Recent posts

#31

Others / Cannot elevate from CentOS 7 t...

Last post by senthil - Sep 29, 2025, 08:18 AM

Symptoms

• Cannot elevate from CentOS 7 to Almalinux 8, the script fails with the following error:
  
  
  Code Select Expand
  Summary: {"STDERR": "Failed to create directory
/var/lib/leapp/el8userspace//sys/fs/selinux: Read only file system
Failed to create directory /var/lib/leapp/el8userspace//sys/fs/selinux: Read-only file
system\nNo matches found for the following disable plugin patterns: subscription-manager
Repository extras is listed more than once in the configuration
Repository extras-source is listed more than once in the configuration
Warning: Package marked by Leapp to upgrade not found in repositories metadata: gpg-
pubkey leapp-upgrade-el7toel8 leapp python2-leapp
Error: Error downloading packages:
psa-autoinstaller-3.64.0-0redhat.8.240819.0536.x86_64: Cannot download, all mirrors were
already tried without success\n", "STDOUT": "Last metadata expiration check: 0:00:18 ago
on Sat Oct 5 00:23:56 2024.
Package python3-inotify-0.9.6-2.centos.7+p18.0.63.0+t240715.1829.x86_64 is already
installed.
Dependencies resolved.

Cause

Missing DNF package.

Resolution

• Connect to the server using SSH.
• Install the package:
  
  Code Select Expand
  # yum install dnf-plugins-core

#32

Others / Is it required to purchase Ple...

Last post by senthil - Sep 29, 2025, 08:11 AM

Question

• Is it required to purchase a separate Plesk technical support subscription for each individual server?
• If I purchase a single support subscription, how many servers would I be able to request support for?

Answer
Plesk Support Subscription enables technical support option specifically for 1 license which is installed to server you wish to get support
for. It is necessary to purchase Plesk technical support subscription for each individual server.

NOTE: Plesk technical support subscription is required for customers who obtained their licenses from a reseller
and wish to contact Plesk technical support team directly.

#33

Others / Vulnerability CVE-2024-38998/C...

Last post by senthil - Sep 29, 2025, 08:02 AM

Situation

Is Plesk affected by CVE-2024-38998 or CVE-2024-38999?

Impact

Plesk is not affected by this vulnerability.

Exploiting these vulnerabilities is only possible if an attacker can inject a custom configuration with the __proto__
option into RequireJS via one of the following functions: config, s.contexts._.configure, or parse. In Plesk, user-
supplied input is not passed to these functions, preventing attackers from exploiting this vulnerability.

Call to Action

No actions are required.

[Question:]

#34

Webserver / The nginx web server fails to ...

Last post by Suhitha - Sep 29, 2025, 07:05 AM

Question: The nginx web server fails to start on a Plesk server: "nginx: [emerg] bind() to 203.0.113.2:443 failed (99: Cannot assign requested address)"


Applicable to:

• Plesk for Linux

Symptoms

• The Plesk panel data (including the web server configuration files) has been recently migrated to a new Plesk server

• The nginx web server fails to start and the following error is shown in the Plesk GUI:

New configuration files for the Apache web server were not created due to the errors in configuration templates: nginx: [emerg] bind() to 203.0.113.2:443 failed (99: Cannot assign requested address) nginx: configuration file /etc/nginx/nginx.conf test failed

• The IP address 203.0.113.2 has been used during the migration, however it is no longer present on the network interfaces of the Linux server

Cause

The file /etc/nginx/plesk.conf.d/server.conf still contains the outdated 203.0.113.2 IP address, which is mentioned in the error message.


Resolution

Force the recreation of the nginx configuration file by following these steps:

1.Connect to the server via SSH as the root user

2.Disable nginx by executing the following command:

# plesk sbin nginxmng -d

3.Enable nginx once again by executing the following command:

# plesk sbin nginxmng -e

After this has been done, the IP address that was causing the issue should not be present within the /etc/nginx/plesk.conf.d/server.conf file any longer.

[Question:]

#35

Webserver / How to configure custom error ...

Last post by Suhitha - Sep 29, 2025, 06:50 AM

Question: How to configure custom error pages for all domains on Plesk server?


Applicable to:

• Plesk Obsidian for Linux

Answer

1.Download linux_default_skeleton.zip from the How to change the default content for newly created domains in Plesk? article.

Unpack it and modify the content of error_docs directory in a desired way.

2.Compress it back into a .zip archive. Please note that .zip archive should contain all the files and directories in its root (cgi-bin, error_docs, httpdocs and so on).

3.Upload this modified archive in Tools & Settings > Virtual Host Template menu, by selecting a file and pressing Send file button.

4.Enable Custom error documents option for domains/subscriptions using one of the following ways:

• For multiple subscriptions on a specific service plan: Enable option Custom error documents option in Service Plans > Example-plan-name > Hosting Parameters and then synchronize plan with subscriptions.
• For all domains at once via CLI:

1.Connect to the server via SSH

2.Run the following command:

Code Select Expand

# for i in plesk bin domain --list; do plesk bin domain -u $i -err_docs true; done
Note: Some websites/applications might use their own custom error pages (by means of rewrite rules), so the changes might not be reflected on some domains.

[Question:]

#36

Webserver / How to manage PHP session temp...

Last post by Suhitha - Sep 29, 2025, 06:43 AM

Question: How to manage PHP session temporary files and how often they are cleaned up from the Plesk server?


Answer

Configuring PHP to clean up these session files automatically is done by editing session.gc_probability, session.gc_divisor, and session.gc_maxlifetime PHP variables:

                                  session.gc_probability

This setting defines the probability that the garbage collection (session cleanup) routine will be executed on each session start. It's expressed as a fraction, where the numerator is session.gc_probability, and the denominator is session.gc_divisor. The default values are 1 and 100, which means there is a 1% chance that the garbage collection will run when a session starts. To increase the probability, you can set it to a higher value. For example, to make it a 10% probability, you can set session.gc_probability to 10.

                                  session.gc_divisor

This setting represents the denominator for the probability calculation mentioned above. The default is 100, meaning that, by default, there's a 1% probability that garbage collection will run. If you set session.gc_probability to 10, you should leave session.gc_divisor as 100.

                                  session.gc_maxlifetime

This setting specifies the maximum lifetime of a session in seconds. Once a session exceeds this time limit, it becomes eligible for garbage collection and will be deleted.


These variables can be edited on different levels: per domain name, per service plan or on a server level. This is where the configuration directives should be put depending on the level:


Per Domain Name

Domains > click on a Domain > PHP > Additional directives


Per Service Plan

Service plans > click on a Service plan > PHP settings > Additional directives


On the Server Level:

Tools & Settings > PHP Settings > required PHP handler > php.ini > Find the variables in question and edit their values

[Question:]

#37

Webserver / Incoming requests from real vi...

Last post by Suhitha - Sep 29, 2025, 06:35 AM

Question: Incoming requests from real visitor IPs are logged as 127.0.0.1 when Apache is listening on localhost


Applicable to:

• Plesk for Linux

Symptoms

• Plesk Obsidian running on a Linux-based operating system
• Incoming requests from real visitor IPs are logged as 127.0.0.1 in the website access logs when Apache is listening on localhost (after update to 18.0.56)

Cause

This behavior is tied to a known bug that has already been resolved in more recent versions of Plesk Obsidian:

• #PPPM-14170 "Websites and web applications that analyze client IP addresses now work properly again when Apache is listening to localhost.

Fixed in:

     

• Plesk Obsidian 18.0.56 Update 2 27 October 2023 (Linux)

Resolution

1. Update your Plesk version to the latest available version by using the steps in this article:
How to update Plesk Obsidian to the latest build?

2. Log into your server via SSH

3. Fix the malfunctioning server configurations after they have been updated to Plesk Obsidian 18.0.56 Update 2 and higher by running the following command:

Warning: Your sites may become inaccessible after executing this command. If that happens, make sure to restart the nginx web server manually

Code Select Expand

# plesk bin apache --listen-on-localhost true

                    If you cannot or do not want to update Plesk, use this workaround

1. Log into your server via SSH

2. Execute the following command in order to stop Apache from listening on localhost:

Warning: Your sites may become inaccessible after executing this command. If that happens, make sure to restart the nginx web server manually

Code Select Expand

# plesk bin apache --listen-on-localhost false

[Question:]

#38

Webserver / Unable to start Apache on Ples...

Last post by Suhitha - Sep 29, 2025, 06:21 AM

Question: Unable to start Apache on Plesk server: Cannot load modules/mod_<name>.so


Applicable to:

• Plesk for Linux

Symptoms

Note: The following symtoms could be presented with different Apache module

• The following error show up while checking apache configuration syntax:

Code Select Expand

# httpd -t
httpd: Syntax error on line 56 of /etc/httpd/conf/httpd.conf: Syntax error on line 36 of /etc/httpd/conf.modules.d/00-base.conf: Cannot load modules/mod_ext_filter.so into server: /etc/httpd/modules/mod_ext_filter.so: cannot open shared object file: No such file or directory

• The mentioned apache module file is missing:

Code Select Expand

ls -l /etc/httpd/modules/mod_ext_filter.so
ls: cannot access /etc/httpd/modules/mod_ext_filter.so: No such file or directory

• Apache fails to start with the following error:

Code Select Expand

# systemctl start httpd
Job for httpd.service failed because the control process exited with error code. See "systemctl status httpd.service" and "journalctl -xe" for details.

• The following error show up in journalctl -xe command

Code Select Expand

-- Unit httpd.service has begun starting up.
httpd[29946]: httpd: Syntax error on line 56 of /etc/httpd/conf/httpd.conf: Syntax error on line 36 of /etc/httpd/conf.modules.d/00-base.conf: Cannot load modules/mod_ext_filter.so into server: /etc/httpd/modules/mod_ext_filter.so: cannot open shared object file: No such file or directory
systemd[1]: httpd.service: main process exited, code=exited, status=1/FAILURE
systemd[1]: Failed to start The Apache HTTP Server.

Cause

Apache module was not removed correctly and orphan reference to it exists in Apache configuration files.


Resolution

1.Connect to the server via SSH

2.Find in which Apache configuration file the module is being referenced, with command:

Code Select Expand

# grep -lR "mod_ext_filter.so" /etc/httpd/
/etc/httpd/conf.modules.d/00-base.conf

Note: Modify the command above accordingly with the corresponding missing module if is different than mod_ext_filter.so

3.Use vi or any text editor of your preference to edit the configuration file with the missing module

Code Select Expand

# vi /etc/httpd/conf.modules.d/00-base.conf
Note: In case the previous step command prompt a different path, use that path instead

4.Comment out the missing module. It should stay this way:

Code Select Expand

#LoadModule ext_filter_module modules/mod_ext_filter.so
5.Restart Apache:

Code Select Expand

# service httpd restart
6.Run the following command:

Code Select Expand

# plesk repair web
7.Verify that Apache syntax is correct:

Code Select Expand

# httpd -t
Syntax OK

[Question:]

#39

Webserver / How to set up nginx to log rea...

Last post by Suhitha - Sep 29, 2025, 05:56 AM

Question: How to set up nginx to log real client IP address when the Plesk server is behind an external proxy server


Applicable to:

• Plesk for Linux

Answer

For one domain (nginx reverse proxy is enabled)

1.Log in to Plesk.

2.Go to Domains > example.com > Hosting & DNS > Apache & nginx Settings.

3.Add the following lines to the additional nginx directives field:

Code Select Expand

set_real_ip_from 192.0.2.2/8;
real_ip_header X-Forwarded-For;
real_ip_recursive on;


• where 192.0.2.2/8 is an IP address of the external proxy server.

4.Apply the changes.


For one domain (nginx reverse proxy is disabled)

1.Log in to Plesk.

2.Go to Domains > example.com > Hosting & DNS > Apache & nginx Settings.

3.Add the following lines to the additional directives for HTTP/HTTPS fields:

Code Select Expand

RemoteIPHeader X-Forwarded-For
RemoteIPInternalProxy 192.0.2.2/8;

• where 192.0.2.2/8 is an IP address of the external proxy server.

4.Apply the changes.


For all domains (nginx reverse proxy is enabled)
 

1.Connect to the server via SSH.

2.Create the file with additional nginx configuration:

# touch /etc/nginx/conf.d/ext_proxy_server.conf

3.Open the file in a text editor. In this example, we are using vi editor:

# vi /etc/nginx/conf.d/ext_proxy_server.conf

4.Add the following lines to the additional nginx directives:

Code Select Expand

set_real_ip_from 192.0.2.2/8;
real_ip_header X-Forwarded-For;
real_ip_recursive on;

• where 192.0.2.2/8 is an IP address of the external proxy server.

5.Save the changes and close the file.

6.Restart nginx service:

# service nginx restart

Related to

Plesk for Linux

[Question:]

#40

Webserver / Login to Plesk from WHMCS redi...

Last post by Suhitha - Sep 29, 2025, 05:47 AM

Question: Login to Plesk from WHMCS redirects to Plesk login page instead of logging in.


Applicable to:

• Plesk for Linux

Symptoms

• SSO connection to Plesk though WHMCS fails. Login to Plesk from WHMCS redirects to Plesk login page back instead of logging in.
• WHMCS is behind Cloudflare,

Cause

User's IP address is changed during a Plesk session. This may happen if user's IP address is dynamic or the domain/hostname which is used as a URL for Plesk is behind Cloudflare.

By default, Plesk does not allow IP address changes during a single session.


Resolution

Allow IP address changes during a single session:

1.Log in to Plesk (use the server IP address if it is not possible to log in using the hostname).

2.Go to Tools & Settings > Session Idle Time and enable the option Allow IP address changes during a single session.

3.Click OK to apply the changes.


                              Command-line instructions

1.Connect to the Plesk server via SSH (Linux) / RDP (Windows Server).

2.Run the following command to enable IP address changes during a single session (on Windows Server, start a command prompt as an Administrator):

Code Select Expand

# plesk bin settings --set disable_check_session_ip=true