Recent posts

[SEL]

#61

Others / Fail2Ban unable to block IP Ad...

Last post by senthil - Sep 27, 2025, 07:21 AM

Applicable to:

• Plesk for Linux

Symptoms

• Users encounter an error when attempting to block an IP address using Fail2Ban in Plesk. The error message states:
  ERROR:__main__Command ['set', 'plesk-one-week-ban', 'banip', '203.0.112.1'] failed with error
  UnknownJailException('plesk-one-week-ban').
  
• Reloading fail2ban gives a permission error.
  
  Code Select Expand
  # fail2ban-client reload
2024-12-05 16:11:12,990 fail2ban [3005163]: ERROR NOK: (13, 'Permission denied')
[Errno 13] Permission denied: '/var/www/vhosts/system/example.com/logs/error_log'
• SELinux context of file seen in error above is httpd_sys_content instead of the correct one, httpd_log_t:
  
  Code Select Expand
  # ls -Z /var/www/vhosts/system/example.com/logs/error_log
system_u:object_r:httpd_sys_content_t:s0
/var/www/vhosts/system/example.com/logs/error_log
• Audit log file /var/log/audit/audit.log shows a denied request regarding fail2ban-server:
  
  Code Select Expand
  # egrep "fail2ban-server" /var/log/audit/audit.log | grep denied
type=AVC msg=audit(1733411472.982:1927101): avc: denied { read } for pid=2373127
comm="fail2ban-server" name="error_log" dev="sda1" ino=230808319
scontext=system_u:system_r:fail2ban_t:s0
tcontext=system_u:object_r:httpd_sys_content_t:s0 tclass=file permissive=0

Cause
The issue is caused by incorrect SELinux context for the log files that Fail2Ban needs to access.

Resolution

• Connect to the server via SSH
• Remove currently installed psa-selinux package via Plesk installer:
  
  Code Select Expand
  # plesk installer --select-product-id plesk --select-release-current --remove-component
selinux
• Reinstall currently installed psa-selinux package via Plesk installer:
  
  Code Select Expand
  # plesk installer --select-product-id plesk --select-release-current --install-component
selinux
• Check again context of affected file, it should look like this:
  
  Code Select Expand
  # ls -Z /var/www/vhosts/system/example.com/logs/error_log
system_u:object_r:httpd_log_t:s0 /var/www/vhosts/system/example.com/logs/error_log

[Question:]

#62

Webserver / Sites are unavailable with HTT...

Last post by Suhitha - Sep 27, 2025, 07:17 AM

Question: Sites are unavailable with HTTP 503 error: failed to make connection to backend: httpd-UDS


Symptoms

Websites hosted on the Plesk server are returning an HTTP 503 Service Unavailable error.
There are the following records in /var/www/vhosts/example.com/logs/error_log file:
[proxy:error] [pid 4322:tid 139839798191872] (2)No such file or directory: AH02454: FCGI: attempt to connect to Unix domain socket /var/www/vhosts/system/example.com/php-fpm.sock (*) failed
[proxy_fcgi:error] [pid 4322:tid 139839798191872] [client 203.0.113.2:58526] AH01079: failed to make connection to backend: httpd-UDS


Cause

Unclear.


Resolution

1. Connect to the server via SSH.
2. Run the following command:

# plesk repair web

[Question:]

#63

Webserver / .NET Core environment variable...

Last post by Suhitha - Sep 27, 2025, 07:13 AM

Question: .NET Core environment variables do not persist on a Plesk server


Applicable to:

• Plesk for Windows

Symptoms

• After saving .NET Core environment variables, they are not recognized or visible through the application.
• Opening the Environment Variables window shows that no variables had been saved or persisted.
• There are no visible errors on Plesk UI or panel.log.

Cause

Missing ASP.net compilation temporary directory for domain.


Resolution

1.Connect to the server via RDP

2.Install the latest Plesk update, if available.

3.Run the following command on an elevated command prompt:

# plesk repair web example.com -y

Note: change example.com to the domain in question.

[Question:]

#64

Webserver / Plesk with LiteSpeed poor perf...

Last post by Suhitha - Sep 27, 2025, 07:09 AM

Question: Plesk with LiteSpeed poor performance: Reached max children process limit, please increase LSAPI_CHILDREN


Applicable to:

• Plesk for Linux

Symptoms

1.Plesk with LiteSpeed extension presents poor performance.

2.The LiteSpeed Apache error log /var/log/httpd/stderr.log shows a lot of errors like below:

Code Select Expand

Child process with pid: 3906418 was killed by signal: 15, core dump: 0
[4015974] Reached max children process limit: 10, extra: 3, current: 13, busy: 13, please increase LSAPI_CHILDREN.
3.Websites' error logs like /var/www/vhosts/system/example.com/logs/error_log show:

Code Select Expand

[INFO] [3840776] [203.0.113.2:52353] Abort request processing by PID:3849827, kill: 1, begin time: 30, sent time: 30, req processed: 0
[INFO] [3904563] [203.0.113.3:55639] Connection idle time: 121 while in state: 7 stream flag: 131072,close!
...
[INFO] [3904563] [203.0.113.4:55639] Lsapi connection state: 3, watching event: 25, respState: 1, m_iTotalPending:0, m_iPacketLeft:0, m_iCurRespHeader:5, req sent for 121 seconds,Total processing time: 121.



Cause

LiteSpeed PHP limits exceeded.


Resolution

• Configure LiteSpeed PHP limits at Extension > LiteSpeed > WebAdmin Console based on the LiteSpeed documentation.

• To get assistance with the configuration, visit LiteSpeed website at https://www.litespeedtech.com/support to check available support options.

[.rpmnew]

#65

Others / After an OS upgrade, where is ...

Last post by senthil - Sep 27, 2025, 07:04 AM

Applicable to:

• Plesk for Linux

Question

After an OS upgrade, where is the old my.cnf file?

Answer

After an upgrade from CentOS to Almalinux, or between Ubuntu versions, for safety purposes a backup file is
created with a different extension and the file will be placed in the same folder.

For CentOS to AlmaLinux conversions, the file may be renamed with the extension .rpmnew. More information
can be found in the log files /var/log/yum.log or /var/log/dnf.log.

For Ubuntu, the file may be renamed with extension .dpkg-old or .dpkg-dist. More information can be found in
the log file /var/log/dpkg.log.

[Question:]

#66

Webserver / Apache fails to start: undefin...

Last post by Suhitha - Sep 27, 2025, 07:03 AM

Question: Apache fails to start: undefined symbol: ap_set_content_type_ex


Applicable to:

• Plesk for Linux

Symptoms

• Apache fails to start with an error like this one:

Code Select Expand

Cannot load /usr/lib/apache2/modules/mod_headers.so into server: /usr/lib/apache2/modules/mod_headers.so: undefined symbol: ap_set_content_type_ex

• LiteSpeed is/was installed on the server

Cause

Incorrect package configuration from LiteSpeed.


Resolution


Reinstall Apache

1.Connect to the server over SSH

2.Reinstall the Apache components

RHEL-based

Code Select Expand

# yum reinstall httpd

Debian-based

Code Select Expand

# apt reinstall apache2 apache2-bin apache2-data apache2-utils
3.Start Apache

Code Select Expand

# systemctl start httpd || systemctl start apache2

Related to

Apache

[Question:]

#67

Webserver / Domain forwarding does not wor...

Last post by Suhitha - Sep 27, 2025, 06:58 AM

Question: Domain forwarding does not work but does not throw any errors


Applicable to:

• Plesk for Linux

Symptoms

• After configuring domain forwarding for a domain, it does not work, but there are no errors in Plesk.

Cause

Domain's DNS records are pointed to another IP address.


Resolution

Point the DNS records to the IP address assigned in Plesk.

[Question:]

#68

Webserver / WordPress instance is marked a...

Last post by Suhitha - Sep 27, 2025, 06:55 AM

Question: WordPress instance is marked as broken on CloudLinux OS: timeout: fork system call failed: Cannot allocate memory


Symptoms

• WordPress instance is marked as broken on CloudLinux OS:

Code Select Expand

timeout: fork system call failed: Cannot allocate memory

• The following errors can be found in the /var/log/plesk/panel.log:

Code Select Expand

ERR [extension/wp-toolkit] An error occurred while executing WP-CLI command for instance: WordPress instance #504 ('http://example.com'). timeout: fork system call failed: Cannot allocate memory
ERR [util_exec] proc_close() failed ['/usr/local/psa/admin/bin/filemng' 'johndoe' 'base64' '/var/www/vhosts/example.com/httpdocs/wp-includes/version.php'] with exit code [1]
ERR [panel] '/usr/local/psa/admin/bin/filemng' 'johndoe' 'base64' '/var/www/vhosts/example.com/httpdocs/wp-includes/version.php' failed with code 1.
stdout:
filemng: Error occured during /usr/bin/base64 command: No child processes
System error 10: No child processes

Cause

Misconfiguration in CageFS.


Resolution

1.Log into Plesk.

2.Go to Extensions > My Extensions > CageFS > Open.

3.Disable the subscription system user (and additional user if it exists) by moving it to Disabled form.

4.Enable the subscription system user back by moving it to Enabled form.

#69

Others / Unable to access a website hos...

Last post by senthil - Sep 27, 2025, 06:54 AM

Symptoms

• Website example. com is unavailable with the following error:
  Sorry, you have been blocked
• When checking the DNS settings, it can be seen traffic is proxied through CloudFlare:
  
  Code Select Expand
  # dig +trace example.com
example.com. 3600 IN NS frank.ns.cloudflare.com.
example.com. 3600 IN NS ingrid.ns.cloudflare.com.
example.com. 300 IN A 188.114.97.12
example.com. 300 IN A 188.114.96.12

Cause
Cloudflare's Web Application Firewall is blocking access to the website.

Resolution
Disable the blocking rule via Cloudflare or Contact Cloudflare Support for further assistance.

[Question:]

#70

Web Application / How to Detach Reserved IPs fro...

Last post by Suhitha - Sep 27, 2025, 06:46 AM

Question: How to Detach Reserved IPs from a Vultr Instance


A guide explaining how to remove a Reserved IP address from a Vultr instance while keeping the IP in your account for future use.

Detaching a Reserved IP from an instance removes the associated public network address from the instance's main network interface. You can detach a Reserved IP to reassign it to another instance, release unused resources, or adjust your network configuration.

Follow this guide to detach Reserved IPs from a Vultr instance using the Vultr Customer Portal, API, CLI, or Terraform.


Vultr Customer Portal


1.Navigate to Products, expand the Network drop-down, and select Reserved IPs from the list of options.

2.Click your target Reserved IP to open its management page.

3.In the Attached to Server section, click Detach to begin removing the Reserved IP from the instance.

4.Click Detach Reserved IP in the confirmation prompt to finalize the detachment.


Vultr API

1.Send a GET request to the List Reserved IPs endpoint and note the target Reserved IP's ID.

Code Select Expand

console


$ curl "https://api.vultr.com/v2/reserved-ips" \
    -X GET \
    -H "Authorization: Bearer ${VULTR_API_KEY}"

2.Send a POST request to the Detach Reserved IP endpoint to detach the Reserved IP from the instance.

Code Select Expand

console


$ curl "https://api.vultr.com/v2/reserved-ips/{reserved-ip-id}/detach" \
    -X POST \
    -H "Authorization: Bearer ${VULTR_API_KEY}"

Vultr CLI

1.List all active Reserved IPs in your Vultr account and note the target Reserved IP's ID.

Code Select Expand

console


$ vultr-cli reserved-ip list

2.Detach the Reserved IP from the instance.

Code Select Expand

console


$ vultr-cli reserved-ip detach <reserved-ip-id>

Terraform

1.Open your Terraform configuration for the existing Reserved IP.

2.Remove instance_id from the vultr_reserved_ip to detach, then apply.

Code Select Expand

terraform


resource "vultr_reserved_ip" "rip" {
    region  = "ewr"
    ip_type = "v4"
    label   = "web-rip"
    # instance_id removed to detach
}

3.Apply the configuration and observe:

Apply complete! Resources: 0 added, 1 changed, 0 destroyed.