Article ID: 118757, created on Mar 23, 2014, last review on Apr 27, 2014

  • Applies to:
  • Plesk Automation 11.1
  • Plesk Automation 11.5
  • Plesk 11.0 for Linux
  • Plesk 11.5 for Linux


While accessing Plesk Panel, I see the "502 Bad Gateway" error.

/var/log/sw-cp-server/error_log contains the following error message:

recv() failed (104: Connection reset by peer) while reading response header from upstream, client:, server: , request: "POST  <some url>


The issue is caused due to a small value being set for fastcgi_buffer in the Plesk Panel configuration file for Nginx.


It is necessary to increase the buffer size inside the file /etc/sw-cp-server/config as follows :


root@hostname:~# grep buffer  /etc/sw-cp-server/config
    fastcgi_buffers 16 16k;
    fastcgi_buffer_size 32k;


root@hostname:~# grep buffer  /etc/sw-cp-server/config
    fastcgi_buffers 32 32k;
    fastcgi_buffer_size 64k;

After editing the file, please restart the below services:

root@hostname:~# /etc/init.d/sw-cp-server restart

root@hostname:~# /etc/init.d/sw-engine restart

a914db3fdc7a53ddcfd1b2db8f5a1b9c 56797cefb1efc9130f7c48a7d1db0f0c 01bc4c8cf5b7f01f815a7ada004154a2 29d1e90fd304f01e6420fbe60f66f838 0a53c5a9ca65a74d37ef5c5eaeb55d7f e0aff7830fa22f92062ee4db78133079 aea4cd7bfd353ad7a1341a257ad4724a c1ecc6010feff26cb42d1d14a7881dd6 33a70544d00d562bbc5b17762c4ed2b3 caea8340e2d186a540518d08602aa065

Email subscription for changes to this article