Problems with server?

[Barney]

Hello,

Hope you are well.

My site went live a few days ago and everyday the website has gone offline and I get 502 Bad Gateway error.

I am paying a lot of money for the hosting so £170 a month, so it should be good spec.

Yesterday they replied with the following information (pasted below) which I do not understand.

Please can you have a read and advise, does KLEO need certain apache/php… and if I make the change they suggest will it break the whole site?

Staff

Hello

Disabling DSO handler means changing the PHP handler to suPHP or Fast CGI. The change can lead to a few permission issues with CMS plugins, themes etc. The details are explained below for each handler:

++++++
1. suPHP: Runs as a CGI module and its major benefits are stronger security and easy permissions. The main benefit of using the suPHP handler is that it isolates one user on the server from the others. So if one account was compromised because of an exploit in one of your PHP scripts, the attacker would only be able to view or modify files owned by that particular user. It’s also among the easiest to use with CMS (Content Management Systems) such as WordPress or Joomla. As typically these applications require the ability to create files on the server, and being that all of your files are owned by just one user.

2. Fast CGI: Run as a CGI module, FCGI also referred to as mod_fcgid, is one of the newest PHP handlers and its major benefits are CPU performance. The main benefit of using the FastCGI handler is that you can use suEXEC just like with suPHP to allow PHP scripts to be executed by the actual user of the PHP script instead of using the Apache “nobody” user. FastCGI also doesn’t require a single PHP process execution per request like suPHP does, so it can be much faster and reduces CPU usage by holding PHP scripts in memory.
   ++++++

Kindly check and confirm with your developer. Let us know if a modification is required.

---

Sonia G.
UH Support

03/07/2018 (12:32)

Client

Hi Sonia,

Thank you for your reply.

That sounds scary! I don’t want to risk the website. In what way can it effect the website?

I have a friend who has some developer knowledge, I could ask them what they think?

(I apologise for my lack of knowledge with the technical side!)

Thanks again for all your help, all the best.

03/07/2018 (12:26)

Staff

Hello

That is great news. Unfortunately, the solution to this problem will be changing Apache to work in worker MPM. However, changing this will require disabling PHP DSO (which is currently rebuild for PHP 5 on your server). Doing so might effect a few websites on your server. So before making any changes I would suggest you have a word with your website developers for all the accounts hosted on this server. We can then proceed with modifying the Apache/PHP parameters.

Snippet:
+++++
~]# httpd -V
Server version: Apache/2.2.31 (Unix)
Server built: Nov 8 2015 09:17:50
Cpanel::Easy::Apache v3.32.5 rev9999
Server’s Module Magic Number: 20051115:40
Server loaded: APR 1.5.1, APR-Util 1.5.4
Compiled using: APR 1.5.1, APR-Util 1.5.4
Architecture: 64-bit
Server MPM: Prefork

~]# /usr/local/cpanel/bin/rebuild_phpconf –current
Available handlers: suphp dso cgi none
DEFAULT PHP: 5
PHP4 SAPI: none
PHP5 SAPI: dso
SUEXEC: enabled
RUID2: enabled
+++++

Do let us know if you we can be of any help here.

Thank you,

---

Sonia G.
UH Support

03/07/2018 (12:12)

Client

Hello Sonia,

I just had a look and the website is back online, thank you very much for your help!

Is there anything I can do to stop that issue repeating at all, or is it a random thing?

Thanks again for your fast resolution, it is much appreciated.

03/07/2018 (12:05)

Staff

Hello

The server load has come down and the website is accessible without any errors displayed or any glitches. The current load on the server is as follows:

+++++
root@cloud [~]# w
11:55:40 up 53 days, 22:57, 1 user, load average: 1.61, 9.40, 14.85
+++++

There were too many httpd processes in sleep status created under the user ‘womenssoccerunit’. These processes were not killed and hence the issue. As a result, I had to restart Nginx service.

Snippet:
+++++
~]# top -u womenssoccerunit

top – 11:43:53 up 53 days, 22:45, 1 user, load average: 16.86, 20.12, 17.74
Tasks: 197 total, 5 running, 191 sleeping, 0 stopped, 1 zombie
Cpu(s): 25.5%us, 12.7%sy, 3.6%ni, 15.3%id, 40.5%wa, 0.0%hi, 0.1%si, 2.2%st
Mem: 10257216k total, 6194232k used, 4062984k free, 241580k buffers
Swap: 0k total, 0k used, 0k free, 587552k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 
4563 womensso 20 0 427m 100m 5156 S 12.3 1.0 0:07.81 httpd 
635 womensso 20 0 443m 116m 5472 R 3.0 1.2 0:14.64 httpd 
3365 womensso 20 0 437m 110m 5408 S 2.3 1.1 0:13.83 httpd 
2725 womensso 20 0 439m 112m 5260 S 2.0 1.1 0:12.22 httpd 
2794 womensso 20 0 438m 111m 5168 S 1.3 1.1 0:11.63 httpd 
4520 womensso 20 0 438m 112m 5348 S 0.3 1.1 0:05.40 httpd 
639 womensso 20 0 439m 112m 5184 S 0.0 1.1 0:17.53 httpd 
1508 womensso 20 0 435m 109m 5308 S 0.0 1.1 0:34.83 httpd 
1542 womensso 20 0 438m 111m 5288 S 0.0 1.1 0:17.86 httpd 
1827 womensso 20 0 438m 111m 5192 S 0.0 1.1 0:17.77 httpd 
1884 womensso 20 0 436m 110m 5160 S 0.0 1.1 0:10.05 httpd 
1950 womensso 20 0 436m 108m 5200 S 0.0 1.1 0:13.91 httpd 
2686 womensso 20 0 437m 110m 5212 S 0.0 1.1 0:15.97 httpd 
3009 womensso 20 0 437m 110m 5172 S 0.0 1.1 0:15.71 httpd 
+++++

Please check now and let us know if you have any further concerns.

---

Sonia G.

 

 

 

[Laura]

Not marked as solution

[Barney]

Not marked as solution

[Radu]

Not marked as solution

[Barney]

Not marked as solution

[Radu]

Not marked as solution

[Barney]

Not marked as solution

[Radu]

Not marked as solution

[Barney]

Not marked as solution

[Radu]

Not marked as solution

[Barney]

Not marked as solution

[Radu]

Not marked as solution

[Author]

Posts