I've been running a magento site using HHVM and Nginx. So far I hadn't experienced any problems and all I got was a very welcomed noticeable performance boost.
However, I've just added a plugin which uses PHP functions which are unsupported by HHVM. The good news is that this plugin is only needed for one page. Bad news is I'm not configuring Nginx right to serve this page only with PHP.
The usual fallback trick used by some people using error directives does not work in this case because the page doesn't throw an error. It just doesn't work if HHVM is enabled.
Instead, I've tried to write many variations of location blocks for that particular page. None worked, and these are the two that I had thought would do the trick.
Is there a way to run PHP only for a specific page?
FAILED SOLUTION 1
location ~ /page/.+\.php${
if (!-e $request_filename) { rewrite / /index.php last; }
fastcgi_pass unix:/var/run/php5-fpm.sock; ##Unix socket
fastcgi_param HTTPS $fastcgi_https;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
include fastcgi_params;
}
location ~ \.(hh|php)$ {
if (!-e $request_filename) { rewrite / /index.php last; }
fastcgi_keep_conn on;
fastcgi_pass unix:/var/run/hhvm/sock;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param HTTPS $fastcgi_https;
include fastcgi_params;
}
FAILED SOLUTION 2 (with nested location)
location ~ /page/ {
location ~ .php$ {
if (!-e $request_filename) { rewrite / /index.php last; }
fastcgi_pass unix:/var/run/php5-fpm.sock; ##Unix socket
fastcgi_param HTTPS $fastcgi_https;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
include fastcgi_params;
}
}
location ~ \.(hh|php)$ {
if (!-e $request_filename) { rewrite / /index.php last; }
fastcgi_keep_conn on;
fastcgi_pass unix:/var/run/hhvm/sock;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param HTTPS $fastcgi_https;
include fastcgi_params;
}
Try using variable condition approach, it works for me, locations starting with /serve-with-php/ are being served with unix:/var/run/php5-fpm.sock, while all the others are being served with 127.0.0.1:9000
server {
root /home/vearutop/test-hhvm;
index index.php index.html index.htm;
error_log /var/log/nginx-error.log error;
charset utf-8;
server_name test-hhvm;
location / {
set $fcgi_worker 127.0.0.1:9000; # hhvm handle
try_files $uri $uri/ /index.php?$query_string;
}
location /serve-with-php/ {
set $fcgi_worker unix:/var/run/php5-fpm.sock; # php-fpm handle
try_files $uri $uri/ /index.php?$query_string;
}
location ~ \.php$ {
fastcgi_split_path_info ^(.+\.php)(/.+)$;
fastcgi_pass $fcgi_worker;
fastcgi_index index.php;
include fastcgi_params;
}
}
Related
After adding various configurations into the Wordpress subdirectory, I can access without any issues the WP homepage, but I'm still not able to access the posts, returning 404.
www.test.com/blog - WP homepage, works perfectly fine
www.test.com/blog/test-post-for-blog - WP Post, 404
Here is the config:
server {
listen 80;
listen [::]:80;
server_name www.test.com;
location ^~ /media {
alias /var/local/test/static/media;
}
location ^~ /icons {
alias /var/local/test/static/icons;
}
location /blog {
alias /var/www/html;
index index.html index.htm index.php;
try_files $uri $uri/ /blog/index.php?$args /blog/index.php?q=$uri&$args;
# try_files $uri $uri/ /blog/index.php?q=$uri$args;
location ~ \.php$ {
try_files $uri =404;
include fastcgi_params;
fastcgi_intercept_errors on;
fastcgi_index index.php;
fastcgi_pass blog-test:9000;
fastcgi_split_path_info ^(/blog)(/.*)$;
fastcgi_param SCRIPT_FILENAME $request_filename;
fastcgi_param PATH_INFO $fastcgi_path_info;
}
}
}
And the log:
2022/02/09 14:47:25 [error] 32#32: *45 open() "/etc/nginx/html/index.php" failed (2: No such file or directory), client: 10.10.10.10, server: www.test.com, request: "GET /blog/test-post-for-blog/ HTTP/1.1", host: "www.test.com", referrer: "https://www.test.com/blog/"
10.10.10.10 - admin [09/Feb/2022:14:47:25 +0000] "GET /blog/test-post-for-blog/ HTTP/1.1" 404 187 "https://www.test.com/blog/" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/98.0.4758.87 Safari/537.36"
Checking the logs, it seems to look into the wrong place, which is /etc/nginx/html/index.php, but if I add the alias in the php location block, then the homepage will stop working as well, which gets me a bit confused.
Currently, I'm not quite sure if the problem is the alias in the blog block ( nginx recommends not using alias together with try_files, apparently due to a bug ), or anything else. If it is indeed the alias directive, I'll try to add root, together with some rewrite rules to avoid modifying the file structure.
It really took me more than it should to figure it out, and still can not actually understand what might be the issue here.
UPDATE 1 :
This is extremely odd. With the following blog config, separating the php location from the blog, the WP posts are accesible, but not the homepage and admin, returning 404:
location /blog {
alias /var/www/html;
try_files $uri $uri/ /index.php?$args;
}
location ~ \.php$ {
alias /var/www/html;
try_files $uri =404;
include fastcgi_params;
fastcgi_intercept_errors on;
fastcgi_index index.php;
fastcgi_pass blog-test:9000;
fastcgi_split_path_info ^(/blog)(/.*)$;
fastcgi_param SCRIPT_FILENAME $request_filename;
fastcgi_param PATH_INFO $fastcgi_path_info;
}
UPDATE 2 :
Now this works ( the blog and the posts can be accesed ) , but the admin seems to be going into a redirect loop:
location /blog {
alias /var/www/html;
index index.php;
try_files $uri $uri/ /blog/index.php?q=$uri&$args;
}
location ~ \.php$ {
alias /var/www/html;
try_files $uri $uri/ /index.php?$args;
include fastcgi_params;
fastcgi_intercept_errors on;
fastcgi_index index.php;
fastcgi_pass blog-test:9000;
fastcgi_param SCRIPT_FILENAME $request_filename;
}
UPDATE 3 & FIX :
This apparently works, but the config seems to be really unpleasant and unoptimised. If I try to nest the php location into the blog block, then the php files will download instead of rendering. If I try to use alias instead of root, some pages will not show resulting in 404. In any case, this seems to be functional:
root /var/www/html;
index index.php;
location /blog {
rewrite ^/blog(.*)$ /$1 break;
try_files $uri $uri/ /blog/index.php?$args;
}
location ~ \.php$ {
rewrite ^/blog(.*)$ $1 break;
fastcgi_pass blog-test:9000;
fastcgi_index index.php;
include fastcgi_params;
fastcgi_split_path_info ^(/blog)(/.*)$;
fastcgi_param SCRIPT_FILENAME $request_filename;
fastcgi_param PATH_INFO $fastcgi_path_info;
}
Outside the /blog dir you might also have php files. There should be a separate block for the wordpress install that has the same settings, but then for the blog dir.
This could be the block for wordpress (tested in Docker)
location /blog {
try_files $uri $uri/ /blog/index.php?$args;
location ~ \.php$ {
fastcgi_split_path_info ^/blog/(.+\.php)(/.+)$;
fastcgi_pass wp:9000;
fastcgi_index index.php;
include fastcgi_params;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;
}
}
So your complete nginx config would look like this:
server {
listen 80;
listen [::]:80;
server_name www.test.com;
root /var/www/html;
index index.php;
location ^~ /media {
alias /var/local/test/static/media;
}
location ^~ /icons {
alias /var/local/test/static/icons;
}
# This is for Wordpress
location /blog {
try_files $uri $uri/ /blog/index.php?$args;
location ~ \.php$ {
fastcgi_split_path_info ^/blog/(.+\.php)(/.+)$;
fastcgi_pass wp:9000;
fastcgi_index index.php;
include fastcgi_params;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;
}
}
# This is for php files in the root. If there is no php to be parsed there you could leave these blocks out.
location / {
try_files $uri $uri/ /index.php?$args;
}
location ~ \.php$ {
fastcgi_split_path_info ^(.+\.php)(/.+)$;
fastcgi_pass wp:9000;
fastcgi_index index.php;
include fastcgi_params;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;
}
}
UPDATE: I added a root and index property to the nginx server config.
UPDATE 3 & FIX :
This apparently works, but the config seems to be really unpleasant and unoptimised. If I try to nest the php location into the blog block, then the php files will download instead of rendering. If I try to use alias instead of root, some pages will not show resulting in 404. In any case, this seems to be functional:
root /var/www/html;
index index.php;
location /blog {
rewrite ^/blog(.*)$ /$1 break;
try_files $uri $uri/ /blog/index.php?$args;
}
location ~ \.php$ {
rewrite ^/blog(.*)$ $1 break;
fastcgi_pass blog-test:9000;
fastcgi_index index.php;
include fastcgi_params;
fastcgi_split_path_info ^(/blog)(/.*)$;
fastcgi_param SCRIPT_FILENAME $request_filename;
fastcgi_param PATH_INFO $fastcgi_path_info;
}
I have site on joomla and subfolder folder with standalone php files.
location / {
try_files $uri $uri/ /index.php?q=$request_uri;
}
location ~ \.php$ {
include fastcgi_params;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param HTTPS on;
fastcgi_index index.php;
fastcgi_pass backend;
fastcgi_read_timeout 90;
}
Now I want to do something like this:
location ~ /forder/(.*)\.php {
try_files $uri /folder/app.php?controller=$1;
}
When I go to https://site/folder/script.php nginx send source code of that file.
Where I go wrong?
Solution for me is to put
if (!-e $request_filename){
rewrite /folder/(.*)\.php /folder/app.php?controller=$1 last;
}
in server section
Here is my nginx's content.
My current access url is http://localhost/lampi/, and I am receiving response 403 forbidden.
My server's documentroot is /Library/WebServer/Documents/.
When I access http://localhost/, it shows ok. I can also see the index.html page's content.
I don't know what the matter is. I have checked top 10 pages in stackoverflow.
server {
server_name localhost;
access_log /var/log/nginx/nginx.host.access.log main;
root /Library/WebServer/Documents/;
location / {
#root html;
index index.html index.htm index.php;
}
location /lampi {
#autoindex on;
if (!-e $request_filename){
rewrite ^/lampi/(.*)$ /lampi/index.php?s=$1 last;
}
}
location ~ \.php$ {
include /usr/local/etc/nginx/fastcgi_params;
fastcgi_pass 127.0.0.1:9000;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME /Library/WebServer/Documents/lampi/$fastcgi_script_name;
}
location ~ [^/]\.php(/|$) {
fastcgi_split_path_info ^(.+?\.php)(/.*)$;
if (!-f $document_root$fastcgi_script_name) {
return 404;
}
fastcgi_param HTTP_PROXY "";
fastcgi_pass 127.0.0.1:9000;
fastcgi_index index.php;
include fastcgi_params;
}
#location /images/ {
# root /usr/local/var/www;
#}
}
There are three potential problems with your configuration file.
The use of if (!-e $request_filename) is causing problems because it checks for the existence of directories, and you should probably be using try_files anyway (see this document for details):
location /lampi {
try_files $uri $uri/ #lampi;
}
location #lampi {
rewrite ^/lampi/(.*)$ /lampi/index.php?s=$1 last;
}
The value of SCRIPT_FILENAME adds an extra /lampi into the pathname. Use either of (both evaluate to the same value in this case):
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param SCRIPT_FILENAME $request_filename;
For example:
location ~ \.php$ {
try_files $uri =404;
include /usr/local/etc/nginx/fastcgi_params;
fastcgi_pass 127.0.0.1:9000;
fastcgi_param SCRIPT_FILENAME $request_filename;
}
The location ~ [^/]\.php(/|$) block is confusing. Normally it is adequate to use either location ~ \.php$ or something like location ~ [^/]\.php(/|$) depending on whether or not your application uses PATH_INFO. Delete the block you are not using. See this document for details.
Recently I set up a virtual machine through vagrant and I'm using nginx as a web server. The OS I'm using for the VM is Ubuntu Linux 12.04. The problem is that any script I write in PHP gets downloaded, instead of being executed on the browser. Seeing as the PHP interpreter is normally installed I figured I'd post here the config file of nginx, as this is where the problem is likely to be found. Being a novice in the world of web development, I can't figure out what is out of the ordinary, can you guys take a look and tell me if you see something wrong? Thanks.
server {
listen 80;
server_name localhost;
root /vagrant/www/web;
error_log /var/log/nginx/error.log;
access_log /var/log/nginx/access.log;
#strip app.php/ prefix if it is present
rewrite ^/app\.php/?(.*)$ /$1 permanent;
location / {
index app.php;
try_files $uri #rewriteapp;
}
location #rewriteapp {
rewrite ^(.*)$ /app.php/$1 last;
}
# pass the PHP scripts to FastCGI server listening socket
location ~ ^/(app|app_dev)\.php(/|$) {
fastcgi_pass unix://var/run/php5-fpm.sock;
fastcgi_keep_conn on;
fastcgi_split_path_info ^(.+\.php)(/.*)$;
include fastcgi_params;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param HTTPS off;
}
# enable global phpMyAdmin
location /phpmyadmin {
root /usr/share/;
index index.php index.html index.htm;
location ~ ^/phpmyadmin/(.+\.php)$ {
try_files $uri =404;
root /usr/share/;
fastcgi_pass unix://var/run/php5-fpm.sock;
fastcgi_index index.php;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
include /etc/nginx/fastcgi_params;
}
location ~* ^/phpmyadmin/(.+\.(jpg|jpeg|gif|css|png|js|ico|html|xml|txt))$ {
root /usr/share/;
}
}
location /phpMyAdmin {
rewrite ^/* /phpmyadmin last;
}
}
Looks like you're having double // to indicate the fastcgi_pass path of php in your location block, try this instead
location ~ \.php$ {
fastcgi_pass unix:/var/run/php5-fpm.sock;
fastcgi_keep_conn on;
fastcgi_split_path_info ^(.+\.php)(/.*)$;
include fastcgi_params;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param HTTPS off;
}
As was my suspicion the problem was in this location block:
location ~ ^/(app|app_dev)\.php(/|$) {
fastcgi_pass unix://var/run/php5-fpm.sock;
fastcgi_keep_conn on;
fastcgi_split_path_info ^(.+\.php)(/.*)$;
include fastcgi_params;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_param HTTPS off;
}
The parameter to the location directive is a regular expression that defines only two possible file names: app.php and app_dev.php, thus making any other filename non-executable. For this to change, one needs to either add the names of one's php scripts to the parameter or change the regular expression to a more inclusive form as per Frederic Henri's suggestion.
I have nginx set up for Rails application that has been working beautifully for me. Now I want to move my Wordpress blog from blog.website.com to website.com/blog, so web crawlers treat it as part of the site.
I created a symbolic link in my Rails app's public/ directory and added the following to my nginx configuration:
# Rails server
server {
root /project/path/current/public;
server_name project.com;
passenger_enabled on;
rails_env production;
client_max_body_size 20m;
if ($http_x_forwarded_proto != 'https') {
return 301 https://$host$request_uri;
}
location /blog {
index index.html index.php;
try_files $uri $uri/ /blog/index.php?q=$uri&$args;
if (!-e $request_filename) {
rewrite ^.+?(/blog/.*.php)$ $1 last;
rewrite ^ /blog/index.php last;
}
}
location ~ \.php$ {
try_files $uri =404;
fastcgi_intercept_errors on;
include fastcgi_params;
fastcgi_index index.php;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_pass unix:/var/run/php5-fpm.sock;
}
}
This works great.
But what I'd like to do is skip the symbolic link part. I'd like nginx to route directly to the wordpress directory. I tried changing the /blog block to:
location ^~ /blog {
root /home/ubuntu/apps/blog;
index index.html index.php;
try_files $uri $uri/ /blog/index.php?q=$uri&$args;
if (!-e $request_filename) {
rewrite ^.+?(/blog/.*.php)$ $1 last;
rewrite ^ /blog/index.php last;
}
}
This works, as in I'm being correctly redirected to the wordpress folder but my server doesn't know what to do with php files and sends them to my browser as files to download.
I tried nesting the \.php$ location inside the /blog location but it results in 404 error.
What's wrong with the following piece and how to fix it?
location ^~ /blog {
root /home/ubuntu/apps/blog;
index index.html index.php;
try_files $uri $uri/ /blog/index.php?q=$uri&$args;
if (!-e $request_filename) {
rewrite ^.+?(/blog/.*.php)$ $1 last;
rewrite ^ /blog/index.php last;
}
location ~ \.php$ {
try_files $uri =404;
fastcgi_intercept_errors on;
include fastcgi_params;
fastcgi_index index.php;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_pass unix:/var/run/php5-fpm.sock;
}
}
I haven't fixed the snippet from the question but I managed to solve the problem with nginx configuration without using a symbolic link by using two server blocks.
server {
location / {
proxy_pass http://localhost:82/;
}
location /blog {
root /var/proj/blog/current;
fastcgi_intercept_errors on;
include fastcgi_params;
fastcgi_index index.php;
fastcgi_param SCRIPT_NAME $fastcgi_script_name;
fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name;
fastcgi_pass unix:/var/run/php5-fpm.sock;
}
}
# Rails server
server {
listen 82;
root /var/proj/site/current/public;
server_name site.com;
passenger_enabled on;
rails_env staging;
client_max_body_size 20m;
# other configuration
}