Difference between revisions of "Nginx"

Revision as of 12:49, 19 February 2013

zh-CN:Nginx Nginx (pronounced "engine X") written by Igor Sysoev (Russia) in 2005, is a free, open-source, high-performance HTTP server and reverse proxy, as well as an IMAP/POP3 proxy server. According to Netcraft's April 2012 Web Server Survey, Nginx now hosts 10.32% of all domains worldwide, while Apache hosts about 65.46%. Nginx is now well known for its stability, rich feature set, simple configuration, and low resource consumption.

Installation

Install package nginx in the official repositories.

For a Ruby on Rails oriented installation, see The Perfect Rails Setup.

Installation in a chroot

Installing Nginx in a chroot adds an additional layer of security. For maximum security the chroot should include only the files needed to run the Nginx server and all files should have the most restrictive permissions possible, e.g., as much as possible should be owned by root, directories such as /usr/bin should be unreadable and unwriteable, etc.

Arch comes with an http user and group by default which will run the server. The chroot will be in /srv/http.

A perl script to create this jail is available at jail.pl gist. It expects to be run as root. You'll need to uncomment a line before it makes any changes.

Create Necessary Devices

Nginx needs /dev/null, /dev/random, and /dev/urandom. To install these in the chroot we create the /dev/ folder and add the devices with mknod. We avoid mounting all of /dev/ to ensure that, even if the chroot is compromised, an attacker must break out of the chroot to access important devices like /dev/sda1.

Tip: See man mknod and

ls -l
/dev/{null,random,urandom}

to better understand the argument to mknod.

# export JAIL=/srv/http
# mkdir $JAIL/dev
# mknod -m 0666 $JAIL/dev/null c 1 3
# mknod -m 0666 $JAIL/dev/random c 1 8
# mknod -m 0444 $JAIL/dev/urandom c 1 9

Create Necessary Folders

Nginx requires a bunch of files to run properly. Before copying them over, create the folders to store them. This assumes your Nginx document root will be /srv/http/www.

# mkdir -p $JAIL/etc/nginx/logs
# mkdir -p $JAIL/usr/{lib,sbin}
# mkdir -p $JAIL/usr/share/nginx
# mkdir -p $JAIL/var/{log,lib}/nginx
# mkdir -p $JAIL/www/cgi-bin
# mkdir -p $JAIL/{run,tmp}
# cd $JAIL; ln -s usr/lib lib

Note: If using a 64 bit kernel you will need to create $JAIL/usr/lib64 and a symbolic link to this directory: cd $JAIL; ln -s usr/lib64 lib64. The 64bit libraries will need to be copied in a later section

Then mount $JAIL/tmp and $JAIL/run as tmpfs's. The size should be limited to ensure an attacker cannot eat all the RAM.

# mount -t tmpfs none $JAIL/run -o 'noexec,size=1M'
# mount -t tmpfs none $JAIL/tmp -o 'noexec,size=100M'

Populate the chroot

First copy over the easy files.

# cp -r /usr/share/nginx/* $JAIL/usr/share/nginx
# cp -r /usr/share/nginx/html/* $JAIL/www
# cp /usr/sbin/nginx $JAIL/usr/sbin/
# cp -r /var/lib/nginx $JAIL/var/lib/nginx

Now copy over required libraries. Use ldd to list them and then copy them all to the correct location. Copying is preferred over hardlinks to ensure that even if an attacker gains write access to the files they cannot destroy or alter the true system files.

$ ldd /usr/sbin/nginx

   linux-vdso.so.1 (0x00007fff5adff000)
   libpthread.so.0 => /usr/lib/libpthread.so.0 (0x00007fdb1a8d3000)
   libcrypt.so.1 => /usr/lib/libcrypt.so.1 (0x00007fdb1a69c000)
   libstdc++.so.6 => /usr/lib/libstdc++.so.6 (0x00007fdb1a399000)
   libm.so.6 => /usr/lib/libm.so.6 (0x00007fdb1a09f000)
   libpcre.so.1 => /usr/lib/libpcre.so.1 (0x00007fdb19e3c000)
   libssl.so.1.0.0 => /usr/lib/libssl.so.1.0.0 (0x00007fdb19bd2000)
   libcrypto.so.1.0.0 => /usr/lib/libcrypto.so.1.0.0 (0x00007fdb197c9000)
   libdl.so.2 => /usr/lib/libdl.so.2 (0x00007fdb195c5000)
   libz.so.1 => /usr/lib/libz.so.1 (0x00007fdb193af000)
   libGeoIP.so.1 => /usr/lib/libGeoIP.so.1 (0x00007fdb1917c000)
   libgcc_s.so.1 => /usr/lib/libgcc_s.so.1 (0x00007fdb18f67000)
   libc.so.6 => /usr/lib/libc.so.6 (0x00007fdb18bc0000)
   /lib/ld-linux-x86-64.so.2 (0x00007fdb1aaef000)

# cp /usr/lib/libpthread.so.0 $JAIL/usr/lib/
# cp /usr/lib/libcrypt.so.1 $JAIL/usr/lib
...
# cp /lib/ld-linux-x86-64.so.2 $JAIL/lib

Copy over some misc. but necessary libraries and system files.

# cp /usr/lib/libnss_* $JAIL/usr/lib
# cp -rfvL /etc/{services,localtime,nsswitch.conf,nscd.conf,protocols,hosts,ld.so.cache,ld.so.conf,resolv.conf,host.conf,nginx} $JAIL/etc

Create restricted user/group files for the chroot. This way only the users needed for the chroot to function exist as far as the chroot knows, and none of the system users/groups are leaked to attackers should they gain access to the chroot.

$JAIL/etc/group

http:x:33:
nobody:x:99:

$JAIL/etc/passwd

http:x:33:33:http:/:/bin/false
nobody:x:99:99:nobody:/:/bin/false

$JAIL/etc/shadow

http:x:14871::::::
nobody:x:14871::::::

$JAIL/etc/gshadow

http:::
nobody:::

# touch $JAIL/etc/shells
# touch $JAIL/run/nginx.pid

Finally make set very restrictive permissions. As much as possible should be owned by root and set unwritable.

# chown -R root:root $JAIL/
 
# chown -R #USER:$GROUP $JAIL/www
# chown -R #USER:$GROUP $JAIL/etc/nginx
# chown -R #USER:$GROUP $JAIL/var/{log,lib}/nginx
# chown #USER:$GROUP $JAIL/run/nginx.pid
 
# find $JAIL/ -gid 0 -uid 0 -type d -print | xargs chmod -rw
# find $JAIL/ -gid 0 -uid 0 -type d -print | xargs chmod +x
# find $JAIL/etc -gid 0 -uid 0 -type f -print | xargs chmod -x
# find $JAIL/usr/sbin -type f -print | xargs chmod ug+rx
# find $JAIL/ -group http -user http -print | xargs chmod o-rwx
# chmod +rw $JAIL/tmp
# chmod +rw $JAIL/run

If your server will bind port 80 (or any port 0-1024), give the chrooted executable permission to bind these ports without root.

# setcap 'cap_net_bind_service=+ep' $JAIL/usr/sbin/nginx

Modify nginx.service to start chroot

The systemd unit must be changed to start up Nginx in the chroot, as the http user, and store the pid file in the chroot

Note: I'm not sure if the pid file needs to be stored in the root.

/usr/lib/systemd/system/nginx.service

 [Unit]
 Description=A high performance web server and a reverse proxy server
 After=syslog.target network.target
 
 [Service]
 Type=forking
 PIDFile=/srv/http/run/nginx.pid
 ExecStartPre=/usr/bin/chroot --userspec=http:http /srv/http /usr/sbin/nginx -t -q -g 'pid /run/nginx.pid; daemon on; master_process on;'
 ExecStart=/usr/bin/chroot --userspec=http:http /srv/http /usr/sbin/nginx -g 'pid /run/nginx.pid; daemon on; master_process on;'
 ExecReload=/usr/bin/chroot --userspec=http:http /srv/http /usr/sbin/nginx -g 'pid /run/nginx.pid; daemon on; master_process on;' -s reload
 ExecStop=/usr/bin/chroot --userspec=http:http /srv/http /usr/sbin/nginx -g 'pid /run/nginx.pid;' -s quit
 
 [Install]
 WantedBy=multi-user.target

Starting Service

To start the Nginx service, run:

# rc.d start nginx

The default served page at http://127.0.0.1 is:

/usr/share/nginx/html/index.html

To enable the Nginx service by default at start-up just add nginx to the DAEMONS array in /etc/rc.conf:

DAEMONS=(ntpd syslog-ng ... nginx)

Using Systemd

To start the Nginx service, run:

# systemctl start nginx

To enable the Nginx service by default at start-up, run:

# systemctl enable nginx

Configuring

You can modify the configuration by editing the files in /etc/nginx/. The main configuration file is located at /etc/nginx/nginx.conf.

More details can be found here: Nginx Configuration Examples.

FastCGI

FastCGI, also FCGI, is a protocol for interfacing interactive programs with a web server. FastCGI is a variation on the earlier Common Gateway Interface (CGI); FastCGI's main aim is to reduce the overhead associated with interfacing the web server and CGI programs, allowing a server to handle more web page requests at once.

FastCGI technology is introduced into Nginx to work with many external tools, i.e.: Perl, PHP and Python. So, you cannot use these unless a FastCGI server has been started.

PHP implementation

There are different ways to run a FastCGI server for PHP. We cover php-fpm, a recommended solution.

Step 1: PHP configuration

The open_basedir in /etc/php/php.ini has to list base directories which contain PHP files, like /srv/http/ and /usr/share/webapps/:

open_basedir = /usr/share/webapps/:/srv/http/:/home/:/tmp/:/usr/share/pear/

Step 2: php-fpm

http://php-fpm.org

Install php-fpm. The configuration file is /etc/php/php-fpm.conf. Enable and start the systemd service.

Step 3: Nginx configuration

Inside each server block serving a PHP web application should appear a location block similar to:

 location ~ \.php$ {
      fastcgi_pass   unix:/run/php-fpm/php-fpm.sock;
      fastcgi_index  index.php;
      include        fastcgi.conf;
 }

You could create /etc/nginx/php.conf and save this configuration there, then when needed just include this file into the server block.

 server = {
     ...
     include  php.conf;
     ...
 }

If you are going to process .html and .htm files with PHP, you should have something like this:

 location ~ \.(php|html|htm)$ {
      fastcgi_pass   unix:/var/run/php-fpm/php-fpm.sock;
      fastcgi_index  index.php;
      include        fastcgi.conf;
 }

Non .php files processing in php-fpm should be explicitly enabled in /etc/php/php-fpm.conf:

 security.limit_extensions = .php .html .htm

You need to restart the php-fpm daemon if you changed the configuration.

# systemctl restart php-fpm

Pay attention to the fastcgi_pass argument, as it must be the TCP or Unix socket defined by the chosen FastCGI server in its config file. The default Unix for php-fpm is

fastcgi_pass unix:/run/php-fpm/php-fpm.sock;

Or you may use the common TCP socket, not default,

fastcgi_pass 127.0.0.1:9000;

Unix domain sockets are however much better.

fastcgi.conf or fastcgi_params are usually included because they hold FastCGI settings for Nginx; the use of the latter is deprecated, though. They come within the Nginx installation.

Finally, if Nginx has been working, run:

# systemctl restart nginx

If you would like to test the FastCGI implementation, create /usr/share/nginx/html/index.php with content

<?php
  phpinfo();
?>

and visit the URL http://127.0.0.1/index.php with your browser.

CGI implementation

This implementation is needed for CGI applications.

Install fcgiwrap.

Systemd

The systemd unit file is currently being discussed on this ArchLinux task page. You may want to examine the unit file yourself to ensure it will work the way you want. Copy the unit file from /usr/lib/systemd/system/fcgiwrap.service to /etc/systemd/system/fcgiwrap.service (and the fcgiwrap.socket unit, if present), and modify the ExecStart line to suit your needs.

If you want to spawn multiple worker threads, it's recommended that you use multiwatch^AUR, which will take care of restarting crashed children. You'll need to use spawn-fcgi to create the unix socket, as multiwatch seems unable to handle the systemd-created socket, even though fcgiwrap itself doesn't have any trouble if invoked directly in the unit file. Here is a unit file that uses multiwatch^AUR. Make sure fcgiwrap.socket is not started or enabled, because it will conflict with this unit:

/etc/systemd/system/fcgiwrap.service

[Unit]
Description=Simple CGI Server
After=nss-user-lookup.target

[Service]
ExecStart=/usr/bin/spawn-fcgi -u http -g http -s /run/fcgiwrap.sock -n -- /usr/bin/multiwatch -f 10 -- /usr/sbin/fcgiwrap
ExecStartPost=/usr/bin/chmod 660 /run/fcgiwrap.sock
PrivateTmp=true
Restart=on-failure

[Install]
WantedBy=multi-user.target

Tweak -f 10 to change the number of children that are spawned.

Warning: The ExecStartPost line is required because of strange behaviour I'm seeing when I use the -M 660 option for spawn-fcgi. The wrong mode is set. This may be a bug?

Initscripts

Using a Unix Domain Socket

The default configuration sets TCP socket 127.0.0.1:9001 for listening. It is possible to alter the configuration to use a Unix domain socket, which will be more efficient.

If you desire to use a Unix domain socket, edit /etc/conf.d/fcgiwrap like this:

SPAWNER='/usr/bin/spawn-fcgi'

FCGI_SOCKET='/run/fcgiwrap.sock'
FCGI_USER='http'
FCGI_GROUP='http'
FCGI_EXTRA_OPTIONS='-M 660'

SPAWNER_ARGS="-u $FCGI_USER -g $FCGI_GROUP -s $FCGI_SOCKET $FCGI_EXTRA_OPTIONS -- /usr/sbin/fcgiwrap"

Start the service:

# rc.d start fcgiwrap

and add fcgiwrap to the DAEMONS array in /etc/rc.conf.

Multiple Workers

You can increase the number of worker threads by adding the -F <num> option to the SPAWNER_ARGS variable in /etc/conf.d/fcgiwrap. Although this option will work, you may find it difficult to monitor the workers, because they have no parent process.

A better way to handle multiple worker threads is to use the multiwatch package from the AUR. After installing the package, perform the following configuration to achieve this:

Add the following line near the top of the /etc/conf.d/fcgiwrap file:

DAEMON='multiwatch'

and modify the SPAWNER_ARGS line:

FCGI_CHILDREN=5
SPAWNER_ARGS="-u $FCGI_USER -g $FCGI_GROUP -s $FCGI_SOCKET $FCGI_EXTRA_OPTIONS -- /usr/bin/multiwatch -f $FCGI_CHILDREN -- /usr/sbin/fcgiwrap"

You'll find that the /etc/rc.d/fcgiwrap script now creates a /run/multiwatch.pid file instead of the old /run/fcgiwrap.pid file. The multiwatch daemon will take care of respawning children that die.

Nginx Configuration

Inside each server block serving a PHP web application should appear a location block similar to:

 location ~ \.cgi$ {
      fastcgi_pass   unix:/run/fcgiwrap.sock;
      include        fastcgi.conf;
 }
   
 location ~ \.pl$ {
      fastcgi_pass   unix:/run/fcgiwrap.sock;
      include        fastcgi.conf;
 }

Troubleshooting

Accessing local IP redirects to localhost

Solution from the Arch Linux forum.

Edit /etc/nginx/nginx.conf and locate the "server_name localhost" line without a # infront of it, and add below:

server_name_in_redirect off;

Default behavior is that nginx redirects any requests to the value given as server_name in the config.

Error: 403 (Permission error)

This is most likely a permission error. Are you sure whatever user configured in the Nginx configuration is able to read the correct files?

If the files are located within a home directory, (e.g. /home/arch/public/webapp) and you are sure the user running Nginx has the right permissions (you can temporarily chmod all the files to 777 in order to determine this), /home/arch might be chmod 750, simply chmod it to 751, and it should work.

If you have changed your document root

If you are sure that permissions are as they should be, make sure that your document root directory is not empty. Try creating index.html in there.

Error: 404 (Pathinfo error)

In some framework (like thinkphp, cakephp) or CMS, they need the pathinfo function.

1. Edit the file /etc/php/php.ini, make sure

cgi.fix_pathinfo=1

2. Edit /etc/nginx/conf/nginx.conf, comment

location ~ \.php$ {
...
}

to

#location ~ \.php$ {
#...
#}

Then add the follows,

location ~ ^(.+\.php)(.*)$ {
  root   /srv/http/nginx;
  fastcgi_pass   unix:/var/run/php-fpm/php-fpm.sock; 	
  #fastcgi_pass   127.0.0.1:9000; #Un-comment this and comment "fastcgi_pass   unix:/var/run/php-fpm/php-fpm.sock;" if you are not using php-fpm.
  fastcgi_index  index.php;
  set $document_root2 $document_root;
  if ($document_root2 ~ "^(.*\\\\).*?[\\\\|\/]\.\.\/(.*)$") { set $document_root2 $1$2; }
  if ($document_root2 ~ "^(.*\\\\).*?[\\\\|\/]\.\.\/(.*)$") {	set $document_root2 $1$2; }
  if ($document_root2 ~ "^(.*\\\\).*?[\\\\|\/]\.\.\/(.*)$") {	set $document_root2 $1$2; }
  if ($document_root2 ~ "^(.*\\\\).*?[\\\\|\/]\.\.\/(.*)$") {	set $document_root2 $1$2; }
  if ($document_root2 ~ "^(.*\\\\).*?[\\\\|\/]\.\.\/(.*)$") {	set $document_root2 $1$2; }
  fastcgi_split_path_info ^(.+\.php)(.*)$;
  fastcgi_param	SCRIPT_FILENAME	$document_root2$fastcgi_script_name;
  fastcgi_param	PATH_INFO	$fastcgi_path_info;
  fastcgi_param	PATH_TRANSLATED	$document_root2$fastcgi_path_info;
  include	fastcgi_params;
  fastcgi_param  DOCUMENT_ROOT      $document_root2;
}

Error: The page you are looking for is temporarily unavailable. Please try again later.

This is because the FastCGI server has not been started, or the socket used has wrong permissions.

Error: No input file specified

Most Likely you do not have the SCRIPT_FILENAME containing the full path to you scripts. If the configuration of nginx (fastcgi_param SCRIPT_FILENAME) is all right, this kind of error means php fail to load the requestd script. Usually it is simply a permissions issue, you can just run php-cgi as root

# spawn-fcgi -a 127.0.0.1 -p 9000 -f /usr/bin/php-cgi

or you should create some group and user to start the php-cgi. For example:

# groupadd www
# useradd -g www www
# chmod +w /srv/www/nginx/html
# chown -R www:www /srv/www/nginx/html
# spawn-fcgi -a 127.0.0.1 -p 9000 -u www -g www -f /usr/bin/php-cgi

Another occasion is that, wrong "root" argument in the "location ~ \.php$" section in nginx.conf, make sure the "root" points to the same directory as it in "location /" in the same server. Or you may just set root as global, do not define it in any location section.

Also keep in mind that your php script path was defined as /srv/www/nginx/html by default using the variable "open_basedir" in /etc/php/php.ini; you can change them if you need.

Also notice that not only php script should have read permission, but also the entire directory structure should have execute permission so that PHP user can traverse the path.