How to find my php-fpm.sock?

I'm running Wordpress with: Nginx + PHP-FPM + APC + W3 Total Cache + PageSpeed.

After 3 days researching and configuring, I succeeded to make it work. I configured PHP-FPM to run via 127.0.0.1:9000. But now I want to configure via Socket.

The problem is that I can't find the socket path in my server. I just found /var/run/php-fpm/php-fpm.pid, but php-fpm.sock was not there.

Running whereis php-fpm the output is:

php-fpm: /usr/sbin/php-fpm /etc/php-fpm.d /etc/php-fpm.conf /usr/share/man/man8/php-fpm.8.gz 

But there isn't any php-fpm.sock there.

How can I find php-fpm.sock?

My specs:

• Amazon Micro EC2
• Linux version 3.4.48-45.46.amzn1.x86_64 Red Hat 4.6.3-2 (I think it's based on CentOS 5)
• PHP 5.3.26 (fpm-fcgi)

6 Answers

I know this is old questions but since I too have the same problem just now and found out the answer, thought I might share it. The problem was due to configuration at pool.d/ directory.

Open

/etc/php5/fpm/pool.d/ 

find

listen = 127.0.0.1:9000 

change to

listen = /var/run/php5-fpm.sock 

Restart both nginx and php5-fpm service afterwards and check if php5-fpm.sock already created.

I faced this same issue on CentOS 7 years later

Posting hoping that it may help others...

Steps:

FIRST, configure the php-fpm settings:

-> systemctl stop php-fpm.service

-> cd /etc/php-fpm.d

-> ls -hal (should see a file)

-> cp (back file up just in case)

-> vi

-> :/listen = (to get to the line we need to change)

-> i (to enter VI's text insertion mode)

-> change from listen = 127.0.0.1:9000 TO listen = /var/run/php-fpm/php-fpm.sock

-> Esc then :/listen.owner (to find it) then i (to change)

-> UNCOMMENT the listen.owner = nobody AND listen.group = nobody lines

-> Hit Esc then type :/user = then i

-> change user = apache TO user = nginx

-> AND change group = apache TO group = nginx

-> Hit Esc then :wq (to save and quit)

-> systemctl start php-fpm.service (now you will have a php-fpm.sock file)

SECOND, you configure your server {} block in your /etc/nginx/nginx.conf file. Then run:systemctl restart nginx.service

FINALLY, create a new .php file in your /usr/share/nginx/html directory for your Nginx server to serve up via the internet browser as a test.

-> vi /usr/share/nginx/html/mytest.php

-> type o

-> <?php echo date("Y/m/d-l"); ?> (PHP page will print date and day in browser)

-> Hit Esc

-> type :wq (to save and quite VI editor)

-> open up a browser and go to: (you should see the date and day printed)

Check the config file, the config path is /etc/php5/fpm/pool.d/, there you'll find the path by config and if you want you can change it.

EDIT:
well you're correct, you need to replace listen = 127.0.0.1:9000 to listen = /var/run/php5-fpm/php5-fpm.sock, then you need to run sudo service php5-fpm restart, and make sure it says that it restarted correctly, if not then make sure that /var/run/ has a folder called php5-fpm, or make it listen to /var/run/php5-fpm.sock cause i don't think the folder inside /var/run is created automatically, i remember i had to edit the start up script to create that folder, otherwise even if you mkdir /var/run/php5-fpm after restart that folder will disappear and the service starting will fail.

Solved in my case, i look at

sudo tail -f /var/log/nginx/error.log 

and error is php5-fpm.sock not found

I look at sudo ls -lah /var/run/

there was no php5-fpm.sock

I edit the  

sudo vim /etc/php5/fpm/pool.d/ 

change

listen = 127.0.0.1:9000 

for

listen = /var/run/php5-fpm.sock 

and reboot

When you look up your php-fpm.conf

example location: cat /usr/src/php/sapi/fpm/php-fpm.conf 

you will see, that you need to configure the PHP FastCGI Process Manager to actually use Unix sockets. Per default, the listen directive` is set up to listen on a TCP socket on one port. If there's no Unix socket defined, you won't find a Unix socket file.

; The address on which to accept FastCGI requests. ; Valid syntaxes are: ; 'ip.add.re.ss:port' - to listen on a TCP socket to a specific IPv4 address on ; a specific port; ; '[ip:6:addr:ess]:port' - to listen on a TCP socket to a specific IPv6 address on ; a specific port; ; 'port' - to listen on a TCP socket to all IPv4 addresses on a ; specific port; ; '[::]:port' - to listen on a TCP socket to all addresses ; (IPv6 and IPv4-mapped) on a specific port; ; '/path/to/unix/socket' - to listen on a unix socket. ; Note: This value is mandatory. listen = 127.0.0.1:9000 

I encounter this issue when I first run LEMP on centos7 refer to this post.

I restart nginx to test the phpinfo page, but get this

is not unreachable now.

Then I use tail -f /var/log/nginx/error.log to see more info. I find is the php-fpm.sock file not exist. Then I reboot the system, everything is OK.

Here may not need to reboot the system as Fath's post, just reload nginx and php-fpm.

restart php-fpm

reload nginx config