Apache or Nginx webservers are restarted frequently: GENERIC: error receiving data -- Resource temporarily unavailable

Symptoms

• Websites are intermittently unavailable. Web browser shows the error immediately upon accessing a website.

• Nginx or Apache service is frequently restarted, below information can be found in /var/log/plesk/rc_actions.log:

  CONFIG_TEXT: INFO: Service: httpd, Action: stop
  INFO: Service: httpd, Action: status
  INFO: Service: httpd, Action: start

• The following records can be found in /var/log/syslog (Debian/Ubuntu), /var/log/messages (CentOS/CloudLinux/RHEL), or /var/log/monit.log:

  CONFIG_TEXT: 'nginx' failed protocol test [generic] at INET[203.0.113.2:80].
  'nginx' trying to restart
  'nginx' stop: /opt/psa/admin/bin/nginx_control
  'nginx' start: /opt/psa/admin/bin/nginx_control

  CONFIG_TEXT: GENERIC: error receiving data -- Resource temporarily unavailable
  'apache' failed protocol test [generic] at INET[203.0.113.2:80].
  'apache' trying to restart
  'apache' stop: /usr/local/psa/admin/bin/websrvmng
  'apache' start: /usr/local/psa/admin/bin/websrvmng
  'apache' connection passed

Cause

Nginx or Apache is being restarted by Watchdog due to the false-negative protocol test.
This is likely to happen due to the short connection timeout value.

Resolution

1. Log in to Plesk.

2. Increase Connection timeout value for nginx in Extensions > My Extensions > Watchdog > Web Proxy Server (Nginx):

   

   Or increase Connection timeout value for Apache in Extensions > My Extensions > Watchdog > Open > Web Server (Apache):