# PHP dead pages with php-fpm and nginx

**URL:** <https://www.ruby-forum.com/t/php-dead-pages-with-php-fpm-and-nginx/206260>\
**Category:** NGINX\
**Created:** [April 26, 2011, 8:40pm UTC](https://www.ruby-forum.com/t/php-dead-pages-with-php-fpm-and-nginx/206260 "2011-04-26T20:40:45Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![luislavena](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/l/0ea827/32.png) [@luislavena](https://www.ruby-forum.com/u/luislavena)\
**Post date:** [April 26, 2011, 8:40pm UTC](https://www.ruby-forum.com/t/php-dead-pages-with-php-fpm-and-nginx/206260/1 "2011-04-26T20:40:45Z")

</div>

Hi all,

I noticed today some weird errors when I tried to refresh a basic php  
page (phpinfo() function).  
The page loads fine when I click refresh several times, then all of the  
sudden I get a dead page, no 404, no Nginx error, nothing. It looks like  
someone disconnected the network cable for a fraction of second. I  
checked both php and nginx logs, nothing is there. I even enabled the  
debug mode to be safe.

I have a very basic configuration:

upstream fastcgi {  
server 192.168.0.2:9000;  
server 192.168.0.3:9000 weight=5;  
server 192.168.0.10:9000 weight=5;  
}

location / {  
try\_files $uri $uri/ /index.php;  
}

location ~ .php$ {  
fastcgi\_pass fastcgi;  
include fastcgi.conf;  
}

Do you know what could generate this issue? The site is closed now so  
only 2-3 people have access to the setup.  
Do you recommend any tools or settings that will allow me to verify if  
the network is shaky?

Posted at Nginx Forum:

> **[PHP dead pages with php-fpm and nginx](https://forum.nginx.org/read.php?2,193878,193878#msg-193878)**
>
> Hi all, I noticed today some weird errors when I tried to refresh a basic php page (phpinfo() function). The page loads fine when I click refresh several times, then all of the sudden I get a dead page, no 404, no Nginx error, nothing. It looks like...

---

<div class="post-metadata">

**Author:** ![teck](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/b2d939/32.png) [@teck](https://www.ruby-forum.com/u/teck)\
**Post date:** [April 26, 2011, 9:07pm UTC](https://www.ruby-forum.com/t/php-dead-pages-with-php-fpm-and-nginx/206260/2 "2011-04-26T21:07:20Z")

</div>

OK, I finally got some error message into logs:

2011/04/26 13:44:39 [error] 5727#0: \*1 upstream timed out (110:  
Connection timed out) while connecting to upstream, client:  
206.53.59.32, server: [sandbox.domain.com](http://sandbox.domain.com), request: “GET /info.php  
HTTP/1.1”, upstream: “fastcgi://192.168.0.10:9000”, host:  
“[sandbox.domain.com](http://sandbox.domain.com)”  
2011/04/26 13:44:57 [error] 5727#0: \*1 connect() failed (113: No route  
to host) while connecting to upstream, client: 206.53.59.32, server:  
[sandbox.domain.com](http://sandbox.domain.com), request: “GET  
/info.php?=PHPE9568F35-D428-11d2-A769-00AA001ACF42 HTTP/1.1”, upstream:  
“fastcgi://192.168.0.10:9000”, host: “[sandbox.domain.com](http://sandbox.domain.com)”, referrer:  
“[http://sandbox.domain.com/info.php](http://sandbox.domain.com/info.php)”

/info.php?=PHPE9568F35-D428-11d2-A769-00AA001ACF42 is the php image we  
see into phpinfo regular page.

Posted at Nginx Forum:

> **[Re: PHP dead pages with php-fpm and nginx](https://forum.nginx.org/read.php?2,193878,193883#msg-193883)**
>
> Hi all, I noticed today some weird errors when I tried to refresh a basic php page (phpinfo() function). The page loads fine when I click refresh several times, then all of the sudden I get a dead page, no 404, no Nginx error, nothing. It looks like...

---

<div class="post-metadata">

**Author:** ![teck](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/b2d939/32.png) [@teck](https://www.ruby-forum.com/u/teck)\
**Post date:** [April 27, 2011, 6:54am UTC](https://www.ruby-forum.com/t/php-dead-pages-with-php-fpm-and-nginx/206260/3 "2011-04-27T06:54:29Z")

</div>

On Tue, Apr 26, 2011 at 03:06:58PM -0400, TECK wrote:

> /info.php?=PHPE9568F35-D428-11d2-A769-00AA001ACF42 HTTP/1.1", upstream:  
> “fastcgi://192.168.0.10:9000”, host: “[sandbox.domain.com](http://sandbox.domain.com)”, referrer:  
> “[http://sandbox.domain.com/info.php](http://sandbox.domain.com/info.php)”
> 
> /info.php?=PHPE9568F35-D428-11d2-A769-00AA001ACF42 is the php image we  
> see into phpinfo regular page.

“No route to host” means that you have some network promblem between  
nginx host and 192.168.0.10.  
“Connection timed out” in this case is a prognostic error of next  
“No route to host” error.

–  
Igor S.

---

<div class="post-metadata">

**Author:** ![teck](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/t/b2d939/32.png) [@teck](https://www.ruby-forum.com/u/teck)\
**Post date:** [April 28, 2011, 8:14pm UTC](https://www.ruby-forum.com/t/php-dead-pages-with-php-fpm-and-nginx/206260/4 "2011-04-28T20:14:14Z")

</div>

Thanks Igor, you are right. It was a network issue.

Posted at Nginx Forum:

> **[Re: PHP dead pages with php-fpm and nginx](https://forum.nginx.org/read.php?2,193878,194322#msg-194322)**
>
> Hi all, I noticed today some weird errors when I tried to refresh a basic php page (phpinfo() function). The page loads fine when I click refresh several times, then all of the sudden I get a dead page, no 404, no Nginx error, nothing. It looks like...