -r“

PROBLEM:

While Login into Webmail it returns the error: “Login failed”

SOLUTION:

Make sure that it is possible to connect to localhost:143 from the command line using telnet. If you get the error as below:

# telnet localhost 143
localhost/143: Name or service not known

then check /etc/hosts file. It must have permissions -rw-r–r– and contain the record for localhost in the following format:

127.0.0.1       localhost localhost.localdomain

Above is from From Parallels Knowledgebase

If /var/log/psa-horde/psa-horde.log shows something like:

FAILED LOGIN xx.xx.xx.xx to localhost:143[imap/notls] as me@mydomain.co.uk
FAILED LOGIN xx.xx.xx.xx to localhost:143[imap/notls] as me@mydomain.co.uk

I have searched regarding this but unable to find any fix so, I just re-installed the IMAP Service on the server. Following are the steps using which I have re-installed IMAP Server:

1) You can find the RPMS used by Plesk in /root/psa/PSA_version directory OR you can download the Tar (Not Autoinstaller) containing RPMS from Parallels website http://www.parallels.com/en/download/ (need registration it’s free)

2) go to the RPM directory which contains courier Imap RPM’s
#cd [path to the PSA RPM Directory]/dist-rpm-CentOS-4.2-i386/base/

3) Re-install courier-imap-3.0.8-cos4.build81070322.16.i586.rpm and psa-courier-imap-add-8.1.1-cos4.build81070322.16.i586.rpm which will automatically restart the services too

[root@vps base]# pwd
/root/plesk/dist-rpm-CentOS-4.2-i386/base
[root@vps base]# rpm -Uvh courier-imap-3.0.8-cos4.build81070322.16.i586.rpm psa-courier-imap-add-8.1.1-cos4.build81070322.16.i586.rpm –force
Preparing… ########################################### [100%]
Reloading configuration: [ OK ]
1:courier-imap ########################################### [ 50%]
Stopping Courier-IMAP server:
Stopping imap [ OK ]
Stopping imap-ssl [ OK ]
Stopping pop3 [ OK ]
Stopping pop3-ssl [ OK ]

Starting Courier-IMAP server:
Starting imapd [ OK ]
Starting imap-ssl [ OK ]
Starting pop3 [ OK ]
Starting pop3-ssl [ OK ]

2:psa-courier-imap-add ########################################### [100%]
Stopping Courier-IMAP server:
Stopping imap [ OK ]
Stopping imap-ssl [ OK ]
Stopping pop3 [ OK ]
Stopping pop3-ssl [ OK ]

Starting Courier-IMAP server:
Starting imapd [ OK ]
Starting imap-ssl [ OK ]
Starting pop3 [ OK ]
Starting pop3-ssl [ OK ]

[root@vps base]# netstat -nap | grep :143
tcp 0 0 0.0.0.0:143 0.0.0.0:* LISTEN 1657/couriertcpd
[root@vps base]# telnet localhost 143
Trying 127.0.0.1…
Connected to localhost.
Escape character is ‘^]’.

After this I was able to login into Webmail without any problem. I hope this will help other to fix the issue too

Recently while changing the name of a domain from Plesk (Linux) it thrown following Error:

No such parameter 'handler-type' for handler '10-spam-MVquwk'
Unable to get type of handler: 10-spam-MVquwk

No such parameter 'handler-type' for handler '10-spam-LztqGr'
Unable to get type of handler: 10-spam-LztqGr

The problem is due to orphaned handlers in /usr/local/psa/handlers Directory. Whenever a domain name has been changed whole mail handler system is checked by Plesk. We have created handler directory copy in /usr/local/psa/handlers.copy and removed problem handlers manually. Then executed mail reconfiguration utility:

# /usr/local/psa/admin/sbin/mchk --with-spam
==> Checking for: mail_spam_restore... ok
==> Checking for: mailsrv_conf_init... ok
==> Checking for: mail_mailbox_restore... ok
==> Checking for: mailsrv_entities_dump... ok
==> Checking for: mail_auth_dump... ok
==> Checking for: mailman_lists_dump... ok
==> Checking for: mail_responder_restore... ok
==> Checking for: mail_drweb_restore... ok
==> Checking for: mail_kav_restore... not exsists
==> Checking for: mail_spf_restore... ok
==> Checking for: mail_dk_restore... ok

After this we were able to rename the domain successfully:

# /usr/local/psa/bin/domain --update supportest123.com -new_name 123testsupport.com
SUCCESS: Update of domain 'supportest123.com' complete.

If the any problem noted due to the removal of Email Handlers, changes can be reverted using the backup directory which has been created before clearing handlers.

Qmail (Default Mail Server for Plesk) on Plesk 9.0 sometime start consuming high CPU and overloading server. This sometime happens due to lot of failed message to ‘postmaster’ which overload Qmail. Following is the header of the failed message:

In order to fix the problem, please execute following command from the Shell of the Server (root):

This will install few micro-updates for Plesk which fix postmaster issues. After that use /usr/local/psa/admin/sbin/mailqueuemng utility to delete messages from qmail queue.

Now the server load will be under control and there will be no failed messages in queue.

Atmail webmail system offers a webmail client, email server platform and mailserver appliance for Linux, Unix and Windows. Plesk 9.x supports Atmail and while sending emails using Atmail every email shows following message in the footer of the email sent.

Msg sent via @Mail - http://atmail.com/

This message is quite irritating and look un-proffestional. This message can be removed by editing AtMail Configuration file which can be located at following location

Plesk Windows: C:\Inetpub\vhosts\webmail\atmail\libs\Atmail\config.php for Win
Plesk Linux: /var/www/atmail/libs/Atmail/Config.php
Normal Location: [Atmail installtion location]/libs/Atmail/Config.php

Search for ‘footer_msg’ option and change the message you want to set OR keep it blank

FROM:
'footer_msg' => 'Msg sent via @Mail - http://atmail.com/',

TO:
'footer_msg' => 'YOUR COMPANY NAME OR ANY TEXT',
OR
'footer_msg' => '',

Restart the webserver.