Overview
I'm trying to host a few legacy PHP apps on Heroku with Apache. They all relied on the following deprecated syntax to parse any unknown file types (without the .php extension) as PHP.
DefaultType application/x-httpd-php
This has been replaced by AddType in Apache 2.4 (Heroku currently uses v2.4.37). Heroku also uses mod_proxy_fcgi to process PHP files via fcgi://heroku-fcgi.
Issue
I have a file foo.test and I want to have it handled by PHP FPM. Taking cues from the docs and the default Apache config provided by Heroku, here's what I've tried:
# .htaccess
<FilesMatch \.test$>
<If "-f %{REQUEST_FILENAME}">
SetHandler proxy:fcgi://heroku-fcgi
</If>
</FilesMatch>
# apache_app.conf (properly loaded via Procfile)
ProxyPassMatch "^/(.*\.test(/.*)?)$" "fcgi://heroku-fcgi/app/$1"
With both of these I get a plain-text 403 Access denied. response from PHP FPM. I'm sure both configs are properly loading and pointing to the FCGI handler because changing the endpoint results in other errors.
My Apache skills are long since rusty and I can't seem to find any good pointers online. The Apache error log is also clean. Any ideas (without the obvious "change all extensions to PHP, you dumbass") would be appreciated!
Fairly obvious solution. PHP FPM has its own configuration with a security.limit_extensions flag. It defaults to .php.
The solution was to unset that value: security.limit_extensions =. This naturally can pose some security threats, but these apps are only going up for static demo.
I was using heroku/heroku-buildpack-php but forked that to update this file. The htaccess FilesMatch should work now but I just ended up placing it into the Apache config file to avoid repetition across the sites I'll be serving.
security.limit_extensions can be customized with a configuration file passed as a Procfile argument.
https://devcenter.heroku.com/articles/custom-php-settings#php-fpm-settings
PHP-FPM settings:
In addition to php_value and php_flag for php.ini specific settings, any pool specific PHP-FPM configuration directives are valid in that configuration file, so you can use it to fine tune PHP-FPM’s behavior.
So you can set up it like the following
Procfile
web: vendor/bin/heroku-php-apache2 -C apache.conf -F fpm_custom.conf web/
apache.conf
<FilesMatch \.test$>
<If "-f %{REQUEST_FILENAME}"> # make sure the file exists so that if not, Apache will show its 404 page and not FPM
SetHandler proxy:fcgi://heroku-fcgi
</If>
</FilesMatch>
fpm_custom.conf
security.limit_extensions = .php .test
Related
I have setup php/apache according to this linode guide: https://www.linode.com/docs/guides/install-php-8-for-apache-and-nginx-on-ubuntu/
This seems to have worked, and I have an apache server that can serve files, and process .php files.
However, I am having the darndest time figuring out how to tell apache to process .html files with .php. Any modification I make to .htaccess (a-la- https://manage.accuwebhosting.com/knowledgebase/2492/Parse-HTML-As-PHP-Using-HTACCESS-File.html and similar) produce no change. I have also set AllowOverride All to the /var/www/ directory in /etc/apache2/apache2.conf. Additionally, most online information on the subject points to earlier versions of .php, with no specific reference to 8.0.
What can get apache to process .html with php 8.0?
You need to have module for this, confirm you have this line in httpd.conf, or add it there:
LoadModule mime_module modules/mod_mime.so
Also check if you have the modules/mod_mime.so file present on your system.
Then find or add module section in httpd.conf:
<IfModule mime_module>
# following line will change mime type of .html file to php
# and they will be handled as such
AddType application/x-httpd-php .html
</IfModule>
Directive AllowOverride All will enable .htaccess files but you need the mime_module enabled too.
And of course restart the apache server after making configuration changes.
Module documentation: here
I am trying to configure everything in order to run simultaneously php5 and php7 on Fedora 27. I am using Remi's guides from here and here, I am able to switch versions in the command line with module load/unload php71/php56
but on the page where I output phpinfo(); I get php version of 7.1. I have also
running php56-php-fpm.service and php71-php-fpm.service running.
What should I check or where to search? Thank you.
php.conf file:
# Redirect to local php-fpm if mod_php (5 or 7) is not available
<IfModule !mod_php5.c>
<IfModule !mod_php7.c>
# Enable http authorization headers
SetEnvIfNoCase ^Authorization$ "(.+)" HTTP_AUTHORIZATION=$1
<FilesMatch \.php$>
SetHandler "proxy:unix:/run/php-fpm/www.sock|fcgi://localhost"
</FilesMatch>
</IfModule>
</IfModule>
also there are php56-php.conf and php71-php.conf files. As I understand mod_php is serving files instead of php-fpm, is there any method to disable it?
Check all SetHandler Directives in /etc/httpd/conf.d/*conf
Each package comes with its configuration file (php##-php.conf), you may have to disable them to be able to set the proper version per vhost / project / directory, or ensure your configuration files are loaded after provided ones.
In a newly setup digitalOcean cloud server (CentOS), I have installed php and Apache. The webserver is running fine:
[root#a2m5cent01 httpd]# service httpd status
httpd (pid 11232) is running...
[root#a2m5cent01 httpd]# php --version | head -1
PHP 5.3.3 (cli) (built: Dec 11 2013 03:29:57)
But browser is showing blank pages (white page) if I try to visit any php page.
Here is what I have done so far to troubleshoot:
Created a page with following content: <?php phpinfo(); ?>. It displays a blank page when viewed from browser.
Just to ensure, apache is pointing to the correct directory, placed a static .html page there, and saw it comes out fine in browser, so apache is working and directory is correct.
In /etc/php.ini, changed display_errors directive to On. Still blank page
In Apache config file (/etc/httpd/conf/httpd.conf) found this line Include conf.d/*.conf. Inside conf.d directory, there is a php.conf file containing the line:
LoadModule php5_module modules/libphp5.so. Ensured that this .so file actually exists in this place.
In the same file I have these two lines as well: AddHandler php5-script .php and AddType text/html .php
Executed the php page from CLI, it works fine - so php is working locally.
Then why is it always shows a blank/white page over the browser? What else am I missing?
EDIT
Based on suggestions from #Nathan,
I checked Apache error log file, could not see any error being
reported there.
My /etc/php.ini says, php error_log is located as syslog. So I checked /var/log/messages but could not find any PHP error message
Next I put some normal HTML in the php file containing phpinfo() call. Interestingly I found that even the normal HTML texts are also not coming. It still produces blank page.
Then I checked Apache access log. Surprise! There is no GET request for any of the PHP files I tried to load in the browser. But GET request for all the non-php files are there with 200 return code.
Apache is not even logging any access request for PHP files. Any idea why would that happen?
check out your phpinfo() script.
<?php
phpinfo();
?>
missing the "php" behind the first "?" will give a blank page
I think your php installation with apache is faulty. Thats why you can not see any php page in your webserver. Clean remove all the existing apps, like httpd,php,php-fpm,php-cli etc. and try to clean isntall in this order
yum install httpd -y
yum install php php-common php-cli php-gd php-curl php-fpm -y
then make sure you restart yout httpd server.
service httpd restart
Install mod_fastcgi:
yum install mod_fastcgi
Start the service:
service php-fpm start
Restart Apache:
service httpd restart
5. Configuration of Apache with PHP-FPM
Open the fastcgi.conf file:
nano /etc/httpd/conf.d/fastcgi.conf
Add this to the end of the file:
<IfModule mod_fastcgi.c>
DirectoryIndex index.html index.shtml index.cgi index.php
AddHandler php5-fcgi .php
Action php5-fcgi /php5-fcgi
Alias /php5-fcgi /usr/lib/cgi-bin/php5-fcgi
FastCgiExternalServer /usr/lib/cgi-bin/php5-fcgi -host 127.0.0.1:9000 -pass-header Authorization
</IfModule>
After that search after "FastCgiWrapper" and make sure it's set to "off" then save the file.
The /usr/lib/cgi-bin/ directory must exist, so we create it:
mkdir /usr/lib/cgi-bin/
If mod_php is installed and enabled, we need to disable it so open the configuration at /etc/httpd/conf.d/php.conf:
nano /etc/httpd/conf.d/php.conf
Comment out the AddHandler and AddType lines so it looks like here:
#
# PHP is an HTML-embedded scripting language which attempts to make it
# easy for developers to write dynamically generated webpages.
#
<IfModule prefork.c>
LoadModule php5_module modules/libphp5.so
</IfModule>
<IfModule worker.c>
LoadModule php5_module modules/libphp5-zts.so
</IfModule>
#
# Cause the PHP interpreter to handle files with a .php extension.
#
#AddHandler php5-script .php
#AddType text/html .php
#
# Add index.php to the list of files that will be served as directory
# indexes.
#
DirectoryIndex index.php
#
# Uncomment the following line to allow PHP to pretty-print .phps
# files as PHP source code:
#
#AddType application/x-httpd-php-source .phps
Save the file and restart Apache:
service httpd restart
Are you navigating to the php file directly? Or are you just going to the directory root?
If the later, Apache might not be recognizing .php as the directory index.
To test, try create a .htaccess file in your web root containing the following line:
DirectoryIndex index.php
I have the same issue...
The problem is in the iptables. (It seems like it)
Try with:
service iptables stop
## check if it stop...
service iptables status
Then try to reload the page again.
If you had other solution please share.
[edit]
Restarting the iptables service is working for me.
Try:
service iptables restart
First of all you should check the permissions of your file.
If you don't grant read-permission to public, Apache produces a blank page without showing any errors.
It's been sometime, but I wanted to come back to this question to update that the issue was with the directory permission setup.
The FPM user I was using didn't have necessary permission to execute the index.php file in the web root.
To avoid these issues in the future, I have created an automated bash script that will automatically create and configure webservers in DigitalOcean boxes. Please take a look here https://github.com/akash-mitra/fairy
This script will automatically,
Installs Nginx
Create virtual server block for nginx
Installs PHP, PHP APC, PHP Curl etc.
Supports PHP Fast Process Manager (php-fpm)
Installs Memcached
Installs Database (MariaDB / MySQL)
Optionally Installs PHP Composer and Laravel
Configures and Strengthens SSH
Activates Firewall
Optionally enables SWAP space in DO server and fixes a locale issue
Since everything looks at its default state, have you checked this part just for confirmation
cat /etc/php.ini | grep log_errors
If log_errors is disabled then enable it and check this below log after restarting httpd.
/var/log/httpd/error_log
--
Also check this part from php configuration.
cat /etc/php.ini | grep error_reporting
This value should be enabled by default so that display_errors will work.
error_reporting = E_ALL
Sorry to repost an old thread...this is important.
I also was having these problems where no html response was outputting
After double-checking php.ini or my apache conf files and was still receiving no output, I later found out that I was suppressing the error of an include / require of a class, with #, which was nested within a constructor function. There was a syntax error in the included file, which stopped all output altogether when errors were thrown.
So, check your handlers first.
If you are storing all your output into vars first and you are including various scripts first that fail you'll have to see those errors. If you suppress file handler errors, you'll get a blank screen if you have a syntax error in the file.
Search your files for all instances of # in your php code. Then turn #include "/path_to/script.php"; to include "/path_to/script.php"; or anything #$foo into $foo as such var might reference a dependency that is causing your script to end with nothing showing in the httpd error log or in the http response.
So my server suddenly stopped parsing PHP (sends raw php to the client). I'm not really sure when it happened, but I just noticed it today. I was messing with some mod_rewrite stuff, but I put it back and it didn't change. Other than that I haven't changed anything (to the best of my knowledge). Ideas? It's an Ubuntu 11.10 server, BTW.
Few things to try...
First:
apachectl -t -D DUMP_MODULES | grep php
You should get something like:
php5_module (shared)
at the very least.
Secondly... how are you restarting apache? Are you sure it is getting killed? I have used an apache init.d script before that would actually run an apachectl configtest to make sure it was error free before stopping and restarting. So, perhaps you need to stop, ensure it is stopped, then start again.
Also, in your config, make sure you have something like:
<IfModule php5_module>
AddType application/x-httpd-php .php
</IfModule>
You don't necessarily need the <IfModule> directive, but doesn't hurt.
And what version of apache are you running?
check:
that PHP files have their executable bit set
that "index.php" is set as one of default index files in your web
root
that php module is loaded by apache (see apache config file, perhaps
near by enabling mod_rewrite module)
Remove the Rewrite setting might be you have return wrong.
And restart the Apache server. Hope that will resolve the issue.
Can refer to
http://www.matthewwittering.co.uk/blog/ubuntu-tips/apache-not-running-php-files.htm
https://askubuntu.com/questions/59272/php-not-working-in-apache2-after-system-upgrade
<IfModule mod_php5.c>
<FilesMatch "\.ph(p3?|tml)$">
SetHandler application/x-httpd-php
</FilesMatch>
<FilesMatch "\.phps$">
SetHandler application/x-httpd-php-source
</FilesMatch>
# To re-enable php in user directories comment the following lines
# (from <IfModule ...> to </IfModule>.) Do NOT set it to On as it
# prevents .htaccess files from disabling it.
<IfModule mod_userdir.c>
<Directory /home/*/public_html>
php_admin_value engine Off
</Directory>
</IfModule>
</IfModule>
I have tried reinstalling PHP. PHP is working, and Apache2 is running. I don't know why it's not opening in a browser and displaying normally.
Just so you know - my httpd.conf is empty - and instead I have everything in apache2.conf. This is because I'm using Ubuntu.
Can you help me? I know it's something simple, but I can't seem to find the answer.
Have you virtual host on this project?
Are you open php file with http://localhost/file.php or directly like file://...../file.php ?
In your apache conf:
<FilesMatch \.php$>
SetHandler application/x-httpd-php
</FilesMatch>
Don't forget to restart apache afterwards. Let me know how it goes.
In my case there was a modification of the /etc/apache2/mods-enabled/php5.conf by module userdir
<IfModule mod_php5.c>
<FilesMatch "\.ph(p3?|tml)$">
SetHandler application/x-httpd-php
</FilesMatch>
<FilesMatch "\.phps$">
SetHandler application/x-httpd-php-source
</FilesMatch>
# To re-enable php in user directories comment the following lines
# (from <IfModule ...> to </IfModule>.) Do NOT set it to On as it
# prevents .htaccess files from disabling it.
<IfModule mod_userdir.c>
<Directory /home/*/public_html>
php_admin_value engine Off
</Directory>
</IfModule>
</IfModule>
The solution is inside the config file, just comment the lines from <IfModule mod_userdir> to </IfModule>.
Try
sudo a2enmod php5
in terminal.
Assuming you are using php 5 :)
If you are using php5 the complete solution would be
sudo apt-get install libapache2-mod-php5
Then
sudo a2enmod php5
I recently had this problem, but only when accessing my site's root (e.g. http://example.com) - it worked as expected when accessing index.php explicitly (e.g. http://example.com/index.php).
The issue was that before creating index.php, I had an index.html file in the web root, but then I renamed it to index.html.bak to 'move it out of the way'. Unfortunately, that doesn't move it out of the way. For whatever reason, Apache will serve that file up with a MIME type of application/x-trash.
(As the /etc/mime.types file contains a line reading "application/x-trash ~ % bak old sik")
Maybe that will help someone else who is scratching their head as I was.
Also, when moving a site and you encounter this, make sure your .htaccess file doesn't specify a different method handler for php. We ran into this moving a wordpress site for a client.
Had same problem - and it was in the .htaccess file I had accidentally downloaded from the live server. Once you've deleted the .htaccess file, you will need to clear your cache to be able to load the .php file in your browser via http://