Nginx, Inc.

NGINX Plus Reference Guide

NGINX Plus - release 6, based on 1.7.11 core
April 8, 2015
Copyright Notice
2012-2015 Nginx, Inc. All rights reserved. NGINX, NGINX Plus and any
Nginx, Inc. product or service name or logo used herein are trademarks of Nginx, Inc.
All other trademarks used herein belong to their respective owners. The trademarks
and logos displayed herein may not be used without the prior written consent of
Nginx, Inc. or their respective owners.
This documentation is provided AS IS and is subject to change without notice
and should not be interpreted as a commitment by Nginx, Inc. This documentation
may not be copied, modified or distributed without authorization of Nginx, Inc. and
may be used only in connection with Nginx, Inc. products and services. Nginx, Inc.
assumes no responsibility or liability for any errors or inaccuracies that may appear
in this documentation.

1
Preface

About NGINX
NGINX (engine x) is a high performance, high concurrency web server
excelling at large scale content delivery, web acceleration and protecting
application containers. Its precise integration with modern operating systems
allows unprecedented levels of efficiency even when running on commodity
hardware.
Nginx, Inc. develops and maintains NGINX open source distribution, and
offers commercial support and professional services for NGINX.

About NGINX Plus

Offers additional features on top of the free open source NGINX version.

Prepared, tested and supported by NGINX core engineering team led by

the original author Igor Sysoev.

For more information

Find more details about NGINX products and support at
http://nginx.com.

For online NGINX documentation visit http://nginx.org/en/docs.

NGINX and NGINX Plus Tutorial and Admin Guide is available here:
http://nginx.com/resources/admin-guide.

For general inquiries, please use: [email protected]

2
Contents

Title 1

Preface 2

Table of Contents 3

1 Core modules 6
1.1 Core functionality . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2 Setting up hashes . . . . . . . . . . . . . . . . . . . . . . . . . . 16
1.3 Connection processing methods . . . . . . . . . . . . . . . . . . 17
1.4 Logging to syslog . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2 HTTP server modules 19

2.1 Module ngx http core module . . . . . . . . . . . . . . . . . . . 19
2.2 Module ngx http access module . . . . . . . . . . . . . . . . . . 55
2.3 Module ngx http addition module . . . . . . . . . . . . . . . . . 57
2.4 Module ngx http auth basic module . . . . . . . . . . . . . . . 59
2.5 Module ngx http auth request module . . . . . . . . . . . . . . 61
2.6 Module ngx http autoindex module . . . . . . . . . . . . . . . . 63
2.7 Module ngx http browser module . . . . . . . . . . . . . . . . . 65
2.8 Module ngx http charset module . . . . . . . . . . . . . . . . . 67
2.9 Module ngx http dav module . . . . . . . . . . . . . . . . . . . 70
2.10 Module ngx http empty gif module . . . . . . . . . . . . . . . . 73
2.11 Module ngx http f4f module . . . . . . . . . . . . . . . . . . . . 74
2.12 Module ngx http fastcgi module . . . . . . . . . . . . . . . . . . 75
2.13 Module ngx http flv module . . . . . . . . . . . . . . . . . . . . 93
2.14 Module ngx http geo module . . . . . . . . . . . . . . . . . . . 94
2.15 Module ngx http geoip module . . . . . . . . . . . . . . . . . . 97
2.16 Module ngx http gunzip module . . . . . . . . . . . . . . . . . 100
2.17 Module ngx http gzip module . . . . . . . . . . . . . . . . . . . 101
2.18 Module ngx http gzip static module . . . . . . . . . . . . . . . 105
2.19 Module ngx http headers module . . . . . . . . . . . . . . . . . 106
2.20 Module ngx http hls module . . . . . . . . . . . . . . . . . . . . 108
2.21 Module ngx http image filter module . . . . . . . . . . . . . . . 112
2.22 Module ngx http index module . . . . . . . . . . . . . . . . . . 115
2.23 Module ngx http limit conn module . . . . . . . . . . . . . . . 116
2.24 Module ngx http limit req module . . . . . . . . . . . . . . . . 119

3
CONTENTS CONTENTS

2.25 Module ngx http log module . . . . . . . . . . . . . . . . . . . . 122

2.26 Module ngx http map module . . . . . . . . . . . . . . . . . . . 126
2.27 Module ngx http memcached module . . . . . . . . . . . . . . . 129
2.28 Module ngx http mp4 module . . . . . . . . . . . . . . . . . . . 133
2.29 Module ngx http perl module . . . . . . . . . . . . . . . . . . . 136
2.30 Module ngx http proxy module . . . . . . . . . . . . . . . . . . 142
2.31 Module ngx http random index module . . . . . . . . . . . . . 167
2.32 Module ngx http realip module . . . . . . . . . . . . . . . . . . 168
2.33 Module ngx http referer module . . . . . . . . . . . . . . . . . . 170
2.34 Module ngx http rewrite module . . . . . . . . . . . . . . . . . 172
2.35 Module ngx http scgi module . . . . . . . . . . . . . . . . . . . 178
2.36 Module ngx http secure link module . . . . . . . . . . . . . . . 193
2.37 Module ngx http session log module . . . . . . . . . . . . . . . 196
2.38 Module ngx http spdy module . . . . . . . . . . . . . . . . . . . 198
2.39 Module ngx http split clients module . . . . . . . . . . . . . . . 200
2.40 Module ngx http ssi module . . . . . . . . . . . . . . . . . . . . 201
2.41 Module ngx http ssl module . . . . . . . . . . . . . . . . . . . . 206
2.42 Module ngx http status module . . . . . . . . . . . . . . . . . . 216
2.43 Module ngx http stub status module . . . . . . . . . . . . . . . 224
2.44 Module ngx http sub module . . . . . . . . . . . . . . . . . . . 226
2.45 Module ngx http upstream module . . . . . . . . . . . . . . . . 228
2.46 Module ngx http upstream conf module . . . . . . . . . . . . . 242
2.47 Module ngx http userid module . . . . . . . . . . . . . . . . . . 246
2.48 Module ngx http uwsgi module . . . . . . . . . . . . . . . . . . 249
2.49 Module ngx http xslt module . . . . . . . . . . . . . . . . . . . 268

3 Stream proxy modules 271

3.1 Module ngx stream core module . . . . . . . . . . . . . . . . . 271
3.2 Module ngx stream proxy module . . . . . . . . . . . . . . . . . 275
3.3 Module ngx stream ssl module . . . . . . . . . . . . . . . . . . 281
3.4 Module ngx stream upstream module . . . . . . . . . . . . . . . 285

4 Mail server modules 291

4.1 Module ngx mail core module . . . . . . . . . . . . . . . . . . . 291
4.2 Module ngx mail auth http module . . . . . . . . . . . . . . . . 296
4.3 Module ngx mail proxy module . . . . . . . . . . . . . . . . . . 300
4.4 Module ngx mail ssl module . . . . . . . . . . . . . . . . . . . . 302
4.5 Module ngx mail imap module . . . . . . . . . . . . . . . . . . 308
4.6 Module ngx mail pop3 module . . . . . . . . . . . . . . . . . . 309
4.7 Module ngx mail smtp module . . . . . . . . . . . . . . . . . . 310

5 Miscellaneous 311
5.1 High Availability support for NGINX Plus . . . . . . . . . . . . 311

A Changelog for NGINX Plus 317

B Legal Notices 320

Nginx, Inc. p.4 of 330

CONTENTS CONTENTS

Index 325

Nginx, Inc. p.5 of 330

Chapter 1

Core modules

1.1 Core functionality

1.1.1 Example Configuration . . . . . . . . . . . . . . . . . . . 7
1.1.2 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 7
accept mutex . . . . . . . . . . . . . . . . . . . . . . . . 7
accept mutex delay . . . . . . . . . . . . . . . . . . . . . 7
daemon . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
debug connection . . . . . . . . . . . . . . . . . . . . . . 8
debug points . . . . . . . . . . . . . . . . . . . . . . . . 8
error log . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
env . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
events . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
include . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
lock file . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
master process . . . . . . . . . . . . . . . . . . . . . . . 10
multi accept . . . . . . . . . . . . . . . . . . . . . . . . . 10
pcre jit . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
pid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
ssl engine . . . . . . . . . . . . . . . . . . . . . . . . . . 11
thread pool . . . . . . . . . . . . . . . . . . . . . . . . . 11
timer resolution . . . . . . . . . . . . . . . . . . . . . . . 12
use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
user . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
worker aio requests . . . . . . . . . . . . . . . . . . . . . 12
worker connections . . . . . . . . . . . . . . . . . . . . . 13
worker cpu affinity . . . . . . . . . . . . . . . . . . . . . 13
worker priority . . . . . . . . . . . . . . . . . . . . . . . 13
worker processes . . . . . . . . . . . . . . . . . . . . . . 14
worker rlimit core . . . . . . . . . . . . . . . . . . . . . . 14
worker rlimit nofile . . . . . . . . . . . . . . . . . . . . . 14
worker rlimit sigpending . . . . . . . . . . . . . . . . . . 14
working directory . . . . . . . . . . . . . . . . . . . . . . 15

6
CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

1.1.1 Example Configuration

user www www ;

w or k er _ pr o ce s se s 2;

error_log / var / log / nginx - error . log info ;

events {
use kqueue ;
w o r k e r _ c o n n e c t i o n s 2048;
}

...

1.1.2 Directives
accept mutex
Syntax: accept_mutex on | off;
Default on
Context: events

If accept_mutex is enabled, worker processes will accept new connections

by turn. Otherwise, all worker processes will be notified about new
connections, and if volume of new connections is low, some of the worker
processes may just waste system resources.

The use of rtsig connection processing method requires accept_mutex to

be enabled.

accept mutex delay

Syntax: accept_mutex_delay time;
Default 500ms
Context: events

If accept mutex is enabled, specifies the maximum time during which a

worker process will try to restart accepting new connections if another worker
process is currently accepting new connections.

daemon
Syntax: daemon on | off;
Default on
Context: main

Determines whether nginx should become a daemon. Mainly used during

development.

Nginx, Inc. p.7 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

debug connection
Syntax: debug_connection address | CIDR | unix:;
Default
Context: events

Enables debugging log for selected client connections. Other connections

will use logging level set by the error log directive. Debugged connections
are specified by IPv4 or IPv6 (1.3.0, 1.2.1) address or network. A connection
may also be specified using a hostname. For connections using UNIX-domain
sockets (1.3.0, 1.2.1), debugging log is enabled by the unix: parameter.

events {
d eb u g_ c on n ec t io n 127.0.0.1;
d eb u g_ c on n ec t io n localhost ;
d eb u g_ c on n ec t io n 192.0.2.0/24;
d eb u g_ c on n ec t io n ::1;
d eb u g_ c on n ec t io n 2001:0 db8 ::/32;
d eb u g_ c on n ec t io n unix :;
...
}

For this directive to work, nginx needs to be built with --with-debug,

see A debugging log.

debug points
Syntax: debug_points abort | stop;
Default
Context: main

This directive is used for debugging.

When internal error is detected, e.g. the leak of sockets on restart of
working processes, enabling debug_points leads to a core file creation (abort)
or to stopping of a process (stop) for further analysis using a system debugger.

error log
Syntax: error_log file | stderr | syslog:server=address[,parameter=value] |
memory:size [debug | info | notice | warn | error | crit | alert |
emerg];
Default logs/error.log error
Context: main, http, stream, server, location

Configures logging. Several logs can be specified on the same level (1.5.2).
The first parameter defines a file that will store the log.
The special value stderr selects the standard error file. Logging to syslog
can be configured by specifying the syslog: prefix. Logging to a cyclic
memory buffer can be configured by specifying the memory: prefix and buffer
size, and is generally used for debugging (1.7.11).

Nginx, Inc. p.8 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

The second parameter determines the level of logging. Log levels above
are listed in the order of increasing severity. Setting a certain log level will
cause all messages of the specified and more severe log levels to be logged. For
example, the default level error will cause error, crit, alert, and emerg
messages to be logged. If this parameter is omitted then error is used.

For debug logging to work, nginx needs to be built with --with-debug,

see A debugging log.

The directive can be specified on the stream level starting from version
1.7.11.

env
Syntax: env variable[=value];
Default TZ
Context: main

By default, nginx removes all environment variables inherited from its

parent process except the TZ variable. This directive allows preserving some
of the inherited variables, changing their values, or creating new environment
variables. These variables are then:
inherited during a live upgrade of an executable file;

used by the ngx http perl module module;

used by worker processes. One should bear in mind that controlling

system libraries in this way is not always possible as it is common for
libraries to check variables only during initialization, well before they can
be set using this directive. An exception from this is an above mentioned
live upgrade of an executable file.
The TZ variable is always inherited and available to the ngx http perl -
module module, unless it is configured explicitly.
Usage example:

env MALLOC_OPTIONS ;
env PERL5LIB =/ data / site / modules ;
env O P E N S S L _ A L L O W _ P R O X Y _ C E R T S =1;

The NGINX environment variable is used internally by nginx and should

not be set directly by the user.

events
Syntax: events { . . . }
Default
Context: main

Nginx, Inc. p.9 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

Provides the configuration file context in which the directives that affect
connection processing are specified.

include
Syntax: include file | mask;
Default
Context: any

Includes another file, or files matching the specified mask, into

configuration. Included files should consist of syntactically correct directives
and blocks.
Usage example:

include mime . types ;

include vhosts /*. conf ;

lock file
Syntax: lock_file file;
Default logs/nginx.lock
Context: main

nginx uses the locking mechanism to implement accept mutex and serialize
access to shared memory. On most systems the locks are implemented using
atomic operations, and this directive is ignored. On other systems the lock
file mechanism is used. This directive specifies a prefix for the names of lock
files.

master process
Syntax: master_process on | off;
Default on
Context: main

Determines whether worker processes are started. This directive is intended

for nginx developers.

multi accept
Syntax: multi_accept on | off;
Default off
Context: events

If multi_accept is disabled, a worker process will accept one new

connection at a time. Otherwise, a worker process will accept all new
connections at a time.

Nginx, Inc. p.10 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

The directive is ignored if kqueue connection processing method is used,

because it reports the number of new connections waiting to be accepted.

The use of rtsig connection processing method automatically enables

multi_accept.

pcre jit
Syntax: pcre_jit on | off;
Default off
Context: main
This directive appeared in version 1.1.12.

Enables or disables the use of just-in-time compilation (PCRE JIT) for

the regular expressions known by the time of configuration parsing.
PCRE JIT can speed up processing of regular expressions significantly.

The JIT is available in PCRE libraries starting from version 8.20 built
with the --enable-jit configuration parameter. When the PCRE library
is built with nginx (--with-pcre=), the JIT support is enabled via the
--with-pcre-jit configuration parameter.

pid
Syntax: pid file;
Default nginx.pid
Context: main

Defines a file that will store the process ID of the main process.

ssl engine
Syntax: ssl_engine device;
Default
Context: main

Defines the name of the hardware SSL accelerator.

thread pool
Syntax: thread_pool name threads=number [max_queue=number];
Default default threads=32 max_queue=65536
Context: main
This directive appeared in version 1.7.11.

Defines named thread pools used for multi-threaded reading and sending
of files without blocking worker processes.
The threads parameter defines the number of threads in the pool.

Nginx, Inc. p.11 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

In the event that all threads in the pool are busy, a new task will wait in
the queue. The max_queue parameter limits the number of tasks allowed to
be waiting in the queue. By default, up to 65536 tasks can wait in the queue.
When the queue overflows, the task is completed with an error.

timer resolution
Syntax: timer_resolution interval;
Default
Context: main

Reduces timer resolution in worker processes, thus reducing the number

of gettimeofday system calls made. By default, gettimeofday is called each
time a kernel event is received. With reduced resolution, gettimeofday is only
called once per specified interval.
Example:

t im e r_ r es o lu t io n 100 ms ;

Internal implementation of the interval depends on the method used:

the EVFILT_TIMER filter if kqueue is used;

timer_create if eventport is used;

setitimer otherwise.

use
Syntax: use method;
Default
Context: events

Specifies the connection processing method to use. There is normally no

need to specify it explicitly, because nginx will by default use the most efficient
method.

user
Syntax: user user [group];
Default nobody nobody
Context: main

Defines user and group credentials used by worker processes. If group is

omitted, a group whose name equals that of user is used.

worker aio requests

Syntax: worker_aio_requests number;
Default 32
Context: events

Nginx, Inc. p.12 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

This directive appeared in versions 1.1.4 and 1.0.7.

When using aio with the epoll connection processing method, sets the
maximum number of outstanding asynchronous I/O operations for a single
worker process.

worker connections
Syntax: worker_connections number;
Default 512
Context: events

Sets the maximum number of simultaneous connections that can be opened

by a worker process.
It should be kept in mind that this number includes all connections (e.g.
connections with proxied servers, among others), not only connections with
clients. Another consideration is that the actual number of simultaneous
connections cannot exceed the current limit on the maximum number of open
files, which can be changed by worker rlimit nofile.

worker cpu affinity

Syntax: worker_cpu_affinity cpumask . . . ;
Default
Context: main

Binds worker processes to the sets of CPUs. Each CPU set is represented
by a bitmask of allowed CPUs. There should be a separate set defined for each
of the worker processes. By default, worker processes are not bound to any
specific CPUs.
For example,

w or k er _ pr o ce s se s 4;
w o r k e r _ c p u _ a f f i n i t y 0001 0010 0100 1000;

binds each worker process to a separate CPU, while

w or k er _ pr o ce s se s 2;
w o r k e r _ c p u _ a f f i n i t y 0101 1010;

binds the first worker process to CPU0/CPU2, and the second worker
process to CPU1/CPU3. The second example is suitable for hyper-threading.

The directive is only available on FreeBSD and Linux.

worker priority
Syntax: worker_priority number;
Default 0
Context: main

Nginx, Inc. p.13 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

Defines the scheduling priority for worker processes like it is done by the
nice command: a negative number means higher priority. Allowed range
normally varies from -20 to 20.
Example:

wo rker _pri orit y -10;

worker processes
Syntax: worker_processes number | auto;
Default 1
Context: main

Defines the number of worker processes.

The optimal value depends on many factors including (but not limited to)
the number of CPU cores, the number of hard disk drives that store data, and
load pattern. When one is in doubt, setting it to the number of available CPU
cores would be a good start (the value auto will try to autodetect it).

The auto parameter is supported starting from versions 1.3.8 and 1.2.5.

worker rlimit core

Syntax: worker_rlimit_core size;
Default
Context: main

Changes the limit on the largest size of a core file (RLIMIT_CORE) for worker
processes. Used to increase the limit without restarting the main process.

worker rlimit nofile

Syntax: worker_rlimit_nofile number;
Default
Context: main

Changes the limit on the maximum number of open files (RLIMIT_NOFILE)

for worker processes. Used to increase the limit without restarting the main
process.

worker rlimit sigpending

Syntax: worker_rlimit_sigpending number;
Default
Context: main

On systems that support rtsig connection processing method, changes the

limit on the number of signals that may be queued (RLIMIT_SIGPENDING)
for worker processes. Used to increase the limit without restarting the main
process.

Nginx, Inc. p.14 of 330

CHAPTER 1. CORE MODULES 1.1. CORE FUNCTIONALITY

working directory
Syntax: working_directory directory;
Default
Context: main

Defines the current working directory for a worker process. It is primarily

used when writing a core-file, in which case a worker process should have write
permission for the specified directory.

Nginx, Inc. p.15 of 330

CHAPTER 1. CORE MODULES 1.2. SETTING UP HASHES

1.2 Setting up hashes

1.2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 16

1.2.1 Overview
To quickly process static sets of data such as server names, map directives
values, MIME types, names of request header strings, nginx uses hash tables.
During the start and each re-configuration nginx selects the minimum possible
sizes of hash tables such that the bucket size that stores keys with identical
hash values does not exceed the configured parameter (hash bucket size). The
size of a table is expressed in buckets. The adjustment is continued until
the table size exceeds the hash max size parameter. Most hashes have the
corresponding directives that allow changing these parameters, for example,
for the server names hash they are server names hash max size and server -
names hash bucket size.
The hash bucket size parameter is aligned to the size that is a multiple of
the processors cache line size. This speeds up key search in a hash on modern
processors by reducing the number of memory accesses. If hash bucket size is
equal to one processors cache line size then the number of memory accesses
during the key search will be two in the worst case first to compute the
bucket address, and second during the key search inside the bucket. Therefore,
if nginx emits the message requesting to increase either hash max size or hash
bucket size then the first parameter should first be increased.

Nginx, Inc. p.16 of 330

CHAPTER 1. CORE MODULES 1.3. CONNECTION PROCESSING METHODS

1.3 Connection processing methods

1.3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 17

1.3.1 Overview
nginx supports a variety of connection processing methods. The availability
of a particular method depends on the platform used. On platforms that
support several methods nginx will normally select the most efficient method
automatically. However, if needed, a connection processing method can be
selected explicitly with the use directive.
The following connection processing methods are supported:

select standard method. The supporting module is built

automatically on platforms that lack more efficient methods. The
--with-select_module and --without-select_module configuration
parameters can be used to forcibly enable or disable the build of this
module.

poll standard method. The supporting module is built automatically

on platforms that lack more efficient methods. The --with-poll_module
and --without-poll_module configuration parameters can be used to
forcibly enable or disable the build of this module.

kqueue efficient method used on FreeBSD 4.1+, OpenBSD 2.9+,

NetBSD 2.0, and Mac OS X.

epoll efficient method used on Linux 2.6+.

Some older distributions like SuSE 8.2 provide patches that add epoll
support to 2.4 kernels.

rtsig real time signals, efficient method used on Linux 2.2.19+. By

default, the system-wide event queue is limited by 1024 signals. On
loaded servers it may become necessary to increase this limit by changing
the /proc/sys/kernel/rtsig-max kernel parameter. However, in Linux
2.6.6-mm2 this parameter is gone, and each process now has its own event
queue. The size of each queue is limited by RLIMIT_SIGPENDING and can
be changed with worker rlimit sigpending.
On queue overflow, nginx discards the queue and falls back to poll
connection processing method until the situation gets back to normal.

/dev/poll efficient method used on Solaris 7 11/99+, HP/UX 11.22+

(eventport), IRIX 6.5.15+, and Tru64 UNIX 5.1A+.

eventport event ports, efficient method used on Solaris 10.

Nginx, Inc. p.17 of 330

CHAPTER 1. CORE MODULES 1.4. LOGGING TO SYSLOG

1.4 Logging to syslog

1.4.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . 18

1.4.1 Overview
The error log and access log directives support logging to syslog. The
following parameters configure logging to syslog:

server=address
Defines the address of a syslog server. The address can be specified as a
domain name, IP address, or a UNIX-domain socket path (specified after
the unix: prefix). With a domain name or IP address, the port can be
specified. If port is not specified, the port 514 is used. If a domain name
resolves to several IP addresses, the first resolved address is used.
facility=string
Sets facility of syslog messages, as defined in RFC 3164. Facility can
be one of kern, user, mail, daemon, auth, intern, lpr,
news, uucp, clock, authpriv, ftp, ntp, audit, alert,
cron, local0..local7. Default is local7.
severity=string
Sets severity of syslog messages for access log, as defined in RFC 3164.
Possible values are the same as for the second parameter (level) of the
error log directive. Default is info.
tag=string
Sets the tag of syslog messages. Default is nginx.

Example syslog configuration:

error_log syslog : server =192.168.1.1 debug ;

access_log syslog : server = unix :/ var / log / nginx . sock ;

access_log syslog : server =[2001: db8 ::1]:12345 , facility = local7 , tag = nginx ,
severity = info combined ;

Logging to syslog is available since version 1.7.1. As part of our

commercial subscription logging to syslog is available since version 1.5.3.

Nginx, Inc. p.18 of 330

Chapter 2

HTTP server modules

2.1 Module ngx http core module

2.1.1 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 21
aio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
alias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
chunked transfer encoding . . . . . . . . . . . . . . . . . 23
client body buffer size . . . . . . . . . . . . . . . . . . . 23
client body in file only . . . . . . . . . . . . . . . . . . . 23
client body in single buffer . . . . . . . . . . . . . . . . 23
client body temp path . . . . . . . . . . . . . . . . . . . 24
client body timeout . . . . . . . . . . . . . . . . . . . . . 24
client header buffer size . . . . . . . . . . . . . . . . . . 24
client header timeout . . . . . . . . . . . . . . . . . . . . 24
client max body size . . . . . . . . . . . . . . . . . . . . 25
connection pool size . . . . . . . . . . . . . . . . . . . . 25
default type . . . . . . . . . . . . . . . . . . . . . . . . . 25
directio . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
directio alignment . . . . . . . . . . . . . . . . . . . . . 26
disable symlinks . . . . . . . . . . . . . . . . . . . . . . 26
error page . . . . . . . . . . . . . . . . . . . . . . . . . . 27
etag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
http . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
if modified since . . . . . . . . . . . . . . . . . . . . . . 28
ignore invalid headers . . . . . . . . . . . . . . . . . . . 28
internal . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
keepalive disable . . . . . . . . . . . . . . . . . . . . . . 29
keepalive requests . . . . . . . . . . . . . . . . . . . . . . 30
keepalive timeout . . . . . . . . . . . . . . . . . . . . . . 30
large client header buffers . . . . . . . . . . . . . . . . . 30
limit except . . . . . . . . . . . . . . . . . . . . . . . . . 31
limit rate . . . . . . . . . . . . . . . . . . . . . . . . . . 31
limit rate after . . . . . . . . . . . . . . . . . . . . . . . 31
lingering close . . . . . . . . . . . . . . . . . . . . . . . . 32

19
CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

lingering time . . . . . . . . . . . . . . . . . . . . . . . . 32
lingering timeout . . . . . . . . . . . . . . . . . . . . . . 32
listen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
location . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
log not found . . . . . . . . . . . . . . . . . . . . . . . . 37
log subrequest . . . . . . . . . . . . . . . . . . . . . . . . 37
max ranges . . . . . . . . . . . . . . . . . . . . . . . . . 37
merge slashes . . . . . . . . . . . . . . . . . . . . . . . . 38
msie padding . . . . . . . . . . . . . . . . . . . . . . . . 38
msie refresh . . . . . . . . . . . . . . . . . . . . . . . . . 38
open file cache . . . . . . . . . . . . . . . . . . . . . . . 39
open file cache errors . . . . . . . . . . . . . . . . . . . . 39
open file cache min uses . . . . . . . . . . . . . . . . . . 39
open file cache valid . . . . . . . . . . . . . . . . . . . . 40
optimize server names . . . . . . . . . . . . . . . . . . . 40
output buffers . . . . . . . . . . . . . . . . . . . . . . . . 40
port in redirect . . . . . . . . . . . . . . . . . . . . . . . 40
postpone output . . . . . . . . . . . . . . . . . . . . . . 40
read ahead . . . . . . . . . . . . . . . . . . . . . . . . . 41
recursive error pages . . . . . . . . . . . . . . . . . . . . 41
request pool size . . . . . . . . . . . . . . . . . . . . . . 41
reset timedout connection . . . . . . . . . . . . . . . . . 41
resolver . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
resolver timeout . . . . . . . . . . . . . . . . . . . . . . . 42
root . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
satisfy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
satisfy any . . . . . . . . . . . . . . . . . . . . . . . . . . 43
send lowat . . . . . . . . . . . . . . . . . . . . . . . . . . 43
send timeout . . . . . . . . . . . . . . . . . . . . . . . . 44
sendfile . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
sendfile max chunk . . . . . . . . . . . . . . . . . . . . . 44
server . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
server name . . . . . . . . . . . . . . . . . . . . . . . . . 45
server name in redirect . . . . . . . . . . . . . . . . . . . 47
server names hash bucket size . . . . . . . . . . . . . . . 47
server names hash max size . . . . . . . . . . . . . . . . 47
server tokens . . . . . . . . . . . . . . . . . . . . . . . . 47
tcp nodelay . . . . . . . . . . . . . . . . . . . . . . . . . 47
tcp nopush . . . . . . . . . . . . . . . . . . . . . . . . . 48
try files . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
types hash bucket size . . . . . . . . . . . . . . . . . . . 50
types hash max size . . . . . . . . . . . . . . . . . . . . 51
underscores in headers . . . . . . . . . . . . . . . . . . . 51
variables hash bucket size . . . . . . . . . . . . . . . . . 51
variables hash max size . . . . . . . . . . . . . . . . . . 51

Nginx, Inc. p.20 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

2.1.2 Embedded Variables . . . . . . . . . . . . . . . . . . . . 52

2.1.1 Directives
aio
Syntax: aio on | off | threads[=pool];
Default off
Context: http, server, location
This directive appeared in version 0.8.11.

Enables or disables the use of asynchronous file I/O (AIO) on FreeBSD and
Linux:

location / video / {
aio on ;
output_buffers 1 64 k ;
}

On FreeBSD, AIO can be used starting from FreeBSD 4.3. AIO can either
be linked statically into a kernel:

options VFS_AIO

or loaded dynamically as a kernel loadable module:

kldload aio

On Linux, AIO can be used starting from kernel version 2.6.22. Also, it is
necessary to enable directio, or otherwise reading will be blocking:

location / video / {
aio on ;
directio 512;
output_buffers 1 128 k ;
}

On Linux, directio can only be used for reading blocks that are aligned on
512-byte boundaries (or 4K for XFS). Files unaligned end is read in blocking
mode. The same holds true for byte range requests and for FLV requests not
from the beginning of a file: reading of unaligned data at the beginning and
end of a file will be blocking.
When both AIO and sendfile are enabled on Linux, AIO is used for files
that are larger than or equal to the size specified in the directio directive, while
sendfile is used for files of smaller sizes or when directio is disabled.

location / video / {
sendfile on ;
aio on ;
directio 8m;
}

Finally, files can be read and sent using multi-threading (1.7.11), without
blocking a worker process:

Nginx, Inc. p.21 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

location / video / {
sendfile on ;
aio threads ;
}

Read and send file operations are offloaded to threads of the specified pool.
If the pool name is omitted, the pool with the name default is used. The
pool name can also be set with variables:

aio threads = pool$disk ;

By default, multi-threading is disabled, it should be enabled with the

--with-threads configuration parameter. Currently, multi-threading is
compatible only with the epoll, kqueue, and eventport methods. Multi-
threaded sending of files is only supported on Linux.
See also the sendfile directive.

alias
Syntax: alias path;
Default
Context: location

Defines a replacement for the specified location. For example, with the
following configuration

location / i / {
alias / data / w3 / images /;
}

on request of /i/top.gif, the file /data/w3/images/top.gif will be

sent.
The path value can contain variables, except $document root and
$realpath root.
If alias is used inside a location defined with a regular expression then
such regular expression should contain captures and alias should refer to these
captures (0.7.40), for example:

location ~ ^/ users /(.+\.(?: gif | jpe ? g | png ) ) $ {

alias / data / w3 / images / $1 ;
}

When location matches the last part of the directives value:

location / images / {
alias / data / w3 / images /;
}

it is better to use the root directive instead:

location / images / {

Nginx, Inc. p.22 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

root / data / w3 ;
}

chunked transfer encoding

Syntax: chunked_transfer_encoding on | off;
Default on
Context: http, server, location

Allows disabling chunked transfer encoding in HTTP/1.1. It may come in

handy when using a software failing to support chunked encoding despite the
standards requirement.

client body buffer size

Syntax: client_body_buffer_size size;
Default 8k|16k
Context: http, server, location

Sets buffer size for reading client request body. In case the request body is
larger than the buffer, the whole body or only its part is written to a temporary
file. By default, buffer size is equal to two memory pages. This is 8K on x86,
other 32-bit platforms, and x86-64. It is usually 16K on other 64-bit platforms.

client body in file only

Syntax: client_body_in_file_only on | clean | off;
Default off
Context: http, server, location

Determines whether nginx should save the entire client request body into
a file. This directive can be used during debugging, or when using the
$request body file variable, or the $r->request body file method of the module
ngx http perl module.
When set to the value on, temporary files are not removed after request
processing.
The value clean will cause the temporary files left after request processing
to be removed.

client body in single buffer

Syntax: client_body_in_single_buffer on | off;
Default off
Context: http, server, location

Determines whether nginx should save the entire client request body in
a single buffer. The directive is recommended when using the $request body
variable, to save the number of copy operations involved.

Nginx, Inc. p.23 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

client body temp path

Syntax: client_body_temp_path path [level1 [level2 [level3]]];
Default client_body_temp
Context: http, server, location

Defines a directory for storing temporary files holding client request bodies.
Up to three-level subdirectory hierarchy can be used under the specified
directory. For example, in the following configuration

c l i e n t _ b o d y _ t e m p _ p a t h / spool / nginx / client_temp 1 2;

a path to a temporary file might look like this:

/ spool / nginx / client_temp / 7 / 4 5 / 0 0 0 0 0 1 2 3 4 5 7

client body timeout

Syntax: client_body_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for reading client request body. The timeout is set only
for a period between two successive read operations, not for the transmission
of the whole request body. If a client does not transmit anything within this
time, the 408 Request Time-out error is returned to the client.

client header buffer size

Syntax: client_header_buffer_size size;
Default 1k
Context: http, server

Sets buffer size for reading client request header. For most requests, a
buffer of 1K bytes is enough. However, if a request includes long cookies, or
comes from a WAP client, it may not fit into 1K. If a request line or a request
header field does not fit into this buffer then larger buffers, configured by the
large client header buffers directive, are allocated.

client header timeout

Syntax: client_header_timeout time;
Default 60s
Context: http, server

Defines a timeout for reading client request header. If a client does not
transmit the entire header within this time, the 408 Request Time-out error
is returned to the client.

Nginx, Inc. p.24 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

client max body size

Syntax: client_max_body_size size;
Default 1m
Context: http, server, location

Sets the maximum allowed size of the client request body, specified in the
Content-Length request header field. If the size in a request exceeds the
configured value, the 413 Request Entity Too Large error is returned to
the client. Please be aware that browsers cannot correctly display this error.
Setting size to 0 disables checking of client request body size.

connection pool size

Syntax: connection_pool_size size;
Default 256
Context: http, server

Allows accurate tuning of per-connection memory allocations. This

directive has minimal impact on performance and should not generally be
used.

default type
Syntax: default_type mime-type;
Default text/plain
Context: http, server, location

Defines the default MIME type of a response. Mapping of file name

extensions to MIME types can be set with the types directive.

directio
Syntax: directio size | off;
Default off
Context: http, server, location
This directive appeared in version 0.7.7.

Enables the use of the O_DIRECT flag (FreeBSD, Linux), the F_NOCACHE flag
(Mac OS X), or the directio function (Solaris), when reading files that are
larger than or equal to the specified size. The directive automatically disables
(0.7.15) the use of sendfile for a given request. It can be useful for serving large
files:

directio 4 m ;

or when using aio on Linux.

Nginx, Inc. p.25 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

directio alignment
Syntax: directio_alignment size;
Default 512
Context: http, server, location
This directive appeared in version 0.8.11.

Sets the alignment for directio. In most cases, a 512-byte alignment is

enough. However, when using XFS under Linux, it needs to be increased to
4K.

disable symlinks
Syntax: disable_symlinks off;
Syntax: disable_symlinks on | if_not_owner [from=part];
Default off
Context: http, server, location
This directive appeared in version 1.1.15.

Determines how symbolic links should be treated when opening files:

off
Symbolic links in the pathname are allowed and not checked. This is the
default behavior.
on
If any component of the pathname is a symbolic link, access to a file is
denied.
if_not_owner
Access to a file is denied if any component of the pathname is a symbolic
link, and the link and object that the link points to have different owners.
from=part
When checking symbolic links (parameters on and if_not_owner), all
components of the pathname are normally checked. Checking of symbolic
links in the initial part of the pathname may be avoided by specifying
additionally the from=part parameter. In this case, symbolic links are
checked only from the pathname component that follows the specified
initial part. If the value is not an initial part of the pathname checked,
the whole pathname is checked as if this parameter was not specified
at all. If the value matches the whole file name, symbolic links are not
checked. The parameter value can contain variables.

Example:

d is a bl e _s y ml i nk s on from = $document_root ;

This directive is only available on systems that have the openat and
fstatat interfaces. Such systems include modern versions of FreeBSD, Linux,
and Solaris.
Parameters on and if_not_owner add a processing overhead.

Nginx, Inc. p.26 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

On systems that do not support opening of directories only for search,

to use these parameters it is required that worker processes have read
permissions for all directories being checked.

The ngx http autoindex module, ngx http random index module, and
ngx http dav module modules currently ignore this directive.

error page
Syntax: error_page code . . . [=[response]] uri;
Default
Context: http, server, location, if in location

Defines the URI that will be shown for the specified errors. error_page
directives are inherited from the previous level only if there are no error_page
directives defined on the current level. A uri value can contain variables.
Example:

error_page 404 /404. html ;

error_page 500 502 503 504 /50 x . html ;

Furthermore, it is possible to change the response code to another using

the =response syntax, for example:

error_page 404 =200 / empty . gif ;

If an error response is processed by a proxied server or a FastCGI/uws-

gi/SCGI server, and the server may return different response codes (e.g., 200,
302, 401 or 404), it is possible to respond with the code it returns:

error_page 404 = /404. php ;

It is also possible to use redirects for error processing:

error_page 403 http :// example . com / forbidden . html ;

error_page 404 =301 http :// example . com / notfound . html ;

In this case, by default, the response code 302 is returned to the client. It
can only be changed to one of the redirect status codes (301, 302, 303, and
307).
If there is no need to change URI during internal redirection it is possible
to pass error processing into a named location:

location / {
error_page 404 = @fallback ;
}

location @fallback {
proxy_pass http :// backend ;
}

Nginx, Inc. p.27 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

If uri processing leads to an error, the status code of the last occurred
error is returned to the client.

etag
Syntax: etag on | off;
Default on
Context: http, server, location
This directive appeared in version 1.3.3.

Enables or disables automatic generation of the ETag response header field

for static resources.

http
Syntax: http { . . . }
Default
Context: main

Provides the configuration file context in which the HTTP server directives
are specified.

if modified since
Syntax: if_modified_since off | exact | before;
Default exact
Context: http, server, location
This directive appeared in version 0.7.24.

Specifies how to compare modification time of a response with the time in

the If-Modified-Since request header field:

off
the If-Modified-Since request header field is ignored (0.7.34);
exact
exact match;
before
modification time of a response is less than or equal to the time in the
If-Modified-Since request header field.

ignore invalid headers

Syntax: ignore_invalid_headers on | off;
Default on
Context: http, server

Controls whether header fields with invalid names should be ignored.

Valid names are composed of English letters, digits, hyphens, and possibly
underscores (as controlled by the underscores in headers directive).

Nginx, Inc. p.28 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

If the directive is specified on the server level, its value is only used if a
server is a default one. The value specified also applies to all virtual servers
listening on the same address and port.

internal
Syntax: internal;
Default
Context: location

Specifies that a given location can only be used for internal requests. For
external requests, the client error 404 Not Found is returned. Internal requests
are the following:

requests redirected by the error page, index, random index, and try files
directives;

requests redirected by the X-Accel-Redirect response header field from

an upstream server;

subrequests formed by the include virtual command of the ngx -

http ssi module module and by the ngx http addition module module
directives;

requests changed by the rewrite directive.

Example:

error_page 404 /404. html ;

location /404. html {

internal ;
}

There is a limit of 10 internal redirects per request to prevent request

processing cycles that can occur in incorrect configurations. If this limit is
reached, the error 500 Internal Server Error is returned. In such cases,
the rewrite or internal redirection cycle message can be seen in the error
log.

keepalive disable
Syntax: keepalive_disable none | browser . . . ;
Default msie6
Context: http, server, location

Disables keep-alive connections with misbehaving browsers. The browser

parameters specify which browsers will be affected. The value msie6 disables
keep-alive connections with old versions of MSIE, once a POST request is
received. The value safari disables keep-alive connections with Safari and

Nginx, Inc. p.29 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Safari-like browsers on Mac OS X and Mac OS X-like operating systems. The

value none enables keep-alive connections with all browsers.

Prior to version 1.1.18, the value safari matched all Safari and Safari-like
browsers on all operating systems, and keep-alive connections with them were
disabled by default.

keepalive requests
Syntax: keepalive_requests number;
Default 100
Context: http, server, location
This directive appeared in version 0.8.0.

Sets the maximum number of requests that can be served through one
keep-alive connection. After the maximum number of requests are made, the
connection is closed.

keepalive timeout
Syntax: keepalive_timeout timeout [header timeout];
Default 75s
Context: http, server, location

The first parameter sets a timeout during which a keep-alive client

connection will stay open on the server side. The zero value disables keep-
alive client connections. The optional second parameter sets a value in the
Keep-Alive: timeout=time response header field. Two parameters may
differ.
The Keep-Alive: timeout=time header field is recognized by Mozilla
and Konqueror. MSIE closes keep-alive connections by itself in about 60
seconds.

large client header buffers

Syntax: large_client_header_buffers number size;
Default 4 8k
Context: http, server

Sets the maximum number and size of buffers used for reading large client
request header. A request line cannot exceed the size of one buffer, or the 414
Request-URI Too Large error is returned to the client. A request header field
cannot exceed the size of one buffer as well, or the 400 Bad Request error is
returned to the client. Buffers are allocated only on demand. By default,
the buffer size is equal to 8K bytes. If after the end of request processing a
connection is transitioned into the keep-alive state, these buffers are released.

Nginx, Inc. p.30 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

limit except
Syntax: limit_except method . . . { . . . }
Default
Context: location

Limits allowed HTTP methods inside a location. The method parameter

can be one of the following: GET, HEAD, POST, PUT, DELETE, MKCOL, COPY, MOVE,
OPTIONS, PROPFIND, PROPPATCH, LOCK, UNLOCK, or PATCH. Allowing the GET
method makes the HEAD method also allowed. Access to other methods can
be limited using the ngx http access module and ngx http auth basic module
modules directives:

limit_except GET {
allow 192 .168 .1.0 /32 ;
deny all ;
}

Please note that this will limit access to all methods except GET and
HEAD.

limit rate
Syntax: limit_rate rate;
Default 0
Context: http, server, location, if in location

Limits the rate of response transmission to a client. The rate is specified

in bytes per second. The zero value disables rate limiting.
The limit is set per a request, and so if a client simultaneously opens two
connections, the overall rate will be twice as much as the specified limit.
Rate limit can also be set in the $limit rate variable. It may be useful in
cases where rate should be limited depending on a certain condition:

server {

if ( $slow ) {
set $limit_rate 4 k ;
}

...
}

Rate limit can also be set in the X-Accel-Limit-Rate header field

of a proxied server response. This capability can be disabled using the
proxy ignore headers, fastcgi ignore headers, uwsgi ignore headers, and scgi -
ignore headers directives.

limit rate after

Syntax: limit_rate_after size;
Default 0
Context: http, server, location, if in location

Nginx, Inc. p.31 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

This directive appeared in version 0.8.0.

Sets the initial amount after which the further transmission of a response
to a client will be rate limited.
Example:

location / flv / {
flv ;
l im i t_ r at e _a f te r 500 k ;
limit_rate 50 k ;
}

lingering close
Syntax: lingering_close off | on | always;
Default on
Context: http, server, location
This directive appeared in versions 1.1.0 and 1.0.6.

Controls how nginx closes client connections.

The default value on instructs nginx to wait for and process additional
data from a client before fully closing a connection, but only if heuristics
suggests that a client may be sending more data.
The value always will cause nginx to unconditionally wait for and process
additional client data.
The value off tells nginx to never wait for more data and close the
connection immediately. This behavior breaks the protocol and should not
be used under normal circumstances.

lingering time
Syntax: lingering_time time;
Default 30s
Context: http, server, location

When lingering close is in effect, this directive specifies the maximum time
during which nginx will process (read and ignore) additional data coming from
a client. After that, the connection will be closed, even if there will be more
data.

lingering timeout
Syntax: lingering_timeout time;
Default 5s
Context: http, server, location

When lingering close is in effect, this directive specifies the maximum

waiting time for more client data to arrive. If data are not received during
this time, the connection is closed. Otherwise, the data are read and ignored,

Nginx, Inc. p.32 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

and nginx starts waiting for more data again. The wait-read-ignore cycle is
repeated, but no longer than specified by the lingering time directive.

listen
Syntax: listen address[:port] [default_server] [ssl] [spdy] [proxy_protocol]
[setfib=number] [fastopen=number] [backlog=number] [rcvbuf=size]
[sndbuf=size] [accept_filter=filter] [deferred] [bind]
[ipv6only=on|off]
[so_keepalive=on|off|[keepidle]:[keepintvl]:[keepcnt]];
Syntax: listen port [default_server] [ssl] [spdy] [proxy_protocol]
[setfib=number] [fastopen=number] [backlog=number] [rcvbuf=size]
[sndbuf=size] [accept_filter=filter] [deferred] [bind]
[ipv6only=on|off]
[so_keepalive=on|off|[keepidle]:[keepintvl]:[keepcnt]];
Syntax: listen unix:path [default_server] [ssl] [spdy] [proxy_protocol]
[backlog=number] [rcvbuf=size] [sndbuf=size] [accept_filter=filter]
[deferred] [bind] [so_keepalive=on|off|[keepidle]:[keepintvl]:[keepcnt]];
Default *:80 | *:8000
Context: server

Sets the address and port for IP, or the path for a UNIX-domain socket on
which the server will accept requests. Both address and port, or only address
or only port can be specified. An address may also be a hostname, for example:

listen 12 7.0. 0.1 :800 0;

listen 127.0.0.1;
listen 8000;
listen *:8000;
listen localhost :8000;

IPv6 addresses (0.7.36) are specified in square brackets:

listen [::]:8000;
listen [::1];

UNIX-domain sockets (0.8.21) are specified with the unix: prefix:

listen unix :/ var / run / nginx . sock ;

If only address is given, the port 80 is used.

If the directive is not present then either *:80 is used if nginx runs with
the superuser privileges, or *:8000 otherwise.
The default_server parameter, if present, will cause the server to become
the default server for the specified address:port pair. If none of the directives
have the default_server parameter then the first server with the address:port
pair will be the default server for this pair.

Nginx, Inc. p.33 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

In versions prior to 0.8.21 this parameter is named simply default.

The ssl parameter (0.7.14) allows specifying that all connections accepted
on this port should work in SSL mode. This allows for a more compact
configuration for the server that handles both HTTP and HTTPS requests.
The spdy parameter (1.3.15) allows accepting SPDY connections on this
port. Normally, for this to work the ssl parameter should be specified as well,
but nginx can also be configured to accept SPDY connections without SSL.
The proxy_protocol parameter (1.5.12) allows specifying that all
connections accepted on this port should use the PROXY protocol.
The listen directive can have several additional parameters specific to
socket-related system calls. These parameters can be specified in any listen
directive, but only once for a given address:port pair.

In versions prior to 0.8.21, they could only be specified in the listen

directive together with the default parameter.

setfib=number
this parameter (0.8.44) sets the associated routing table, FIB (the
SO_SETFIB option) for the listening socket. This currently works only
on FreeBSD.
fastopen=number
enables TCP Fast Open for the listening socket (1.5.8) and limits
the maximum length for the queue of connections that have not yet
completed the three-way handshake.

Do not enable this feature unless the server can handle receiving the
same SYN packet with data more than once.

backlog=number
sets the backlog parameter in the listen call that limits the maximum
length for the queue of pending connections. By default, backlog is set
to -1 on FreeBSD and Mac OS X, and to 511 on other platforms.
rcvbuf=size
sets the receive buffer size (the SO_RCVBUF option) for the listening socket.
sndbuf=size
sets the send buffer size (the SO_SNDBUF option) for the listening socket.
accept_filter=filter
sets the name of accept filter (the SO_ACCEPTFILTER option) for the
listening socket that filters incoming connections before passing them
to accept. This works only on FreeBSD and NetBSD 5.0+. Possible
values are dataready and httpready.
deferred
instructs to use a deferred accept (the TCP_DEFER_ACCEPT socket option)
on Linux.

Nginx, Inc. p.34 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

bind
instructs to make a separate bind call for a given address:port pair.
This is useful because if there are several listen directives with the
same port but different addresses, and one of the listen directives
listens on all addresses for the given port (*:port), nginx will bind
only to *:port. It should be noted that the getsockname system call
will be made in this case to determine the address that accepted the
connection. If the setfib, backlog, rcvbuf, sndbuf, accept_filter,
deferred, ipv6only, or so_keepalive parameters are used then for a
given address:port pair a separate bind call will always be made.
ipv6only=on|off
this parameter (0.7.42) determines (via the IPV6_V6ONLY socket option)
whether an IPv6 socket listening on a wildcard address [::] will
accept only IPv6 connections or both IPv6 and IPv4 connections. This
parameter is turned on by default. It can only be set once on start.

Prior to version 1.3.4, if this parameter was omitted then the operating
systems settings were in effect for the socket.

so_keepalive=on|off|[keepidle]:[keepintvl]:[keepcnt]
this parameter (1.1.11) configures the TCP keepalive behavior for the
listening socket. If this parameter is omitted then the operating systems
settings will be in effect for the socket. If it is set to the value on, the
SO_KEEPALIVE option is turned on for the socket. If it is set to the
value off, the SO_KEEPALIVE option is turned off for the socket. Some
operating systems support setting of TCP keepalive parameters on a per-
socket basis using the TCP_KEEPIDLE, TCP_KEEPINTVL, and TCP_KEEPCNT
socket options. On such systems (currently, Linux 2.4+, NetBSD 5+,
and FreeBSD 9.0-STABLE), they can be configured using the keepidle,
keepintvl, and keepcnt parameters. One or two parameters may be
omitted, in which case the system default setting for the corresponding
socket option will be in effect. For example,
so_keepalive =30 m ::10

will set the idle timeout (TCP_KEEPIDLE) to 30 minutes, leave the probe
interval (TCP_KEEPINTVL) at its system default, and set the probes count
(TCP_KEEPCNT) to 10 probes.
Example:

listen 127.0.0.1 default_server accept_filter = dataready backlog =1024;

location
Syntax: location [ = | ~ | ~* | ^~ ] uri { . . . }
Syntax: location @name { . . . }
Default
Context: server, location

Nginx, Inc. p.35 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Sets configuration depending on a request URI.

The matching is performed against a normalized URI, after decoding
the text encoded in the %XX form, resolving references to relative path
components . and .., and possible compression of two or more adjacent
slashes into a single slash.
A location can either be defined by a prefix string, or by a regular
expression. Regular expressions are specified with the preceding ~*
modifier (for case-insensitive matching), or the ~ modifier (for case-sensitive
matching). To find location matching a given request, nginx first checks
locations defined using the prefix strings (prefix locations). Among them,
the location with the longest matching prefix is selected and remembered.
Then regular expressions are checked, in the order of their appearance in the
configuration file. The search of regular expressions terminates on the first
match, and the corresponding configuration is used. If no match with a regular
expression is found then the configuration of the prefix location remembered
earlier is used.
location blocks can be nested, with some exceptions mentioned below.
For case-insensitive operating systems such as Mac OS X and Cygwin,
matching with prefix strings ignores a case (0.7.7). However, comparison is
limited to one-byte locales.
Regular expressions can contain captures (0.7.40) that can later be used in
other directives.
If the longest matching prefix location has the ^~ modifier then regular
expressions are not checked.
Also, using the = modifier it is possible to define an exact match of
URI and location. If an exact match is found, the search terminates. For
example, if a / request happens frequently, defining location = / will
speed up the processing of these requests, as search terminates right after the
first comparison. Such a location cannot obviously contain nested locations.

In versions from 0.7.1 to 0.8.41, if a request matched the prefix location

without the = and ^~ modifiers, the search also terminated and regular
expressions were not checked.

Lets illustrate the above by an example:

location = / {
[ configuration A ]
}

location / {
[ configuration B ]
}

location / documents / {
[ configuration C ]
}

location ^~ / images / {
[ configuration D ]
}

Nginx, Inc. p.36 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

location ~* \.( gif | jpg | jpeg ) $ {

[ configuration E ]
}

The / request will match configuration A, the /index.html request will

match configuration B, the /documents/document.html request will match
configuration C, the /images/1.gif request will match configuration D, and
the /documents/1.jpg request will match configuration E.
The @ prefix defines a named location. Such a location is not used for
a regular request processing, but instead used for request redirection. They
cannot be nested, and cannot contain nested locations.
If a location is defined by a prefix string that ends with the slash character,
and requests are processed by one of proxy pass, fastcgi pass, uwsgi pass,
scgi pass, or memcached pass, then the special processing is performed. In
response to a request with URI equal to this string, but without the trailing
slash, a permanent redirect with the code 301 will be returned to the requested
URI with the slash appended. If this is not desired, an exact match of the URI
and location could be defined like this:

location / user / {
proxy_pass http :// user . example . com ;
}

location = / user {
proxy_pass http :// login . example . com ;
}

log not found

Syntax: log_not_found on | off;
Default on
Context: http, server, location

Enables or disables logging of errors about not found files into error log.

log subrequest
Syntax: log_subrequest on | off;
Default off
Context: http, server, location

Enables or disables logging of subrequests into access log.

max ranges
Syntax: max_ranges number;
Default
Context: http, server, location
This directive appeared in version 1.1.2.

Nginx, Inc. p.37 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Limits the maximum allowed number of ranges in byte-range requests.

Requests that exceed the limit are processed as if there were no byte ranges
specified. By default, the number of ranges is not limited. The zero value
disables the byte-range support completely.

merge slashes
Syntax: merge_slashes on | off;
Default on
Context: http, server

Enables or disables compression of two or more adjacent slashes in a URI

into a single slash.
Note that compression is essential for the correct matching of prefix
string and regular expression locations. Without it, the //scripts/one.php
request would not match

location / scripts / {
...
}

and might be processed as a static file. So it gets converted to

/scripts/one.php.
Turning the compression off can become necessary if a URI contains
base64-encoded names, since base64 uses the / character internally. However,
for security considerations, it is better to avoid turning the compression off.
If the directive is specified on the server level, its value is only used if a
server is a default one. The value specified also applies to all virtual servers
listening on the same address and port.

msie padding
Syntax: msie_padding on | off;
Default on
Context: http, server, location

Enables or disables adding comments to responses for MSIE clients with

status greater than 400 to increase the response size to 512 bytes.

msie refresh
Syntax: msie_refresh on | off;
Default off
Context: http, server, location

Enables or disables issuing refreshes instead of redirects for MSIE clients.

Nginx, Inc. p.38 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

open file cache

Syntax: open_file_cache off;
Syntax: open_file_cache max=N [inactive=time];
Default off
Context: http, server, location

Configures a cache that can store:

open file descriptors, their sizes and modification times;

information on existence of directories;

file lookup errors, such as file not found, no read permission, and so
on.

Caching of errors should be enabled separately by the open file cache -

errors directive.

The directive has the following parameters:

max
sets the maximum number of elements in the cache; on cache overflow
the least recently used (LRU) elements are removed;
inactive
defines a time after which an element is removed from the cache if it has
not been accessed during this time; by default, it is 60 seconds;
off
disables the cache.

Example:

op en_f ile_ cach e max =1000 inactive =20 s ;

open_file_cache_valid 30 s ;
open_file_cache_min_uses 2;
open_file_cache_errors on ;

open file cache errors

Syntax: open_file_cache_errors on | off;
Default off
Context: http, server, location

Enables or disables caching of file lookup errors by open file cache.

open file cache min uses

Syntax: open_file_cache_min_uses number;
Default 1
Context: http, server, location

Nginx, Inc. p.39 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Sets the minimum number of file accesses during the period configured by
the inactive parameter of the open file cache directive, required for a file
descriptor to remain open in the cache.

open file cache valid

Syntax: open_file_cache_valid time;
Default 60s
Context: http, server, location

Sets a time after which open file cache elements should be validated.

optimize server names

Syntax: optimize_server_names on | off;
Default off
Context: http, server

This directive is obsolete. The server name in redirect directive should be

used instead.

output buffers
Syntax: output_buffers number size;
Default 1 32k
Context: http, server, location

Sets the number and size of the buffers used for reading a response from a
disk.

port in redirect
Syntax: port_in_redirect on | off;
Default on
Context: http, server, location

Enables or disables specifying the port in redirects issued by nginx.

The use of the primary server name in redirects is controlled by the server -
name in redirect directive.

postpone output
Syntax: postpone_output size;
Default 1460
Context: http, server, location

If possible, the transmission of client data will be postponed until nginx

has at least size bytes of data to send. The zero value disables postponing data
transmission.

Nginx, Inc. p.40 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

read ahead
Syntax: read_ahead size;
Default 0
Context: http, server, location

Sets the amount of pre-reading for the kernel when working with file.
On Linux, the posix_fadvise(0, 0, 0, POSIX_FADV_SEQUENTIAL) sys-
tem call is used, and so the size parameter is ignored.
On FreeBSD, the fcntl(O_READAHEAD, size) system call, supported since
FreeBSD 9.0-CURRENT, is used. FreeBSD 7 has to be patched.

recursive error pages

Syntax: recursive_error_pages on | off;
Default off
Context: http, server, location

Enables or disables doing several redirects using the error page directive.
The number of such redirects is limited.

request pool size

Syntax: request_pool_size size;
Default 4k
Context: http, server

Allows accurate tuning of per-request memory allocations. This directive

has minimal impact on performance and should not generally be used.

reset timedout connection

Syntax: reset_timedout_connection on | off;
Default off
Context: http, server, location

Enables or disables resetting timed out connections. The reset is performed

as follows. Before closing a socket, the SO_LINGER option is set on it with a
timeout value of 0. When the socket is closed, TCP RST is sent to the client,
and all memory occupied by this socket is released. This helps avoid keeping
an already closed socket with filled buffers in a FIN WAIT1 state for a long
time.
It should be noted that timed out keep-alive connections are closed
normally.

resolver
Syntax: resolver address . . . [valid=time] [ipv6=on|off];
Default
Context: http, server, location

Nginx, Inc. p.41 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Configures name servers used to resolve names of upstream servers into

addresses, for example:

resolver 127.0.0.1 [::1]:5353;

An address can be specified as a domain name or IP address, and an

optional port (1.3.1, 1.2.2). If port is not specified, the port 53 is used. Name
servers are queried in a round-robin fashion.

Before version 1.1.7, only a single name server could be configured.

Specifying name servers using IPv6 addresses is supported starting from
versions 1.3.1 and 1.2.2.

By default, nginx will look up both IPv4 and IPv6 addresses while resolving.
If looking up of IPv6 addresses is not desired, the ipv6=off parameter can be
specified.

Resolving of names into IPv6 addresses is supported starting from version

1.5.8.

By default, nginx caches answers using the TTL value of a response. An

optional valid parameter allows overriding it:

resolver 127.0.0.1 [::1]:5353 valid =30 s ;

Before version 1.1.9, tuning of caching time was not possible, and nginx
always cached answers for the duration of 5 minutes.

resolver timeout
Syntax: resolver_timeout time;
Default 30s
Context: http, server, location

Sets a timeout for name resolution, for example:

r es o lv e r_ t im e ou t 5 s ;

root
Syntax: root path;
Default html
Context: http, server, location, if in location

Sets the root directory for requests. For example, with the following
configuration

Nginx, Inc. p.42 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

location / i / {
root / data / w3 ;
}

The /data/w3/i/top.gif file will be sent in response to the /i/top.gif

request.
The path value can contain variables, except $document root and
$realpath root.
A path to the file is constructed by merely adding a URI to the value of
the root directive. If a URI has to be modified, the alias directive should be
used.

satisfy
Syntax: satisfy all | any;
Default all
Context: http, server, location

Allows access if all (all) or at least one (any) of the ngx http -
access module, ngx http auth basic module or ngx http auth request module
modules allow access.
Example:

location / {
satisfy any ;

allow 192 .168 .1.0 /32 ;

deny all ;

auth_basic " closed site ";

a u t h _ b a s i c _ u s e r _ f i l e conf / htpasswd ;
}

satisfy any
Syntax: satisfy_any on | off;
Default off
Context: http, server, location

This directive has been replaced by the any parameter of the satisfy
directive.

send lowat
Syntax: send_lowat size;
Default 0
Context: http, server, location

If the directive is set to a non-zero value, nginx will try to minimize the
number of send operations on client sockets by using either NOTE_LOWAT flag

Nginx, Inc. p.43 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

of the kqueue method or the SO_SNDLOWAT socket option. In both cases the
specified size is used.
This directive is ignored on Linux, Solaris, and Windows.

send timeout
Syntax: send_timeout time;
Default 60s
Context: http, server, location

Sets a timeout for transmitting a response to the client. The timeout is set
only between two successive write operations, not for the transmission of the
whole response. If the client does not receive anything within this time, the
connection is closed.

sendfile
Syntax: sendfile on | off;
Default off
Context: http, server, location, if in location

Enables or disables the use of sendfile.

Starting from nginx 0.8.12 and FreeBSD 5.2.1, aio can be used to pre-load
data for sendfile:

location / video / {
sendfile on ;
tcp_nopush on ;
aio on ;
}

In this configuration, sendfile is called with the SF_NODISKIO flag which

causes it not to block on disk I/O, but, instead, report back that the data are
not in memory. nginx then initiates an asynchronous data load by reading one
byte. On the first read, the FreeBSD kernel loads the first 128K bytes of a file
into memory, although next reads will only load data in 16K chunks. This can
be changed using the read ahead directive.

Before version 1.7.11, pre-loading could be enabled with aio sendfile;.

sendfile max chunk

Syntax: sendfile_max_chunk size;
Default 0
Context: http, server, location

When set to a non-zero value, limits the amount of data that can be
transferred in a single sendfile call. Without the limit, one fast connection
may seize the worker process entirely.

Nginx, Inc. p.44 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

server
Syntax: server { . . . }
Default
Context: http

Sets configuration for a virtual server. There is no clear separation between

IP-based (based on the IP address) and name-based (based on the Host request
header field) virtual servers. Instead, the listen directives describe all addresses
and ports that should accept connections for the server, and the server name
directive lists all server names. Example configurations are provided in the
How nginx processes a request document.

server name
Syntax: server_name name . . . ;
Default ""
Context: server

Sets names of a virtual server, for example:

server {
server_name example . com www . example . com ;
}

The first name becomes the primary server name.

Server names can include an asterisk (*) replacing the first or last part
of a name:

server {
server_name example . com *. example . com www . example .*;
}

Such names are called wildcard names.

The first two of the names mentioned above can be combined in one:

server {
server_name . example . com ;
}

It is also possible to use regular expressions in server names, preceding the

name with a tilde (~):

server {
server_name www . example . com ~^ www \ d +\. example \. com$ ;
}

Regular expressions can contain captures (0.7.40) that can later be used in
other directives:

server {
server_name ~^( www \.) ?(.+) $ ;

location / {

Nginx, Inc. p.45 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

root / sites / $2 ;
}
}

server {
server_name _ ;

location / {
root / sites / default ;
}
}

Named captures in regular expressions create variables (0.8.25) that can

later be used in other directives:

server {
server_name ~^( www \.) ?(? < domain >.+) $ ;

location / {
root / sites / $domain ;
}
}

server {
server_name _ ;

location / {
root / sites / default ;
}
}

If the directives parameter is set to $hostname (0.9.4), the machines

hostname is inserted.
It is also possible to specify an empty server name (0.7.11):

server {
server_name www . example . com "";
}

It allows this server to process requests without the Host header field
instead of the default server for the given address:port pair. This is the
default setting.

Before 0.8.48, the machines hostname was used by default.

During searching for a virtual server by name, if the name matches more
than one of the specified variants, (e.g. both a wildcard name and regular
expression match), the first matching variant will be chosen, in the following
order of priority:

1. the exact name

2. the longest wildcard name starting with an asterisk, e.g.

*.example.com

3. the longest wildcard name ending with an asterisk, e.g. mail.*

Nginx, Inc. p.46 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

4. the first matching regular expression (in order of appearance in the

configuration file)

Detailed description of server names is provided in a separate Server names

document.

server name in redirect

Syntax: server_name_in_redirect on | off;
Default off
Context: http, server, location

Enables or disables the use of the primary server name, specified by the
server name directive, in redirects issued by nginx. When the use of the
primary server name is disabled, the name from the Host request header field
is used. If this field is not present, the IP address of the server is used.
The use of a port in redirects is controlled by the port in redirect directive.

server names hash bucket size

Syntax: server_names_hash_bucket_size size;
Default 32|64|128
Context: http

Sets the bucket size for the server names hash tables. The default value
depends on the size of the processors cache line. The details of setting up
hash tables are provided in a separate document.

server names hash max size

Syntax: server_names_hash_max_size size;
Default 512
Context: http

Sets the maximum size of the server names hash tables. The details of
setting up hash tables are provided in a separate document.

server tokens
Syntax: server_tokens on | off;
Default on
Context: http, server, location

Enables or disables emitting nginx version in error messages and in the

Server response header field.

tcp nodelay
Syntax: tcp_nodelay on | off;
Default on
Context: http, server, location

Nginx, Inc. p.47 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Enables or disables the use of the TCP_NODELAY option. The option is

enabled only when a connection is transitioned into the keep-alive state.

tcp nopush
Syntax: tcp_nopush on | off;
Default off
Context: http, server, location

Enables or disables the use of the TCP_NOPUSH socket option on FreeBSD

or the TCP_CORK socket option on Linux. The options are enabled only when
sendfile is used. Enabling the option allows
sending the response header and the beginning of a file in one packet, on
Linux and FreeBSD 4.*;

sending a file in full packets.

try files
Syntax: try_files file . . . uri;
Syntax: try_files file . . . =code;
Default
Context: server, location

Checks the existence of files in the specified order and uses the first found
file for request processing; the processing is performed in the current context.
The path to a file is constructed from the file parameter according to the root
and alias directives. It is possible to check directorys existence by specifying
a slash at the end of a name, e.g. $uri/. If none of the files were found,
an internal redirect to the uri specified in the last parameter is made. For
example:

location / images / {
try_files $uri / images / default . gif ;
}

location = / images / default . gif {

expires 30 s ;
}

The last parameter can also point to a named location, as shown in

examples below. Starting from version 0.7.51, the last parameter can also
be a code:

location / {
try_files $uri $uri / index . html $uri . html =404;
}

Example in proxying Mongrel:

location / {
try_files / system / maintenance . html

Nginx, Inc. p.48 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

$uri $uri / index . html $uri . html

@mongrel ;
}

location @mongrel {
proxy_pass http :// mongrel ;
}

Example for Drupal/FastCGI:

location / {
try_files $uri $uri / @drupal ;
}

location ~ \. php$ {
try_files $uri @drupal ;

fastcgi_pass ...;

fastcgi_param SCR IPT_ FIL ENAM E / path / t o $ f a s t c g i _ s c r i p t _ n a m e ;

fastcgi_param SCRIPT_NAME $fastcgi_script_name ;
fastcgi_param QUERY_STRING $args ;

... other fastcgi_param s

}

location @drupal {
fastcgi_pass ...;

fastcgi_param SCR IPT_ FIL ENAM E / path / to / index . php ;

fastcgi_param SCRIPT_NAME / index . php ;
fastcgi_param QUERY_STRING q = $uri & $args ;

... other fastcgi_param s

}

In the following example,

location / {
try_files $uri $uri / @drupal ;
}

the try_files directive is equivalent to

location / {
error_page 404 = @drupal ;
log_not_found off ;
}

And here,

location ~ \. php$ {
try_files $uri @drupal ;

fastcgi_pass ...;

fastcgi_param SCR IPT_ FIL ENAM E / path / t o $ f a s t c g i _ s c r i p t _ n a m e ;

...
}

try_files checks the existence of the PHP file before passing the request
to the FastCGI server.

Nginx, Inc. p.49 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Example for Wordpress and Joomla:

location / {
try_files $uri $uri / @wordpress ;
}

location ~ \. php$ {
try_files $uri @wordpress ;

fastcgi_pass ...;

fastcgi_param SCR IPT_ FIL ENAM E / path / t o $ f a s t c g i _ s c r i p t _ n a m e ;

... other fastcgi_param s
}

location @wordpress {
fastcgi_pass ...;

fastcgi_param SCR IPT_ FIL ENAM E / path / to / index . php ;

... other fastcgi_param s
}

types
Syntax: types { . . . }
Default text/html html; image/gif gif; image/jpeg jpg;
Context: http, server, location

Maps file name extensions to MIME types of responses. Extensions are

case-insensitive. Several extensions can be mapped to one type, for example:

types {
application / octet - stream bin exe dll ;
application / octet - stream deb ;
application / octet - stream dmg ;
}

A sufficiently full mapping table is distributed with nginx in the conf/

mime.types file.
To make a particular location emit the application/octet-stream
MIME type for all requests, the following configuration can be used:

location / download / {
types { }
default_type application / octet - stream ;
}

types hash bucket size

Syntax: types_hash_bucket_size size;
Default 64
Context: http, server, location

Sets the bucket size for the types hash tables. The details of setting up
hash tables are provided in a separate document.

Nginx, Inc. p.50 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

Prior to version 1.5.13, the default value depended on the size of the
processors cache line.

types hash max size

Syntax: types_hash_max_size size;
Default 1024
Context: http, server, location

Sets the maximum size of the types hash tables. The details of setting up
hash tables are provided in a separate document.

underscores in headers
Syntax: underscores_in_headers on | off;
Default off
Context: http, server

Enables or disables the use of underscores in client request header fields.

When the use of underscores is disabled, request header fields whose names
contain underscores are marked as invalid and become subject to the ignore -
invalid headers directive.
If the directive is specified on the server level, its value is only used if a
server is a default one. The value specified also applies to all virtual servers
listening on the same address and port.

variables hash bucket size

Syntax: variables_hash_bucket_size size;
Default 64
Context: http

Sets the bucket size for the variables hash table. The details of setting up
hash tables are provided in a separate document.

variables hash max size

Syntax: variables_hash_max_size size;
Default 1024
Context: http

Sets the maximum size of the variables hash table. The details of setting
up hash tables are provided in a separate document.

Prior to version 1.5.13, the default value was 512.

Nginx, Inc. p.51 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

2.1.2 Embedded Variables

The ngx_http_core_module module supports embedded variables with
names matching the Apache Server variables. First of all, these are variables
representing client request header fields, such as $http user agent, $http cookie,
and so on. Also there are other variables:

$arg name
argument name in the request line
$args
arguments in the request line
$binary remote addr
client address in a binary form, values length is always 4 bytes
$body bytes sent
number of bytes sent to a client, not counting the response header; this
variable is compatible with the %B parameter of the mod_log_config
Apache module
$bytes sent
number of bytes sent to a client (1.3.8, 1.2.5)
$connection
connection serial number (1.3.8, 1.2.5)
$connection requests
current number of requests made through a connection (1.3.8, 1.2.5)
$content length
Content-Length request header field
$content type
Content-Type request header field
$cookie name
the name cookie
$document root
root or alias directives value for the current request
$document uri
same as $uri
$host
in this order of precedence: host name from the request line, or host
name from the Host request header field, or the server name matching a
request
$hostname
host name
$http name
arbitrary request header field; the last part of a variable name is the field
name converted to lower case with dashes replaced by underscores
$https
on if connection operates in SSL mode, or an empty string otherwise
$is args
? if a request line has arguments, or an empty string otherwise

Nginx, Inc. p.52 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

$limit rate
setting this variable enables response rate limiting; see limit rate
$msec
current time in seconds with the milliseconds resolution (1.3.9, 1.2.6)
$nginx version
nginx version
$pid
PID of the worker process
$pipe
p if request was pipelined, . otherwise (1.3.12, 1.2.7)
$proxy protocol addr
client address from the PROXY protocol header, or an empty string
otherwise (1.5.12)
The PROXY protocol must be previously enabled by setting the
proxy_protocol parameter in the listen directive.
$query string
same as $args
$realpath root
an absolute pathname corresponding to the root or alias directives value
for the current request, with all symbolic links resolved to real paths
$remote addr
client address
$remote port
client port
$remote user
user name supplied with the Basic authentication
$request
full original request line
$request body
request body
The variables value is made available in locations processed by the
proxy pass, fastcgi pass, uwsgi pass, and scgi pass directives.
$request body file
name of a temporary file with the request body
At the end of processing, the file needs to be removed. To always write
the request body to a file, client body in file only needs to be enabled.
When the name of a temporary file is passed in a proxied request or in
a request to a FastCGI/uwsgi/SCGI server, passing the request body
should be disabled by the proxy pass request body off, fastcgi pass -
request body off, uwsgi pass request body off, or scgi pass request -
body off directives, respectively.
$request completion
OK if a request has completed, or an empty string otherwise
$request filename
file path for the current request, based on the root or alias directives,

Nginx, Inc. p.53 of 330

CHAPTER 2. HTTP SERVER MODULES 2.1. MODULE NGX HTTP CORE MODULE

and the request URI

$request length
request length (including request line, header, and request body) (1.3.12,
1.2.7)
$request method
request method, usually GET or POST
$request time
request processing time in seconds with a milliseconds resolution (1.3.9,
1.2.6); time elapsed since the first bytes were read from the client
$request uri
full original request URI (with arguments)
$scheme
request scheme, http or https
$sent http name
arbitrary response header field; the last part of a variable name is the
field name converted to lower case with dashes replaced by underscores
$server addr
an address of the server which accepted a request
Computing a value of this variable usually requires one system call. To
avoid a system call, the listen directives must specify addresses and use
the bind parameter.
$server name
name of the server which accepted a request
$server port
port of the server which accepted a request
$server protocol
request protocol, usually HTTP/1.0 or HTTP/1.1
$status
response status (1.3.2, 1.2.2)
$tcpinfo rtt, $tcpinfo rttvar, $tcpinfo snd cwnd, $tcpinfo rcv space
information about the client TCP connection; available on systems that
support the TCP_INFO socket option
$time iso8601
local time in the ISO 8601 standard format (1.3.12, 1.2.7)
$time local
local time in the Common Log Format (1.3.12, 1.2.7)
$uri
current URI in request, normalized
The value of $uri may change during request processing, e.g. when doing
internal redirects, or when using index files.

Nginx, Inc. p.54 of 330

CHAPTER 2. HTTP SERVER MODULES 2.2. MODULE NGX HTTP ACCESS MODULE

2.2 Module ngx http access module

2.2.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 55
2.2.2 Example Configuration . . . . . . . . . . . . . . . . . . . 55
2.2.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 55
allow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
deny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

2.2.1 Summary
The ngx_http_access_module module allows limiting access to certain
client addresses.
Access can also be limited by password or by the result of subrequest.
Simultaneous limitation of access by address and by password is controlled by
the satisfy directive.

2.2.2 Example Configuration

location / {
deny 192.168.1.1;
allow 192 .168 .1.0 /24 ;
allow 10.1.1.0/16;
allow 2001:0 db8 ::/32;
deny all ;
}

The rules are checked in sequence until the first match is found. In
this example, access is allowed only for IPv4 networks 10.1.1.0/16 and
192.168.1.0/24 excluding the address 192.168.1.1, and for IPv6 network
2001:0db8::/32. In case of a lot of rules, the use of the ngx http geo module
module variables is preferable.

2.2.3 Directives
allow
Syntax: allow address | CIDR | unix: | all;
Default
Context: http, server, location, limit except

Allows access for the specified network or address. If the special value
unix: is specified (1.5.1), allows access for all UNIX-domain sockets.

deny
Syntax: deny address | CIDR | unix: | all;
Default
Context: http, server, location, limit except

Nginx, Inc. p.55 of 330

CHAPTER 2. HTTP SERVER MODULES 2.2. MODULE NGX HTTP ACCESS MODULE

Denies access for the specified network or address. If the special value
unix: is specified (1.5.1), denies access for all UNIX-domain sockets.

Nginx, Inc. p.56 of 330

CHAPTER 2. HTTP SERVER MODULES 2.3. MODULE NGX HTTP ADDITION MODULE

2.3 Module ngx http addition module

2.3.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.3.2 Example Configuration . . . . . . . . . . . . . . . . . . . 57
2.3.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 57
add before body . . . . . . . . . . . . . . . . . . . . . . 57
add after body . . . . . . . . . . . . . . . . . . . . . . . 57
addition types . . . . . . . . . . . . . . . . . . . . . . . . 57

2.3.1 Summary
The ngx_http_addition_module module is a filter that adds text before
and after a response. This module is not built by default, it should be enabled
with the --with-http_addition_module configuration parameter.

2.3.2 Example Configuration

location / {
ad d_be fore _bod y / before_action ;
add_after_body / after_action ;
}

2.3.3 Directives
add before body
Syntax: add_before_body uri;
Default
Context: http, server, location

Adds the text returned as a result of processing a given subrequest before

the response body. An empty string ("") as a parameter cancels addition
inherited from the previous configuration level.

add after body

Syntax: add_after_body uri;
Default
Context: http, server, location

Adds the text returned as a result of processing a given subrequest after the
response body. An empty string ("") as a parameter cancels addition inherited
from the previous configuration level.

addition types
Syntax: addition_types mime-type . . . ;
Default text/html
Context: http, server, location

Nginx, Inc. p.57 of 330

CHAPTER 2. HTTP SERVER MODULES 2.3. MODULE NGX HTTP ADDITION MODULE

This directive appeared in version 0.7.9.

Allows adding text in responses with the specified MIME types, in addition
to text/html. The special value * matches any MIME type (0.8.29).

Nginx, Inc. p.58 of 330

CHAPTER 2. HTTP SERVER MODULES 2.4. MODULE NGX HTTP AUTH BASIC MODULE

2.4 Module ngx http auth basic module

2.4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 59
2.4.2 Example Configuration . . . . . . . . . . . . . . . . . . . 59
2.4.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 59
auth basic . . . . . . . . . . . . . . . . . . . . . . . . . . 59
auth basic user file . . . . . . . . . . . . . . . . . . . . . 59

2.4.1 Summary
The ngx_http_auth_basic_module module allows limiting access to
resources by validating the user name and password using the HTTP Basic
Authentication protocol.
Access can also be limited by address or by the result of subrequest.
Simultaneous limitation of access by address and by password is controlled
by the satisfy directive.

2.4.2 Example Configuration

location / {
auth_basic " closed site ";
a u t h _ b a s i c _ u s e r _ f i l e conf / htpasswd ;
}

2.4.3 Directives
auth basic
Syntax: auth_basic string | off;
Default off
Context: http, server, location, limit except

Enables validation of user name and password using the HTTP Basic
Authentication protocol. The specified parameter is used as a realm.
Parameter value can contain variables (1.3.10, 1.2.7). The special value off
allows cancelling the effect of the auth_basic directive inherited from the
previous configuration level.

auth basic user file

Syntax: auth_basic_user_file file;
Default
Context: http, server, location, limit except

Specifies a file that keeps user names and passwords, in the following format:

# comment
name1 : password1
name2 : password2 : comment

Nginx, Inc. p.59 of 330

CHAPTER 2. HTTP SERVER MODULES 2.4. MODULE NGX HTTP AUTH BASIC MODULE

name3 : password3

The following password types are supported:

encrypted with the crypt function; can be generated using the

htpasswd utility from the Apache HTTP Server distribution or the
openssl passwd command;

hashed with the Apache variant of the MD5-based password algorithm

(apr1); can be generated with the same tools;

specified by the {scheme}data syntax (1.0.3+) as described in RFC

2307; currently implemented schemes include PLAIN (an example one,
should not be used), SHA (1.3.13) (plain SHA-1 hashing, should not be
used) and SSHA (salted SHA-1 hashing, used by some software packages,
notably OpenLDAP and Dovecot).

Support for SHA scheme was added only to aid in migration from other
web servers. It should not be used for new passwords, since unsalted
SHA-1 hashing that it employs is vulnerable to rainbow table attacks.

Nginx, Inc. p.60 of 330

CHAPTER 2. HTTP SERVER MODULES 2.5. MODULE NGX HTTP AUTH REQUEST MODULE

2.5 Module ngx http auth request module

2.5.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 61
2.5.2 Example Configuration . . . . . . . . . . . . . . . . . . . 61
2.5.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 61
auth request . . . . . . . . . . . . . . . . . . . . . . . . . 61
auth request set . . . . . . . . . . . . . . . . . . . . . . 62

2.5.1 Summary
The ngx_http_auth_request_module module (1.5.4+) implements client
authorization based on the result of a subrequest. If the subrequest returns a
2xx response code, the access is allowed. If it returns 401 or 403, the access is
denied with the corresponding error code. Any other response code returned
by the subrequest is considered an error.
For the 401 error, the client also receives the WWW-Authenticate header
from the subrequest response.
This module is not built by default, it should be enabled with the
--with-http_auth_request_module configuration parameter.
The module may be combined with other access modules, such as ngx -
http access module and ngx http auth basic module, via the satisfy directive.

Before version 1.7.3, responses to authorization subrequests could not be

cached (using proxy cache, proxy store, etc.).

2.5.2 Example Configuration

location / private / {
auth_request / auth ;
...
}

location = / auth {
proxy_pass ...
p r o x y _ p a s s _ r e q u e s t _ b o d y off ;
p ro x y_ s et _ he a de r Content - Length "";
p ro x y_ s et _ he a de r X - Original - URI $request_uri ;
}

2.5.3 Directives
auth request
Syntax: auth_request uri | off;
Default off
Context: http, server, location

Enables authorization based on the result of a subrequest and sets the URI
to which the subrequest will be sent.

Nginx, Inc. p.61 of 330

CHAPTER 2. HTTP SERVER MODULES 2.5. MODULE NGX HTTP AUTH REQUEST MODULE

auth request set

Syntax: auth_request_set variable value;
Default
Context: http, server, location

Sets the request variable to the given value after the authorization request
completes. The value may contain variables from the authorization request,
such as $upstream http *.

Nginx, Inc. p.62 of 330

CHAPTER 2. HTTP SERVER MODULES 2.6. MODULE NGX HTTP AUTOINDEX MODULE

2.6 Module ngx http autoindex module

2.6.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 63
2.6.2 Example Configuration . . . . . . . . . . . . . . . . . . . 63
2.6.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 63
autoindex . . . . . . . . . . . . . . . . . . . . . . . . . . 63
autoindex exact size . . . . . . . . . . . . . . . . . . . . 63
autoindex format . . . . . . . . . . . . . . . . . . . . . . 63
autoindex localtime . . . . . . . . . . . . . . . . . . . . . 64

2.6.1 Summary
The ngx_http_autoindex_module module processes requests ending with
the slash character (/) and produces a directory listing. Usually a request
is passed to the ngx_http_autoindex_module module when the ngx http -
index module module cannot find an index file.

2.6.2 Example Configuration

location / {
autoindex on ;
}

2.6.3 Directives
autoindex
Syntax: autoindex on | off;
Default off
Context: http, server, location

Enables or disables the directory listing output.

autoindex exact size

Syntax: autoindex_exact_size on | off;
Default on
Context: http, server, location

For the HTML format, specifies whether exact file sizes should be output in
the directory listing, or rather rounded to kilobytes, megabytes, and gigabytes.

autoindex format
Syntax: autoindex_format html | xml | json | jsonp;
Default html
Context: http, server, location
This directive appeared in version 1.7.9.

Nginx, Inc. p.63 of 330

CHAPTER 2. HTTP SERVER MODULES 2.6. MODULE NGX HTTP AUTOINDEX MODULE

Sets the format of a directory listing.

When the JSONP format is used, the name of a callback function is set
with the callback request argument. If the argument is missing or has an
empty value, then the JSON format is used.

autoindex localtime
Syntax: autoindex_localtime on | off;
Default off
Context: http, server, location

For the HTML format, specifies whether times in the directory listing
should be output in the local time zone or UTC.

Nginx, Inc. p.64 of 330

CHAPTER 2. HTTP SERVER MODULES 2.7. MODULE NGX HTTP BROWSER MODULE

2.7 Module ngx http browser module

2.7.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 65
2.7.2 Example Configuration . . . . . . . . . . . . . . . . . . . 65
2.7.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 66
ancient browser . . . . . . . . . . . . . . . . . . . . . . . 66
ancient browser value . . . . . . . . . . . . . . . . . . . 66
modern browser . . . . . . . . . . . . . . . . . . . . . . . 66
modern browser value . . . . . . . . . . . . . . . . . . . 66

2.7.1 Summary
The ngx_http_browser_module module creates variables whose values
depend on the value of the User-Agent request header field:
$modern browser
equals the value set by the modern browser value directive, if a browser
was identified as modern;
$ancient browser
equals the value set by the ancient browser value directive, if a browser
was identified as ancient;
$msie
equals 1 if a browser was identified as MSIE of any version.

2.7.2 Example Configuration

Choosing an index file:

m o d e r n _ b r o w s e r _ v a l u e " modern .";

modern_browser msie 5.5;

modern_browser gecko 1.0.0;
modern_browser opera 9.0;
modern_browser safari 413;
modern_browser konqueror 3.0;

index index . $ { modern_browser } html index . html ;

Redirection for old browsers:

modern_browser msie 5.0;

modern_browser gecko 0.9.1;
modern_browser opera 8.0;
modern_browser safari 413;
modern_browser konqueror 3.0;

modern_browser unlisted ;

an cien t_br owse r Links Lynx netscape4 ;

if ( $ an c ie n t_ b ro w se r ) {
rewrite ^ / ancient . html ;
}

Nginx, Inc. p.65 of 330

CHAPTER 2. HTTP SERVER MODULES 2.7. MODULE NGX HTTP BROWSER MODULE

2.7.3 Directives
ancient browser
Syntax: ancient_browser string . . . ;
Default
Context: http, server, location

If any of the specified substrings is found in the User-Agent request header

field, the browser will be considered ancient. The special string netscape4
corresponds to the regular expression ^Mozilla/[1-4].

ancient browser value

Syntax: ancient_browser_value string;
Default 1
Context: http, server, location

Sets a value for the $ancient browser variables.

modern browser
Syntax: modern_browser browser version;
Syntax: modern_browser unlisted;
Default
Context: http, server, location

Specifies a version starting from which a browser is considered modern.

A browser can be any one of the following: msie, gecko (browsers based on
Mozilla), opera, safari, or konqueror.
Versions can be specified in the following formats: X, X.X, X.X.X, or
X.X.X.X. The maximum values for each of the format are 4000, 4000.99,
4000.99.99, and 4000.99.99.99, respectively.
The special value unlisted specifies to consider a browser as modern
if it was not listed by the modern_browser and ancient browser directives.
Otherwise such a browser is considered ancient. If a request does not provide
the User-Agent field in the header, the browser is treated as not being listed.

modern browser value

Syntax: modern_browser_value string;
Default 1
Context: http, server, location

Sets a value for the $modern browser variables.

Nginx, Inc. p.66 of 330

CHAPTER 2. HTTP SERVER MODULES 2.8. MODULE NGX HTTP CHARSET MODULE

2.8 Module ngx http charset module

2.8.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 67
2.8.2 Example Configuration . . . . . . . . . . . . . . . . . . . 67
2.8.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 67
charset . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
charset map . . . . . . . . . . . . . . . . . . . . . . . . . 68
charset types . . . . . . . . . . . . . . . . . . . . . . . . 69
override charset . . . . . . . . . . . . . . . . . . . . . . . 69
source charset . . . . . . . . . . . . . . . . . . . . . . . . 69

2.8.1 Summary
The ngx_http_charset_module module adds the specified charset to the
Content-Type response header field. In addition, the module can convert data
from one charset to another, with some limitations:

conversion is performed one way from server to client,

only single-byte charsets can be converted

or single-byte charsets to/from UTF-8.

2.8.2 Example Configuration

include conf / koi - win ;

charset windows -1251;

source_charset koi8 - r ;

2.8.3 Directives
charset
Syntax: charset charset | off;
Default off
Context: http, server, location, if in location

Adds the specified charset to the Content-Type response header field.

If this charset is different from the charset specified in the source charset
directive, a conversion is performed.
The parameter off cancels the addition of charset to the Content-Type
response header field.
A charset can be defined with a variable:

charset $charset ;

Nginx, Inc. p.67 of 330

CHAPTER 2. HTTP SERVER MODULES 2.8. MODULE NGX HTTP CHARSET MODULE

In such a case, all possible values of a variable need to be present in

the configuration at least once in the form of the charset map, charset, or
source charset directives. For utf-8, windows-1251, and koi8-r charsets,
it is sufficient to include the files conf/koi-win, conf/koi-utf, and conf
/win-utf into configuration. For other charsets, simply making a fictitious
conversion table works, for example:

charset_map iso -8859 -5 _ { }

In addition, a charset can be set in the X-Accel-Charset response header

field. This capability can be disabled using the proxy ignore headers, fastcgi -
ignore headers, uwsgi ignore headers, and scgi ignore headers directives.

charset map
Syntax: charset_map charset1 charset2 { . . . }
Default
Context: http

Describes the conversion table from one charset to another. A reverse

conversion table is built using the same data. Character codes are given in
hexadecimal. Missing characters in the range 80-FF are replaced with ?.
When converting from UTF-8, characters missing in a one-byte charset are
replaced with &#XXXX;.
Example:

charset_map koi8 - r windows -1251 {

C0 FE ; # small yu
C1 E0 ; # small a
C2 E1 ; # small b
C3 F6 ; # small ts
...
}

When describing a conversion table to UTF-8, codes for the UTF-8 charset
should be given in the second column, for example:

charset_map koi8 - r utf -8 {

C0 D18E ; # small yu
C1 D0B0 ; # small a
C2 D0B1 ; # small b
C3 D186 ; # small ts
...
}

Full conversion tables from koi8-r to windows-1251, and from koi8-r and
windows-1251 to utf-8 are provided in the distribution files conf/koi-win,
conf/koi-utf, and conf/win-utf.

Nginx, Inc. p.68 of 330

CHAPTER 2. HTTP SERVER MODULES 2.8. MODULE NGX HTTP CHARSET MODULE

charset types
Syntax: charset_types mime-type . . . ;
Default text/html text/xml text/plain text/vnd.wap.wml
application/javascript application/rss+xml
Context: http, server, location
This directive appeared in version 0.7.9.

Enables module processing in responses with the specified MIME types

in addition to text/html. The special value * matches any MIME type
(0.8.29).

Until version 1.5.4, application/x-javascript was used as the default

MIME type instead of application/javascript.

override charset
Syntax: override_charset on | off;
Default off
Context: http, server, location, if in location

Determines whether a conversion should be performed for answers received

from a proxied or a FastCGI/uwsgi/SCGI server when the answers already
carry a charset in the Content-Type response header field. If conversion is
enabled, a charset specified in the received response is used as a source charset.

It should be noted that if a response is received in a subrequest then the

conversion from the response charset to the main request charset is always
performed, regardless of the override_charset directive setting.

source charset
Syntax: source_charset charset;
Default
Context: http, server, location, if in location

Defines the source charset of a response. If this charset is different from

the charset specified in the charset directive, a conversion is performed.

Nginx, Inc. p.69 of 330

CHAPTER 2. HTTP SERVER MODULES 2.9. MODULE NGX HTTP DAV MODULE

2.9 Module ngx http dav module

2.9.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 70
2.9.2 Example Configuration . . . . . . . . . . . . . . . . . . . 70
2.9.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 70
dav access . . . . . . . . . . . . . . . . . . . . . . . . . . 70
dav methods . . . . . . . . . . . . . . . . . . . . . . . . 71
create full put path . . . . . . . . . . . . . . . . . . . . . 71
min delete depth . . . . . . . . . . . . . . . . . . . . . . 71

2.9.1 Summary
The ngx_http_dav_module module is intended for file management
automation via the WebDAV protocol. The module processes HTTP and
WebDAV methods PUT, DELETE, MKCOL, COPY, and MOVE.
This module is not built by default, it should be enabled with the
--with-http_dav_module configuration parameter.

WebDAV clients that require additional WebDAV methods to operate will

not work with this module.

2.9.2 Example Configuration

location / {
root / data / www ;

c l i e n t _ b o d y _ t e m p _ p a t h / data / client_temp ;

dav_methods PUT DELETE MKCOL COPY MOVE ;

create_full_put_path on ;
dav_access group : rw all : r ;

limit_except GET {
allow 192 .168 .1.0 /32 ;
deny all ;
}
}

2.9.3 Directives
dav access
Syntax: dav_access users:permissions . . . ;
Default user:rw
Context: http, server, location

Sets access permissions for newly created files and directories, e.g.:

dav_access user : rw group : rw all : r ;

Nginx, Inc. p.70 of 330

CHAPTER 2. HTTP SERVER MODULES 2.9. MODULE NGX HTTP DAV MODULE

If any group or all access permissions are specified then user permissions
may be omitted:

dav_access group : rw all : r ;

dav methods
Syntax: dav_methods off | method . . . ;
Default off
Context: http, server, location

Allows the specified HTTP and WebDAV methods. The parameter off
denies all methods processed by this module. The following methods are
supported: PUT, DELETE, MKCOL, COPY, and MOVE.
A file uploaded with the PUT method is first written to a temporary file,
and then the file is renamed. Starting from version 0.8.9, temporary files and
the persistent store can be put on different file systems. However, be aware
that in this case a file is copied across two file systems instead of the cheap
renaming operation. It is thus recommended that for any given location both
saved files and a directory holding temporary files, set by the client body -
temp path directive, are put on the same file system.
When creating a file with the PUT method, it is possible to specify the
modification date by passing it in the Date header field.

create full put path

Syntax: create_full_put_path on | off;
Default off
Context: http, server, location

The WebDAV specification only allows creating files in already existing

directories. This directive allows creating all needed intermediate directories.

min delete depth

Syntax: min_delete_depth number;
Default 0
Context: http, server, location

Allows the DELETE method to remove files provided that the number of
elements in a request path is not less than the specified number. For example,
the directive

m in _ de l et e _d e pt h 4;

allows removing files on requests

/ users /00/00/ name

/ users /00/00/ name / pic . jpg
/ users /00/00/ page . html

Nginx, Inc. p.71 of 330

CHAPTER 2. HTTP SERVER MODULES 2.9. MODULE NGX HTTP DAV MODULE

and denies the removal of

/ users /00/00

Nginx, Inc. p.72 of 330

CHAPTER 2. HTTP SERVER MODULES 2.10. MODULE NGX HTTP EMPTY GIF MODULE

2.10 Module ngx http empty gif module

2.10.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 73
2.10.2 Example Configuration . . . . . . . . . . . . . . . . . . . 73
2.10.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 73
empty gif . . . . . . . . . . . . . . . . . . . . . . . . . . 73

2.10.1 Summary
The ngx_http_empty_gif_module module emits single-pixel transparent
GIF.

2.10.2 Example Configuration

location = / _ . gif {
empty_gif ;
}

2.10.3 Directives
empty gif
Syntax: empty_gif;
Default
Context: location

Turns on module processing in a surrounding location.

Nginx, Inc. p.73 of 330

CHAPTER 2. HTTP SERVER MODULES 2.11. MODULE NGX HTTP F4F MODULE

2.11 Module ngx http f4f module

2.11.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 74
2.11.2 Example Configuration . . . . . . . . . . . . . . . . . . . 74
2.11.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 74
f4f . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
f4f buffer size . . . . . . . . . . . . . . . . . . . . . . . . 74

2.11.1 Summary
The ngx_http_f4f_module module provides server-side support for Adobe
HTTP Dynamic Streaming (HDS).
This module implements handling of HTTP Dynamic Streaming requests
in the /videoSeg1-Frag1 form extracting the needed fragment from the
videoSeg1.f4f file using the videoSeg1.f4x index file. This module is an
alternative to the Adobes f4f module (HTTP Origin Module) for Apache.
Usual pre-processing with Adobes f4fpackager is required, see relevant
documentation for details.

This module is available as part of our commercial subscription.

2.11.2 Example Configuration

location / video / {
f4f ;
...
}

2.11.3 Directives
f4f
Syntax: f4f;
Default
Context: location

Turns on module processing in the surrounding location.

f4f buffer size

Syntax: f4f_buffer_size size;
Default 512k
Context: http, server, location

Sets the size of the buffer used for reading the .f4x index file.

Nginx, Inc. p.74 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

2.12 Module ngx http fastcgi module

2.12.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 76
2.12.2 Example Configuration . . . . . . . . . . . . . . . . . . . 76
2.12.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 76
fastcgi bind . . . . . . . . . . . . . . . . . . . . . . . . . 76
fastcgi buffer size . . . . . . . . . . . . . . . . . . . . . . 76
fastcgi buffering . . . . . . . . . . . . . . . . . . . . . . . 77
fastcgi buffers . . . . . . . . . . . . . . . . . . . . . . . . 77
fastcgi busy buffers size . . . . . . . . . . . . . . . . . . 77
fastcgi cache . . . . . . . . . . . . . . . . . . . . . . . . . 78
fastcgi cache bypass . . . . . . . . . . . . . . . . . . . . 78
fastcgi cache key . . . . . . . . . . . . . . . . . . . . . . 78
fastcgi cache lock . . . . . . . . . . . . . . . . . . . . . . 78
fastcgi cache lock age . . . . . . . . . . . . . . . . . . . 79
fastcgi cache lock timeout . . . . . . . . . . . . . . . . . 79
fastcgi cache methods . . . . . . . . . . . . . . . . . . . 79
fastcgi cache min uses . . . . . . . . . . . . . . . . . . . 79
fastcgi cache path . . . . . . . . . . . . . . . . . . . . . 79
fastcgi cache purge . . . . . . . . . . . . . . . . . . . . . 80
fastcgi cache revalidate . . . . . . . . . . . . . . . . . . . 81
fastcgi cache use stale . . . . . . . . . . . . . . . . . . . 81
fastcgi cache valid . . . . . . . . . . . . . . . . . . . . . 82
fastcgi catch stderr . . . . . . . . . . . . . . . . . . . . . 82
fastcgi connect timeout . . . . . . . . . . . . . . . . . . 83
fastcgi force ranges . . . . . . . . . . . . . . . . . . . . . 83
fastcgi hide header . . . . . . . . . . . . . . . . . . . . . 83
fastcgi ignore client abort . . . . . . . . . . . . . . . . . 83
fastcgi ignore headers . . . . . . . . . . . . . . . . . . . 84
fastcgi index . . . . . . . . . . . . . . . . . . . . . . . . . 84
fastcgi intercept errors . . . . . . . . . . . . . . . . . . . 84
fastcgi keep conn . . . . . . . . . . . . . . . . . . . . . . 85
fastcgi limit rate . . . . . . . . . . . . . . . . . . . . . . 85
fastcgi max temp file size . . . . . . . . . . . . . . . . . 85
fastcgi next upstream . . . . . . . . . . . . . . . . . . . 85
fastcgi next upstream timeout . . . . . . . . . . . . . . . 86
fastcgi next upstream tries . . . . . . . . . . . . . . . . . 86
fastcgi no cache . . . . . . . . . . . . . . . . . . . . . . . 87
fastcgi param . . . . . . . . . . . . . . . . . . . . . . . . 87
fastcgi pass . . . . . . . . . . . . . . . . . . . . . . . . . 88
fastcgi pass header . . . . . . . . . . . . . . . . . . . . . 88
fastcgi read timeout . . . . . . . . . . . . . . . . . . . . 88
fastcgi pass request body . . . . . . . . . . . . . . . . . 88
fastcgi pass request headers . . . . . . . . . . . . . . . . 89
fastcgi request buffering . . . . . . . . . . . . . . . . . . 89
fastcgi send lowat . . . . . . . . . . . . . . . . . . . . . . 89

Nginx, Inc. p.75 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi send timeout . . . . . . . . . . . . . . . . . . . . 89

fastcgi split path info . . . . . . . . . . . . . . . . . . . 89
fastcgi store . . . . . . . . . . . . . . . . . . . . . . . . . 90
fastcgi store access . . . . . . . . . . . . . . . . . . . . . 91
fastcgi temp file write size . . . . . . . . . . . . . . . . . 91
fastcgi temp path . . . . . . . . . . . . . . . . . . . . . . 91
2.12.4 Parameters Passed to a FastCGI Server . . . . . . . . . . 92
2.12.5 Embedded Variables . . . . . . . . . . . . . . . . . . . . 92

2.12.1 Summary
The ngx_http_fastcgi_module module allows passing requests to a
FastCGI server.

2.12.2 Example Configuration

location / {
fastcgi_pass localhost :9000;
fastcgi_index index . php ;

fastcgi_param SCR IPT_ FIL ENAM E / home / www / scripts /

php$fastcgi_script_name ;
fastcgi_param QUERY_STRING $query_string ;
fastcgi_param REQUEST_METHOD $re que st_m etho d ;
fastcgi_param CONTENT_TYPE $content_type ;
fastcgi_param CONTENT_LENGTH $co nte nt_l engt h ;
}

2.12.3 Directives
fastcgi bind
Syntax: fastcgi_bind address | off;
Default
Context: http, server, location
This directive appeared in version 0.8.22.

Makes outgoing connections to a FastCGI server originate from the

specified local IP address. Parameter value can contain variables (1.3.12).
The special value off (1.3.12) cancels the effect of the fastcgi_bind directive
inherited from the previous configuration level, which allows the system to
auto-assign the local IP address.

fastcgi buffer size

Syntax: fastcgi_buffer_size size;
Default 4k|8k
Context: http, server, location

Nginx, Inc. p.76 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

Sets the size of the buffer used for reading the first part of the response
received from the FastCGI server. This part usually contains a small response
header. By default, the buffer size is equal to the size of one buffer set by the
fastcgi buffers directive. It can be made smaller, however.

fastcgi buffering
Syntax: fastcgi_buffering on | off;
Default on
Context: http, server, location
This directive appeared in version 1.5.6.

Enables or disables buffering of responses from the FastCGI server.

When buffering is enabled, nginx receives a response from the FastCGI
server as soon as possible, saving it into the buffers set by the fastcgi buffer -
size and fastcgi buffers directives. If the whole response does not fit into
memory, a part of it can be saved to a temporary file on the disk. Writing
to temporary files is controlled by the fastcgi max temp file size and fastcgi -
temp file write size directives.
When buffering is disabled, the response is passed to a client synchronously,
immediately as it is received. nginx will not try to read the whole response
from the FastCGI server. The maximum size of the data that nginx can receive
from the server at a time is set by the fastcgi buffer size directive.
Buffering can also be enabled or disabled by passing yes or no in the
X-Accel-Buffering response header field. This capability can be disabled
using the fastcgi ignore headers directive.

fastcgi buffers
Syntax: fastcgi_buffers number size;
Default 8 4k|8k
Context: http, server, location

Sets the number and size of the buffers used for reading a response from
the FastCGI server, for a single connection. By default, the buffer size is equal
to one memory page. This is either 4K or 8K, depending on a platform.

fastcgi busy buffers size

Syntax: fastcgi_busy_buffers_size size;
Default 8k|16k
Context: http, server, location

When buffering of responses from the FastCGI server is enabled, limits the
total size of buffers that can be busy sending a response to the client while the
response is not yet fully read. In the meantime, the rest of the buffers can be
used for reading the response and, if needed, buffering part of the response to
a temporary file. By default, size is limited by the size of two buffers set by
the fastcgi buffer size and fastcgi buffers directives.

Nginx, Inc. p.77 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi cache
Syntax: fastcgi_cache zone | off;
Default off
Context: http, server, location

Defines a shared memory zone used for caching. The same zone can be
used in several places. Parameter value can contain variables (1.7.9). The off
parameter disables caching inherited from the previous configuration level.

fastcgi cache bypass

Syntax: fastcgi_cache_bypass string . . . ;
Default
Context: http, server, location

Defines conditions under which the response will not be taken from a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be taken from the cache:

f a s t c g i _ c a c h e _ b y p a s s $coo kie_ noc ache $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

f a s t c g i _ c a c h e _ b y p a s s $http_pragma $http_authorization ;

Can be used along with the fastcgi no cache directive.

fastcgi cache key

Syntax: fastcgi_cache_key string;
Default
Context: http, server, location

Defines a key for caching, for example

f a s t c g i _ c a c h e _ k e y localhost :9000 $request_uri ;

fastcgi cache lock

Syntax: fastcgi_cache_lock on | off;
Default off
Context: http, server, location
This directive appeared in version 1.1.12.

When enabled, only one request at a time will be allowed to populate a new
cache element identified according to the fastcgi cache key directive by passing
a request to a FastCGI server. Other requests of the same cache element will
either wait for a response to appear in the cache or the cache lock for this
element to be released, up to the time set by the fastcgi cache lock timeout
directive.

Nginx, Inc. p.78 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi cache lock age

Syntax: fastcgi_cache_lock_age time;
Default 5s
Context: http, server, location
This directive appeared in version 1.7.8.

If the last request passed to the FastCGI server for populating a new cache
element has not completed for the specified time, one more request may be
passed to the FastCGI server.

fastcgi cache lock timeout

Syntax: fastcgi_cache_lock_timeout time;
Default 5s
Context: http, server, location
This directive appeared in version 1.1.12.

Sets a timeout for fastcgi cache lock. When the time expires, the request
will be passed to the FastCGI server, however, the response will not be cached.

Before 1.7.8, the response could be cached.

fastcgi cache methods

Syntax: fastcgi_cache_methods GET | HEAD | POST . . . ;
Default GET HEAD
Context: http, server, location
This directive appeared in version 0.7.59.

If the client request method is listed in this directive then the response will
be cached. GET and HEAD methods are always added to the list, though
it is recommended to specify them explicitly. See also the fastcgi no cache
directive.

fastcgi cache min uses

Syntax: fastcgi_cache_min_uses number;
Default 1
Context: http, server, location

Sets the number of requests after which the response will be cached.

fastcgi cache path

Syntax: fastcgi_cache_path path [levels=levels] [use_temp_path=on|off]
keys_zone=name:size [inactive=time] [max_size=size]
[loader_files=number] [loader_sleep=time]
[loader_threshold=time];
Default
Context: http

Nginx, Inc. p.79 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

Sets the path and other parameters of a cache. Cache data are stored in
files. Both the key and file name in a cache are a result of applying the MD5
function to the proxied URL.
The levels parameter defines hierarchy levels of a cache. For example, in
the following configuration

f a s t c g i _ c a c h e _ p a t h / data / nginx / cache levels =1:2 keys_zone = one :10 m ;

file names in a cache will look like this:

/ data / nginx / cache /c /29 / b 7 f 5 4 b 2 d f 7 7 7 3 7 2 2 d 3 8 2 f 4 8 0 9 d 6 5 0 29c

A cached response is first written to a temporary file, and then the file is
renamed. Starting from version 0.8.9, temporary files and the cache can be put
on different file systems. However, be aware that in this case a file is copied
across two file systems instead of the cheap renaming operation. It is thus
recommended that for any given location both cache and a directory holding
temporary files are put on the same file system. A directory for temporary
files is set based on the use_temp_path parameter (1.7.10). If this parameter
is omitted or set to the value on, the directory set by the fastcgi temp path
directive for the given location will be used. If the value is set to off, temporary
files will be put directly in the cache directory.
In addition, all active keys and information about data are stored in a
shared memory zone, whose name and size are configured by the keys_zone
parameter. One megabyte zone can store about 8 thousand keys.
Cached data that are not accessed during the time specified by the
inactive parameter get removed from the cache regardless of their freshness.
By default, inactive is set to 10 minutes.
The special cache manager process monitors the maximum cache size set
by the max_size parameter. When this size is exceeded, it removes the least
recently used data.
A minute after the start the special cache loader process is activated. It
loads information about previously cached data stored on file system into a
cache zone. The loading is done in iterations. During one iteration no more
than loader_files items are loaded (by default, 100). Besides, the duration of
one iteration is limited by the loader_threshold parameter (by default, 200
milliseconds). Between iterations, a pause configured by the loader_sleep
parameter (by default, 50 milliseconds) is made.

fastcgi cache purge

Syntax: fastcgi_cache_purgestring . . . ;
Default
Context: http, server, location
This directive appeared in version 1.5.7.

Defines conditions under which the request will be considered a cache purge
request. If at least one value of the string parameters is not empty and

Nginx, Inc. p.80 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

is not equal to 0 then the cache entry with a corresponding cache key is
removed. The result of successful operation is indicated by returning the 204
No Content response.
If the cache key of a purge request ends with an asterisk (*), all cache
entries matching the wildcard key will be removed from the cache.
Example configuration:

f a s t c g i _ c a c h e _ p a t h / data / nginx / cache keys_zone = cache_zone :10 m ;

map $re ques t_m etho d $purge_method {

PURGE 1;
default 0;
}

server {
...
location / {
fastcgi_pass backend ;
fastcgi_cache cache_zone ;
fastcgi_cache_key $uri ;
fastcgi_cache_purge $purge_method ;
}
}

This functionality is available as part of our commercial subscription.

fastcgi cache revalidate

Syntax: fastcgi_cache_revalidate on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.7.

Enables revalidation of expired cache items using conditional requests with

the If-Modified-Since and If-None-Match header fields.

fastcgi cache use stale

Syntax: fastcgi_cache_use_stale error | timeout | invalid_header |
updating | http_500 | http_503 | http_403 | http_404 | off . . . ;
Default off
Context: http, server, location

Determines in which cases a stale cached response can be used when an

error occurs during communication with the FastCGI server. The directives
parameters match the parameters of the fastcgi next upstream directive.
Additionally, the updating parameter permits using a stale cached response
if it is currently being updated. This allows minimizing the number of accesses
to FastCGI servers when updating cached data.
To minimize the number of accesses to FastCGI servers when populating a
new cache element, the fastcgi cache lock directive can be used.

Nginx, Inc. p.81 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi cache valid

Syntax: fastcgi_cache_valid [code . . . ] time;
Default
Context: http, server, location

Sets caching time for different response codes. For example, the following
directives

f a s t c g i _ c a c h e _ v a l i d 200 302 10 m ;
f a s t c g i _ c a c h e _ v a l i d 404 1m;

set 10 minutes of caching for responses with codes 200 and 302 and 1 minute
for responses with code 404.
If only caching time is specified

fastcgi_cache_valid 5m;

then only 200, 301, and 302 responses are cached.

In addition, the any parameter can be specified to cache any responses:

f a s t c g i _ c a c h e _ v a l i d 200 302 10 m ;
f a s t c g i _ c a c h e _ v a l i d 301 1h;
f a s t c g i _ c a c h e _ v a l i d any 1m;

Parameters of caching can also be set directly in the response header. This
has higher priority than setting of caching time using the directive.
The X-Accel-Expires header field sets caching time of a response in
seconds. The zero value disables caching for a response. If the value
starts with the @ prefix, it sets an absolute time in seconds since Epoch,
up to which the response may be cached.

If the header does not include the X-Accel-Expires field, parameters of

caching may be set in the header fields Expires or Cache-Control.

If the header includes the Set-Cookie field, such a response will not be
cached.

If the header includes the Vary field with the special value *, such a
response will not be cached (1.7.7). If the header includes the Vary field
with another value, such a response will be cached taking into account
the corresponding request header fields (1.7.7).
Processing of one or more of these response header fields can be disabled using
the fastcgi ignore headers directive.

fastcgi catch stderr

Syntax: fastcgi_catch_stderr string;
Default
Context: http, server, location

Nginx, Inc. p.82 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

Sets a string to search for in the error stream of a response received from
a FastCGI server. If the string is found then it is considered that the FastCGI
server has returned an invalid response. This allows handling application errors
in nginx, for example:

location / php {
fastcgi_pass backend :9000;
...
f a s t c g i _ c a t c h _ s t d e r r " PHP Fatal error ";
f a s t c g i _ n e x t _ u p s t r e a m error timeout invalid_header ;
}

fastcgi connect timeout

Syntax: fastcgi_connect_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for establishing a connection with a FastCGI server. It

should be noted that this timeout cannot usually exceed 75 seconds.

fastcgi force ranges

Syntax: fastcgi_force_ranges on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.7.

Enables byte-range support for both cached and uncached responses from
the FastCGI server regardless of the Accept-Ranges field in these responses.

fastcgi hide header

Syntax: fastcgi_hide_header field;
Default
Context: http, server, location

By default, nginx does not pass the header fields Status and X-Accel-...
from the response of a FastCGI server to a client. The fastcgi_hide_header
directive sets additional fields that will not be passed. If, on the contrary, the
passing of fields needs to be permitted, the fastcgi pass header directive can
be used.

fastcgi ignore client abort

Syntax: fastcgi_ignore_client_abort on | off;
Default off
Context: http, server, location

Determines whether the connection with a FastCGI server should be closed

when a client closes the connection without waiting for a response.

Nginx, Inc. p.83 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi ignore headers

Syntax: fastcgi_ignore_headers field . . . ;
Default
Context: http, server, location

Disables processing of certain response header fields from the FastCGI

server. The following fields can be ignored: X-Accel-Redirect,
X-Accel-Expires, X-Accel-Limit-Rate (1.1.6), X-Accel-Buffering (1.1.6),
X-Accel-Charset (1.1.6), Expires, Cache-Control, Set-Cookie (0.8.44),
and Vary (1.7.7).
If not disabled, processing of these header fields has the following effect:

X-Accel-Expires, Expires, Cache-Control, Set-Cookie, and Vary set

the parameters of response caching;

X-Accel-Redirect performs an internal redirect to the specified URI;

X-Accel-Limit-Rate sets the rate limit for transmission of a response

to a client;

X-Accel-Buffering enables or disables buffering of a response;

X-Accel-Charset sets the desired charset of a response.

fastcgi index
Syntax: fastcgi_index name;
Default
Context: http, server, location

Sets a file name that will be appended after a URI that ends with a slash, in
the value of the $fastcgi script name variable. For example, with these settings

fastcgi_index index . php ;

fastcgi_param SCR IPT _FIL ENAM E / home / www / scripts / p h p $ f a s t c g i _ s c r i p t _ n a m e ;

and the /page.php request, the SCRIPT_FILENAME parameter will be

equal to /home/www/scripts/php/page.php, and with the / request it
will be equal to /home/www/scripts/php/index.php.

fastcgi intercept errors

Syntax: fastcgi_intercept_errors on | off;
Default off
Context: http, server, location

Determines whether FastCGI server responses with codes greater than or

equal to 300 should be passed to a client or be redirected to nginx for processing
with the error page directive.

Nginx, Inc. p.84 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi keep conn

Syntax: fastcgi_keep_conn on | off;
Default off
Context: http, server, location
This directive appeared in version 1.1.4.

By default, a FastCGI server will close a connection right after sending the
response. However, when this directive is set to the value on, nginx will instruct
a FastCGI server to keep connections open. This is necessary, in particular,
for keepalive connections to FastCGI servers to function.

fastcgi limit rate

Syntax: fastcgi_limit_rate rate;
Default 0
Context: http, server, location
This directive appeared in version 1.7.7.

Limits the speed of reading the response from the FastCGI server. The
rate is specified in bytes per second. The zero value disables rate limiting. The
limit is set per a request, and so if nginx simultaneously opens two connections
to the FastCFI server, the overall rate will be twice as much as the specified
limit. The limitation works only if buffering of responses from the FastCGI
server is enabled.

fastcgi max temp file size

Syntax: fastcgi_max_temp_file_size size;
Default 1024m
Context: http, server, location

When buffering of responses from the FastCGI server is enabled, and the
whole response does not fit into the buffers set by the fastcgi buffer size and
fastcgi buffers directives, a part of the response can be saved to a temporary
file. This directive sets the maximum size of the temporary file. The size of
data written to the temporary file at a time is set by the fastcgi temp file -
write size directive.
The zero value disables buffering of responses to temporary files.

This restriction does not apply to responses that will be cached or stored
on disk.

fastcgi next upstream

Syntax: fastcgi_next_upstream error | timeout | invalid_header | http_500
| http_503 | http_403 | http_404 | off . . . ;
Default error timeout
Context: http, server, location

Nginx, Inc. p.85 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

Specifies in which cases a request should be passed to the next server:

error
an error occurred while establishing a connection with the server, passing
a request to it, or reading the response header;
timeout
a timeout has occurred while establishing a connection with the server,
passing a request to it, or reading the response header;
invalid_header
a server returned an empty or invalid response;
http_500
a server returned a response with the code 500;
http_503
a server returned a response with the code 503;
http_403
a server returned a response with the code 403;
http_404
a server returned a response with the code 404;
off
disables passing a request to the next server.

One should bear in mind that passing a request to the next server is only
possible if nothing has been sent to a client yet. That is, if an error or timeout
occurs in the middle of the transferring of a response, fixing this is impossible.
The directive also defines what is considered an unsuccessful attempt
of communication with a server. The cases of error, timeout and
invalid_header are always considered unsuccessful attempts, even if they
are not specified in the directive. The cases of http_500 and http_503 are
considered unsuccessful attempts only if they are specified in the directive. The
cases of http_403 and http_404 are never considered unsuccessful attempts.
Passing a request to the next server can be limited by the number of tries
and by time.

fastcgi next upstream timeout

Syntax: fastcgi_next_upstream_timeout time;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the time allowed to pass a request to the next server. The 0 value
turns off this limitation.

fastcgi next upstream tries

Syntax: fastcgi_next_upstream_tries number;
Default 0
Context: http, server, location

Nginx, Inc. p.86 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

This directive appeared in version 1.7.5.

Limits the number of possible tries for passing a request to the next server.
The 0 value turns off this limitation.

fastcgi no cache
Syntax: fastcgi_no_cache string . . . ;
Default
Context: http, server, location

Defines conditions under which the response will not be saved to a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be saved:

f as t cg i _n o _c a ch e $ cook ie_n ocac he $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

f as t cg i _n o _c a ch e $http_pragma $http_authorization ;

Can be used along with the fastcgi cache bypass directive.

fastcgi param
Syntax: fastcgi_param parameter value [if_not_empty];
Default
Context: http, server, location

Sets a parameter that should be passed to the FastCGI server. The

value can contain text, variables, and their combination. These directives
are inherited from the previous level if and only if there are no fastcgi_param
directives defined on the current level.
The following example shows the minimum required settings for PHP:

fastcgi_param SCR IPT _FIL ENAM E / home / www / scripts / p h p $ f a s t c g i _ s c r i p t _ n a m e ;

fastcgi_param QUERY_STRING $query_string ;

The SCRIPT_FILENAME parameter is used in PHP for determining the script

name, and the QUERY_STRING parameter is used to pass request parameters.
For scripts that process POST requests, the following three parameters are
also required:

fastcgi_param REQUEST_METHOD $r eque st_m etho d ;

fastcgi_param CONTENT_TYPE $content_type ;
fastcgi_param CONTENT_LENGTH $c onte nt_l engt h ;

If PHP was built with the --enable-force-cgi-redirect configuration

parameter, the REDIRECT_STATUS parameter should also be passed with the
value 200:

fastcgi_param RED IRE CT_S TATU S 200;

If a directive is specified with if_not_empty (1.1.11) then such a parameter

will not be passed to the server until its value is not empty:

Nginx, Inc. p.87 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi_param HTTPS $https if_not_empty ;

fastcgi pass
Syntax: fastcgi_pass address;
Default
Context: location, if in location

Sets the address of a FastCGI server. The address can be specified as a

domain name or IP address, and an optional port:

fastcgi_pass localhost :9000;

or as a UNIX-domain socket path:

fastcgi_pass unix :/ tmp / fastcgi . socket ;

If a domain name resolves to several addresses, all of them will be used

in a round-robin fashion. In addition, an address can be specified as a server
group.

fastcgi pass header

Syntax: fastcgi_pass_header field;
Default
Context: http, server, location

Permits passing otherwise disabled header fields from a FastCGI server to

a client.

fastcgi read timeout

Syntax: fastcgi_read_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for reading a response from the FastCGI server. The
timeout is set only between two successive read operations, not for the
transmission of the whole response. If the FastCGI server does not transmit
anything within this time, the connection is closed.

fastcgi pass request body

Syntax: fastcgi_pass_request_body on | off;
Default on
Context: http, server, location

Indicates whether the original request body is passed to the FastCGI server.
See also the fastcgi pass request headers directive.

Nginx, Inc. p.88 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi pass request headers

Syntax: fastcgi_pass_request_headers on | off;
Default on
Context: http, server, location

Indicates whether the header fields of the original request are passed to the
FastCGI server. See also the fastcgi pass request body directive.

fastcgi request buffering

Syntax: fastcgi_request_buffering on | off;
Default on
Context: http, server, location
This directive appeared in version 1.7.11.

Enables or disables buffering of a client request body.

When buffering is enabled, the entire request body is read from the client
before sending the request to a FastCGI server.
When buffering is disabled, the request body is sent to the FastCGI server
immediately as it is received. In this case, the request cannot be passed to the
next server if nginx already started sending the request body.

fastcgi send lowat

Syntax: fastcgi_send_lowat size;
Default 0
Context: http, server, location

If the directive is set to a non-zero value, nginx will try to minimize the
number of send operations on outgoing connections to a FastCGI server by
using either NOTE_LOWAT flag of the kqueue method, or the SO_SNDLOWAT socket
option, with the specified size.
This directive is ignored on Linux, Solaris, and Windows.

fastcgi send timeout

Syntax: fastcgi_send_timeout time;
Default 60s
Context: http, server, location

Sets a timeout for transmitting a request to the FastCGI server. The

timeout is set only between two successive write operations, not for the
transmission of the whole request. If the FastCGI server does not receive
anything within this time, the connection is closed.

fastcgi split path info

Syntax: fastcgi_split_path_info regex;
Default
Context: location

Nginx, Inc. p.89 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

Defines a regular expression that captures a value for the $fastcgi path info
variable. The regular expression should have two captures: the first becomes
a value of the $fastcgi script name variable, the second becomes a value of the
$fastcgi path info variable. For example, with these settings

location ~ ^(.+\. php ) (.*) $ {

fastcgi_split_path_info ^(.+\. php ) (.*) $ ;
fastcgi_param SCR IPT_ FIL ENAM E / path / to / p h p $ f a s t c g i _ s c r i p t _ n a m e ;
fastcgi_param PATH_INFO $fastcgi_path_info ;

and the /show.php/article/0001 request, the SCRIPT_FILENAME

parameter will be equal to /path/to/php/show.php, and the PATH_INFO
parameter will be equal to /article/0001.

fastcgi store
Syntax: fastcgi_store on | off | string;
Default off
Context: http, server, location

Enables saving of files to a disk. The on parameter saves files with paths
corresponding to the directives alias or root. The off parameter disables saving
of files. In addition, the file name can be set explicitly using the string with
variables:

fastcgi_store / data / ww w $o r ig i na l _u r i ;

The modification time of files is set according to the received

Last-Modified response header field. The response is first written to a
temporary file, and then the file is renamed. Starting from version 0.8.9,
temporary files and the persistent store can be put on different file systems.
However, be aware that in this case a file is copied across two file systems
instead of the cheap renaming operation. It is thus recommended that for any
given location both saved files and a directory holding temporary files, set by
the fastcgi temp path directive, are put on the same file system.
This directive can be used to create local copies of static unchangeable files,
e.g.:

location / images / {
root / data / www ;
error_page 404 = / fetch$uri ;
}

location / fetch / {
internal ;

fastcgi_pass backend :9000;

...

fastcgi_store on ;
f a s t c g i _ s t o r e _ a c c e s s user : rw group : rw all : r ;
fastcgi_temp_path / data / temp ;

alias / data / www /;

}

Nginx, Inc. p.90 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

fastcgi store access

Syntax: fastcgi_store_access users:permissions . . . ;
Default user:rw
Context: http, server, location

Sets access permissions for newly created files and directories, e.g.:

f a s t c g i _ s t o r e _ a c c e s s user : rw group : rw all : r ;

If any group or all access permissions are specified then user permissions
may be omitted:

f a s t c g i _ s t o r e _ a c c e s s group : rw all : r ;

fastcgi temp file write size

Syntax: fastcgi_temp_file_write_size size;
Default 8k|16k
Context: http, server, location

Limits the size of data written to a temporary file at a time, when buffering
of responses from the FastCGI server to temporary files is enabled. By default,
size is limited by two buffers set by the fastcgi buffer size and fastcgi buffers
directives. The maximum size of a temporary file is set by the fastcgi max -
temp file size directive.

fastcgi temp path

Syntax: fastcgi_temp_path path [level1 [level2 [level3]]];
Default fastcgi_temp
Context: http, server, location

Defines a directory for storing temporary files with data received from
FastCGI servers. Up to three-level subdirectory hierarchy can be used
underneath the specified directory. For example, in the following configuration

f a s t c g i _ t e m p _ p a t h / spool / nginx / fastcgi_temp 1 2;

a temporary file might look like this:

/ spool / nginx / fastcgi_temp /7 /45 /00000123457

See also the use_temp_path parameter of the fastcgi cache path directive.

Nginx, Inc. p.91 of 330

CHAPTER 2. HTTP SERVER MODULES 2.12. MODULE NGX HTTP FASTCGI MODULE

2.12.4 Parameters Passed to a FastCGI Server

HTTP request header fields are passed to a FastCGI server as parameters.
In applications and scripts running as FastCGI servers, these parameters are
usually made available as environment variables. For example, the User-Agent
header field is passed as the HTTP_USER_AGENT parameter. In addition to
HTTP request header fields, it is possible to pass arbitrary parameters using
the fastcgi param directive.

2.12.5 Embedded Variables

The ngx_http_fastcgi_module module supports embedded variables that
can be used to set parameters using the fastcgi param directive:

$fastcgi script name

request URI or, if a URI ends with a slash, request URI with an index
file name configured by the fastcgi index directive appended to it. This
variable can be used to set the SCRIPT_FILENAME and PATH_TRANSLATED
parameters that determine the script name in PHP. For example, for the
/info/ request with the following directives

fastcgi_index index . php ;

fastcgi_param SCR IPT _FIL ENAM E / home / www / scripts /
php$fastcgi_script_name ;

the SCRIPT_FILENAME parameter will be equal to

/home/www/scripts/php/info/index.php.
When using the fastcgi split path info directive, the $fastcgi script name
variable equals the value of the first capture set by the directive.
$fastcgi path info
the value of the second capture set by the fastcgi split path info
directive. This variable can be used to set the PATH_INFO parameter.

Nginx, Inc. p.92 of 330

CHAPTER 2. HTTP SERVER MODULES 2.13. MODULE NGX HTTP FLV MODULE

2.13 Module ngx http flv module

2.13.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 93
2.13.2 Example Configuration . . . . . . . . . . . . . . . . . . . 93
2.13.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 93
flv . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

2.13.1 Summary
The ngx_http_flv_module module provides pseudo-streaming server-side
support for Flash Video (FLV) files.
It handles requests with the start argument in the request URIs query
string specially, by sending back the contents of a file starting from the
requested byte offset and with the prepended FLV header.
This module is not built by default, it should be enabled with the
--with-http_flv_module configuration parameter.

2.13.2 Example Configuration

location ~ \. flv$ {
flv ;
}

2.13.3 Directives
flv
Syntax: flv;
Default
Context: location

Turns on module processing in a surrounding location.

Nginx, Inc. p.93 of 330

CHAPTER 2. HTTP SERVER MODULES 2.14. MODULE NGX HTTP GEO MODULE

2.14 Module ngx http geo module

2.14.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 94
2.14.2 Example Configuration . . . . . . . . . . . . . . . . . . . 94
2.14.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 94
geo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

2.14.1 Summary
The ngx_http_geo_module module creates variables with values depending
on the client IP address.

2.14.2 Example Configuration

geo $geo {
default 0;

127.0.0.1 2;
192.168.1.0/24 1;
10.1.0.0/16 1;

::1 2;
2001:0 db8 ::/32 1;
}

2.14.3 Directives
geo
Syntax: geo [$address] $variable { . . . }
Default
Context: http

Describes the dependency of values of the specified variable on the client

IP address. By default, the address is taken from the $remote addr variable,
but it can also be taken from another variable (0.7.27), for example:

geo $ a rg _ re mo t e_ a dd r $geo {
...;
}

Since variables are evaluated only when used, the mere existence of even
a large number of declared geo variables does not cause any extra costs for
request processing.

If the value of a variable does not represent a valid IP address then the
255.255.255.255 address is used.
Addresses are specified either as prefixes in CIDR notation (including
individual addresses) or as ranges (0.7.23).

Nginx, Inc. p.94 of 330

CHAPTER 2. HTTP SERVER MODULES 2.14. MODULE NGX HTTP GEO MODULE

IPv6 prefixes are supported starting from versions 1.3.10 and 1.2.7.

The following special parameters are also supported:

delete
deletes the specified network (0.7.23).
default
a value set to the variable if the client address does not match any of
the specified addresses. When addresses are specified in CIDR notation,
0.0.0.0/0 and ::/0 can be used instead of default. When default
is not specified, the default value will be an empty string.
include
includes a file with addresses and values. There can be several inclusions.
proxy
defines trusted addresses (0.8.7, 0.7.63). When a request comes from a
trusted address, an address from the X-Forwarded-For request header
field will be used instead. In contrast to the regular addresses, trusted
addresses are checked sequentially.

Trusted IPv6 addresses are supported starting from versions 1.3.0 and
1.2.1.

proxy_recursive
enables recursive address search (1.3.0, 1.2.1). If recursive search is
disabled then instead of the original client address that matches one
of the trusted addresses, the last address sent in X-Forwarded-For will
be used. If recursive search is enabled then instead of the original client
address that matches one of the trusted addresses, the last non-trusted
address sent in X-Forwarded-For will be used.
ranges
indicates that addresses are specified as ranges (0.7.23). This parameter
should be the first. To speed up loading of a geo base, addresses should
be put in ascending order.

Example:

geo $country {
default ZZ ;
include conf / geo . conf ;
delete 127.0.0.0/16;
proxy 192.168.100.0/24;
proxy 2001:0 db8 ::/32;

127.0.0.0/24 US ;
127.0.0.1/32 RU ;
10.1.0.0/16 RU ;
192.168.1.0/24 UK ;
}

The conf/geo.conf file could contain the following lines:

Nginx, Inc. p.95 of 330

CHAPTER 2. HTTP SERVER MODULES 2.14. MODULE NGX HTTP GEO MODULE

10.2.0.0/16 RU ;
192.168.2.0/24 RU ;

A value of the most specific match is used. For example, for the 127.0.0.1
address the value RU will be chosen, not US.
Example with ranges:

geo $country {
ranges ;
default ZZ ;
127.0.0.0 -127.0.0.0 US ;
127.0.0.1 -127.0.0.1 RU ;
127.0.0.1 -127.0.0.255 US ;
10.1.0.0 -10.1.255.255 RU ;
192.168.1.0 -192.168.1.255 UK ;
}

Nginx, Inc. p.96 of 330

CHAPTER 2. HTTP SERVER MODULES 2.15. MODULE NGX HTTP GEOIP MODULE

2.15 Module ngx http geoip module

2.15.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 97
2.15.2 Example Configuration . . . . . . . . . . . . . . . . . . . 97
2.15.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 97
geoip country . . . . . . . . . . . . . . . . . . . . . . . . 97
geoip city . . . . . . . . . . . . . . . . . . . . . . . . . . 98
geoip org . . . . . . . . . . . . . . . . . . . . . . . . . . 99
geoip proxy . . . . . . . . . . . . . . . . . . . . . . . . . 99
geoip proxy recursive . . . . . . . . . . . . . . . . . . . . 99

2.15.1 Summary
The ngx_http_geoip_module module (0.8.6+) creates variables with
values depending on the client IP address, using the precompiled MaxMind
databases.
When using the databases with IPv6 support (1.3.12, 1.2.7), IPv4 addresses
are looked up as IPv4-mapped IPv6 addresses.
This module is not built by default, it should be enabled with the
--with-http_geoip_module configuration parameter.

This module requires the MaxMind GeoIP library.

2.15.2 Example Configuration

http {
geoip_country GeoIP . dat ;
geoip_city GeoLiteCity . dat ;
geoip_proxy 192.168.100.0/24;
geoip_proxy 2001:0 db8 ::/32;
geoip_proxy_recursive on ;
...

2.15.3 Directives
geoip country
Syntax: geoip_country file;
Default
Context: http

Specifies a database used to determine the country depending on the client

IP address. The following variables are available when using this database:
$geoip country code
two-letter country code, for example, RU, US.
$geoip country code3
three-letter country code, for example, RUS, USA.

Nginx, Inc. p.97 of 330

CHAPTER 2. HTTP SERVER MODULES 2.15. MODULE NGX HTTP GEOIP MODULE

$geoip country name

country name, for example, Russian Federation, United States.

geoip city
Syntax: geoip_city file;
Default
Context: http

Specifies a database used to determine the country, region, and city

depending on the client IP address. The following variables are available when
using this database:

$geoip area code

telephone area code (US only).

This variable may contain outdated information since the corresponding

database field is deprecated.

$geoip city continent code

two-letter continent code, for example, EU, NA.
$geoip city country code
two-letter country code, for example, RU, US.
$geoip city country code3
three-letter country code, for example, RUS, USA.
$geoip city country name
country name, for example, Russian Federation, United States.
$geoip dma code
DMA region code in US (also known as metro code), according to the
geotargeting in Google AdWords API.
$geoip latitude
latitude.
$geoip longitude
longitude.
$geoip region
two-symbol country region code (region, territory, state, province, federal
land and the like), for example, 48, DC.
$geoip region name
country region name (region, territory, state, province, federal land and
the like), for example, Moscow City, District of Columbia.
$geoip city
city name, for example, Moscow, Washington.
$geoip postal code
postal code.

Nginx, Inc. p.98 of 330

CHAPTER 2. HTTP SERVER MODULES 2.15. MODULE NGX HTTP GEOIP MODULE

geoip org
Syntax: geoip_org file;
Default
Context: http
This directive appeared in version 1.0.3.

Specifies a database used to determine the organization depending on the

client IP address. The following variable is available when using this database:

$geoip org
organization name, for example, The University of Melbourne.

geoip proxy
Syntax: geoip_proxy address | CIDR;
Default
Context: http
This directive appeared in versions 1.3.0 and 1.2.1.

Defines trusted addresses. When a request comes from a trusted address, an

address from the X-Forwarded-For request header field will be used instead.

geoip proxy recursive

Syntax: geoip_proxy_recursive on | off;
Default off
Context: http
This directive appeared in versions 1.3.0 and 1.2.1.

If recursive search is disabled then instead of the original client address that
matches one of the trusted addresses, the last address sent in X-Forwarded-For
will be used. If recursive search is enabled then instead of the original client
address that matches one of the trusted addresses, the last non-trusted address
sent in X-Forwarded-For will be used.

Nginx, Inc. p.99 of 330

CHAPTER 2. HTTP SERVER MODULES 2.16. MODULE NGX HTTP GUNZIP MODULE

2.16 Module ngx http gunzip module

2.16.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 100
2.16.2 Example Configuration . . . . . . . . . . . . . . . . . . . 100
2.16.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 100
gunzip . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
gunzip buffers . . . . . . . . . . . . . . . . . . . . . . . . 100

2.16.1 Summary
The ngx_http_gunzip_module module is a filter that decompresses
responses with Content-Encoding: gzip for clients that do not support
gzip encoding method. The module will be useful when it is desirable to
store data compressed to save space and reduce I/O costs.
This module is not built by default, it should be enabled with the
--with-http_gunzip_module configuration parameter.

2.16.2 Example Configuration

location / storage / {
gunzip on ;
...
}

2.16.3 Directives
gunzip
Syntax: gunzip on | off;
Default off
Context: http, server, location

Enables or disables decompression of gzipped responses for clients that lack

gzip support. If enabled, the following directives are also taken into account
when determining if clients support gzip: gzip http version, gzip proxied, and
gzip disable. See also the gzip vary directive.

gunzip buffers
Syntax: gunzip_buffers number size;
Default 32 4k|16 8k
Context: http, server, location

Sets the number and size of buffers used to decompress a response. By

default, the buffer size is equal to one memory page. This is either 4K or 8K,
depending on a platform.

Nginx, Inc. p.100 of 330

CHAPTER 2. HTTP SERVER MODULES 2.17. MODULE NGX HTTP GZIP MODULE

2.17 Module ngx http gzip module

2.17.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 101
2.17.2 Example Configuration . . . . . . . . . . . . . . . . . . . 101
2.17.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 101
gzip . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
gzip buffers . . . . . . . . . . . . . . . . . . . . . . . . . 101
gzip comp level . . . . . . . . . . . . . . . . . . . . . . . 102
gzip disable . . . . . . . . . . . . . . . . . . . . . . . . . 102
gzip min length . . . . . . . . . . . . . . . . . . . . . . . 102
gzip http version . . . . . . . . . . . . . . . . . . . . . . 102
gzip proxied . . . . . . . . . . . . . . . . . . . . . . . . . 103
gzip types . . . . . . . . . . . . . . . . . . . . . . . . . . 103
gzip vary . . . . . . . . . . . . . . . . . . . . . . . . . . 104
2.17.4 Embedded Variables . . . . . . . . . . . . . . . . . . . . 104

2.17.1 Summary
The ngx_http_gzip_module module is a filter that compresses responses
using the gzip method. This often helps to reduce the size of transmitted
data by half or even more.

2.17.2 Example Configuration

gzip on ;
gz ip_m in_l engt h 1000;
gzip_proxied expired no - cache no - store private auth ;
gzip_types text / plain application / xml ;

The $gzip ratio variable can be used to log the achieved compression ratio.

2.17.3 Directives
gzip
Syntax: gzip on | off;
Default off
Context: http, server, location, if in location

Enables or disables gzipping of responses.

gzip buffers
Syntax: gzip_buffers number size;
Default 32 4k|16 8k
Context: http, server, location

Nginx, Inc. p.101 of 330

CHAPTER 2. HTTP SERVER MODULES 2.17. MODULE NGX HTTP GZIP MODULE

Sets the number and size of buffers used to compress a response. By default,
the buffer size is equal to one memory page. This is either 4K or 8K, depending
on a platform.

Until version 0.7.28, four 4K or 8K buffers were used by default.

gzip comp level

Syntax: gzip_comp_level level;
Default 1
Context: http, server, location

Sets a gzip compression level of a response. Acceptable values are in the

range from 1 to 9.

gzip disable
Syntax: gzip_disable regex . . . ;
Default
Context: http, server, location
This directive appeared in version 0.6.23.

Disables gzipping of responses for requests with User-Agent header fields

matching any of the specified regular expressions.
The special mask msie6 (0.7.12) corresponds to the regular expres-
sion MSIE [4-6]\., but works faster. Starting from version 0.8.11,
MSIE 6.0; ...SV1 is excluded from this mask.

gzip min length

Syntax: gzip_min_length length;
Default 20
Context: http, server, location

Sets the minimum length of a response that will be gzipped. The length is
determined only from the Content-Length response header field.

gzip http version

Syntax: gzip_http_version 1.0 | 1.1;
Default 1.1
Context: http, server, location

Sets the minimum HTTP version of a request required to compress a

response.

Nginx, Inc. p.102 of 330

CHAPTER 2. HTTP SERVER MODULES 2.17. MODULE NGX HTTP GZIP MODULE

gzip proxied
Syntax: gzip_proxied off | expired | no-cache | no-store | private |
no_last_modified | no_etag | auth | any . . . ;
Default off
Context: http, server, location

Enables or disables gzipping of responses for proxied requests depending on

the request and response. The fact that the request is proxied is determined
by the presence of the Via request header field. The directive accepts multiple
parameters:

off
disables compression for all proxied requests, ignoring other parameters;
expired
enables compression if a response header includes the Expires field with
a value that disables caching;
no-cache
enables compression if a response header includes the Cache-Control
field with the no-cache parameter;
no-store
enables compression if a response header includes the Cache-Control
field with the no-store parameter;
private
enables compression if a response header includes the Cache-Control
field with the private parameter;
no_last_modified
enables compression if a response header does not include the
Last-Modified field;
no_etag
enables compression if a response header does not include the ETag field;
auth
enables compression if a request header includes the Authorization field;
any
enables compression for all proxied requests.

gzip types
Syntax: gzip_types mime-type . . . ;
Default text/html
Context: http, server, location

Enables gzipping of responses for the specified MIME types in addition

to text/html. The special value * matches any MIME type (0.8.29).
Responses with the text/html type are always compressed.

Nginx, Inc. p.103 of 330

CHAPTER 2. HTTP SERVER MODULES 2.17. MODULE NGX HTTP GZIP MODULE

gzip vary
Syntax: gzip_vary on | off;
Default off
Context: http, server, location

Enables or disables inserting the Vary: Accept-Encoding response

header field if the directives gzip, gzip static, or gunzip are active.

2.17.4 Embedded Variables

$gzip ratio
achieved compression ratio, computed as the ratio between the original
and compressed response sizes.

Nginx, Inc. p.104 of 330

CHAPTER 2. HTTP SERVER MODULES 2.18. MODULE NGX HTTP GZIP STATIC MODULE

2.18 Module ngx http gzip static module

2.18.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 105
2.18.2 Example Configuration . . . . . . . . . . . . . . . . . . . 105
2.18.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 105
gzip static . . . . . . . . . . . . . . . . . . . . . . . . . . 105

2.18.1 Summary
The ngx_http_gzip_static_module module allows sending precom-
pressed files with the .gz filename extension instead of regular files.
This module is not built by default, it should be enabled with the
--with-http_gzip_static_module configuration parameter.

2.18.2 Example Configuration

gzip_static on ;
gzip_proxied expired no - cache no - store private auth ;

2.18.3 Directives
gzip static
Syntax: gzip_static on | off | always;
Default off
Context: http, server, location

Enables (on) or disables (off) checking the existence of precompressed

files. The following directives are also taken into account: gzip http version,
gzip proxied, gzip disable, and gzip vary.
With the always value (1.3.6), gzipped file is used in all cases, without
checking if the client supports it. It is useful if there are no uncompressed files
on the disk anyway or the ngx http gunzip module is used.
The files can be compressed using the gzip command, or any other
compatible one. It is recommended that the modification date and time of
original and compressed files be the same.

Nginx, Inc. p.105 of 330

CHAPTER 2. HTTP SERVER MODULES 2.19. MODULE NGX HTTP HEADERS MODULE

2.19 Module ngx http headers module

2.19.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 106
2.19.2 Example Configuration . . . . . . . . . . . . . . . . . . . 106
2.19.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 106
add header . . . . . . . . . . . . . . . . . . . . . . . . . 106
expires . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

2.19.1 Summary
The ngx_http_headers_module module allows adding the Expires and
Cache-Control header fields, and arbitrary fields, to a response header.

2.19.2 Example Configuration

expires 24 h ;
expires modified +24 h ;
expires @24h ;
expires 0;
expires -1;
expires epoch ;
expires $expires ;
add_header Cache - Control private ;

2.19.3 Directives
add header
Syntax: add_header name value [always];
Default
Context: http, server, location, if in location

Adds the specified field to a response header provided that the response
code equals 200, 201, 204, 206, 301, 302, 303, 304, or 307. A value can contain
variables.
There could be several add_header directives. These directives are
inherited from the previous level if and only if there are no add_header
directives defined on the current level.
If the always parameter is specified (1.7.5), the header field will be added
regardless of the response code.

expires
Syntax: expires [modified] time;
Syntax: expires epoch | max | off;
Default off
Context: http, server, location, if in location

Enables or disables adding or modifying the Expires and Cache-Control

response header fields provided that the response code equals 200, 201, 204,

Nginx, Inc. p.106 of 330

CHAPTER 2. HTTP SERVER MODULES 2.19. MODULE NGX HTTP HEADERS MODULE

206, 301, 302, 303, 304, or 307. A parameter can be a positive or negative
time.
A time in the Expires field is computed as a sum of the current time
and time specified in the directive. If the modified parameter is used (0.7.0,
0.6.32) then time is computed as a sum of the files modification time and time
specified in the directive.
In addition, it is possible to specify a time of the day using the @ prefix
(0.7.9, 0.6.34):

expires @15h30m ;

The epoch parameter corresponds to the absolute time

Thu, 01 Jan 1970 00:00:01 GMT. The contents of the Cache-Control
field depends on the sign of the specified time:

time is negative Cache-Control: no-cache.

time is positive or zero Cache-Control: max-age=t, where t is a

time specified in the directive, in seconds.

The max parameter sets Expires to the value

Thu, 31 Dec 2037 23:55:55 GMT, and Cache-Control to 10 years.
The off parameter disables adding or modifying the Expires and
Cache-Control response header fields.
The last parameter value can contain variables (1.7.9):

map $ s e n t _ h t t p _ c o n t e n t _ t y p e $expires {
default off ;
application / pdf 42 d ;
~ image / max ;
}

expires $expires ;

Nginx, Inc. p.107 of 330

CHAPTER 2. HTTP SERVER MODULES 2.20. MODULE NGX HTTP HLS MODULE

2.20 Module ngx http hls module

2.20.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 108
2.20.2 Example Configuration . . . . . . . . . . . . . . . . . . . 108
2.20.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 109
hls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
hls buffers . . . . . . . . . . . . . . . . . . . . . . . . . . 109
hls forward args . . . . . . . . . . . . . . . . . . . . . . . 109
hls fragment . . . . . . . . . . . . . . . . . . . . . . . . . 110
hls mp4 buffer size . . . . . . . . . . . . . . . . . . . . . 110
hls mp4 max buffer size . . . . . . . . . . . . . . . . . . 110

2.20.1 Summary
The ngx_http_hls_module module provides HTTP Live Streaming (HLS)
server-side support for H.264/AAC files. Such files typically have the .mp4,
.m4v, or .m4a filename extensions.
nginx supports two URIs for each MP4 file:

The playlist URI that ends with .m3u8 and accepts the optional len
argument that defines the fragment length in seconds;

The fragment URI that ends with .ts and accepts start and end
arguments that define fragment boundaries in seconds.

This module is available as part of our commercial subscription.

2.20.2 Example Configuration

location / video / {
hls ;
hls_fragment 5s;
hls_buffers 10 10 m ;
hls_mp4_buffer_size 1m;
hls_mp4_max_buffer_size 5m;
alias / var / video /;
}

With this configuration, the following URIs are supported for the /var
/video/test.mp4 file:

http :// hls . example . com / video / test . mp4 . m3u8 ? len =8.000
http :// hls . example . com / video / test . mp4 . ts ? start =1.000& end =2.200

Nginx, Inc. p.108 of 330

CHAPTER 2. HTTP SERVER MODULES 2.20. MODULE NGX HTTP HLS MODULE

2.20.3 Directives
hls
Syntax: hls;
Default
Context: location

Turns on HLS streaming in the surrounding location.

hls buffers
Syntax: hls_buffers number size;
Default 8 2m
Context: http, server, location

Sets the maximum number and size of buffers that are used for reading and
writing data frames.

hls forward args

Syntax: hls_forward_args on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.12.

Adds arguments from a playlist request to URIs of fragments. This may

be useful for performing client authorization at the moment of requesting a
fragment, or when protecting an HLS stream with the ngx http secure link -
module module.
For example, if a client requests a playlist
http://example.com/hls/test.mp4.m3u8?a=1&b=2, the arguments a=1
and b=2 will be added to URIs of fragments after the arguments start and
end:

# EXTM3U
# EXT -X - VERSION :3
# EXT -X - TARGETDURATION :15
# EXT -X - PLAYLIST - TYPE : VOD

# EXTINF :9.333 ,
test . mp4 . ts ? start =0.000& end =9.333& a =1& b =2
# EXTINF :7.167 ,
test . mp4 . ts ? start =9.333& end =16.500& a =1& b =2
# EXTINF :5.416 ,
test . mp4 . ts ? start =16.500& end =21.916& a =1& b =2
# EXTINF :5.500 ,
test . mp4 . ts ? start =21.916& end =27.416& a =1& b =2
# EXTINF :15.167 ,
test . mp4 . ts ? start =27.416& end =42.583& a =1& b =2
# EXTINF :9.626 ,
test . mp4 . ts ? start =42.583& end =52.209& a =1& b =2

# EXT -X - ENDLIST

If an HLS stream is protected with the ngx http secure link module
module, $uri should not be used in the secure link md5 expression because

Nginx, Inc. p.109 of 330

CHAPTER 2. HTTP SERVER MODULES 2.20. MODULE NGX HTTP HLS MODULE

this will cause errors when requesting the fragments. Base URI should be used
instead of $uri ($hls uri in the example):

http {
...

map $uri $hls_uri {

~^(? < base_uri >.*) . m3u8$ $base_uri ;
~^(? < base_uri >.*) . ts$ $base_uri ;
default $uri ;
}

server {
...

location / hls {
hls ;
h ls _ fo r wa r d_ a rg s on ;

alias / var / videos ;

secure_link $arg_md5 , $arg_expires ;

se cure _lin k_md 5 " $ s e c u r e _ l i n k _ e x p i r e s $ h l s _ u r i $ r e m o t e _ a d d r
secret ";

if ( $secure_link = "") {
return 403;
}

if ( $secure_link = "0") {
return 410;
}
}
}
}

hls fragment
Syntax: hls_fragment time;
Default 5s
Context: http, server, location

Defines the default fragment length for playlist URIs requested without the
len argument.

hls mp4 buffer size

Syntax: hls_mp4_buffer_size size;
Default 512k
Context: http, server, location

Sets the initial size of the buffer used for processing MP4 files.

hls mp4 max buffer size

Syntax: hls_mp4_max_buffer_size size;
Default 10m
Context: http, server, location

Nginx, Inc. p.110 of 330

CHAPTER 2. HTTP SERVER MODULES 2.20. MODULE NGX HTTP HLS MODULE

During metadata processing, a larger buffer may become necessary. Its size
cannot exceed the specified size, or else nginx will return the server error 500
Internal Server Error, and log the following message:

"/ some / movie / file . mp4 " mp4 moov atom is too large :
12583268 , you may want to increase h l s _ m p 4 _ m a x _ b u f f e r _ s i z e

Nginx, Inc. p.111 of 330

CHAPTER 2. HTTP SERVER MODULES 2.21. MODULE NGX HTTP IMAGE FILTER MODULE

2.21 Module ngx http image filter module

2.21.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 112
2.21.2 Example Configuration . . . . . . . . . . . . . . . . . . . 112
2.21.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 112
image filter . . . . . . . . . . . . . . . . . . . . . . . . . 112
image filter buffer . . . . . . . . . . . . . . . . . . . . . . 113
image filter interlace . . . . . . . . . . . . . . . . . . . . 113
image filter jpeg quality . . . . . . . . . . . . . . . . . . 114
image filter sharpen . . . . . . . . . . . . . . . . . . . . 114
image filter transparency . . . . . . . . . . . . . . . . . . 114

2.21.1 Summary
The ngx_http_image_filter_module module (0.7.54+) is a filter that
transforms images in JPEG, GIF, and PNG formats.
This module is not built by default, it should be enabled with the
--with-http_image_filter_module configuration parameter.

This module utilizes the libgd library. It is recommended to use the latest
available version of the library.

2.21.2 Example Configuration

location / img / {
proxy_pass http :// backend ;
image_filter resize 150 100;
image_filter rotate 90;
error_page 415 = / empty ;
}

location = / empty {
empty_gif ;
}

2.21.3 Directives
image filter
Syntax: image_filter off;
Syntax: image_filter test;
Syntax: image_filter size;
Syntax: image_filter rotate 90 | 180 | 270;
Syntax: image_filter resize width height;
Syntax: image_filter crop width height;
Default off
Context: location

Sets the type of transformation to perform on images:

Nginx, Inc. p.112 of 330

CHAPTER 2. HTTP SERVER MODULES 2.21. MODULE NGX HTTP IMAGE FILTER MODULE

off
turns off module processing in a surrounding location.
test
ensures that responses are images in either JPEG, GIF, or PNG format.
Otherwise, the 415 Unsupported Media Type error is returned.
size
outputs information about images in a JSON format, e.g.:

{ " img " : { " width ": 100 , " height ": 100 , " type ": " gif " } }

In case of an error, the output is as follows:

{}

rotate 90|180|270
rotates images counter-clockwise by the specified number of degrees.
Parameter value can contain variables. This mode can be used either
alone or along with the resize and crop transformations.
resize width height
proportionally reduces an image to the specified sizes. To reduce by
only one dimension, another dimension can be specified as -. In case
of an error, the server will return code 415 Unsupported Media Type.
Parameter values can contain variables. When used along with the
rotate parameter, the rotation happens after reduction.
crop width height
proportionally reduces an image to the larger side size and crops
extraneous edges by another side. To reduce by only one dimension,
another dimension can be specified as -. In case of an error, the server
will return code 415 Unsupported Media Type. Parameter values can
contain variables. When used along with the rotate parameter, the
rotation happens before reduction.

image filter buffer

Syntax: image_filter_buffer size;
Default 1M
Context: http, server, location

Sets the maximum size of the buffer used for reading images. When the
size is exceeded the server returns error 415 Unsupported Media Type.

image filter interlace

Syntax: image_filter_interlace on | off;
Default off
Context: http, server, location
This directive appeared in version 1.3.15.

Nginx, Inc. p.113 of 330

CHAPTER 2. HTTP SERVER MODULES 2.21. MODULE NGX HTTP IMAGE FILTER MODULE

If enabled, final images will be interlaced. For JPEG, final images will be
in progressive JPEG format.

image filter jpeg quality

Syntax: image_filter_jpeg_quality quality;
Default 75
Context: http, server, location

Sets the desired quality of the transformed JPEG images. Acceptable values
are in the range from 1 to 100. Lesser values usually imply both lower image
quality and less data to transfer. The maximum recommended value is 95.
Parameter value can contain variables.

image filter sharpen

Syntax: image_filter_sharpen percent;
Default 0
Context: http, server, location

Increases sharpness of the final image. The sharpness percentage can

exceed 100. The zero value disables sharpening. Parameter value can contain
variables.

image filter transparency

Syntax: image_filter_transparency on|off;
Default on
Context: http, server, location

Defines whether transparency should be preserved when transforming

GIF images or PNG images with colors specified by a palette. The loss
of transparency results in images of a better quality. The alpha channel
transparency in PNG is always preserved.

Nginx, Inc. p.114 of 330

CHAPTER 2. HTTP SERVER MODULES 2.22. MODULE NGX HTTP INDEX MODULE

2.22 Module ngx http index module

2.22.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 115
2.22.2 Example Configuration . . . . . . . . . . . . . . . . . . . 115
2.22.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 115
index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

2.22.1 Summary
The ngx_http_index_module module processes requests ending with the
slash character (/). Such requests can also be processed by the ngx http -
autoindex module and ngx http random index module modules.

2.22.2 Example Configuration

location / {
index index . $geo . html index . html ;
}

2.22.3 Directives
index
Syntax: index file . . . ;
Default index.html
Context: http, server, location

Defines files that will be used as an index. The file name can contain
variables. Files are checked in the specified order. The last element of the list
can be a file with an absolute path. Example:

index index . $geo . html index .0. html / index . html ;

It should be noted that using an index file causes an internal redirect, and
the request can be processed in a different location. For example, with the
following configuration:

location = / {
index index . html ;
}

location / {
...
}

a / request will actually be processed in the second location as

/index.html.

Nginx, Inc. p.115 of 330

CHAPTER 2. HTTP SERVER MODULES 2.23. MODULE NGX HTTP LIMIT CONN MODULE

2.23 Module ngx http limit conn module

2.23.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 116
2.23.2 Example Configuration . . . . . . . . . . . . . . . . . . . 116
2.23.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 116
limit conn . . . . . . . . . . . . . . . . . . . . . . . . . . 116
limit conn log level . . . . . . . . . . . . . . . . . . . . . 117
limit conn status . . . . . . . . . . . . . . . . . . . . . . 117
limit conn zone . . . . . . . . . . . . . . . . . . . . . . . 117
limit zone . . . . . . . . . . . . . . . . . . . . . . . . . . 118

2.23.1 Summary
The ngx_http_limit_conn_module module is used to limit the number of
connections per the defined key, in particular, the number of connections from
a single IP address.
Not all connections are counted. A connection is counted only if it has a
request processed by the server and the whole request header has already been
read.

2.23.2 Example Configuration

http {
li mit_ conn _zon e $ b i n a r y _ r e m o t e _ a d d r zone = addr :10 m ;

...

server {

...

location / download / {
limit_conn addr 1;
}

2.23.3 Directives
limit conn
Syntax: limit_conn zone number;
Default
Context: http, server, location

Sets the shared memory zone and the maximum allowed number of
connections for a given key value. When this limit is exceeded, the server
will return the 503 Service Temporarily Unavailable error in reply to a
request. For example, the directives

li mit_ conn _zon e $ b i n a r y _ r e m o t e _ a d d r zone = addr :10 m ;

server {

Nginx, Inc. p.116 of 330

CHAPTER 2. HTTP SERVER MODULES 2.23. MODULE NGX HTTP LIMIT CONN MODULE

location / download / {
limit_conn addr 1;
}

allow only one connection per an IP address at a time.

In SPDY, each concurrent request is considered a separate connection.

When several limit_conn directives are specified, any configured limit

will apply. For example, the following configuration will limit the number
of connections to the server per a client IP and, at the same time, the total
number of connections to the virtual host:

li mit_ conn _zon e $ b i n a r y _ r e m o t e _ a d d r zone = perip :10 m ;

li mit_ conn _zon e $server_name zone = perserver :10 m ;

server {
...
limit_conn perip 10;
limit_conn perserver 100;
}

These directives are inherited from the previous level if and only if there
are no limit_conn directives on the current level.

limit conn log level

Syntax: limit_conn_log_level info | notice | warn | error;
Default error
Context: http, server, location
This directive appeared in version 0.8.18.

Sets the desired logging level for cases when the server limits the number
of connections.

limit conn status

Syntax: limit_conn_status code;
Default 503
Context: http, server, location
This directive appeared in version 1.3.15.

Sets the status code to return in response to rejected requests.

limit conn zone

Syntax: limit_conn_zone key zone=name:size;
Default
Context: http

Sets parameters for a shared memory zone that will keep states for various
keys. In particular, the state includes the current number of connections. The
key can contain text, variables, and their combination. Requests with an empty
key value are not accounted.

Nginx, Inc. p.117 of 330

CHAPTER 2. HTTP SERVER MODULES 2.23. MODULE NGX HTTP LIMIT CONN MODULE

Prior to version 1.7.6, a key could contain exactly one variable.

Usage example:

li mit_ conn _zon e $ b i n a r y _ r e m o t e _ a d d r zone = addr :10 m ;

Here, a client IP address serves as a key. Note that instead of $remote addr,
the $binary remote addr variable is used here. The $remote addr variables
size can vary from 7 to 15 bytes. The stored state occupies either 32 or 64
bytes of memory on 32-bit platforms and always 64 bytes on 64-bit platforms.
The $binary remote addr variables size is always 4 bytes. The stored state
always occupies 32 bytes on 32-bit platforms and 64 bytes on 64-bit platforms.
One megabyte zone can keep about 32 thousand 32-byte states or about 16
thousand 64-byte states. If the zone storage is exhausted, the server will return
the 503 Service Temporarily Unavailable error to all further requests.

limit zone
Syntax: limit_zone name $variable size;
Default
Context: http

This directive was made obsolete in version 1.1.8 and was removed in
version 1.7.6. An equivalent limit conn zone directive with a changed syntax
should be used instead:

limit_conn_zone $variable zone=name:size;

Nginx, Inc. p.118 of 330

CHAPTER 2. HTTP SERVER MODULES 2.24. MODULE NGX HTTP LIMIT REQ MODULE

2.24 Module ngx http limit req module

2.24.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 119
2.24.2 Example Configuration . . . . . . . . . . . . . . . . . . . 119
2.24.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 119
limit req . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
limit req log level . . . . . . . . . . . . . . . . . . . . . . 120
limit req status . . . . . . . . . . . . . . . . . . . . . . . 120
limit req zone . . . . . . . . . . . . . . . . . . . . . . . . 120

2.24.1 Summary
The ngx_http_limit_req_module module (0.7.21) is used to limit the
request processing rate per a defined key, in particular, the processing rate
of requests coming from a single IP address. The limitation is done using the
leaky bucket method.

2.24.2 Example Configuration

http {
limit_req_zone $ b i n a r y _ r e m o t e _ a d d r zone = one :10 m rate =1 r / s ;

...

server {

...

location / search / {
limit_req zone = one burst =5;
}

2.24.3 Directives
limit req
Syntax: limit_req zone=name [burst=number] [nodelay];
Default
Context: http, server, location

Sets the shared memory zone and the maximum burst size of requests. If the
requests rate exceeds the rate configured for a zone, their processing is delayed
such that requests are processed at a defined rate. Excessive requests are
delayed until their number exceeds the maximum burst size in which case the
request is terminated with an error 503 Service Temporarily Unavailable.
By default, the maximum burst size is equal to zero. For example, the
directives

limit_req_zone $ b i n a r y _ r e m o t e _ a d d r zone = one :10 m rate =1 r / s ;

server {

Nginx, Inc. p.119 of 330

CHAPTER 2. HTTP SERVER MODULES 2.24. MODULE NGX HTTP LIMIT REQ MODULE

location / search / {
limit_req zone = one burst =5;
}

allow not more than 1 request per second at an average, with bursts not
exceeding 5 requests.
If delaying of excessive requests while requests are being limited is not
desired, the parameter nodelay should be used:

limit_req zone = one burst =5 nodelay ;

limit req log level

Syntax: limit_req_log_level info | notice | warn | error;
Default error
Context: http, server, location
This directive appeared in version 0.8.18.

Sets the desired logging level for cases when the server refuses to process
requests due to rate exceeding, or delays request processing. Logging
level for delays is one point less than for refusals; for example, if
limit_req_log_level notice is specified, delays are logged with the info
level.

limit req status

Syntax: limit_req_status code;
Default 503
Context: http, server, location
This directive appeared in version 1.3.15.

Sets the status code to return in response to rejected requests.

limit req zone

Syntax: limit_req_zone key zone=name:size rate=rate;
Default
Context: http

Sets parameters for a shared memory zone that will keep states for various
keys. In particular, the state stores the current number of excessive requests.
The key can contain text, variables, and their combination. Requests with an
empty key value are not accounted.

Prior to version 1.7.6, a key could contain exactly one variable.

Usage example:

limit_req_zone $ b i n a r y _ r e m o t e _ a d d r zone = one :10 m rate =1 r / s ;

Nginx, Inc. p.120 of 330

CHAPTER 2. HTTP SERVER MODULES 2.24. MODULE NGX HTTP LIMIT REQ MODULE

Here, the states are kept in a 10 megabyte zone one, and an average
request processing rate for this zone cannot exceed 1 request per second.
A client IP address serves as a key. Note that instead of $remote addr, the
$binary remote addr variable is used here, that allows decreasing the state size
down to 64 bytes. One megabyte zone can keep about 16 thousand 64-byte
states. If the zone storage is exhausted, the server will return the 503 Service
Temporarily Unavailable error to all further requests.
The rate is specified in requests per second (r/s). If a rate of less than one
request per second is desired, it is specified in request per minute (r/m). For
example, half-request per second is 30r/m.

Nginx, Inc. p.121 of 330

CHAPTER 2. HTTP SERVER MODULES 2.25. MODULE NGX HTTP LOG MODULE

2.25 Module ngx http log module

2.25.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 122
2.25.2 Example Configuration . . . . . . . . . . . . . . . . . . . 122
2.25.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 122
access log . . . . . . . . . . . . . . . . . . . . . . . . . . 122
log format . . . . . . . . . . . . . . . . . . . . . . . . . . 124
open log file cache . . . . . . . . . . . . . . . . . . . . . 125

2.25.1 Summary
The ngx_http_log_module module writes request logs in the specified
format.
Requests are logged in the context of a location where processing ends.
It may be different from the original location, if an internal redirect happens
during request processing.

2.25.2 Example Configuration

log_format compression $remote_addr - $remote_user [ $time_local ]

" $request " $status $bytes_sent
" $http_referer " " $ h tt p _u s er _ ag e nt " " $gzip_ratio " ;

access_log / spool / logs / nginx - access . log compression buffer =32 k ;

2.25.3 Directives
access log
Syntax: access_log path [format [buffer=size [flush=time]] [if=condition]];
Syntax: access_log path format gzip[=level] [buffer=size] [flush=time]
[if=condition];
Syntax: access_log syslog:server=address[,parameter=value] [format
[if=condition]];
Syntax: access_log off;
Default logs/access.log combined
Context: http, server, location, if in location, limit except

Sets the path, format, and configuration for a buffered log write. Several
logs can be specified on the same level. Logging to syslog can be configured
by specifying the syslog: prefix in the first parameter. The special value
off cancels all access_log directives on the current level. If the format is not
specified then the predefined combined format is used.
If either the buffer or gzip (1.3.10, 1.2.7) parameter is used, writes to log
will be buffered.

Nginx, Inc. p.122 of 330

CHAPTER 2. HTTP SERVER MODULES 2.25. MODULE NGX HTTP LOG MODULE

The buffer size must not exceed the size of an atomic write to a disk file.
For FreeBSD this size is unlimited.

When buffering is enabled, the data will be written to the file:

if the next log line does not fit into the buffer;

if the buffered data is older than specified by the flush parameter (1.3.10,
1.2.7);

when a worker process is re-opening log files or is shutting down.

If the gzip parameter is used, then the buffered data will be compressed
before writing to the file. The compression level can be set between 1 (fastest,
less compression) and 9 (slowest, best compression). By default, the buffer
size is equal to 64K bytes, and the compression level is set to 1. Since the data
is compressed in atomic blocks, the log file can be decompressed or read by
zcat at any time.
Example:

access_log / path / to / log . gz combined gzip flush =5 m ;

For gzip compression to work, nginx must be built with the zlib library.

The file path can contain variables (0.7.6+), but such logs have some
constraints:

the user whose credentials are used by worker processes should have
permissions to create files in a directory with such logs;

buffered writes do not work;

the file is opened and closed for each log write. However, since the
descriptors of frequently used files can be stored in a cache, writing to
the old file can continue during the time specified by the open log file -
cache directives valid parameter

during each log write the existence of the requests root directory is
checked, and if it does not exist the log is not created. It is thus a good
idea to specify both root and access_log on the same level:

server {
root / spool / vhost / data / $host ;
access_log / spool / vhost / logs / $host ;
...

The if parameter (1.7.0) enables conditional logging. A request will not

be logged if the condition evaluates to 0 or an empty string. In the following
example, the requests with response codes 2xx and 3xx will not be logged:

Nginx, Inc. p.123 of 330

CHAPTER 2. HTTP SERVER MODULES 2.25. MODULE NGX HTTP LOG MODULE

map $status $loggable {

~^[23] 0;
default 1;
}

access_log / path / to / access . log combined if = $loggable ;

log format
Syntax: log_format name string . . . ;
Default combined "..."
Context: http

Specifies log format.

The log format can contain common variables, and variables that exist only
at the time of a log write:
$bytes sent
the number of bytes sent to a client
$connection
connection serial number
$connection requests
the current number of requests made through a connection (1.1.18)
$msec
time in seconds with a milliseconds resolution at the time of the log write
$pipe
p if request was pipelined, . otherwise
$request length
request length (including request line, header, and request body)
$request time
request processing time in seconds with a milliseconds resolution; time
elapsed between the first bytes were read from the client and the log
write after the last bytes were sent to the client
$status
response status
$time iso8601
local time in the ISO 8601 standard format
$time local
local time in the Common Log Format

In the modern nginx versions variables $status (1.3.2, 1.2.2), $bytes -

sent (1.3.8, 1.2.5), $connection (1.3.8, 1.2.5), $connection requests (1.3.8,
1.2.5), $msec (1.3.9, 1.2.6), $request time (1.3.9, 1.2.6), $pipe (1.3.12, 1.2.7),
$request length (1.3.12, 1.2.7), $time iso8601 (1.3.12, 1.2.7), and $time local
(1.3.12, 1.2.7) are also available as common variables.

Header lines sent to a client have the prefix sent_http_, for example,
$sent http content range.

Nginx, Inc. p.124 of 330

CHAPTER 2. HTTP SERVER MODULES 2.25. MODULE NGX HTTP LOG MODULE

The configuration always includes the predefined combined format:

log_format combined $remote_addr - $remote_user [ $time_local ]

" $request " $status $ b od y _b y te s _s e nt
" $http_referer " " $ h tt p _u s er _ ag e nt " ;

open log file cache

Syntax: open_log_file_cache max=N [inactive=time] [min_uses=N]
[valid=time];
Syntax: open_log_file_cache off;
Default off
Context: http, server, location

Defines a cache that stores the file descriptors of frequently used logs whose
names contain variables. The directive has the following parameters:

max
sets the maximum number of descriptors in a cache; if the cache becomes
full the least recently used (LRU) descriptors are closed
inactive
sets the time after which the cached descriptor is closed if there were no
access during this time; by default, 10 seconds
min_uses
sets the minimum number of file uses during the time defined by the
inactive parameter to let the descriptor stay open in a cache; by default,
1
valid
sets the time after which it should be checked that the file still exists
with the same name; by default, 60 seconds
off
disables caching

Usage example:

o p e n _ l o g _ f i l e _ c a c h e max =1000 inactive =20 s valid =1 m min_uses =2;

Nginx, Inc. p.125 of 330

CHAPTER 2. HTTP SERVER MODULES 2.26. MODULE NGX HTTP MAP MODULE

2.26 Module ngx http map module

2.26.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 126
2.26.2 Example Configuration . . . . . . . . . . . . . . . . . . . 126
2.26.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 126
map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
map hash bucket size . . . . . . . . . . . . . . . . . . . . 128
map hash max size . . . . . . . . . . . . . . . . . . . . . 128

2.26.1 Summary
The ngx_http_map_module module creates variables whose values depend
on values of other variables.

2.26.2 Example Configuration

map $http_host $name {

hostnames ;

default 0;

example . com 1;
*. example . com 1;
example . org 2;
*. example . org 2;
. example . net 3;
wap .* 4;
}

map $ h tt p _u se r _a g en t $mobile {
default 0;
"~ Opera Mini " 1;
}

2.26.3 Directives
map
Syntax: map string $variable { . . . }
Default
Context: http

Creates a new variable whose value depends on values of one or more of

the source variables specified in the first parameter.

Before version 0.9.0 only a single variable could be specified in the first
parameter.

Since variables are evaluated only when they are used, the mere
declaration even of a large number of map variables does not add any extra
costs to request processing.

Nginx, Inc. p.126 of 330

CHAPTER 2. HTTP SERVER MODULES 2.26. MODULE NGX HTTP MAP MODULE

Parameters inside the map block specify a mapping between source and
resulting values.
Source values are specified as strings or regular expressions (0.9.6).
A regular expression should either start from the ~ symbol for a case-
sensitive matching, or from the ~* symbols (1.0.4) for case-insensitive
matching. A regular expression can contain named and positional captures
that can later be used in other directives along with the resulting variable.
If a source value matches one of the names of special parameters described
below, it should be prefixed with the \ symbol.
The resulting value can be a string or another variable (0.9.0).
The directive also supports three special parameters:

default value
sets the resulting value if the source value matches none of the specified
variants. When default is not specified, the default resulting value will
be an empty string.
hostnames
indicates that source values can be hostnames with a prefix or suffix
mask:

*. example . com 1;
example .* 1;

The following two records

example . com 1;
*. example . com 1;

can be combined:

. example . com 1;

This parameter should be specified before the list of values.

include file
includes a file with values. There can be several inclusions.

If the source value matches more than one of the specified variants, e.g.
both a mask and a regular expression match, the first matching variant will be
chosen, in the following order of priority:

1. string value without a mask

2. longest string value with a prefix mask, e.g. *.example.com

3. longest string value with a suffix mask, e.g. mail.*

4. first matching regular expression (in order of appearance in a

configuration file)

5. default value

Nginx, Inc. p.127 of 330

CHAPTER 2. HTTP SERVER MODULES 2.26. MODULE NGX HTTP MAP MODULE

map hash bucket size

Syntax: map_hash_bucket_size size;
Default 32|64|128
Context: http

Sets the bucket size for the map variables hash tables. Default value
depends on the processors cache line size. The details of setting up hash
tables are provided in a separate document.

map hash max size

Syntax: map_hash_max_size size;
Default 2048
Context: http

Sets the maximum size of the map variables hash tables. The details of
setting up hash tables are provided in a separate document.

Nginx, Inc. p.128 of 330

CHAPTER 2. HTTP SERVER MODULES 2.27. MODULE NGX HTTP MEMCACHED MODULE

2.27 Module ngx http memcached module

2.27.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 129
2.27.2 Example Configuration . . . . . . . . . . . . . . . . . . . 129
2.27.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 129
memcached bind . . . . . . . . . . . . . . . . . . . . . . 129
memcached buffer size . . . . . . . . . . . . . . . . . . . 130
memcached connect timeout . . . . . . . . . . . . . . . . 130
memcached force ranges . . . . . . . . . . . . . . . . . . 130
memcached gzip flag . . . . . . . . . . . . . . . . . . . . 130
memcached next upstream . . . . . . . . . . . . . . . . . 130
memcached next upstream timeout . . . . . . . . . . . . 131
memcached next upstream tries . . . . . . . . . . . . . . 131
memcached pass . . . . . . . . . . . . . . . . . . . . . . 131
memcached read timeout . . . . . . . . . . . . . . . . . . 132
memcached send timeout . . . . . . . . . . . . . . . . . 132
2.27.4 Embedded Variables . . . . . . . . . . . . . . . . . . . . 132

2.27.1 Summary
The ngx_http_memcached_module module is used to obtain responses from
a memcached server. The key is set in the $memcached key variable. A
response should be put in memcached in advance by means external to nginx.

2.27.2 Example Configuration

server {
location / {
set $memcached_key " $uri ? $args ";
memcached_pass host :11211;
error_page 404 502 504 = @fallback ;
}

location @fallback {
proxy_pass http :// backend ;
}
}

2.27.3 Directives
memcached bind
Syntax: memcached_bind address | off;
Default
Context: http, server, location
This directive appeared in version 0.8.22.

Makes outgoing connections to a memcached server originate from the

specified local IP address. Parameter value can contain variables (1.3.12). The
special value off (1.3.12) cancels the effect of the memcached_bind directive

Nginx, Inc. p.129 of 330

CHAPTER 2. HTTP SERVER MODULES 2.27. MODULE NGX HTTP MEMCACHED MODULE

inherited from the previous configuration level, which allows the system to
auto-assign the local IP address.

memcached buffer size

Syntax: memcached_buffer_size size;
Default 4k|8k
Context: http, server, location

Sets the size of the buffer used for reading the response received from the
memcached server. The response is passed to the client synchronously, as soon
as it is received.

memcached connect timeout

Syntax: memcached_connect_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for establishing a connection with a memcached server.

It should be noted that this timeout cannot usually exceed 75 seconds.

memcached force ranges

Syntax: memcached_force_ranges on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.7.

Enables byte-range support for both cached and uncached responses from
the memcached server regardless of the Accept-Ranges field in these responses.

memcached gzip flag

Syntax: memcached_gzip_flag flag;
Default
Context: http, server, location
This directive appeared in version 1.3.6.

Enables the test for the flag presence in the memcached server response
and sets the Content-Encoding response header field to gzip if the flag is
set.

memcached next upstream

Syntax: memcached_next_upstream error | timeout | invalid_response |
not_found | off . . . ;
Default error timeout
Context: http, server, location

Specifies in which cases a request should be passed to the next server:

Nginx, Inc. p.130 of 330

CHAPTER 2. HTTP SERVER MODULES 2.27. MODULE NGX HTTP MEMCACHED MODULE

error
an error occurred while establishing a connection with the server, passing
a request to it, or reading the response header;
timeout
a timeout has occurred while establishing a connection with the server,
passing a request to it, or reading the response header;
invalid_response
a server returned an empty or invalid response;
not_found
a response was not found on the server;
off
disables passing a request to the next server.

One should bear in mind that passing a request to the next server is only
possible if nothing has been sent to a client yet. That is, if an error or timeout
occurs in the middle of the transferring of a response, fixing this is impossible.
The directive also defines what is considered an unsuccessful attempt
of communication with a server. The cases of error, timeout and
invalid_header are always considered unsuccessful attempts, even if they
are not specified in the directive. The case of not_found is never considered
an unsuccessful attempt.
Passing a request to the next server can be limited by the number of tries
and by time.

memcached next upstream timeout

Syntax: memcached_next_upstream_timeout time;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the time allowed to pass a request to the next server. The 0 value
turns off this limitation.

memcached next upstream tries

Syntax: memcached_next_upstream_tries number;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the number of possible tries for passing a request to the next server.
The 0 value turns off this limitation.

memcached pass
Syntax: memcached_pass address;
Default
Context: location, if in location

Nginx, Inc. p.131 of 330

CHAPTER 2. HTTP SERVER MODULES 2.27. MODULE NGX HTTP MEMCACHED MODULE

Sets the memcached server address. The address can be specified as a

domain name or an address, and a port:

memcached_pass localhost :11211;

or as a UNIX-domain socket path:

memcached_pass unix :/ tmp / memcached . socket ;

If a domain name resolves to several addresses, all of them will be used

in a round-robin fashion. In addition, an address can be specified as a server
group.

memcached read timeout

Syntax: memcached_read_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for reading a response from the memcached server.

The timeout is set only between two successive read operations, not for the
transmission of the whole response. If the memcached server does not transmit
anything within this time, the connection is closed.

memcached send timeout

Syntax: memcached_send_timeout time;
Default 60s
Context: http, server, location

Sets a timeout for transmitting a request to the memcached server. The

timeout is set only between two successive write operations, not for the
transmission of the whole request. If the memcached server does not receive
anything within this time, the connection is closed.

2.27.4 Embedded Variables

$memcached key
Defines a key for obtaining response from a memcached server.

Nginx, Inc. p.132 of 330

CHAPTER 2. HTTP SERVER MODULES 2.28. MODULE NGX HTTP MP4 MODULE

2.28 Module ngx http mp4 module

2.28.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2.28.2 Example Configuration . . . . . . . . . . . . . . . . . . . 134
2.28.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 134
mp4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
mp4 buffer size . . . . . . . . . . . . . . . . . . . . . . . 134
mp4 max buffer size . . . . . . . . . . . . . . . . . . . . 134
mp4 limit rate . . . . . . . . . . . . . . . . . . . . . . . 135
mp4 limit rate after . . . . . . . . . . . . . . . . . . . . 135

2.28.1 Summary
The ngx_http_mp4_module module provides pseudo-streaming server-side
support for MP4 files. Such files typically have the .mp4, .m4v, or .m4a
filename extensions.
Pseudo-streaming works in alliance with a compatible Flash player. The
player sends an HTTP request to the server with the start time specified in
the query string argument (named simply start and specified in seconds), and
the server responds with the stream such that its start position corresponds to
the requested time, for example:

http :// example . com / ele phan ts_d ream . mp4 ? start =238.88

This allows performing a random seeking at any time, or starting playback

in the middle of the timeline.
To support seeking, H.264-based formats store metadata in a so-called
moov atom. It is a part of the file that holds the index information for
the whole file.
To start playback, the player first needs to read metadata. This is done
by sending a special request with the start=0 argument. A lot of encoding
software insert the metadata at the end of the file. This is suboptimal for
pseudo-streaming, because the player has to download the entire file before
starting playback. If the metadata are located at the beginning of the file,
it is enough for nginx to simply start sending back the file contents. If the
metadata are located at the end of the file, nginx must read the entire file and
prepare a new stream so that the metadata come before the media data. This
involves some CPU, memory, and disk I/O overhead, so it is a good idea to
prepare an original file for pseudo-streaming in advance, rather than having
nginx do this on every such request.
The module also supports the end argument of an HTTP request (1.5.13)
which sets the end point of playback. The end argument can be specified with
the start argument or separately:

http :// example . com / ele phan ts_d ream . mp4 ? start =238.88& end =555.55

Nginx, Inc. p.133 of 330

CHAPTER 2. HTTP SERVER MODULES 2.28. MODULE NGX HTTP MP4 MODULE

For a matching request with a non-zero start or end argument, nginx will
read the metadata from the file, prepare the stream with the requested time
range, and send it to the client. This has the same overhead as described
above.
If a matching request does not include the start and end arguments, there
is no overhead, and the file is sent simply as a static resource. Some players
also support byte-range requests, and thus do not require this module.
This module is not built by default, it should be enabled with the
--with-http_mp4_module configuration parameter.

If a third-party mp4 module was previously used, it should be disabled.

A similar pseudo-streaming support for FLV files is provided by the ngx -

http flv module module.

2.28.2 Example Configuration

location / video / {
mp4 ;
mp 4_bu ffer _siz e 1m;
mp4_max_buffer_size 5m;
mp4_limit_rate on ;
mp4_limit_rate_after 30 s ;
}

2.28.3 Directives
mp4
Syntax: mp4;
Default
Context: location

Turns on module processing in a surrounding location.

mp4 buffer size

Syntax: mp4_buffer_size size;
Default 512K
Context: http, server, location

Sets the initial size of the buffer used for processing MP4 files.

mp4 max buffer size

Syntax: mp4_max_buffer_size size;
Default 10M
Context: http, server, location

Nginx, Inc. p.134 of 330

CHAPTER 2. HTTP SERVER MODULES 2.28. MODULE NGX HTTP MP4 MODULE

During metadata processing, a larger buffer may become necessary. Its size
cannot exceed the specified size, or else nginx will return the 500 Internal
Server Error server error, and log the following message:

"/ some / movie / file . mp4 " mp4 moov atom is too large :
12583268 , you may want to increase m p 4 _ m a x _ b u f f e r _ s i z e

mp4 limit rate

Syntax: mp4_limit_rate on | off | factor;
Default off
Context: http, server, location

Limits the rate of response transmission to a client. The rate is limited

based on the average bitrate of the MP4 file served. To calculate the rate, the
bitrate is multiplied by the specified factor. The special value on corresponds
to the factor of 1.1. The special value off disables rate limiting. The limit
is set per a request, and so if a client simultaneously opens two connections,
the overall rate will be twice as much as the specified limit.

This directive is available as part of our commercial subscription.

mp4 limit rate after

Syntax: mp4_limit_rate_after time;
Default 60s
Context: http, server, location

Sets the initial amount of media data (measured in playback time) after
which the further transmission of the response to a client will be rate limited.

This directive is available as part of our commercial subscription.

Nginx, Inc. p.135 of 330

CHAPTER 2. HTTP SERVER MODULES 2.29. MODULE NGX HTTP PERL MODULE

2.29 Module ngx http perl module

2.29.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 136
2.29.2 Known Bugs . . . . . . . . . . . . . . . . . . . . . . . . . 136
2.29.3 Example Configuration . . . . . . . . . . . . . . . . . . . 137
2.29.4 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 137
perl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
perl modules . . . . . . . . . . . . . . . . . . . . . . . . 138
perl require . . . . . . . . . . . . . . . . . . . . . . . . . 138
perl set . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
2.29.5 Calling Perl from SSI . . . . . . . . . . . . . . . . . . . . 138
2.29.6 The $r Request Object Methods . . . . . . . . . . . . . . 138

2.29.1 Summary
The ngx_http_perl_module module is used to implement location and
variable handlers in Perl and insert Perl calls into SSI.
This module is not built by default, it should be enabled with the
--with-http_perl_module configuration parameter.

This module requires Perl version 5.6.1 or higher. The C compiler should
be compatible with the one used to build Perl.

2.29.2 Known Bugs

The module is experimental, caveat emptor applies.
In order for Perl to recompile the modified modules during reconfiguration,
it should be built with the -Dusemultiplicity=yes or -Dusethreads=yes
parameters. Also, to make Perl leak less memory at run time, it should be
built with the -Dusemymalloc=no parameter. To check the values of these
parameters in an already built Perl (preferred values are specified in the
example), run:

$ perl -V : us emul tipl ici ty -V : usemymalloc

us emul tipl icit y = define ;
usemymalloc = n ;

Note that after rebuilding Perl with the new -Dusemultiplicity=yes or

-Dusethreads=yes parameters, all binary Perl modules will have to be rebuilt
as well they will just stop working with the new Perl.
There is a possibility that the main process and then worker processes
will grow in size after every reconfiguration. If the main process grows to an
unacceptable size, the live upgrade procedure can be applied without changing
the executable file.
While the Perl module is performing a long-running operation, such as
resolving a domain name, connecting to another server, or querying a database,
other requests assigned to the current worker process will not be processed. It

Nginx, Inc. p.136 of 330

CHAPTER 2. HTTP SERVER MODULES 2.29. MODULE NGX HTTP PERL MODULE

is thus recommended to perform only such operations that have predictable

and short execution time, such as accessing the local file system.

2.29.3 Example Configuration

http {

perl_modules perl / lib ;

perl_require hello . pm ;

perl_set $msie6

sub {
my $r = shift ;
my $ua = $r - > header_in (" User - Agent ") ;

return "" if $ua =~ / Opera /;

return "1" if $ua =~ / MSIE [6 -9]\.\ d +/;
return "";
}

server {
location / {
perl hello :: handler ;
}
}

The perl/lib/hello.pm module:

package hello ;

use nginx ;

sub handler {
my $r = shift ;

$r - > s e nd _ ht t p_ h ea d er (" text / html ") ;

return OK if $r - > header_only ;

$r - > print (" hello !\ n < br / >") ;

if ( - f $r - > filename or -d _ ) {
$r - > print ( $r - > uri , " exists !\ n ") ;
}

return OK ;
}

1;
__END__

2.29.4 Directives
perl
Syntax: perl module::function|sub { . . . };
Default
Context: location, limit except

Nginx, Inc. p.137 of 330

CHAPTER 2. HTTP SERVER MODULES 2.29. MODULE NGX HTTP PERL MODULE

Sets a Perl handler for the given location.

perl modules
Syntax: perl_modules path;
Default
Context: http

Sets an additional path for Perl modules.

perl require
Syntax: perl_require module;
Default
Context: http

Defines the name of a module that will be loaded during each

reconfiguration. Several perl_require directives can be present.

perl set
Syntax: perl_set $variable module::function|sub { . . . };
Default
Context: http

Installs a Perl handler for the specified variable.

2.29.5 Calling Perl from SSI

An SSI command calling Perl has the following format:

<! - -# perl sub ="module::function" arg ="parameter1" arg ="parameter2" ...

-->

2.29.6 The $r Request Object Methods

$r->args
returns request arguments.
$r->filename
returns a filename corresponding to the request URI.
$r->has_request_body(handler)
returns 0 if there is no body in a request. If there is a body, the specified
handler is set for the request and 1 is returned. After reading the request
body, nginx will call the specified handler. Note that the handler function
should be passed by reference. Example:

package hello ;

use nginx ;

Nginx, Inc. p.138 of 330

CHAPTER 2. HTTP SERVER MODULES 2.29. MODULE NGX HTTP PERL MODULE

sub handler {
my $r = shift ;

if ( $r - > request_method ne " POST ") {

return DECLINED ;
}

if ( $r - > ha s _r e qu e st _ bo d y (&post ) ) {
return OK ;
}

return HT T P_ B AD_R E QU E ST ;
}

sub post {
my $r = shift ;

$r - > s e nd _ ht t p_ hea d er ;

$r - > print (" request_body : \"" , $r - > request_body , "\" < br / >") ;
$r - > print (" r e q u e s t _ b o d y _ f i l e : \"" , $r - > request_body_file , "\" < br
/ >\ n ") ;

return OK ;
}

1;

__END__

$r->allow_ranges
enables the use of byte ranges when sending responses.
$r->discard_request_body
instructs nginx to discard the request body.
$r->header_in(field)
returns the value of the specified client request header field.
$r->header_only
determines whether the whole response or only its header should be sent
to the client.
$r->header_out(field, value)
sets a value for the specified response header field.
$r->internal_redirect(uri)
does an internal redirect to the specified uri. An actual redirect happens
after the Perl handler execution is completed.

Redirections to named locations are currently not supported.

$r->log_error(errno, message)
writes the specified message into the error log. If errno is non-zero, an
error code and its description will be appended to the message.
$r->print(text, ...)
passes data to a client.
$r->request_body
returns the client request body if it has not been written to a temporary
file. To ensure that the client request body is in memory, its size should

Nginx, Inc. p.139 of 330

CHAPTER 2. HTTP SERVER MODULES 2.29. MODULE NGX HTTP PERL MODULE

be limited by client max body size, and a sufficient buffer size should be
set using client body buffer size.
$r->request_body_file
returns the name of the file with the client request body. After the
processing, the file should be removed. To always write a request body
to a file, client body in file only should be enabled.
$r->request_method
returns the client request HTTP method.
$r->remote_addr
returns the client IP address.
$r->flush
immediately sends data to the client.
$r->sendfile(name[, offset[, length]])
sends the specified file content to the client. Optional parameters specify
the initial offset and length of the data to be transmitted. The actual
data transmission happens after the Perl handler has completed.
$r->send_http_header([type])
sends the response header to the client. The optional type parameter sets
the value of the Content-Type response header field. If the value is an
empty string, the Content-Type header field will not be sent.
$r->status(code)
sets a response code.
$r->sleep(milliseconds, handler)
sets the specified handler and stops request processing for the specified
time. In the meantime, nginx continues to process other requests. After
the specified time has elapsed, nginx will call the installed handler. Note
that the handler function should be passed by reference. In order to pass
data between handlers, $r->variable() should be used. Example:

package hello ;

use nginx ;

sub handler {
my $r = shift ;

$r - > d i s c a r d _ r e q u e s t _ b o d y ;
$r - > variable (" var " , " OK ") ;
$r - > sleep (1000 , &next ) ;

return OK ;
}

sub next {
my $r = shift ;

$r - > s e nd _ ht t p_ hea d er ;
$r - > print ( $r - > variable (" var ") ) ;

return OK ;
}

1;

__END__

Nginx, Inc. p.140 of 330

CHAPTER 2. HTTP SERVER MODULES 2.29. MODULE NGX HTTP PERL MODULE

$r->unescape(text)
decodes a text encoded in the %XX form.
$r->uri
returns a request URI.
$r->variable(name[, value])
returns or sets the value of the specified variable. Variables are local to
each request.

Nginx, Inc. p.141 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

2.30 Module ngx http proxy module

2.30.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 143
2.30.2 Example Configuration . . . . . . . . . . . . . . . . . . . 143
2.30.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 143
proxy bind . . . . . . . . . . . . . . . . . . . . . . . . . 143
proxy buffer size . . . . . . . . . . . . . . . . . . . . . . 144
proxy buffering . . . . . . . . . . . . . . . . . . . . . . . 144
proxy buffers . . . . . . . . . . . . . . . . . . . . . . . . 144
proxy busy buffers size . . . . . . . . . . . . . . . . . . . 145
proxy cache . . . . . . . . . . . . . . . . . . . . . . . . . 145
proxy cache bypass . . . . . . . . . . . . . . . . . . . . . 145
proxy cache key . . . . . . . . . . . . . . . . . . . . . . . 145
proxy cache lock . . . . . . . . . . . . . . . . . . . . . . 146
proxy cache lock age . . . . . . . . . . . . . . . . . . . . 146
proxy cache lock timeout . . . . . . . . . . . . . . . . . 146
proxy cache methods . . . . . . . . . . . . . . . . . . . . 146
proxy cache min uses . . . . . . . . . . . . . . . . . . . . 147
proxy cache path . . . . . . . . . . . . . . . . . . . . . . 147
proxy cache purge . . . . . . . . . . . . . . . . . . . . . 148
proxy cache revalidate . . . . . . . . . . . . . . . . . . . 148
proxy cache use stale . . . . . . . . . . . . . . . . . . . . 149
proxy cache valid . . . . . . . . . . . . . . . . . . . . . . 149
proxy connect timeout . . . . . . . . . . . . . . . . . . . 150
proxy cookie domain . . . . . . . . . . . . . . . . . . . . 150
proxy cookie path . . . . . . . . . . . . . . . . . . . . . 151
proxy force ranges . . . . . . . . . . . . . . . . . . . . . 152
proxy headers hash bucket size . . . . . . . . . . . . . . 152
proxy headers hash max size . . . . . . . . . . . . . . . 152
proxy hide header . . . . . . . . . . . . . . . . . . . . . 152
proxy http version . . . . . . . . . . . . . . . . . . . . . 152
proxy ignore client abort . . . . . . . . . . . . . . . . . . 153
proxy ignore headers . . . . . . . . . . . . . . . . . . . . 153
proxy intercept errors . . . . . . . . . . . . . . . . . . . 153
proxy limit rate . . . . . . . . . . . . . . . . . . . . . . . 153
proxy max temp file size . . . . . . . . . . . . . . . . . . 154
proxy method . . . . . . . . . . . . . . . . . . . . . . . . 154
proxy next upstream . . . . . . . . . . . . . . . . . . . . 154
proxy next upstream timeout . . . . . . . . . . . . . . . 155
proxy next upstream tries . . . . . . . . . . . . . . . . . 155
proxy no cache . . . . . . . . . . . . . . . . . . . . . . . 156
proxy pass . . . . . . . . . . . . . . . . . . . . . . . . . . 156
proxy pass header . . . . . . . . . . . . . . . . . . . . . 157
proxy read timeout . . . . . . . . . . . . . . . . . . . . . 158
proxy pass request body . . . . . . . . . . . . . . . . . . 158
proxy pass request headers . . . . . . . . . . . . . . . . 158

Nginx, Inc. p.142 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy redirect . . . . . . . . . . . . . . . . . . . . . . . . 158

proxy request buffering . . . . . . . . . . . . . . . . . . . 160
proxy send lowat . . . . . . . . . . . . . . . . . . . . . . 160
proxy send timeout . . . . . . . . . . . . . . . . . . . . . 160
proxy set body . . . . . . . . . . . . . . . . . . . . . . . 161
proxy set header . . . . . . . . . . . . . . . . . . . . . . 161
proxy ssl certificate . . . . . . . . . . . . . . . . . . . . . 162
proxy ssl certificate key . . . . . . . . . . . . . . . . . . 162
proxy ssl ciphers . . . . . . . . . . . . . . . . . . . . . . 162
proxy ssl crl . . . . . . . . . . . . . . . . . . . . . . . . . 162
proxy ssl name . . . . . . . . . . . . . . . . . . . . . . . 162
proxy ssl password file . . . . . . . . . . . . . . . . . . . 163
proxy ssl server name . . . . . . . . . . . . . . . . . . . 163
proxy ssl session reuse . . . . . . . . . . . . . . . . . . . 163
proxy ssl protocols . . . . . . . . . . . . . . . . . . . . . 163
proxy ssl trusted certificate . . . . . . . . . . . . . . . . 163
proxy ssl verify . . . . . . . . . . . . . . . . . . . . . . . 164
proxy ssl verify depth . . . . . . . . . . . . . . . . . . . 164
proxy store . . . . . . . . . . . . . . . . . . . . . . . . . 164
proxy store access . . . . . . . . . . . . . . . . . . . . . 165
proxy temp file write size . . . . . . . . . . . . . . . . . 165
proxy temp path . . . . . . . . . . . . . . . . . . . . . . 166
2.30.4 Embedded Variables . . . . . . . . . . . . . . . . . . . . 166

2.30.1 Summary
The ngx_http_proxy_module module allows passing requests to another
server.

2.30.2 Example Configuration

location / {
proxy_pass http :// localhost :8000;
p ro x y_ s et _ he a de r Host $host ;
p ro x y_ s et _ he a de r X - Real - IP $remote_addr ;
}

2.30.3 Directives
proxy bind
Syntax: proxy_bind address | off;
Default
Context: http, server, location
This directive appeared in version 0.8.22.

Nginx, Inc. p.143 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

Makes outgoing connections to a proxied server originate from the specified

local IP address. Parameter value can contain variables (1.3.12). The special
value off (1.3.12) cancels the effect of the proxy_bind directive inherited
from the previous configuration level, which allows the system to auto-assign
the local IP address.

proxy buffer size

Syntax: proxy_buffer_size size;
Default 4k|8k
Context: http, server, location

Sets the size of the buffer size used for reading the first part of the response
received from the proxied server. This part usually contains a small response
header. By default, the buffer size is equal to the size of one buffer set by the
proxy buffers directive. It can be made smaller, however.

proxy buffering
Syntax: proxy_buffering on | off;
Default on
Context: http, server, location

Enables or disables buffering of responses from the proxied server.

When buffering is enabled, nginx receives a response from the proxied server
as soon as possible, saving it into the buffers set by the proxy buffer size and
proxy buffers directives. If the whole response does not fit into memory, a part
of it can be saved to a temporary file on the disk. Writing to temporary files
is controlled by the proxy max temp file size and proxy temp file write size
directives.
When buffering is disabled, the response is passed to a client synchronously,
immediately as it is received. nginx will not try to read the whole response
from the proxied server. The maximum size of the data that nginx can receive
from the server at a time is set by the proxy buffer size directive.
Buffering can also be enabled or disabled by passing yes or no in the
X-Accel-Buffering response header field. This capability can be disabled
using the proxy ignore headers directive.

proxy buffers
Syntax: proxy_buffers number size;
Default 8 4k|8k
Context: http, server, location

Sets the number and size of the buffers used for reading a response from
the proxied server, for a single connection. By default, the buffer size is equal
to one memory page. This is either 4K or 8K, depending on a platform.

Nginx, Inc. p.144 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy busy buffers size

Syntax: proxy_busy_buffers_size size;
Default 8k|16k
Context: http, server, location

When buffering of responses from the proxied server is enabled, limits the
total size of buffers that can be busy sending a response to the client while the
response is not yet fully read. In the meantime, the rest of the buffers can be
used for reading the response and, if needed, buffering part of the response to
a temporary file. By default, size is limited by the size of two buffers set by
the proxy buffer size and proxy buffers directives.

proxy cache
Syntax: proxy_cache zone | off;
Default off
Context: http, server, location

Defines a shared memory zone used for caching. The same zone can be
used in several places. Parameter value can contain variables (1.7.9). The off
parameter disables caching inherited from the previous configuration level.

proxy cache bypass

Syntax: proxy_cache_bypass string . . . ;
Default
Context: http, server, location

Defines conditions under which the response will not be taken from a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be taken from the cache:

p r o x y _ c a c h e _ b y p a s s $ coo kie_ noca che $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

p r o x y _ c a c h e _ b y p a s s $http_pragma $http_authorization ;

Can be used along with the proxy no cache directive.

proxy cache key

Syntax: proxy_cache_key string;
Default $scheme$proxy_host$request_uri
Context: http, server, location

Defines a key for caching, for example

pr oxy_ cach e_ke y " $ h o s t $ r e q u e s t _ u r i $cookie_user ";

By default, the directives value is close to the string

pr oxy_ cach e_ke y $ s c h e m e $ p r o x y _ h o s t $ u r i $ i s _ a r g s $ a r g s ;

Nginx, Inc. p.145 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy cache lock

Syntax: proxy_cache_lock on | off;
Default off
Context: http, server, location
This directive appeared in version 1.1.12.

When enabled, only one request at a time will be allowed to populate a new
cache element identified according to the proxy cache key directive by passing
a request to a proxied server. Other requests of the same cache element will
either wait for a response to appear in the cache or the cache lock for this
element to be released, up to the time set by the proxy cache lock timeout
directive.

proxy cache lock age

Syntax: proxy_cache_lock_age time;
Default 5s
Context: http, server, location
This directive appeared in version 1.7.8.

If the last request passed to the proxied server for populating a new cache
element has not completed for the specified time, one more request may be
passed to the proxied server.

proxy cache lock timeout

Syntax: proxy_cache_lock_timeout time;
Default 5s
Context: http, server, location
This directive appeared in version 1.1.12.

Sets a timeout for proxy cache lock. When the time expires, the request
will be passed to the proxied server, however, the response will not be cached.

Before 1.7.8, the response could be cached.

proxy cache methods

Syntax: proxy_cache_methods GET | HEAD | POST . . . ;
Default GET HEAD
Context: http, server, location
This directive appeared in version 0.7.59.

If the client request method is listed in this directive then the response will
be cached. GET and HEAD methods are always added to the list, though it is
recommended to specify them explicitly. See also the proxy no cache directive.

Nginx, Inc. p.146 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy cache min uses

Syntax: proxy_cache_min_uses number;
Default 1
Context: http, server, location

Sets the number of requests after which the response will be cached.

proxy cache path

Syntax: proxy_cache_path path [levels=levels] [use_temp_path=on|off]
keys_zone=name:size [inactive=time] [max_size=size]
[loader_files=number] [loader_sleep=time]
[loader_threshold=time];
Default
Context: http

Sets the path and other parameters of a cache. Cache data are stored in
files. The file name in a cache is a result of applying the MD5 function to
the cache key. The levels parameter defines hierarchy levels of a cache. For
example, in the following configuration

p ro x y_ c ac h e_ p at h / data / nginx / cache levels =1:2 keys_zone = one :10 m ;

file names in a cache will look like this:

/ data / nginx / cache /c /29 / b 7 f 5 4 b 2 d f 7 7 7 3 7 2 2 d 3 8 2 f 4 8 0 9 d 6 5 0 29c

A cached response is first written to a temporary file, and then the file
is renamed. Starting from version 0.8.9, temporary files and the cache can
be put on different file systems. However, be aware that in this case a file
is copied across two file systems instead of the cheap renaming operation. It
is thus recommended that for any given location both cache and a directory
holding temporary files are put on the same file system. The directory for
temporary files is set based on the use_temp_path parameter (1.7.10). If this
parameter is omitted or set to the value on, the directory set by the proxy -
temp path directive for the given location will be used. If the value is set to
off, temporary files will be put directly in the cache directory.
In addition, all active keys and information about data are stored in a
shared memory zone, whose name and size are configured by the keys_zone
parameter. One megabyte zone can store about 8 thousand keys.
Cached data that are not accessed during the time specified by the
inactive parameter get removed from the cache regardless of their freshness.
By default, inactive is set to 10 minutes.
The special cache manager process monitors the maximum cache size set
by the max_size parameter. When this size is exceeded, it removes the least
recently used data.
A minute after the start the special cache loader process is activated. It
loads information about previously cached data stored on file system into a

Nginx, Inc. p.147 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

cache zone. The loading is done in iterations. During one iteration no more
than loader_files items are loaded (by default, 100). Besides, the duration of
one iteration is limited by the loader_threshold parameter (by default, 200
milliseconds). Between iterations, a pause configured by the loader_sleep
parameter (by default, 50 milliseconds) is made.

proxy cache purge

Syntax: proxy_cache_purgestring . . . ;
Default
Context: http, server, location
This directive appeared in version 1.5.7.

Defines conditions under which the request will be considered a cache purge
request. If at least one value of the string parameters is not empty and
is not equal to 0 then the cache entry with a corresponding cache key is
removed. The result of successful operation is indicated by returning the 204
No Content response.
If the cache key of a purge request ends with an asterisk (*), all cache
entries matching the wildcard key will be removed from the cache.
Example configuration:

p ro x y_ c ac h e_ p at h / data / nginx / cache keys_zone = cache_zone :10 m ;

map $re ques t_m etho d $purge_method {

PURGE 1;
default 0;
}

server {
...
location / {
proxy_pass http :// backend ;
proxy_cache cache_zone ;
pro xy_ cach e_ke y $uri ;
p r o x y _ c a c h e _ p u r g e $purge_method ;
}
}

This functionality is available as part of our commercial subscription.

proxy cache revalidate

Syntax: proxy_cache_revalidate on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.7.

Enables revalidation of expired cache items using conditional requests with

the If-Modified-Since and If-None-Match header fields.

Nginx, Inc. p.148 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy cache use stale

Syntax: proxy_cache_use_stale error | timeout | invalid_header | updating
| http_500 | http_502 | http_503 | http_504 | http_403 | http_404 |
off . . . ;
Default off
Context: http, server, location

Determines in which cases a stale cached response can be used when an

error occurs during communication with the proxied server. The directives
parameters match the parameters of the proxy next upstream directive.
Additionally, the updating parameter permits using a stale cached response
if it is currently being updated. This allows minimizing the number of accesses
to proxied servers when updating cached data.
To minimize the number of accesses to proxied servers when populating a
new cache element, the proxy cache lock directive can be used.

proxy cache valid

Syntax: proxy_cache_valid [code . . . ] time;
Default
Context: http, server, location

Sets caching time for different response codes. For example, the following
directives

p r o x y _ c a c h e _ v a l i d 200 302 10 m ;
p r o x y _ c a c h e _ v a l i d 404 1m;

set 10 minutes of caching for responses with codes 200 and 302 and 1 minute
for responses with code 404.
If only caching time is specified

proxy_cache_valid 5m;

then only 200, 301, and 302 responses are cached.

In addition, the any parameter can be specified to cache any responses:

p r o x y _ c a c h e _ v a l i d 200 302 10 m ;
p r o x y _ c a c h e _ v a l i d 301 1h;
p r o x y _ c a c h e _ v a l i d any 1m;

Parameters of caching can also be set directly in the response header. This
has higher priority than setting of caching time using the directive.
The X-Accel-Expires header field sets caching time of a response in
seconds. The zero value disables caching for a response. If the value
starts with the @ prefix, it sets an absolute time in seconds since Epoch,
up to which the response may be cached.
If the header does not include the X-Accel-Expires field, parameters of
caching may be set in the header fields Expires or Cache-Control.

Nginx, Inc. p.149 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

If the header includes the Set-Cookie field, such a response will not be
cached.

If the header includes the Vary field with the special value *, such a
response will not be cached (1.7.7). If the header includes the Vary field
with another value, such a response will be cached taking into account
the corresponding request header fields (1.7.7).

Processing of one or more of these response header fields can be disabled using
the proxy ignore headers directive.

proxy connect timeout

Syntax: proxy_connect_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for establishing a connection with a proxied server. It

should be noted that this timeout cannot usually exceed 75 seconds.

proxy cookie domain

Syntax: proxy_cookie_domain off;
Syntax: proxy_cookie_domain domain replacement;
Default off
Context: http, server, location
This directive appeared in version 1.1.15.

Sets a text that should be changed in the domain attribute of the

Set-Cookie header fields of a proxied server response. Suppose a
proxied server returned the Set-Cookie header field with the attribute
domain=localhost. The directive

p r o x y _ c o o k i e _ d o m a i n localhost example . org ;

will rewrite this attribute to domain=example.org.

A dot at the beginning of the domain and replacement strings and the
domain attribute is ignored. Matching is case-insensitive.
The domain and replacement strings can contain variables:

p r o x y _ c o o k i e _ d o m a i n www . $host $host ;

The directive can also be specified using regular expressions. In this case,
domain should start from the ~ symbol. A regular expression can contain
named and positional captures, and replacement can reference them:

p r o x y _ c o o k i e _ d o m a i n ~\.(? P < sl_domain >[ -0 -9 a - z ]+\.[ a - z ]+) $ $sl_domain ;

There could be several proxy_cookie_domain directives:

Nginx, Inc. p.150 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

p r o x y _ c o o k i e _ d o m a i n localhost example . org ;

p r o x y _ c o o k i e _ d o m a i n ~\.([ a - z ]+\.[ a - z ]+) $ $1 ;

The off parameter cancels the effect of all proxy_cookie_domain

directives on the current level:

p r o x y _ c o o k i e _ d o m a i n off ;
p r o x y _ c o o k i e _ d o m a i n localhost example . org ;
p r o x y _ c o o k i e _ d o m a i n www . example . org example . org ;

proxy cookie path

Syntax: proxy_cookie_path off;
Syntax: proxy_cookie_path path replacement;
Default off
Context: http, server, location
This directive appeared in version 1.1.15.

Sets a text that should be changed in the path attribute of the Set-Cookie
header fields of a proxied server response. Suppose a proxied server returned
the Set-Cookie header field with the attribute path=/two/some/uri/. The
directive

p r o x y _ c o o k i e _ p a t h / two / /;

will rewrite this attribute to path=/some/uri/.

The path and replacement strings can contain variables:

p r o x y _ c o o k i e _ p a t h $uri / some$uri ;

The directive can also be specified using regular expressions. In this case,
path should either start from the ~ symbol for a case-sensitive matching, or
from the ~* symbols for case-insensitive matching. The regular expression
can contain named and positional captures, and replacement can reference
them:

p r o x y _ c o o k i e _ p a t h ~*^/ user /([^/]+) / u / $1 ;

There could be several proxy_cookie_path directives:

p r o x y _ c o o k i e _ p a t h / one / /;
p r o x y _ c o o k i e _ p a t h / / two /;

The off parameter cancels the effect of all proxy_cookie_path directives

on the current level:

p r o x y _ c o o k i e _ p a t h off ;
p r o x y _ c o o k i e _ p a t h / two / /;
p r o x y _ c o o k i e _ p a t h ~*^/ user /([^/]+) / u / $1 ;

Nginx, Inc. p.151 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy force ranges

Syntax: proxy_force_ranges on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.7.

Enables byte-range support for both cached and uncached responses from
the proxied server regardless of the Accept-Ranges field in these responses.

proxy headers hash bucket size

Syntax: proxy_headers_hash_bucket_size size;
Default 64
Context: http, server, location

Sets the bucket size for hash tables used by the proxy hide header and
proxy set header directives. The details of setting up hash tables are provided
in a separate document.

proxy headers hash max size

Syntax: proxy_headers_hash_max_size size;
Default 512
Context: http, server, location

Sets the maximum size of hash tables used by the proxy hide header and
proxy set header directives. The details of setting up hash tables are provided
in a separate document.

proxy hide header

Syntax: proxy_hide_header field;
Default
Context: http, server, location

By default, nginx does not pass the header fields Date, Server, X-Pad,
and X-Accel-... from the response of a proxied server to a client. The
proxy_hide_header directive sets additional fields that will not be passed.
If, on the contrary, the passing of fields needs to be permitted, the proxy -
pass header directive can be used.

proxy http version

Syntax: proxy_http_version 1.0 | 1.1;
Default 1.0
Context: http, server, location
This directive appeared in version 1.1.4.

Sets the HTTP protocol version for proxying. By default, version 1.0 is
used. Version 1.1 is recommended for use with keepalive connections.

Nginx, Inc. p.152 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy ignore client abort

Syntax: proxy_ignore_client_abort on | off;
Default off
Context: http, server, location

Determines whether the connection with a proxied server should be closed

when a client closes the connection without waiting for a response.

proxy ignore headers

Syntax: proxy_ignore_headers field . . . ;
Default
Context: http, server, location

Disables processing of certain response header fields from the proxied server.
The following fields can be ignored: X-Accel-Redirect, X-Accel-Expires,
X-Accel-Limit-Rate (1.1.6), X-Accel-Buffering (1.1.6), X-Accel-Charset
(1.1.6), Expires, Cache-Control, Set-Cookie (0.8.44), and Vary (1.7.7).
If not disabled, processing of these header fields has the following effect:
X-Accel-Expires, Expires, Cache-Control, Set-Cookie, and Vary set
the parameters of response caching;

X-Accel-Redirect performs an internal redirect to the specified URI;

X-Accel-Limit-Rate sets the rate limit for transmission of a response

to a client;

X-Accel-Buffering enables or disables buffering of a response;

X-Accel-Charset sets the desired charset of a response.

proxy intercept errors

Syntax: proxy_intercept_errors on | off;
Default off
Context: http, server, location

Determines whether proxied responses with codes greater than or equal to

300 should be passed to a client or be redirected to nginx for processing with
the error page directive.

proxy limit rate

Syntax: proxy_limit_rate rate;
Default 0
Context: http, server, location
This directive appeared in version 1.7.7.

Limits the speed of reading the response from the proxied server. The rate
is specified in bytes per second. The zero value disables rate limiting. The limit

Nginx, Inc. p.153 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

is set per a request, and so if nginx simultaneously opens two connections to

the proxied server, the overall rate will be twice as much as the specified limit.
The limitation works only if buffering of responses from the proxied server is
enabled.

proxy max temp file size

Syntax: proxy_max_temp_file_size size;
Default 1024m
Context: http, server, location

When buffering of responses from the proxied server is enabled, and the
whole response does not fit into the buffers set by the proxy buffer size and
proxy buffers directives, a part of the response can be saved to a temporary file.
This directive sets the maximum size of the temporary file. The size of data
written to the temporary file at a time is set by the proxy temp file write size
directive.
The zero value disables buffering of responses to temporary files.

This restriction does not apply to responses that will be cached or stored
on disk.

proxy method
Syntax: proxy_method method;
Default
Context: http, server, location

Specifies the HTTP method to use in requests forwarded to the proxied

server instead of the method from the client request.

proxy next upstream

Syntax: proxy_next_upstream error | timeout | invalid_header | http_500 |
http_502 | http_503 | http_504 | http_403 | http_404 | off . . . ;
Default error timeout
Context: http, server, location

Specifies in which cases a request should be passed to the next server:

error
an error occurred while establishing a connection with the server, passing
a request to it, or reading the response header;
timeout
a timeout has occurred while establishing a connection with the server,
passing a request to it, or reading the response header;
invalid_header
a server returned an empty or invalid response;

Nginx, Inc. p.154 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

http_500
a server returned a response with the code 500;
http_502
a server returned a response with the code 502;
http_503
a server returned a response with the code 503;
http_504
a server returned a response with the code 504;
http_403
a server returned a response with the code 403;
http_404
a server returned a response with the code 404;
off
disables passing a request to the next server.

One should bear in mind that passing a request to the next server is only
possible if nothing has been sent to a client yet. That is, if an error or timeout
occurs in the middle of the transferring of a response, fixing this is impossible.
The directive also defines what is considered an unsuccessful attempt
of communication with a server. The cases of error, timeout and
invalid_header are always considered unsuccessful attempts, even if they
are not specified in the directive. The cases of http_500, http_502, http_503
and http_504 are considered unsuccessful attempts only if they are specified
in the directive. The cases of http_403 and http_404 are never considered
unsuccessful attempts.
Passing a request to the next server can be limited by the number of tries
and by time.

proxy next upstream timeout

Syntax: proxy_next_upstream_timeout time;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the time allowed to pass a request to the next server. The 0 value
turns off this limitation.

proxy next upstream tries

Syntax: proxy_next_upstream_tries number;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the number of possible tries for passing a request to the next server.
The 0 value turns off this limitation.

Nginx, Inc. p.155 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy no cache
Syntax: proxy_no_cache string . . . ;
Default
Context: http, server, location

Defines conditions under which the response will not be saved to a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be saved:

proxy_no_cache $ cook ie_n oca che $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

proxy_no_cache $http_pragma $http_authorization ;

Can be used along with the proxy cache bypass directive.

proxy pass
Syntax: proxy_pass URL;
Default
Context: location, if in location, limit except

Sets the protocol and address of a proxied server and an optional URI to
which a location should be mapped. As a protocol, http or https can be
specified. The address can be specified as a domain name or IP address, and
an optional port:

proxy_pass http :// localhost :8000/ uri /;

or as a UNIX-domain socket path specified after the word unix and

enclosed in colons:

proxy_pass http :// unix :/ tmp / backend . socket :/ uri /;

If a domain name resolves to several addresses, all of them will be used

in a round-robin fashion. In addition, an address can be specified as a server
group.
A request URI is passed to the server as follows:

If the proxy_pass directive is specified with a URI, then when a request

is passed to the server, the part of a normalized request URI matching
the location is replaced by a URI specified in the directive:

location / name / {
proxy_pass http ://127.0.0.1/ remote /;
}

If proxy_pass is specified without a URI, the request URI is passed to

the server in the same form as sent by a client when the original request is
processed, or the full normalized request URI is passed when processing
the changed URI:

Nginx, Inc. p.156 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

location / some / path / {

proxy_pass http ://127.0.0.1;
}

Before version 1.1.12, if proxy_pass is specified without a URI, the

original request URI might be passed instead of the changed URI in
some cases.

In some cases, the part of a request URI to be replaced cannot be

determined:

When location is specified using a regular expression.

In this case, the directive should be specified without a URI.

When the URI is changed inside a proxied location using the rewrite
directive, and this same configuration will be used to process a request
(break):

location / name / {
rewrite / name /([^/]+) / users ? name = $1 break ;
proxy_pass http ://127.0.0.1;
}

In this case, the URI specified in the directive is ignored and the full
changed request URI is passed to the server.

A server name, its port and the passed URI can also be specified using
variables:

proxy_pass http :// $host$uri ;

or even like this:

proxy_pass $request ;

In this case, the server name is searched among the described server groups,
and, if not found, is determined using a resolver.
WebSocket proxying requires special configuration and is supported since
version 1.3.13.

proxy pass header

Syntax: proxy_pass_header field;
Default
Context: http, server, location

Permits passing otherwise disabled header fields from a proxied server to a

client.

Nginx, Inc. p.157 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy read timeout

Syntax: proxy_read_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for reading a response from the proxied server. The
timeout is set only between two successive read operations, not for the
transmission of the whole response. If the proxied server does not transmit
anything within this time, the connection is closed.

proxy pass request body

Syntax: proxy_pass_request_body on | off;
Default on
Context: http, server, location

Indicates whether the original request body is passed to the proxied server.

location /x - accel - redirect - here / {

proxy_method GET ;
p r o x y _ p a s s _ r e q u e s t _ b o d y off ;
p ro x y_ s et _ he a de r Content - Length "";

proxy_pass ...
}

See also the proxy set header and proxy pass request headers directives.

proxy pass request headers

Syntax: proxy_pass_request_headers on | off;
Default on
Context: http, server, location

Indicates whether the header fields of the original request are passed to the
proxied server.

location /x - accel - redirect - here / {

proxy_method GET ;
p r o x y _ p a s s _ r e q u e s t _ h e a d e r s off ;
p r o x y _ p a s s _ r e q u e s t _ b o d y off ;

proxy_pass ...
}

See also the proxy set header and proxy pass request body directives.

proxy redirect
Syntax: proxy_redirect default;
Syntax: proxy_redirect off;
Syntax: proxy_redirect redirect replacement;
Default default
Context: http, server, location

Nginx, Inc. p.158 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

Sets the text that should be changed in the Location and Refresh header
fields of a proxied server response. Suppose a proxied server returned the
header field Location: http://localhost:8000/two/some/uri/. The
directive

proxy_redirect http :// localhost :8000/ two / http :// frontend / one /;

will rewrite this string toLocation: http://frontend/one/some/uri/.

A server name may be omitted in the replacement string:

proxy_redirect http :// localhost :8000/ two / /;

then the primary servers name and port, if different from 80, will be
inserted.
The default replacement specified by the default parameter uses the
parameters of the location and proxy pass directives. Hence, the two
configurations below are equivalent:

location / one / {
proxy_pass http :// upstream : port / two /;
proxy_redirect default ;

location / one / {
proxy_pass http :// upstream : port / two /;
proxy_redirect http :// upstream : port / two / / one /;

The default parameter is not permitted if proxy pass is specified using

variables.
A replacement string can contain variables:

proxy_redirect http :// localhost :8000/ http :// $host : $server_port /;

A redirect can also contain (1.1.11) variables:

proxy_redirect http :// $proxy_host :8000/ /;

The directive can be specified (1.1.11) using regular expressions. In this

case, redirect should either start with the ~ symbol for a case-sensitive
matching, or with the ~* symbols for case-insensitive matching. The regular
expression can contain named and positional captures, and replacement can
reference them:

proxy_redirect ~^( http ://[^:]+) :\ d +(/.+) $ $1$2 ;

proxy_redirect ~*/ user /([^/]+) /(.+) $ http :// $1 . example . com / $2 ;

There could be several proxy_redirect directives:

proxy_redirect default ;
proxy_redirect http :// localhost :8000/ /;
proxy_redirect http :// www . example . com / /;

Nginx, Inc. p.159 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

The off parameter cancels the effect of all proxy_redirect directives on

the current level:

proxy_redirect off ;
proxy_redirect default ;
proxy_redirect http :// localhost :8000/ /;
proxy_redirect http :// www . example . com / /;

Using this directive, it is also possible to add host names to relative redirects
issued by a proxied server:

proxy_redirect / /;

proxy request buffering

Syntax: proxy_request_buffering on | off;
Default on
Context: http, server, location
This directive appeared in version 1.7.11.

Enables or disables buffering of a client request body.

When buffering is enabled, the entire request body is read from the client
before sending the request to a proxied server.
When buffering is disabled, the request body is sent to the proxied server
immediately as it is received. In this case, the request cannot be passed to the
next server if nginx already started sending the request body.
When HTTP/1.1 chunked transfer encoding is used to send the original
request body, the request body will be buffered regardless of the directive
value unless HTTP/1.1 is enabled for proxying.

proxy send lowat

Syntax: proxy_send_lowat size;
Default 0
Context: http, server, location

If the directive is set to a non-zero value, nginx will try to minimize the
number of send operations on outgoing connections to a proxied server by
using either NOTE_LOWAT flag of the kqueue method, or the SO_SNDLOWAT socket
option, with the specified size.
This directive is ignored on Linux, Solaris, and Windows.

proxy send timeout

Syntax: proxy_send_timeout time;
Default 60s
Context: http, server, location

Sets a timeout for transmitting a request to the proxied server. The timeout
is set only between two successive write operations, not for the transmission of

Nginx, Inc. p.160 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

the whole request. If the proxied server does not receive anything within this
time, the connection is closed.

proxy set body

Syntax: proxy_set_body value;
Default
Context: http, server, location

Allows redefining the request body passed to the proxied server. The value
can contain text, variables, and their combination.

proxy set header

Syntax: proxy_set_header field value;
Default Host $proxy_host
Default Connection close
Context: http, server, location

Allows redefining or appending fields to the request header passed to the

proxied server. The value can contain text, variables, and their combinations.
These directives are inherited from the previous level if and only if there are
no proxy_set_header directives defined on the current level. By default, only
two fields are redefined:

p ro x y_ s et _ he a de r Host $proxy_host ;
p ro x y_ s et _ he a de r Connection close ;

An unchanged Host request header field can be passed like this:

p ro x y_ s et _ he a de r Host $http_host ;

However, if this field is not present in a client request header then nothing
will be passed. In such a case it is better to use the $host variable - its value
equals the server name in the Host request header field or the primary server
name if this field is not present:

p ro x y_ s et _ he a de r Host $host ;

In addition, the server name can be passed together with the port of the
proxied server:

p ro x y_ s et _ he a de r Host $host : $proxy_port ;

If the value of a header field is an empty string then this field will not be
passed to a proxied server:

p ro x y_ s et _ he a de r Accept - Encoding "";

Nginx, Inc. p.161 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy ssl certificate

Syntax: proxy_ssl_certificate file;
Default
Context: http, server, location
This directive appeared in version 1.7.8.

Specifies a file with the certificate in the PEM format used for
authentication to a proxied HTTPS server.

proxy ssl certificate key

Syntax: proxy_ssl_certificate_key file;
Default
Context: http, server, location
This directive appeared in version 1.7.8.

Specifies a file with the secret key in the PEM format used for
authentication to a proxied HTTPS server.

proxy ssl ciphers

Syntax: proxy_ssl_ciphers ciphers;
Default DEFAULT
Context: http, server, location
This directive appeared in version 1.5.6.

Specifies the enabled ciphers for requests to a proxied HTTPS server. The
ciphers are specified in the format understood by the OpenSSL library.
The full list can be viewed using the openssl ciphers command.

proxy ssl crl

Syntax: proxy_ssl_crl file;
Default
Context: http, server, location
This directive appeared in version 1.7.0.

Specifies a file with revoked certificates (CRL) in the PEM format used to
verify the certificate of the proxied HTTPS server.

proxy ssl name

Syntax: proxy_ssl_name name;
Default $proxy_host
Context: http, server, location
This directive appeared in version 1.7.0.

Allows to override the server name used to verify the certificate of the
proxied HTTPS server and to be passed through SNI when establishing a
connection with the proxied HTTPS server.

Nginx, Inc. p.162 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

By default, the host part of the proxy pass URL is used.

proxy ssl password file

Syntax: proxy_ssl_password_file file;
Default
Context: http, server, location
This directive appeared in version 1.7.8.

Specifies a file with passphrases for secret keys where each passphrase is
specified on a separate line. Passphrases are tried in turn when loading the
key.

proxy ssl server name

Syntax: proxy_ssl_server_name on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.0.

Enables or disables passing of the server name through TLS Server Name
Indication extension (SNI, RFC 6066) when establishing a connection with the
proxied HTTPS server.

proxy ssl session reuse

Syntax: proxy_ssl_session_reuse on | off;
Default on
Context: http, server, location

Determines whether SSL sessions can be reused when working with the
proxied server. If the errors SSL3_GET_FINISHED:digest check failed
appear in the logs, try disabling session reuse.

proxy ssl protocols

Syntax: proxy_ssl_protocols [SSLv2] [SSLv3] [TLSv1] [TLSv1.1] [TLSv1.2];
Default SSLv3 TLSv1 TLSv1.1 TLSv1.2
Context: http, server, location
This directive appeared in version 1.5.6.

Enables the specified protocols for requests to a proxied HTTPS server.

proxy ssl trusted certificate

Syntax: proxy_ssl_trusted_certificate file;
Default
Context: http, server, location
This directive appeared in version 1.7.0.

Nginx, Inc. p.163 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

Specifies a file with trusted CA certificates in the PEM format used to

verify the certificate of the proxied HTTPS server.

proxy ssl verify

Syntax: proxy_ssl_verify on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.0.

Enables or disables verification of the proxied HTTPS server certificate.

proxy ssl verify depth

Syntax: proxy_ssl_verify_depth number;
Default 1
Context: http, server, location
This directive appeared in version 1.7.0.

Sets the verification depth in the proxied HTTPS server certificates chain.

proxy store
Syntax: proxy_store on | off | string;
Default off
Context: http, server, location

Enables saving of files to a disk. The on parameter saves files with paths
corresponding to the directives alias or root. The off parameter disables saving
of files. In addition, the file name can be set explicitly using the string with
variables:

proxy_store / data / w w w$ or i gi n al _ ur i ;

The modification time of files is set according to the received

Last-Modified response header field. The response is first written to a
temporary file, and then the file is renamed. Starting from version 0.8.9,
temporary files and the persistent store can be put on different file systems.
However, be aware that in this case a file is copied across two file systems
instead of the cheap renaming operation. It is thus recommended that for any
given location both saved files and a directory holding temporary files, set by
the proxy temp path directive, are put on the same file system.
This directive can be used to create local copies of static unchangeable files,
e.g.:

location / images / {
root / data / www ;
error_page 404 = / fetch$uri ;
}

location / fetch / {
internal ;

Nginx, Inc. p.164 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy_pass http :// backend /;

proxy_store on ;
proxy_store_access user : rw group : rw all : r ;
pr oxy_ temp _pat h / data / temp ;

alias / data / www /;

}

or like this:

location / images / {
root / data / www ;
error_page 404 = @fetch ;
}

location @fetch {
internal ;

proxy_pass http :// backend ;

proxy_store on ;
proxy_store_access user : rw group : rw all : r ;
pr oxy_ temp _pat h / data / temp ;

root / data / www ;

}

proxy store access

Syntax: proxy_store_access users:permissions . . . ;
Default user:rw
Context: http, server, location

Sets access permissions for newly created files and directories, e.g.:

p r o x y _ s t o r e _ a c c e s s user : rw group : rw all : r ;

If any group or all access permissions are specified then user permissions
may be omitted:

p r o x y _ s t o r e _ a c c e s s group : rw all : r ;

proxy temp file write size

Syntax: proxy_temp_file_write_size size;
Default 8k|16k
Context: http, server, location

Limits the size of data written to a temporary file at a time, when buffering
of responses from the proxied server to temporary files is enabled. By default,
size is limited by two buffers set by the proxy buffer size and proxy buffers
directives. The maximum size of a temporary file is set by the proxy max -
temp file size directive.

Nginx, Inc. p.165 of 330

CHAPTER 2. HTTP SERVER MODULES 2.30. MODULE NGX HTTP PROXY MODULE

proxy temp path

Syntax: proxy_temp_path path [level1 [level2 [level3]]];
Default proxy_temp
Context: http, server, location

Defines a directory for storing temporary files with data received from
proxied servers. Up to three-level subdirectory hierarchy can be used
underneath the specified directory. For example, in the following configuration

pr oxy_ temp _pat h / spool / nginx / proxy_temp 1 2;

a temporary file might look like this:

/ spool / nginx / proxy_temp /7 /45 /00000123457

See also the use_temp_path parameter of the proxy cache path directive.

2.30.4 Embedded Variables

The ngx_http_proxy_module module supports embedded variables that
can be used to compose headers using the proxy set header directive:

$proxy host
name and port of a proxied server as specified in the proxy pass directive;

$proxy port
port of a proxied server as specified in the proxy pass directive, or the
protocols default port;
$proxy add x forwarded for
the X-Forwarded-For client request header field with the $remote addr
variable appended to it, separated by a comma. If the
X-Forwarded-For field is not present in the client request header, the
$proxy add x forwarded for variable is equal to the $remote addr vari-
able.

Nginx, Inc. p.166 of 330

CHAPTER 2. HTTP SERVER MODULES 2.31. MODULE NGX HTTP RANDOM INDEX MODULE

2.31 Module ngx http random index module

2.31.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 167
2.31.2 Example Configuration . . . . . . . . . . . . . . . . . . . 167
2.31.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 167
random index . . . . . . . . . . . . . . . . . . . . . . . . 167

2.31.1 Summary
The ngx_http_random_index_module module processes requests ending
with the slash character (/) and picks a random file in a directory to serve
as an index file. The module is processed before the ngx http index module
module.
This module is not built by default, it should be enabled with the
--with-http_random_index_module configuration parameter.

2.31.2 Example Configuration

location / {
random_index on ;
}

2.31.3 Directives
random index
Syntax: random_index on | off;
Default off
Context: location

Enables or disables module processing in a surrounding location.

Nginx, Inc. p.167 of 330

CHAPTER 2. HTTP SERVER MODULES 2.32. MODULE NGX HTTP REALIP MODULE

2.32 Module ngx http realip module

2.32.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 168
2.32.2 Example Configuration . . . . . . . . . . . . . . . . . . . 168
2.32.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 168
set real ip from . . . . . . . . . . . . . . . . . . . . . . . 168
real ip header . . . . . . . . . . . . . . . . . . . . . . . . 168
real ip recursive . . . . . . . . . . . . . . . . . . . . . . . 169

2.32.1 Summary
The ngx_http_realip_module module is used to change the client address
to the one sent in the specified header field.
This module is not built by default, it should be enabled with the
--with-http_realip_module configuration parameter.

2.32.2 Example Configuration

s et _ re a l_ i p_ f ro m 1 92. 168. 1.0/ 24;

s et _ re a l_ i p_ f ro m 192.168.2.1;
s et _ re a l_ i p_ f ro m 2001:0 db8 ::/32;
real_ip_header X - Forwarded - For ;
real_ip_recursive on ;

2.32.3 Directives
set real ip from
Syntax: set_real_ip_from address | CIDR | unix:;
Default
Context: http, server, location

Defines trusted addresses that are known to send correct replacement

addresses. If the special value unix: is specified, all UNIX-domain sockets
will be trusted.

IPv6 addresses are supported starting from versions 1.3.0 and 1.2.1.

real ip header
Syntax: real_ip_header field | X-Real-IP | X-Forwarded-For |
proxy_protocol;
Default X-Real-IP
Context: http, server, location

Defines a request header field used to send the address for a replacement.
The proxy_protocol parameter (1.5.12) changes the client address to
the one from the PROXY protocol header. The PROXY protocol must be

Nginx, Inc. p.168 of 330

CHAPTER 2. HTTP SERVER MODULES 2.32. MODULE NGX HTTP REALIP MODULE

previously enabled by setting the proxy_protocol parameter in the listen

directive.

real ip recursive
Syntax: real_ip_recursive on | off;
Default off
Context: http, server, location
This directive appeared in versions 1.3.0 and 1.2.1.

If recursive search is disabled, the original client address that matches one of
the trusted addresses is replaced by the last address sent in the request header
field defined by the real ip header directive. If recursive search is enabled, the
original client address that matches one of the trusted addresses is replaced by
the last non-trusted address sent in the request header field.

Nginx, Inc. p.169 of 330

CHAPTER 2. HTTP SERVER MODULES 2.33. MODULE NGX HTTP REFERER MODULE

2.33 Module ngx http referer module

2.33.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 170
2.33.2 Example Configuration . . . . . . . . . . . . . . . . . . . 170
2.33.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 170
referer hash bucket size . . . . . . . . . . . . . . . . . . 170
referer hash max size . . . . . . . . . . . . . . . . . . . . 170
valid referers . . . . . . . . . . . . . . . . . . . . . . . . 171
2.33.4 Embedded Variables . . . . . . . . . . . . . . . . . . . . 171

2.33.1 Summary
The ngx_http_referer_module module is used to block access to a site
for requests with invalid values in the Referer header field. It should be kept
in mind that fabricating a request with an appropriate Referer field value is
quite easy, and so the intended purpose of this module is not to block such
requests thoroughly but to block the mass flow of requests sent by regular
browsers. It should also be taken into consideration that regular browsers may
not send the Referer field even for valid requests.

2.33.2 Example Configuration

valid_referers none blocked server_names

*. example . com example .* www . example . org / galleries /
~\. google \.;

if ( $ in v al i d_ r ef e re r ) {
return 403;
}

2.33.3 Directives
referer hash bucket size
Syntax: referer_hash_bucket_size size;
Default 64
Context: server, location
This directive appeared in version 1.0.5.

Sets the bucket size for the valid referers hash tables. The details of setting
up hash tables are provided in a separate document.

referer hash max size

Syntax: referer_hash_max_size size;
Default 2048
Context: server, location
This directive appeared in version 1.0.5.

Nginx, Inc. p.170 of 330

CHAPTER 2. HTTP SERVER MODULES 2.33. MODULE NGX HTTP REFERER MODULE

Sets the maximum size of the valid referers hash tables. The details of
setting up hash tables are provided in a separate document.

valid referers
Syntax: valid_referers none | blocked | server_names | string . . . ;
Default
Context: server, location

Specifies the Referer request header field values that will cause the
embedded $invalid referer variable to be set to an empty string. Otherwise,
the variable will be set to 1. Search for a match is case-insensitive.
Parameters can be as follows:

none
the Referer field is missing in the request header;
blocked
the Referer field is present in the request header, but its value has been
deleted by a firewall or proxy server; such values are strings that do not
start with http:// or https://;
server_names
the Referer request header field contains one of the server names;
arbitrary string
defines a server name and an optional URI prefix. A server name can
have an * at the beginning or end. During the checking, the servers
port in the Referer field is ignored;
regular expression
the first symbol should be a~. It should be noted that an expression will
be matched against the text starting after the http:// or https://.

Example:

valid_referers none blocked server_names

*. example . com example .* www . example . org / galleries /
~\. google \.;

2.33.4 Embedded Variables

$invalid referer
Empty string, if the Referer request header field value is considered
valid, otherwise 1.

Nginx, Inc. p.171 of 330

CHAPTER 2. HTTP SERVER MODULES 2.34. MODULE NGX HTTP REWRITE MODULE

2.34 Module ngx http rewrite module

2.34.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 172
2.34.2 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 172
break . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172
if . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
return . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
rewrite . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174
rewrite log . . . . . . . . . . . . . . . . . . . . . . . . . . 175
set . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
uninitialized variable warn . . . . . . . . . . . . . . . . . 176
2.34.3 Internal Implementation . . . . . . . . . . . . . . . . . . 176

2.34.1 Summary
The ngx_http_rewrite_module module is used to change request
URI using regular expressions, return redirects, and conditionally select
configurations.
The ngx_http_rewrite_module module directives are processed in the
following order:
the directives of this module specified on the server level are executed
sequentially;
repeatedly:
a location is searched based on a request URI;
the directives of this module specified inside the found location are
executed sequentially;
the loop is repeated if a request URI was rewritten, but not more
than 10 times.

2.34.2 Directives
break
Syntax: break;
Default
Context: server, location, if

Stops processing the current set of ngx_http_rewrite_module directives.

If a directive is specified inside the location, further processing of the
request continues in this location.
Example:

if ( $slow ) {
limit_rate 10 k ;
break ;
}

Nginx, Inc. p.172 of 330

CHAPTER 2. HTTP SERVER MODULES 2.34. MODULE NGX HTTP REWRITE MODULE

if
Syntax: if (condition) { . . . }
Default
Context: server, location

The specified condition is evaluated. If true, this module directives specified

inside the braces are executed, and the request is assigned the configuration
inside the if directive. Configurations inside the if directives are inherited
from the previous configuration level.
A condition may be any of the following:
a variable name; false if the value of a variable is an empty string or 0;

Before version 1.0.1, any string starting with 0 was considered a false
value.

comparison of a variable with a string using the = and != operators;

matching of a variable against a regular expression using the ~ (for case-

sensitive matching) and ~* (for case-insensitive matching) operators.
Regular expressions can contain captures that are made available for
later reuse in the $1..$9 variables. Negative operators !~ and !~* are
also available. If a regular expression includes the } or ; characters,
the whole expressions should be enclosed in single or double quotes.

checking of a file existence with the -f and !-f operators;

checking of a directory existence with the -d and !-d operators;

checking of a file, directory, or symbolic link existence with the -e and

!-e operators;

checking for an executable file with the -x and !-x operators.

Examples:

if ( $ ht t p_ u se r _a g en t ~ MSIE ) {
rewrite ^(.*) $ / msie / $1 break ;
}

if ( $http_cookie ~* " id =([^;]+) (?:;| $ ) ") {

set $id $1 ;
}

if ( $req uest _me thod = POST ) {

return 405;
}

if ( $slow ) {
limit_rate 10 k ;
}

if ( $ in v al i d_ r ef e re r ) {
return 403;
}

Nginx, Inc. p.173 of 330

CHAPTER 2. HTTP SERVER MODULES 2.34. MODULE NGX HTTP REWRITE MODULE

A value of the $invalid referer embedded variable is set by the valid -

referers directive.

return
Syntax: return code [text];
Syntax: return code URL;
Syntax: return URL;
Default
Context: server, location, if

Stops processing and returns the specified code to a client. The non-
standard code 444 closes a connection without sending a response header.
Starting from version 0.8.42, it is possible to specify either a redirect URL
(for codes 301, 302, 303, and 307), or the response body text (for other codes).
A response body text and redirect URL can contain variables. As a special
case, a redirect URL can be specified as a URI local to this server, in which
case the full redirect URL is formed according to the request scheme ($scheme)
and the server name in redirect and port in redirect directives.
In addition, a URL for temporary redirect with the code 302 can be specified
as the sole parameter. Such a parameter should start with the http://,
https://, or $scheme string. A URL can contain variables.

Only the following codes could be returned before version 0.7.51: 204,
400, 402 406, 408, 410, 411, 413, 416, and 500 504.

The code 307 was not treated as a redirect until versions 1.1.16 and 1.0.13.

See also the error page directive.

rewrite
Syntax: rewrite regex replacement [flag];
Default
Context: server, location, if

If the specified regular expression matches a request URI, URI is changed

as specified in the replacement string. The rewrite directives are executed
sequentially in order of their appearance in the configuration file. It is possible
to terminate further processing of the directives using flags. If a replacement
string starts with http:// or https://, the processing stops and the
redirect is returned to a client.
An optional flag parameter can be one of:

last
stops processing the current set of ngx_http_rewrite_module directives
and starts a search for a new location matching the changed URI;

Nginx, Inc. p.174 of 330

CHAPTER 2. HTTP SERVER MODULES 2.34. MODULE NGX HTTP REWRITE MODULE

break
stops processing the current set of ngx_http_rewrite_module directives
as with the break directive;
redirect
returns a temporary redirect with the 302 code; used if a replacement
string does not start with http:// or https://;
permanent
returns a permanent redirect with the 301 code.

The full redirect URL is formed according to the request scheme ($scheme)
and the server name in redirect and port in redirect directives.
Example:

server {
...
rewrite ^(/ download /.*) / media /(.*) \..* $ $1 / mp3 / $2 . mp3 last ;
rewrite ^(/ download /.*) / audio /(.*) \..* $ $1 / mp3 / $2 . ra last ;
return 403;
...
}

But if these directives are put inside the /download/ location, the last
flag should be replaced by break, or otherwise nginx will make 10 cycles and
return the 500 error:

location / download / {
rewrite ^(/ download /.*) / media /(.*) \..* $ $1 / mp3 / $2 . mp3 break ;
rewrite ^(/ download /.*) / audio /(.*) \..* $ $1 / mp3 / $2 . ra break ;
return 403;
}

If a replacement string includes the new request arguments, the previous

request arguments are appended after them. If this is undesired, putting a
question mark at the end of a replacement string avoids having them appended,
for example:

rewrite ^/ users /(.*) $ / show ? user = $1 ? last ;

If a regular expression includes the } or ; characters, the whole

expressions should be enclosed in single or double quotes.

rewrite log
Syntax: rewrite_log on | off;
Default off
Context: http, server, location, if

Enables or disables logging of ngx_http_rewrite_module module

directives processing results into the error log at the notice level.

Nginx, Inc. p.175 of 330

CHAPTER 2. HTTP SERVER MODULES 2.34. MODULE NGX HTTP REWRITE MODULE

set
Syntax: set $variable value;
Default
Context: server, location, if

Sets a value for the specified variable. The value can contain text, variables,
and their combination.

uninitialized variable warn

Syntax: uninitialized_variable_warn on | off;
Default on
Context: http, server, location, if

Controls whether warnings about uninitialized variables are logged.

2.34.3 Internal Implementation

The ngx_http_rewrite_module module directives are compiled at the
configuration stage into internal instructions that are interpreted during
request processing. An interpreter is a simple virtual stack machine.
For example, the directives

location / download / {
if ( $forbidden ) {
return 403;
}

if ( $slow ) {
limit_rate 10 k ;
}

rewrite ^/( download /.*) / media /(.*) \..* $ / $1 / mp3 / $2 . mp3 break ;
}

will be translated into these instructions:

variable $forbidden
check against zero
return 403
end of code
variable $slow
check against zero
match of regular expression
copy "/"
copy $1
copy "/ mp3 /"
copy $2
copy ". mp3 "
end of regular expression
end of code

Note that there are no instructions for the limit rate directive above as it is
unrelated to the ngx_http_rewrite_module module. A separate configuration
is created for the if block. If the condition holds true, a request is assigned
this configuration where limit_rate equals to 10k.

Nginx, Inc. p.176 of 330

CHAPTER 2. HTTP SERVER MODULES 2.34. MODULE NGX HTTP REWRITE MODULE

The directive

rewrite ^/( download /.*) / media /(.*) \..* $ / $1 / mp3 / $2 . mp3 break ;

can be made smaller by one instruction if the first slash in the regular
expression is put inside the parentheses:

rewrite ^(/ download /.*) / media /(.*) \..* $ $1 / mp3 / $2 . mp3 break ;

The corresponding instructions will then look like this:

match of regular expression

copy $1
copy "/ mp3 /"
copy $2
copy ". mp3 "
end of regular expression
end of code

Nginx, Inc. p.177 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

2.35 Module ngx http scgi module

2.35.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 179
2.35.2 Example Configuration . . . . . . . . . . . . . . . . . . . 179
2.35.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 179
scgi bind . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
scgi buffer size . . . . . . . . . . . . . . . . . . . . . . . 179
scgi buffering . . . . . . . . . . . . . . . . . . . . . . . . 179
scgi buffers . . . . . . . . . . . . . . . . . . . . . . . . . 180
scgi busy buffers size . . . . . . . . . . . . . . . . . . . . 180
scgi cache . . . . . . . . . . . . . . . . . . . . . . . . . . 180
scgi cache bypass . . . . . . . . . . . . . . . . . . . . . . 180
scgi cache key . . . . . . . . . . . . . . . . . . . . . . . . 181
scgi cache lock . . . . . . . . . . . . . . . . . . . . . . . 181
scgi cache lock age . . . . . . . . . . . . . . . . . . . . . 181
scgi cache lock timeout . . . . . . . . . . . . . . . . . . 181
scgi cache methods . . . . . . . . . . . . . . . . . . . . . 182
scgi cache min uses . . . . . . . . . . . . . . . . . . . . . 182
scgi cache path . . . . . . . . . . . . . . . . . . . . . . . 182
scgi cache purge . . . . . . . . . . . . . . . . . . . . . . 183
scgi cache revalidate . . . . . . . . . . . . . . . . . . . . 184
scgi cache use stale . . . . . . . . . . . . . . . . . . . . . 184
scgi cache valid . . . . . . . . . . . . . . . . . . . . . . . 184
scgi connect timeout . . . . . . . . . . . . . . . . . . . . 185
scgi force ranges . . . . . . . . . . . . . . . . . . . . . . 185
scgi hide header . . . . . . . . . . . . . . . . . . . . . . . 185
scgi ignore client abort . . . . . . . . . . . . . . . . . . . 186
scgi ignore headers . . . . . . . . . . . . . . . . . . . . . 186
scgi intercept errors . . . . . . . . . . . . . . . . . . . . 186
scgi limit rate . . . . . . . . . . . . . . . . . . . . . . . . 187
scgi max temp file size . . . . . . . . . . . . . . . . . . . 187
scgi next upstream . . . . . . . . . . . . . . . . . . . . . 187
scgi next upstream timeout . . . . . . . . . . . . . . . . 188
scgi next upstream tries . . . . . . . . . . . . . . . . . . 188
scgi no cache . . . . . . . . . . . . . . . . . . . . . . . . 188
scgi param . . . . . . . . . . . . . . . . . . . . . . . . . . 189
scgi pass . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
scgi pass header . . . . . . . . . . . . . . . . . . . . . . . 190
scgi read timeout . . . . . . . . . . . . . . . . . . . . . . 190
scgi pass request body . . . . . . . . . . . . . . . . . . . 190
scgi pass request headers . . . . . . . . . . . . . . . . . . 190
scgi request buffering . . . . . . . . . . . . . . . . . . . . 190
scgi send timeout . . . . . . . . . . . . . . . . . . . . . . 191
scgi store . . . . . . . . . . . . . . . . . . . . . . . . . . 191
scgi store access . . . . . . . . . . . . . . . . . . . . . . . 192
scgi temp file write size . . . . . . . . . . . . . . . . . . 192

Nginx, Inc. p.178 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

scgi temp path . . . . . . . . . . . . . . . . . . . . . . . 192

2.35.1 Summary
The ngx_http_scgi_module module allows passing requests to an SCGI
server.

2.35.2 Example Configuration

location / {
include scgi_params ;
scgi_pass localhost :9000;
}

2.35.3 Directives
scgi bind
Syntax: scgi_bind address | off;
Default
Context: http, server, location

Makes outgoing connections to an SCGI server originate from the specified

local IP address. Parameter value can contain variables (1.3.12). The special
value off (1.3.12) cancels the effect of the scgi_bind directive inherited from
the previous configuration level, which allows the system to auto-assign the
local IP address.

scgi buffer size

Syntax: scgi_buffer_size size;
Default 4k|8k
Context: http, server, location

Sets the size of the buffer used for reading the first part of the response
received from the SCGI server. This part usually contains a small response
header. By default, the buffer size is equal to the size of one buffer set by the
scgi buffers directive. It can be made smaller, however.

scgi buffering
Syntax: scgi_buffering on | off;
Default on
Context: http, server, location

Enables or disables buffering of responses from the SCGI server.

When buffering is enabled, nginx receives a response from the SCGI server
as soon as possible, saving it into the buffers set by the scgi buffer size and
scgi buffers directives. If the whole response does not fit into memory, a part

Nginx, Inc. p.179 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

of it can be saved to a temporary file on the disk. Writing to temporary

files is controlled by the scgi max temp file size and scgi temp file write size
directives.
When buffering is disabled, the response is passed to a client synchronously,
immediately as it is received. nginx will not try to read the whole response
from the SCGI server. The maximum size of the data that nginx can receive
from the server at a time is set by the scgi buffer size directive.
Buffering can also be enabled or disabled by passing yes or no in the
X-Accel-Buffering response header field. This capability can be disabled
using the scgi ignore headers directive.

scgi buffers
Syntax: scgi_buffers number size;
Default 8 4k|8k
Context: http, server, location

Sets the number and size of the buffers used for reading a response from
the SCGI server, for a single connection. By default, the buffer size is equal to
one memory page. This is either 4K or 8K, depending on a platform.

scgi busy buffers size

Syntax: scgi_busy_buffers_size size;
Default 8k|16k
Context: http, server, location

When buffering of responses from the SCGI server is enabled, limits the
total size of buffers that can be busy sending a response to the client while the
response is not yet fully read. In the meantime, the rest of the buffers can be
used for reading the response and, if needed, buffering part of the response to
a temporary file. By default, size is limited by the size of two buffers set by
the scgi buffer size and scgi buffers directives.

scgi cache
Syntax: scgi_cache zone | off;
Default off
Context: http, server, location

Defines a shared memory zone used for caching. The same zone can be
used in several places. Parameter value can contain variables (1.7.9). The off
parameter disables caching inherited from the previous configuration level.

scgi cache bypass

Syntax: scgi_cache_bypass string . . . ;
Default
Context: http, server, location

Nginx, Inc. p.180 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

Defines conditions under which the response will not be taken from a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be taken from the cache:

s c g i _ c a c h e _ b y p a s s $co okie _noc ach e $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

s c g i _ c a c h e _ b y p a s s $http_pragma $http_authorization ;

Can be used along with the scgi no cache directive.

scgi cache key

Syntax: scgi_cache_key string;
Default
Context: http, server, location

Defines a key for caching, for example

scgi_cache_key localhost :9000 $request_uri ;

scgi cache lock

Syntax: scgi_cache_lock on | off;
Default off
Context: http, server, location
This directive appeared in version 1.1.12.

When enabled, only one request at a time will be allowed to populate a new
cache element identified according to the scgi cache key directive by passing a
request to an SCGI server. Other requests of the same cache element will either
wait for a response to appear in the cache or the cache lock for this element
to be released, up to the time set by the scgi cache lock timeout directive.

scgi cache lock age

Syntax: scgi_cache_lock_age time;
Default 5s
Context: http, server, location
This directive appeared in version 1.7.8.

If the last request passed to the SCGI server for populating a new cache
element has not completed for the specified time, one more request may be
passed to the SCGI server.

scgi cache lock timeout

Syntax: scgi_cache_lock_timeout time;
Default 5s
Context: http, server, location
This directive appeared in version 1.1.12.

Nginx, Inc. p.181 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

Sets a timeout for scgi cache lock. When the time expires, the request will
be passed to the SCGI server, however, the response will not be cached.

Before 1.7.8, the response could be cached.

scgi cache methods

Syntax: scgi_cache_methods GET | HEAD | POST . . . ;
Default GET HEAD
Context: http, server, location

If the client request method is listed in this directive then the response will
be cached. GET and HEAD methods are always added to the list, though it is
recommended to specify them explicitly. See also the scgi no cache directive.

scgi cache min uses

Syntax: scgi_cache_min_uses number;
Default 1
Context: http, server, location

Sets the number of requests after which the response will be cached.

scgi cache path

Syntax: scgi_cache_path path [levels=levels] [use_temp_path=on|off]
keys_zone=name:size [inactive=time] [max_size=size]
[loader_files=number] [loader_sleep=time]
[loader_threshold=time];
Default
Context: http

Sets the path and other parameters of a cache. Cache data are stored in
files. The file name in a cache is a result of applying the MD5 function to
the cache key. The levels parameter defines hierarchy levels of a cache. For
example, in the following configuration

sc gi_c ache _pat h / data / nginx / cache levels =1:2 keys_zone = one :10 m ;

file names in a cache will look like this:

/ data / nginx / cache /c /29 / b 7 f 5 4 b 2 d f 7 7 7 3 7 2 2 d 3 8 2 f 4 8 0 9 d 6 5 0 29c

A cached response is first written to a temporary file, and then the file is
renamed. Starting from version 0.8.9, temporary files and the cache can be put
on different file systems. However, be aware that in this case a file is copied
across two file systems instead of the cheap renaming operation. It is thus
recommended that for any given location both cache and a directory holding
temporary files are put on the same file system. A directory for temporary

Nginx, Inc. p.182 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

files is set based on the use_temp_path parameter (1.7.10). If this parameter is

omitted or set to the value on, the directory set by the scgi temp path directive
for the given location will be used. If the value is set to off, temporary files
will be put directly in the cache directory.
In addition, all active keys and information about data are stored in a
shared memory zone, whose name and size are configured by the keys_zone
parameter. One megabyte zone can store about 8 thousand keys.
Cached data that are not accessed during the time specified by the
inactive parameter get removed from the cache regardless of their freshness.
By default, inactive is set to 10 minutes.
The special cache manager process monitors the maximum cache size set
by the max_size parameter. When this size is exceeded, it removes the least
recently used data.
A minute after the start the special cache loader process is activated. It
loads information about previously cached data stored on file system into a
cache zone. The loading is done in iterations. During one iteration no more
than loader_files items are loaded (by default, 100). Besides, the duration of
one iteration is limited by the loader_threshold parameter (by default, 200
milliseconds). Between iterations, a pause configured by the loader_sleep
parameter (by default, 50 milliseconds) is made.

scgi cache purge

Syntax: scgi_cache_purgestring . . . ;
Default
Context: http, server, location
This directive appeared in version 1.5.7.

Defines conditions under which the request will be considered a cache purge
request. If at least one value of the string parameters is not empty and
is not equal to 0 then the cache entry with a corresponding cache key is
removed. The result of successful operation is indicated by returning the 204
No Content response.
If the cache key of a purge request ends with an asterisk (*), all cache
entries matching the wildcard key will be removed from the cache.
Example configuration:

sc gi_c ache _pat h / data / nginx / cache keys_zone = cache_zone :10 m ;

map $re ques t_m etho d $purge_method {

PURGE 1;
default 0;
}

server {
...
location / {
scgi_pass backend ;
scgi_cache cache_zone ;
scgi_cache_key $uri ;
s cg i _c a ch e _p ur g e $purge_method ;
}
}

Nginx, Inc. p.183 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

This functionality is available as part of our commercial subscription.

scgi cache revalidate

Syntax: scgi_cache_revalidate on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.7.

Enables revalidation of expired cache items using conditional requests with

the If-Modified-Since and If-None-Match header fields.

scgi cache use stale

Syntax: scgi_cache_use_stale error | timeout | invalid_header | updating
| http_500 | http_503 | http_403 | http_404 | off . . . ;
Default off
Context: http, server, location

Determines in which cases a stale cached response can be used when an

error occurs during communication with the SCGI server. The directives
parameters match the parameters of the scgi next upstream directive.
Additionally, the updating parameter permits using a stale cached response
if it is currently being updated. This allows minimizing the number of accesses
to SCGI servers when updating cached data.
To minimize the number of accesses to SCGI servers when populating a
new cache element, the scgi cache lock directive can be used.

scgi cache valid

Syntax: scgi_cache_valid [code . . . ] time;
Default
Context: http, server, location

Sets caching time for different response codes. For example, the following
directives

s cg i _c a ch e _v a li d 200 302 10 m ;
s cg i _c a ch e _v a li d 404 1m;

set 10 minutes of caching for responses with codes 200 and 302 and 1 minute
for responses with code 404.
If only caching time is specified

s cg i _c a ch e _v a li d 5 m ;

then only 200, 301, and 302 responses are cached.

In addition, the any parameter can be specified to cache any responses:

Nginx, Inc. p.184 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

s cg i _c a ch e _v a li d 200 302 10 m ;
s cg i _c a ch e _v a li d 301 1h;
s cg i _c a ch e _v a li d any 1m;

Parameters of caching can also be set directly in the response header. This
has higher priority than setting of caching time using the directive.

The X-Accel-Expires header field sets caching time of a response in

seconds. The zero value disables caching for a response. If the value
starts with the @ prefix, it sets an absolute time in seconds since Epoch,
up to which the response may be cached.

If the header does not include the X-Accel-Expires field, parameters of

caching may be set in the header fields Expires or Cache-Control.

If the header includes the Set-Cookie field, such a response will not be
cached.

If the header includes the Vary field with the special value *, such a
response will not be cached (1.7.7). If the header includes the Vary field
with another value, such a response will be cached taking into account
the corresponding request header fields (1.7.7).

Processing of one or more of these response header fields can be disabled using
the scgi ignore headers directive.

scgi connect timeout

Syntax: scgi_connect_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for establishing a connection with an SCGI server. It

should be noted that this timeout cannot usually exceed 75 seconds.

scgi force ranges

Syntax: scgi_force_ranges on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.7.

Enables byte-range support for both cached and uncached responses from
the SCGI server regardless of the Accept-Ranges field in these responses.

scgi hide header

Syntax: scgi_hide_header field;
Default
Context: http, server, location

Nginx, Inc. p.185 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

By default, nginx does not pass the header fields Status and X-Accel-...
from the response of an SCGI server to a client. The scgi_hide_header
directive sets additional fields that will not be passed. If, on the contrary, the
passing of fields needs to be permitted, the scgi pass header directive can be
used.

scgi ignore client abort

Syntax: scgi_ignore_client_abort on | off;
Default off
Context: http, server, location

Determines whether the connection with an SCGI server should be closed

when a client closes the connection without waiting for a response.

scgi ignore headers

Syntax: scgi_ignore_headers field . . . ;
Default
Context: http, server, location

Disables processing of certain response header fields from the SCGI server.
The following fields can be ignored: X-Accel-Redirect, X-Accel-Expires,
X-Accel-Limit-Rate (1.1.6), X-Accel-Buffering (1.1.6), X-Accel-Charset
(1.1.6), Expires, Cache-Control, Set-Cookie (0.8.44), and Vary (1.7.7).
If not disabled, processing of these header fields has the following effect:

X-Accel-Expires, Expires, Cache-Control, Set-Cookie, and Vary set

the parameters of response caching;

X-Accel-Redirect performs an internal redirect to the specified URI;

X-Accel-Limit-Rate sets the rate limit for transmission of a response

to a client;

X-Accel-Buffering enables or disables buffering of a response;

X-Accel-Charset sets the desired charset of a response.

scgi intercept errors

Syntax: scgi_intercept_errors on | off;
Default off
Context: http, server, location

Determines whether an SCGI server responses with codes greater than or

equal to 300 should be passed to a client or be redirected to nginx for processing
with the error page directive.

Nginx, Inc. p.186 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

scgi limit rate

Syntax: scgi_limit_rate rate;
Default 0
Context: http, server, location
This directive appeared in version 1.7.7.

Limits the speed of reading the response from the SCGI server. The rate is
specified in bytes per second. The zero value disables rate limiting. The limit
is set per a request, and so if nginx simultaneously opens two connections to
the SCGI server, the overall rate will be twice as much as the specified limit.
The limitation works only if buffering of responses from the SCGI server is
enabled.

scgi max temp file size

Syntax: scgi_max_temp_file_size size;
Default 1024m
Context: http, server, location

When buffering of responses from the SCGI server is enabled, and the
whole response does not fit into the buffers set by the scgi buffer size and
scgi buffers directives, a part of the response can be saved to a temporary file.
This directive sets the maximum size of the temporary file. The size of data
written to the temporary file at a time is set by the scgi temp file write size
directive.
The zero value disables buffering of responses to temporary files.

This restriction does not apply to responses that will be cached or stored
on disk.

scgi next upstream

Syntax: scgi_next_upstream error | timeout | invalid_header | http_500 |
http_503 | http_403 | http_404 | off . . . ;
Default error timeout
Context: http, server, location

Specifies in which cases a request should be passed to the next server:

error
an error occurred while establishing a connection with the server, passing
a request to it, or reading the response header;
timeout
a timeout has occurred while establishing a connection with the server,
passing a request to it, or reading the response header;
invalid_header
a server returned an empty or invalid response;

Nginx, Inc. p.187 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

http_500
a server returned a response with the code 500;
http_503
a server returned a response with the code 503;
http_403
a server returned a response with the code 403;
http_404
a server returned a response with the code 404;
off
disables passing a request to the next server.

One should bear in mind that passing a request to the next server is only
possible if nothing has been sent to a client yet. That is, if an error or timeout
occurs in the middle of the transferring of a response, fixing this is impossible.
The directive also defines what is considered an unsuccessful attempt
of communication with a server. The cases of error, timeout and
invalid_header are always considered unsuccessful attempts, even if they
are not specified in the directive. The cases of http_500 and http_503 are
considered unsuccessful attempts only if they are specified in the directive. The
cases of http_403 and http_404 are never considered unsuccessful attempts.
Passing a request to the next server can be limited by the number of tries
and by time.

scgi next upstream timeout

Syntax: scgi_next_upstream_timeout time;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the time allowed to pass a request to the next server. The 0 value
turns off this limitation.

scgi next upstream tries

Syntax: scgi_next_upstream_tries number;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the number of possible tries for passing a request to the next server.
The 0 value turns off this limitation.

scgi no cache
Syntax: scgi_no_cache string . . . ;
Default
Context: http, server, location

Nginx, Inc. p.188 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

Defines conditions under which the response will not be saved to a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be saved:

scgi_no_cache $co oki e_no cach e $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

scgi_no_cache $http_pragma $http_authorization ;

Can be used along with the scgi cache bypass directive.

scgi param
Syntax: scgi_param parameter value [if_not_empty];
Default
Context: http, server, location

Sets a parameter that should be passed to the SCGI server. The value can
contain text, variables, and their combination. These directives are inherited
from the previous level if and only if there are no scgi_param directives defined
on the current level.
Standard CGI environment variables should be provided as SCGI headers,
see the scgi_params file provided in the distribution:

location / {
include scgi_params ;
...
}

If a directive is specified with if_not_empty (1.1.11) then such a parameter

will not be passed to the server until its value is not empty:

scgi_param HTTPS $https if_not_empty ;

scgi pass
Syntax: scgi_pass address;
Default
Context: location, if in location

Sets the address of an SCGI server. The address can be specified as a

domain name or IP address, and an optional port:

scgi_pass localhost :9000;

or as a UNIX-domain socket path:

scgi_pass unix :/ tmp / scgi . socket ;

If a domain name resolves to several addresses, all of them will be used

in a round-robin fashion. In addition, an address can be specified as a server
group.

Nginx, Inc. p.189 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

scgi pass header

Syntax: scgi_pass_header field;
Default
Context: http, server, location

Permits passing otherwise disabled header fields from an SCGI server to a

client.

scgi read timeout

Syntax: scgi_read_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for reading a response from the SCGI server. The timeout
is set only between two successive read operations, not for the transmission of
the whole response. If the SCGI server does not transmit anything within this
time, the connection is closed.

scgi pass request body

Syntax: scgi_pass_request_body on | off;
Default on
Context: http, server, location

Indicates whether the original request body is passed to the SCGI server.
See also the scgi pass request headers directive.

scgi pass request headers

Syntax: scgi_pass_request_headers on | off;
Default on
Context: http, server, location

Indicates whether the header fields of the original request are passed to the
SCGI server. See also the scgi pass request body directive.

scgi request buffering

Syntax: scgi_request_buffering on | off;
Default on
Context: http, server, location
This directive appeared in version 1.7.11.

Enables or disables buffering of a client request body.

When buffering is enabled, the entire request body is read from the client
before sending the request to an SCGI server.
When buffering is disabled, the request body is sent to the SCGI server
immediately as it is received. In this case, the request cannot be passed to the
next server if nginx already started sending the request body.

Nginx, Inc. p.190 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

When HTTP/1.1 chunked transfer encoding is used to send the original

request body, the request body will be buffered regardless of the directive
value.

scgi send timeout

Syntax: scgi_send_timeout time;
Default 60s
Context: http, server, location

Sets a timeout for transmitting a request to the SCGI server. The timeout
is set only between two successive write operations, not for the transmission
of the whole request. If the SCGI server does not receive anything within this
time, the connection is closed.

scgi store
Syntax: scgi_store on | off | string;
Default off
Context: http, server, location

Enables saving of files to a disk. The on parameter saves files with paths
corresponding to the directives alias or root. The off parameter disables saving
of files. In addition, the file name can be set explicitly using the string with
variables:

scgi_store / data / w ww $ or i gi n al _ ur i ;

The modification time of files is set according to the received

Last-Modified response header field. The response is first written to a
temporary file, and then the file is renamed. Starting from version 0.8.9,
temporary files and the persistent store can be put on different file systems.
However, be aware that in this case a file is copied across two file systems
instead of the cheap renaming operation. It is thus recommended that for any
given location both saved files and a directory holding temporary files, set by
the scgi temp path directive, are put on the same file system.
This directive can be used to create local copies of static unchangeable files,
e.g.:

location / images / {
root / data / www ;
error_page 404 = / fetch$uri ;
}

location / fetch / {
internal ;

scgi_pass backend :9000;

...

scgi_store on ;
s c g i _ s t o r e _ a c c e s s user : rw group : rw all : r ;
scgi_temp_path / data / temp ;

Nginx, Inc. p.191 of 330

CHAPTER 2. HTTP SERVER MODULES 2.35. MODULE NGX HTTP SCGI MODULE

alias / data / www /;

}

scgi store access

Syntax: scgi_store_access users:permissions . . . ;
Default user:rw
Context: http, server, location

Sets access permissions for newly created files and directories, e.g.:

s c g i _ s t o r e _ a c c e s s user : rw group : rw all : r ;

If any group or all access permissions are specified then user permissions
may be omitted:

s c g i _ s t o r e _ a c c e s s group : rw all : r ;

scgi temp file write size

Syntax: scgi_temp_file_write_size size;
Default 8k|16k
Context: http, server, location

Limits the size of data written to a temporary file at a time, when buffering
of responses from the SCGI server to temporary files is enabled. By default, size
is limited by two buffers set by the scgi buffer size and scgi buffers directives.
The maximum size of a temporary file is set by the scgi max temp file size
directive.

scgi temp path

Syntax: scgi_temp_path path [level1 [level2 [level3]]];
Default scgi_temp
Context: http, server, location

Defines a directory for storing temporary files with data received from SCGI
servers. Up to three-level subdirectory hierarchy can be used underneath the
specified directory. For example, in the following configuration

scgi_temp_path / spool / nginx / scgi_temp 1 2;

a temporary file might look like this:

/ spool / nginx / scgi_temp /7 /45 /00000123457

See also the use_temp_path parameter of the scgi cache path directive.

Nginx, Inc. p.192 of 330

CHAPTER 2. HTTP SERVER MODULES 2.36. MODULE NGX HTTP SECURE LINK MODULE

2.36 Module ngx http secure link module

2.36.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 193
2.36.2 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 193
secure link . . . . . . . . . . . . . . . . . . . . . . . . . . 193
secure link md5 . . . . . . . . . . . . . . . . . . . . . . . 194
secure link secret . . . . . . . . . . . . . . . . . . . . . . 194
2.36.3 Embedded Variables . . . . . . . . . . . . . . . . . . . . 195

2.36.1 Summary
The ngx_http_secure_link_module module (0.7.18) is used to check
authenticity of requested links, protect resources from unauthorized access,
and limit link lifetime.
The authenticity of a requested link is verified by comparing the checksum
value passed in a request with the value computed for the request. If a link has
a limited lifetime and the time has expired, the link is considered outdated.
The status of these checks is made available in the $secure link variable.
The module provides two alternative operation modes. The first mode is
enabled by the secure link secret directive and is used to check authenticity
of requested links as well as protect resources from unauthorized access.
The second mode (0.8.50) is enabled by the secure link and secure link md5
directives and is also used to limit lifetime of links.
This module is not built by default, it should be enabled with the
--with-http_secure_link_module configuration parameter.

2.36.2 Directives
secure link
Syntax: secure_link expression;
Default
Context: http, server, location

Defines a string with variables from which the checksum value and lifetime
of a link will be extracted.
Variables used in an expression are usually associated with a request; see
example below.
The checksum value extracted from the string is compared with the MD5
hash value of the expression defined by the secure link md5 directive. If the
checksums are different, the $secure link variable is set to an empty string.
If the checksums are the same, the link lifetime is checked. If the link has a
limited lifetime and the time has expired, the $secure link variable is set to
0. Otherwise, it is set to 1. The MD5 hash value passed in a request is
encoded in base64url.
If a link has a limited lifetime, the expiration time is set in seconds
since Epoch (Thu, 01 Jan 1970 00:00:00 GMT). The value is specified in the
expression after the MD5 hash, and is separated by a comma. The expiration

Nginx, Inc. p.193 of 330

CHAPTER 2. HTTP SERVER MODULES 2.36. MODULE NGX HTTP SECURE LINK MODULE

time passed in a request is available through the $secure link expires variable
for a use in the secure link md5 directive. If the expiration time is not specified,
a link has the unlimited lifetime.

secure link md5

Syntax: secure_link_md5 expression;
Default
Context: http, server, location

Defines an expression for which the MD5 hash value will be computed and
compared with the value passed in a request.
The expression should contain the secured part of a link (resource) and a
secret ingredient. If the link has a limited lifetime, the expression should also
contain $secure link expires.
To prevent unauthorized access, the expression may contain some
information about the client, such as its address and browser version.
Example:

location / s / {
secure_link $arg_md5 , $arg_expires ;
se cure _lin k_md 5 " $ s e c u r e _ l i n k _ e x p i r e s $ u r i $ r e m o t e _ a d d r secret ";

if ( $secure_link = "") {
return 403;
}

if ( $secure_link = "0") {
return 410;
}

...
}

The /s/link?md5=_e4Nc3iduzkWRm01TBBNYw&expires=2147483647 link

restricts access to /s/link for the client with the IP address 127.0.0.1. The
link also has the limited lifetime until January 19, 2038 (GMT).
On UNIX, the md5 request argument value can be obtained as:

echo -n 2147483647/ s / link127 .0.0.1 secret | \

openssl md5 - binary | openssl base64 | tr +/ -_ | tr -d =

secure link secret

Syntax: secure_link_secret word;
Default
Context: location

Defines a secret word used to check authenticity of requested links.

The full URI of a requested link looks as follows:

/prefix/hash/link

Nginx, Inc. p.194 of 330

CHAPTER 2. HTTP SERVER MODULES 2.36. MODULE NGX HTTP SECURE LINK MODULE

where hash is a hexadecimal representation of the MD5 hash computed for

the concatenation of the link and secret word, and prefix is an arbitrary string
without slashes.
If the requested link passes the authenticity check, the $secure link variable
is set to the link extracted from the request URI. Otherwise, the $secure link
variable is set to an empty string.
Example:

location / p / {
s e c u r e _ l i n k _ s e c r e t secret ;

if ( $secure_link = "") {
return 403;
}

rewrite ^ / secure / $secure_link ;

}

location / secure / {
internal ;
}

A request of /p/5e814704a28d9bc1914ff19fa0c4a00a/link will be

internally redirected to /secure/link.
On UNIX, the hash value for this example can be obtained as:

echo -n linksecret | openssl md5 - hex

2.36.3 Embedded Variables

$secure link
The status of a link check. The specific value depends on the selected
operation mode.
$secure link expires
The lifetime of a link passed in a request; intended to be used only in
the secure link md5 directive.

Nginx, Inc. p.195 of 330

CHAPTER 2. HTTP SERVER MODULES 2.37. MODULE NGX HTTP SESSION LOG MODULE

2.37 Module ngx http session log module

2.37.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 196
2.37.2 Example Configuration . . . . . . . . . . . . . . . . . . . 196
2.37.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 196
session log format . . . . . . . . . . . . . . . . . . . . . 196
session log zone . . . . . . . . . . . . . . . . . . . . . . . 196
session log . . . . . . . . . . . . . . . . . . . . . . . . . . 197
2.37.4 Embedded Variables . . . . . . . . . . . . . . . . . . . . 197

2.37.1 Summary
The ngx_http_session_log_module module enables logging sessions (that
is, aggregates of multiple HTTP requests) instead of individual HTTP requests.

This module is available as part of our commercial subscription.

2.37.2 Example Configuration

The following configuration sets up a session log and maps requests to
sessions according to the request client address and User-Agent request header
field:

s es s io n _l o g_ z on e / path / to / log format = combined

zone = one :1 m timeout =30 s
md5 = $ b i n a r y _ r e m o t e _ a d d r $ h t t p _ u s e r _ a g e n t ;

location / media / {
session_log one ;
}

2.37.3 Directives
session log format
Syntax: session_log_format name string . . . ;
Default combined "..."
Context: http

Specifies the output format of a log. The value of the $body bytes sent
variable is aggregated across all requests in a session. The values of all other
variables available for logging correspond to the first request in a session.

session log zone

Syntax: session_log_zone path zone=name:size [format=format]
[timeout=time] [id=id] [md5=md5] ;
Default
Context: http

Nginx, Inc. p.196 of 330

CHAPTER 2. HTTP SERVER MODULES 2.37. MODULE NGX HTTP SESSION LOG MODULE

Sets the path to a log file and configures the shared memory zone that is
used to store currently active sessions.
A session is considered active for as long as the time elapsed since the last
request in the session does not exceed the specified timeout (by default, 30
seconds). Once a session is no longer active, it is written to the log.
The id parameter identifies the session to which a request is mapped. The
id parameter is set to the hexadecimal representation of an MD5 hash (for
example, obtained from a cookie using variables). If this parameter is not
specified or does not represent the valid MD5 hash, nginx computes the MD5
hash from the value of the md5 parameter and creates a new session using this
hash. Both the id and md5 parameters can contain variables.
The format parameter sets the custom session log format configured by
the session log format directive. If format is not specified, the predefined
combined format is used.

session log
Syntax: session_log name | off;
Default off
Context: http, server, location

Enables the use of the specified session log. The special value off cancels
all session_log directives inherited from the previous configuration level.

2.37.4 Embedded Variables

The ngx_http_session_log_module module supports two embedded
variables:

$session log id
current session ID;
$session log binary id
current session ID in binary form (16 bytes).

Nginx, Inc. p.197 of 330

CHAPTER 2. HTTP SERVER MODULES 2.38. MODULE NGX HTTP SPDY MODULE

2.38 Module ngx http spdy module

2.38.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 198
2.38.2 Known Bugs . . . . . . . . . . . . . . . . . . . . . . . . . 198
2.38.3 Example Configuration . . . . . . . . . . . . . . . . . . . 198
2.38.4 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 199
spdy chunk size . . . . . . . . . . . . . . . . . . . . . . . 199
spdy headers comp . . . . . . . . . . . . . . . . . . . . . 199
2.38.5 Embedded Variables . . . . . . . . . . . . . . . . . . . . 199

2.38.1 Summary
The ngx_http_spdy_module module provides experimental support for
SPDY. Currently, draft 3.1 of SPDY protocol is implemented.

Before version 1.5.10, draft 2 of SPDY protocol was implemented.

This module is not built by default, it should be enabled with the

--with-http_spdy_module configuration parameter.

2.38.2 Known Bugs

The module is experimental, caveat emptor applies.
Current implementation of SPDY protocol does not support server push.
In versions prior to 1.5.9, responses in SPDY connections could not be rate
limited.
Buffering of a client request body cannot be disabled regardless
of proxy request buffering, fastcgi request buffering, uwsgi request buffering,
and scgi request buffering directive values.

2.38.3 Example Configuration

server {
listen 443 ssl spdy ;

ss l_ce rtif icat e server . crt ;

s s l _ c e r t i f i c a t e _ k e y server . key ;
...
}

Note that in order to accept both HTTPS and SPDY connections

simultaneously on the same port, OpenSSL library used should support Next
Protocol Negotiation TLS extension, available since OpenSSL version 1.0.1.

Nginx, Inc. p.198 of 330

CHAPTER 2. HTTP SERVER MODULES 2.38. MODULE NGX HTTP SPDY MODULE

2.38.4 Directives
spdy chunk size
Syntax: spdy_chunk_size size;
Default 8k
Context: http, server, location
This directive appeared in version 1.5.9.

Sets the maximum size of chunks into which the response body is sliced. A
too low value results in higher overhead. A too high value impairs prioritization
due to HOL blocking.

spdy headers comp

Syntax: spdy_headers_comp level;
Default 0
Context: http, server

Sets the header compression level of a response in a range from 1 (fastest,

less compression) to 9 (slowest, best compression). The special value 0 turns
off the header compression.

2.38.5 Embedded Variables

The ngx_http_spdy_module module supports the following embedded
variables:

$spdy
SPDY protocol version for SPDY connections, or an empty string
otherwise;
$spdy request priority
request priority for SPDY connections, or an empty string otherwise.

Nginx, Inc. p.199 of 330

CHAPTER 2. HTTP SERVER MODULES 2.39. MODULE NGX HTTP SPLIT CLIENTS MODULE

2.39 Module ngx http split clients module

2.39.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 200
2.39.2 Example Configuration . . . . . . . . . . . . . . . . . . . 200
2.39.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 200
split clients . . . . . . . . . . . . . . . . . . . . . . . . . 200

2.39.1 Summary
The ngx_http_split_clients_module module creates variables suitable
for A/B testing, also known as split testing.

2.39.2 Example Configuration

http {
split_clients " $ { remote_addr } AAA " $variant {
0.5% . one ;
2.0% . two ;
* "";
}

server {
location / {
index index$ { variant }. html ;

2.39.3 Directives
split clients
Syntax: split_clients string $variable { . . . }
Default
Context: http

Creates a variable for A/B testing, for example:

split_clients " $ { remote_addr } AAA " $variant {

0.5% . one ;
2.0% . two ;
* "";
}

The value of the original string is hashed using MurmurHash2. In the

example given, hash values from 0 to 21474835 (0.5%) correspond to the value
".one" of the $variant variable, hash values from 21474836 to 107374180 (2%)
correspond to the value ".two", and hash values from 107374181 to 4294967295
correspond to the value "" (an empty string).

Nginx, Inc. p.200 of 330

CHAPTER 2. HTTP SERVER MODULES 2.40. MODULE NGX HTTP SSI MODULE

2.40 Module ngx http ssi module

2.40.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 201
2.40.2 Example Configuration . . . . . . . . . . . . . . . . . . . 201
2.40.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 201
ssi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
ssi last modified . . . . . . . . . . . . . . . . . . . . . . 201
ssi min file chunk . . . . . . . . . . . . . . . . . . . . . . 202
ssi silent errors . . . . . . . . . . . . . . . . . . . . . . . 202
ssi types . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
ssi value length . . . . . . . . . . . . . . . . . . . . . . . 202
2.40.4 SSI Commands . . . . . . . . . . . . . . . . . . . . . . . 202
2.40.5 Embedded Variables . . . . . . . . . . . . . . . . . . . . 205

2.40.1 Summary
The ngx_http_ssi_module module is a filter that processes SSI (Server
Side Includes) commands in responses passing through it. Currently, the list
of supported SSI commands is incomplete.

2.40.2 Example Configuration

location / {
ssi on ;
...
}

2.40.3 Directives
ssi
Syntax: ssi on | off;
Default off
Context: http, server, location, if in location

Enables or disables processing of SSI commands in responses.

ssi last modified

Syntax: ssi_last_modified on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.1.

Allows preserving the Last-Modified header field from the original

response during SSI processing to facilitate response caching.
By default, the header field is removed as contents of the response are
modified during processing and may contain dynamically generated elements
or parts that are changed independently of the original response.

Nginx, Inc. p.201 of 330

CHAPTER 2. HTTP SERVER MODULES 2.40. MODULE NGX HTTP SSI MODULE

ssi min file chunk

Syntax: ssi_min_file_chunk size;
Default 1k
Context: http, server, location

Sets the minimum size for parts of a response stored on disk, starting from
which it makes sense to send them using sendfile.

ssi silent errors

Syntax: ssi_silent_errors on | off;
Default off
Context: http, server, location

If enabled, suppresses the output of the

[an error occurred while processing the directive] string if an
error occurred during SSI processing.

ssi types
Syntax: ssi_types mime-type . . . ;
Default text/html
Context: http, server, location

Enables processing of SSI commands in responses with the specified MIME

types in addition to text/html. The special value * matches any MIME
type (0.8.29).

ssi value length

Syntax: ssi_value_length length;
Default 256
Context: http, server, location

Sets the maximum length of parameter values in SSI commands.

2.40.4 SSI Commands

SSI commands have the following generic format:

<! - -# command parameter1 = value1 parameter2 = value2 ... -->

The following commands are supported:

block
Defines a block that can be used as a stub in the include command. The
block can contain other SSI commands. The command has the following
parameter:
name
block name.

Nginx, Inc. p.202 of 330

CHAPTER 2. HTTP SERVER MODULES 2.40. MODULE NGX HTTP SSI MODULE

Example:

<! - -# block name =" one " -->

stub
<! - -# endblock -->

config
Sets some parameters used during SSI processing, namely:
errmsg
a string that is output if an error occurs during SSI processing. By
default, the following string is output:

[ an error occurred while processing the directive ]

timefmt
a format string passed to the strftime function used to output date
and time. By default, the following format is used:

"% A , %d -% b -% Y % H :% M :% S % Z "

The %s format is suitable to output time in seconds.

echo
Outputs the value of a variable. The command has the following
parameters:
var
the variable name.
encoding
the encoding method. Possible values include none, url, and
entity. By default, entity is used.
default
a non-standard parameter that sets a string to be output if a variable
is undefined. By default, none is output. The command

<! - -# echo var =" name " default ="no " -->

replaces the following sequence of commands:

<! - -# if expr =" $name " - - > <! - -# echo var =" name " - - > <! - -#
else -->no <! - -# endif -->

if
Performs a conditional inclusion. The following commands are
supported:

<! - -# if expr ="..." -->

...
<! - -# elif expr ="..." -->
...

Nginx, Inc. p.203 of 330

CHAPTER 2. HTTP SERVER MODULES 2.40. MODULE NGX HTTP SSI MODULE

<! - -# else -->

...
<! - -# endif -->

Only one level of nesting is currently supported. The command has the
following parameter:
expr
expression. An expression can be:
variable existence check:

<! - -# if expr =" $name " -->

comparison of a variable with a text:

<! - -# if expr =" $name = text" -->

<! - -# if expr =" $name != text" -->

comparison of a variable with a regular expression:

<! - -# if expr =" $name = /text/" -->

<! - -# if expr =" $name != /text/" -->

If a text contains variables, their values are substituted. A regular

expression can contain positional and named captures that can later
be used through variables, for example:

<! - -# if expr =" $name = /(.+) @ (? P < domain >.+) /" -->
<! - -# echo var ="1" -->
<! - -# echo var =" domain " -->
<! - -# endif -->

include
Includes the result of another request into a response. The command has
the following parameters:
file
specifies an included file, for example:

<! - -# include file =" footer . html " -->

virtual
specifies an included request, for example:

<! - -# include virtual ="/ remote / body . php ? argument = value " -->

Several requests specified on one page and processed by proxied

or FastCGI/uwsgi/SCGI servers run in parallel. If sequential
processing is desired, the wait parameter should be used.

Nginx, Inc. p.204 of 330

CHAPTER 2. HTTP SERVER MODULES 2.40. MODULE NGX HTTP SSI MODULE

stub
a non-standard parameter that names the block whose content will
be output if the included request results in an empty body or if an
error occurs during the request processing, for example:

<! - -# block name =" one " - - >& nbsp ; <! - -# endblock -->
<! - -# include virtual ="/ remote / body . php ? argument = value " stub ="
one " -->

The replacement block content is processed in the included request

context.
wait
a non-standard parameter that instructs to wait for a request to
fully complete before continuing with SSI processing, for example:

<! - -# include virtual ="/ remote / body . php ? argument = value " wait ="
yes " -->

set
a non-standard parameter that instructs to write a successful result
of request processing to the specified variable, for example:

<! - -# include virtual ="/ remote / body . php ? argument = value " set ="
one " -->

It should be noted that only the results of responses obtained using

the ngx http proxy module, ngx http memcached module, ngx -
http fastcgi module (1.5.6), ngx http uwsgi module (1.5.6), and
ngx http scgi module (1.5.6) modules can be written into variables.
set
Sets a value of a variable. The command has the following parameters:
var
the variable name.
value
the variable value. If an assigned value contains variables, their
values are substituted.

2.40.5 Embedded Variables

The ngx_http_ssi_module module supports two embedded variables:
$date local
current time in the local time zone. The format is set by the config
command with the timefmt parameter.
$date gmt
current time in GMT. The format is set by the config command with
the timefmt parameter.

Nginx, Inc. p.205 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

2.41 Module ngx http ssl module

2.41.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 206
2.41.2 Example Configuration . . . . . . . . . . . . . . . . . . . 206
2.41.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 207
ssl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
ssl buffer size . . . . . . . . . . . . . . . . . . . . . . . . 207
ssl certificate . . . . . . . . . . . . . . . . . . . . . . . . 208
ssl certificate key . . . . . . . . . . . . . . . . . . . . . . 208
ssl ciphers . . . . . . . . . . . . . . . . . . . . . . . . . . 208
ssl client certificate . . . . . . . . . . . . . . . . . . . . . 209
ssl crl . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
ssl dhparam . . . . . . . . . . . . . . . . . . . . . . . . . 209
ssl ecdh curve . . . . . . . . . . . . . . . . . . . . . . . . 209
ssl password file . . . . . . . . . . . . . . . . . . . . . . . 209
ssl prefer server ciphers . . . . . . . . . . . . . . . . . . 210
ssl protocols . . . . . . . . . . . . . . . . . . . . . . . . . 210
ssl session cache . . . . . . . . . . . . . . . . . . . . . . 210
ssl session ticket key . . . . . . . . . . . . . . . . . . . . 211
ssl session tickets . . . . . . . . . . . . . . . . . . . . . . 211
ssl session timeout . . . . . . . . . . . . . . . . . . . . . 212
ssl stapling . . . . . . . . . . . . . . . . . . . . . . . . . 212
ssl stapling file . . . . . . . . . . . . . . . . . . . . . . . 212
ssl stapling responder . . . . . . . . . . . . . . . . . . . 212
ssl stapling verify . . . . . . . . . . . . . . . . . . . . . . 213
ssl trusted certificate . . . . . . . . . . . . . . . . . . . . 213
ssl verify client . . . . . . . . . . . . . . . . . . . . . . . 213
ssl verify depth . . . . . . . . . . . . . . . . . . . . . . . 213
2.41.4 Error Processing . . . . . . . . . . . . . . . . . . . . . . 214
2.41.5 Embedded Variables . . . . . . . . . . . . . . . . . . . . 214

2.41.1 Summary
The ngx_http_ssl_module module provides the necessary support for
HTTPS.
This module is not built by default, it should be enabled with the
--with-http_ssl_module configuration parameter.

This module requires the OpenSSL library.

2.41.2 Example Configuration

To reduce the processor load it is recommended to

set the number of worker processes equal to the number of processors,

Nginx, Inc. p.206 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

enable keep-alive connections,

enable the shared session cache,
disable the built-in session cache,
and possibly increase the session lifetime (by default, 5 minutes):

worker_processes auto;

http {

...

server {
listen 443 ssl ;
keepalive_timeout 70;

ssl_protocols TLSv1 TLSv1 .1 TLSv1 .2;

ssl_ciphers AES128 - SHA : AES256 - SHA : RC4 - SHA : DES - CBC3 - SHA : RC4
- MD5 ;
ssl _ce rtif icat e / usr / local / nginx / conf / cert . pem ;
s s l _ c e r t i f i c a t e _ k e y / usr / local / nginx / conf / cert . key ;
ssl_session_cache shared:SSL:10m;
ssl_session_timeout 10m;

...
}

2.41.3 Directives
ssl
Syntax: ssl on | off;
Default off
Context: http, server

Enables the HTTPS protocol for the given virtual server.

It is recommended to use the ssl parameter of the listen directive instead

of this directive.

ssl buffer size

Syntax: ssl_buffer_size size;
Default 16k
Context: http, server
This directive appeared in version 1.5.9.

Sets the size of the buffer used for sending data.

By default, the buffer size is 16k, which corresponds to minimal overhead
when sending big responses. To minimize Time To First Byte it may be
beneficial to use smaller values, for example:

ss l_bu ffer _siz e 4 k ;

Nginx, Inc. p.207 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

ssl certificate
Syntax: ssl_certificate file;
Default
Context: http, server

Specifies a file with the certificate in the PEM format for the given virtual
server. If intermediate certificates should be specified in addition to a primary
certificate, they should be specified in the same file in the following order: the
primary certificate comes first, then the intermediate certificates. A secret key
in the PEM format may be placed in the same file.
It should be kept in mind that due to the HTTPS protocol limitations
virtual servers should listen on different IP addresses:

server {
listen 1 9 2. 1 68 . 1. 1 :4 4 3;
server_name one . example . com ;
ss l_ce rtif icat e / usr / local / nginx / conf / one . example . com . cert ;
...
}

server {
listen 1 9 2. 1 68 . 1. 2 :4 4 3;
server_name two . example . com ;
ss l_ce rtif icat e / usr / local / nginx / conf / two . example . com . cert ;
...
}

otherwise the first servers certificate will be issued for the second site.

ssl certificate key

Syntax: ssl_certificate_key file;
Default
Context: http, server

Specifies a file with the secret key in the PEM format for the given virtual
server.

ssl ciphers
Syntax: ssl_ciphers ciphers;
Default HIGH:!aNULL:!MD5
Context: http, server

Specifies the enabled ciphers. The ciphers are specified in the format
understood by the OpenSSL library, for example:

ssl_ciphers ALL :! aNULL :! EXPORT56 : RC4 + RSA :+ HIGH :+ MEDIUM :+ LOW :+ SSLv2 :+ EXP ;

The full list can be viewed using the openssl ciphers command.

Nginx, Inc. p.208 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

The previous versions of nginx used different ciphers by default.

ssl client certificate

Syntax: ssl_client_certificate file;
Default
Context: http, server

Specifies a file with trusted CA certificates in the PEM format used to

verify client certificates and OCSP responses if ssl stapling is enabled.
The list of certificates will be sent to clients. If this is not desired, the
ssl trusted certificate directive can be used.

ssl crl
Syntax: ssl_crl file;
Default
Context: http, server
This directive appeared in version 0.8.7.

Specifies a file with revoked certificates (CRL) in the PEM format used to
verify client certificates.

ssl dhparam
Syntax: ssl_dhparam file;
Default
Context: http, server
This directive appeared in version 0.7.2.

Specifies a file with DH parameters for EDH ciphers.

ssl ecdh curve

Syntax: ssl_ecdh_curve curve;
Default prime256v1
Context: http, server
This directive appeared in versions 1.1.0 and 1.0.6.

Specifies a curve for ECDHE ciphers.

ssl password file

Syntax: ssl_password_file file;
Default
Context: http, server
This directive appeared in version 1.7.3.

Specifies a file with passphrases for secret keys where each passphrase is
specified on a separate line. Passphrases are tried in turn when loading the
key.

Nginx, Inc. p.209 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

Example:

http {
s s l _ p a s s w o r d _ f i l e / etc / keys / global . pass ;
...

server {
server_name www1 . example . com ;
s s l _ c e r t i f i c a t e _ k e y / etc / keys / first . key ;
}

server {
server_name www2 . example . com ;

# named pipe can also be used instead of a file

s s l _ p a s s w o r d _ f i l e / etc / keys / fifo ;
s s l _ c e r t i f i c a t e _ k e y / etc / keys / second . key ;
}
}

ssl prefer server ciphers

Syntax: ssl_prefer_server_ciphers on | off;
Default off
Context: http, server

Specifies that server ciphers should be preferred over client ciphers when
using the SSLv3 and TLS protocols.

ssl protocols
Syntax: ssl_protocols [SSLv2] [SSLv3] [TLSv1] [TLSv1.1] [TLSv1.2];
Default SSLv3 TLSv1 TLSv1.1 TLSv1.2
Context: http, server

Enables the specified protocols. The TLSv1.1 and TLSv1.2 parameters

work only when the OpenSSL library of version 1.0.1 or higher is used.

The TLSv1.1 and TLSv1.2 parameters are supported starting from

versions 1.1.13 and 1.0.12, so when the OpenSSL version 1.0.1 or higher is
used on older nginx versions, these protocols work, but cannot be disabled.

ssl session cache

Syntax: ssl_session_cache off | none | [builtin[:size]] [shared:name:size];
Default none
Context: http, server

Sets the types and sizes of caches that store session parameters. A cache
can be of any of the following types:

off
the use of a session cache is strictly prohibited: nginx explicitly tells a
client that sessions may not be reused.

Nginx, Inc. p.210 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

none
the use of a session cache is gently disallowed: nginx tells a client that
sessions may be reused, but does not actually store session parameters
in the cache.
builtin
a cache built in OpenSSL; used by one worker process only. The cache
size is specified in sessions. If size is not given, it is equal to 20480
sessions. Use of the built-in cache can cause memory fragmentation.
shared
a cache shared between all worker processes. The cache size is specified
in bytes; one megabyte can store about 4000 sessions. Each shared cache
should have an arbitrary name. A cache with the same name can be used
in several virtual servers.
Both cache types can be used simultaneously, for example:

s s l _ s e s s i o n _ c a c h e builtin :1000 shared : SSL :10 m ;

but using only shared cache without the built-in cache should be more
efficient.

ssl session ticket key

Syntax: ssl_session_ticket_key file;
Default
Context: http, server
This directive appeared in version 1.5.7.

Sets a file with the secret key used to encrypt and decrypt TLS session
tickets. The directive is necessary if the same key has to be shared between
multiple servers. By default, a randomly generated key is used.
If several keys are specified, only the first key is used to encrypt TLS session
tickets. This allows configuring key rotation, for example:

s s l _ s e s s i o n _ t i c k e t _ k e y current . key ;
s s l _ s e s s i o n _ t i c k e t _ k e y previous . key ;

The file must contain 48 bytes of random data and can be created using
the following command:

openssl rand 48 > ticket . key

ssl session tickets

Syntax: ssl_session_tickets on | off;
Default on
Context: http, server
This directive appeared in version 1.5.9.

Enables or disables session resumption through TLS session tickets.

Nginx, Inc. p.211 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

ssl session timeout

Syntax: ssl_session_timeout time;
Default 5m
Context: http, server

Specifies a time during which a client may reuse the session parameters
stored in a cache.

ssl stapling
Syntax: ssl_stapling on | off;
Default off
Context: http, server
This directive appeared in version 1.3.7.

Enables or disables stapling of OCSP responses by the server. Example:

ssl_stapling on ;
resolver 192.0.2.1;

For the OCSP stapling to work, the certificate of the server certificate
issuer should be known. If the ssl certificate file does not contain intermediate
certificates, the certificate of the server certificate issuer should be present in
the ssl trusted certificate file.
For a resolution of the OCSP responder hostname, the resolver directive
should also be specified.

ssl stapling file

Syntax: ssl_stapling_file file;
Default
Context: http, server
This directive appeared in version 1.3.7.

When set, the stapled OCSP response will be taken from the specified file
instead of querying the OCSP responder specified in the server certificate.
The file should be in the DER format as produced by the openssl ocsp
command.

ssl stapling responder

Syntax: ssl_stapling_responder url;
Default
Context: http, server
This directive appeared in version 1.3.7.

Overrides the URL of the OCSP responder specified in the Authority

Information Access certificate extension.
Only http:// OCSP responders are supported:

Nginx, Inc. p.212 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

s s l _ s t a p l i n g _ r e s p o n d e r http :// ocsp . example . com /;

ssl stapling verify

Syntax: ssl_stapling_verify on | off;
Default off
Context: http, server
This directive appeared in version 1.3.7.

Enables or disables verification of OCSP responses by the server.

For verification to work, the certificate of the server certificate issuer, the
root certificate, and all intermediate certificates should be configured as trusted
using the ssl trusted certificate directive.

ssl trusted certificate

Syntax: ssl_trusted_certificate file;
Default
Context: http, server
This directive appeared in version 1.3.7.

Specifies a file with trusted CA certificates in the PEM format used to

verify client certificates and OCSP responses if ssl stapling is enabled.
In contrast to the certificate set by ssl client certificate, the list of these
certificates will not be sent to clients.

ssl verify client

Syntax: ssl_verify_client on | off | optional | optional_no_ca;
Default off
Context: http, server

Enables verification of client certificates. The verification result is stored

in the $ssl client verify variable.
The optional parameter (0.8.7+) requests the client certificate and verifies
it if the certificate is present.
The optional_no_ca parameter (1.3.8, 1.2.5) requests the client certificate
but does not require it to be signed by a trusted CA certificate. This is intended
for the use in cases when a service that is external to nginx performs the actual
certificate verification. The contents of the certificate is accessible through the
$ssl client cert variable.

ssl verify depth

Syntax: ssl_verify_depth number;
Default 1
Context: http, server

Nginx, Inc. p.213 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

Sets the verification depth in the client certificates chain.

2.41.4 Error Processing

The ngx_http_ssl_module module supports several non-standard error
codes that can be used for redirects using the error page directive:

495
an error has occurred during the client certificate verification;
496
a client has not presented the required certificate;
497
a regular request has been sent to the HTTPS port.

The redirection happens after the request is fully parsed and the variables,
such as $request uri, $uri, $args and others, are available.

2.41.5 Embedded Variables

The ngx_http_ssl_module module supports several embedded variables:

$ssl cipher
returns the string of ciphers used for an established SSL connection;
$ssl client cert
returns the client certificate in the PEM format for an established SSL
connection, with each line except the first prepended with the tab
character; this is intended for the use in the proxy set header directive;
$ssl client fingerprint
returns the SHA1 fingerprint of the client certificate for an established
SSL connection (1.7.1);
$ssl client raw cert
returns the client certificate in the PEM format for an established SSL
connection;
$ssl client serial
returns the serial number of the client certificate for an established SSL
connection;
$ssl client s dn
returns the subject DN string of the client certificate for an established
SSL connection;
$ssl client i dn
returns the issuer DN string of the client certificate for an established
SSL connection;
$ssl client verify
returns the result of client certificate verification: SUCCESS, FAILED,
and NONE if a certificate was not present;
$ssl protocol
returns the protocol of an established SSL connection;

Nginx, Inc. p.214 of 330

CHAPTER 2. HTTP SERVER MODULES 2.41. MODULE NGX HTTP SSL MODULE

$ssl server name

returns the server name requested through SNI (1.7.0);
$ssl session id
returns the session identifier of an established SSL connection;
$ssl session reused
returns r if an SSL session was reused, or . otherwise (1.5.11).

Nginx, Inc. p.215 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

2.42 Module ngx http status module

2.42.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 216
2.42.2 Example Configuration . . . . . . . . . . . . . . . . . . . 216
2.42.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 217
status . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
status format . . . . . . . . . . . . . . . . . . . . . . . . 217
status zone . . . . . . . . . . . . . . . . . . . . . . . . . 218
2.42.4 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
2.42.5 Compatibility . . . . . . . . . . . . . . . . . . . . . . . . 223

2.42.1 Summary
The ngx_http_status_module module provides access to various status
information.

This module is available as part of our commercial subscription.

2.42.2 Example Configuration

http {
upstream backend {
zone http_backend 64 k ;

server backend1 . example . com weight =5;

server backend2 . example . com ;
}

p ro x y_ c ac h e_ p at h / data / nginx / cache_backend keys_zone =cache_backend :10 m ;

server {
server_name backend . example . com ;

location / {
proxy_pass http :// backend ;
proxy_cache cache_backend ;

health_check ;
}

status_zone server_backend;
}

server {
listen 127.0.0.1;

location / upstream_conf {
upstream_conf ;
}

location / status {
status ;
}

location = / status . html {

}
}
}

Nginx, Inc. p.216 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

stream {
upstream backend {
zone stream_backend 64 k ;

server backend1 . example . com :12345 weight =5;

server backend2 . example . com :12345;
}

server {
listen 12 7. 0 .0 . 1: 1 23 4 5;
proxy_pass backend ;
status_zone server_backend;
health_check ;
}
}

Examples of status requests with this configuration:

http ://127.0.0.1/ status

http ://127.0.0.1/ status / nginx_version
http ://127.0.0.1/ status / caches / cache_backend
http ://127.0.0.1/ status / upstreams
http ://127.0.0.1/ status / upstreams / backend
http ://127.0.0.1/ status / upstreams / backend /1
http ://127.0.0.1/ status / upstreams / backend /1/ weight
http ://127.0.0.1/ status / stream
http ://127.0.0.1/ status / stream / upstreams
http ://127.0.0.1/ status / stream / upstreams / backend
http ://127.0.0.1/ status / stream / upstreams / backend /1
http ://127.0.0.1/ status / stream / upstreams / backend /1/ weight

The simple monitoring page is shipped with this distribution, accessible

as /status.html in the default configuration. It requires the locations
/status and /status.html to be configured as shown above.

2.42.3 Directives
status
Syntax: status;
Default
Context: location

The status information will be accessible from the surrounding location.

status format
Syntax: status_format json;
Syntax: status_format jsonp [callback];
Default json
Context: http, server, location

By default, status information is output in the JSON format.

Alternatively, data may be output as JSONP. The callback parameter
specifies the name of a callback function. The value can contain variables.
If parameter is omitted, or the computed value is an empty string, then
ngx_status_jsonp_callback is used.

Nginx, Inc. p.217 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

status zone
Syntax: status_zone zone;
Default
Context: server

Enables collection of virtual http or stream (1.7.11) server status

information in the specified zone. Several servers may share the same zone.

2.42.4 Data
The following status information is provided:

version
Version of the provided data set. The current version is 5.
nginx_version
Version of nginx.
address
The address of the server that accepted status request.
generation
The total number of configuration reloads.
load_timestamp
Time of the last reload of configuration, in milliseconds since Epoch.
timestamp
Current time in milliseconds since Epoch.
processes
respawned
The total number of abnormally terminated and respawned child
processes.
connections
accepted
The total number of accepted client connections.
dropped
The total number of dropped client connections.
active
The current number of active client connections.
idle
The current number of idle client connections.
requests
total
The total number of client requests.
current
The current number of client requests.

Nginx, Inc. p.218 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

server_zones
For each status zone:
processing
The number of client requests that are currently being processed.
requests
The total number of client requests received from clients.
responses
total
The total number of responses sent to clients.
1xx, 2xx, 3xx, 4xx, 5xx
The number of responses with status codes 1xx, 2xx, 3xx, 4xx,
and 5xx.
received
The total number of bytes received from clients.
sent
The total number of bytes sent to clients.
upstreams
For each server in the dynamically configurable group, the following data
are provided:
id
The ID of the server.
server
An address of the server.
backup
A boolean value indicating whether the server is a backup server.
weight
Weight of the server.
state
Current state, which may be one of up, draining, down,
unavail, or unhealthy.
active
The current number of active connections.
max_conns
The max conns limit for the server.
requests
The total number of client requests forwarded to this server.
responses
total
The total number of responses obtained from this server.
1xx, 2xx, 3xx, 4xx, 5xx
The number of responses with status codes 1xx, 2xx, 3xx, 4xx,
and 5xx.
sent
The total number of bytes sent to this server.

Nginx, Inc. p.219 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

received
The total number of bytes received from this server.
fails
The total number of unsuccessful attempts to communicate with
the server.
unavail
How many times the server became unavailable for client requests
(state unavail) due to the number of unsuccessful attempts
reaching the max fails threshold.
health_checks
checks
The total number of health check requests made.
fails
The number of failed health checks.
unhealthy
How many times the server became unhealthy (state
unhealthy).
last_passed
Boolean indicating if the last health check request was
successful and passed tests.
downtime
Total time the server was in the unavail and unhealthy states.
downstart
The time (in milliseconds since Epoch) when the server became
unavail or unhealthy.
selected
The time (in milliseconds since Epoch) when the server was last
selected to process a request (1.7.5).
header_time
The average time to get the response header from the server (1.7.10).
The field is available when using the least time load balancing
method.
response_time
The average time to get the full response from the server (1.7.10).
The field is available when using the least time load balancing
method.
caches
For each cache (configured by proxy cache path and the likes):
size
The current size of the cache.
max_size
The limit on the maximum size of the cache specified in the
configuration.
cold

Nginx, Inc. p.220 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

A boolean value indicating whether the cache loader process is

still loading data from disk into the cache.
hit, stale, updating, revalidated
responses
The total number of responses read from the cache (hits, or
stale responses due to proxy cache use stale and the likes).
bytes
The total number of bytes read from the cache.
miss, expired, bypass
responses
The total number of responses not taken from the cache (misses,
expires, or bypasses due to proxy cache bypass and the likes).
bytes
The total number of bytes read from the proxied server.
responses_written
The total number of responses written to the cache.
bytes_written
The total number of bytes written to the cache.
stream
server_zones
For each status zone:
processing
The number of client connections that are currently being
processed.
connections
The total number of connections accepted from clients.
received
The total number of bytes received from clients.
sent
The total number of bytes sent to clients.
upstreams
For each server in the dynamically configurable group, the following
data are provided:
id
The ID of the server.
server
An address of the server.
backup
A boolean value indicating whether the server is a backup
server.
weight
Weight of the server.
state

Nginx, Inc. p.221 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

Current state, which may be one of up, down, unavail, or

unhealthy.
active
The current number of connections.
connections
The total number of client connections forwarded to this server.
connect_time
The average time to connect to the upstream server. The field
is available when using the least time load balancing method.
first_byte_time
The average time to receive the first byte of data. The field is
available when using the least time load balancing method.
response_time
The average time to receive the last byte of data. The field is
available when using the least time load balancing method.
sent
The total number of bytes sent to this server.
received
The total number of bytes received from this server.
fails
The total number of unsuccessful attempts to communicate
with the server.
unavail
How many times the server became unavailable for client con-
nections (state unavail) due to the number of unsuccessful
attempts reaching the max fails threshold.
health_checks
checks
The total number of health check requests made.
fails
The number of failed health checks.
unhealthy
How many times the server became unhealthy (state
unhealthy).
last_passed
Boolean indicating if the last health check request was
successful and passed tests.
downtime
Total time the server was in the unavail and unhealthy
states.
downstart
The time (in milliseconds since Epoch) when the server became
unavail or unhealthy.
selected
The time (in milliseconds since Epoch) when the server was last

Nginx, Inc. p.222 of 330

CHAPTER 2. HTTP SERVER MODULES 2.42. MODULE NGX HTTP STATUS MODULE

selected to process a connection.

2.42.5 Compatibility
The keepalive field of an upstream server was removed in version 5.

The stream status data were added in version 5.

The generation field was added in version 5.

The respawned field in processes was added in version 5.

The header time and response time fields in upstreams were added in
version 5.

The selected field in upstreams was added in version 4.

The draining state in upstreams was added in version 4.

The id and max conns fields in upstreams were added in version 3.

The revalidated field in caches was added in version 3.

The server zones, caches, and load timestamp status data were added in
version 2.

Nginx, Inc. p.223 of 330

CHAPTER 2. HTTP SERVER MODULES 2.43. MODULE NGX HTTP STUB STATUS MODULE

2.43 Module ngx http stub status module

2.43.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 224
2.43.2 Example Configuration . . . . . . . . . . . . . . . . . . . 224
2.43.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 224
stub status . . . . . . . . . . . . . . . . . . . . . . . . . 224
2.43.4 Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
2.43.5 Embedded Variables . . . . . . . . . . . . . . . . . . . . 225

2.43.1 Summary
The ngx_http_stub_status_module module provides access to basic
status information.
This module is not built by default, it should be enabled with the
--with-http_stub_status_module configuration parameter.

2.43.2 Example Configuration

location / basic_status {
stub_status ;
}

This configuration creates a simple web page with basic status data which
may look like as follows:

Active connections : 291

server accepts handled requests
16630948 16630948 31070465
Reading : 6 Writing : 179 Waiting : 106

2.43.3 Directives
stub status
Syntax: stub_status;
Default
Context: server, location

The basic status information will be accessible from the surrounding

location.

In versions prior to 1.7.5, the directive required an arbitrary argument.

2.43.4 Data
The following status information is provided:

Nginx, Inc. p.224 of 330

CHAPTER 2. HTTP SERVER MODULES 2.43. MODULE NGX HTTP STUB STATUS MODULE

Active connections
The current number of active client connections including Waiting
connections.
accepts
The total number of accepted client connections.
handled
The total number of handled connections. Generally, the parameter value
is the same as accepts unless some resource limits have been reached
(for example, the worker connections limit).
requests
The total number of client requests.
Reading
The current number of connections where nginx is reading the request
header.
Writing
The current number of connections where nginx is writing the response
back to the client.
Waiting
The current number of idle client connections waiting for a request.

2.43.5 Embedded Variables

The ngx_http_stub_status_module module supports the following
embedded variables (1.3.14):

$connections active
same as the Active connections value;
$connections reading
same as the Reading value;
$connections writing
same as the Writing value;
$connections waiting
same as the Waiting value.

Nginx, Inc. p.225 of 330

CHAPTER 2. HTTP SERVER MODULES 2.44. MODULE NGX HTTP SUB MODULE

2.44 Module ngx http sub module

2.44.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 226
2.44.2 Example Configuration . . . . . . . . . . . . . . . . . . . 226
2.44.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 226
sub filter . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
sub filter last modified . . . . . . . . . . . . . . . . . . . 226
sub filter once . . . . . . . . . . . . . . . . . . . . . . . . 227
sub filter types . . . . . . . . . . . . . . . . . . . . . . . 227

2.44.1 Summary
The ngx_http_sub_module module is a filter that modifies a response by
replacing one specified string by another.
This module is not built by default, it should be enabled with the
--with-http_sub_module configuration parameter.

2.44.2 Example Configuration

location / {
sub_filter </ head >
</ head > < script language =" javascript " src =" $script " > </ script > ;
su b_fi lter _onc e on ;
}

2.44.3 Directives
sub filter
Syntax: sub_filter string replacement;
Default
Context: http, server, location

Sets a string to replace and a replacement string. The string to replace is

matched ignoring the case. The replacement string can contain variables.

sub filter last modified

Syntax: sub_filter_last_modified on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.1.

Allows preserving the Last-Modified header field from the original

response during replacement to facilitate response caching.
By default, the header field is removed as contents of the response are
modified during processing.

Nginx, Inc. p.226 of 330

CHAPTER 2. HTTP SERVER MODULES 2.44. MODULE NGX HTTP SUB MODULE

sub filter once

Syntax: sub_filter_once on | off;
Default on
Context: http, server, location

Indicates whether to look for a string to replace once or several times.

sub filter types

Syntax: sub_filter_types mime-type . . . ;
Default text/html
Context: http, server, location

Enables string replacement in responses with the specified MIME types

in addition to text/html. The special value * matches any MIME type
(0.8.29).

Nginx, Inc. p.227 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

2.45 Module ngx http upstream module

2.45.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 228
2.45.2 Example Configuration . . . . . . . . . . . . . . . . . . . 228
2.45.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 229
upstream . . . . . . . . . . . . . . . . . . . . . . . . . . 229
server . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
hash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
ip hash . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
keepalive . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
least conn . . . . . . . . . . . . . . . . . . . . . . . . . . 234
least time . . . . . . . . . . . . . . . . . . . . . . . . . . 234
health check . . . . . . . . . . . . . . . . . . . . . . . . . 234
match . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
sticky . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
sticky cookie insert . . . . . . . . . . . . . . . . . . . . . 240
2.45.4 Embedded Variables . . . . . . . . . . . . . . . . . . . . 240

2.45.1 Summary
The ngx_http_upstream_module module is used to define groups of servers
that can be referenced by the proxy pass, fastcgi pass, uwsgi pass, scgi pass,
and memcached pass directives.

2.45.2 Example Configuration

upstream backend {
server backend1 . example . com weight =5;
server backend2 . example . com :8080;
server unix :/ tmp / backend3 ;

server backup1 . example . com :8080 backup ;

server backup2 . example . com :8080 backup ;
}

server {
location / {
proxy_pass http ://backend ;
}
}

Dynamically configurable group, available as part of our

commercial subscription:

resolver 10.0.0.1;

upstream dynamic {
zone up s tr e am _ dy n am i c 64 k ;

server backend1 . example . com weight =5;

server backend2 . example . com :8080 fail_timeout =5 s slow_start =30 s ;

Nginx, Inc. p.228 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

server 192.0.2.1 max_fails =3;

server backend3 . example . com resolve ;

server backup1 . example . com :8080 backup ;

server backup2 . example . com :8080 backup ;
}

server {
location / {
proxy_pass http ://dynamic ;
health_check ;
}
}

2.45.3 Directives
upstream
Syntax: upstream name { . . . }
Default
Context: http

Defines a group of servers. Servers can listen on different ports. In addition,

servers listening on TCP and UNIX-domain sockets can be mixed.
Example:

upstream backend {
server backend1 . example . com weight =5;
server 127.0.0.1:8080 max_fails =3 fail_timeout =30 s ;
server unix :/ tmp / backend3 ;

server backup1 . example . com backup ;

}

By default, requests are distributed between the servers using a weighted

round-robin balancing method. In the above example, each 7 requests will
be distributed as follows: 5 requests go to backend1.example.com and one
request to each of the second and third servers. If an error occurs during
communication with a server, the request will be passed to the next server,
and so on until all of the functioning servers will be tried. If a successful
response could not be obtained from any of the servers, the client will receive
the result of the communication with the last server.

server
Syntax: server address [parameters];
Default
Context: upstream

Defines the address and other parameters of a server. The address can
be specified as a domain name or IP address, with an optional port, or as
a UNIX-domain socket path specified after the unix: prefix. If a port is
not specified, the port 80 is used. A domain name that resolves to several IP
addresses defines multiple servers at once.

Nginx, Inc. p.229 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

The following parameters can be defined:

weight=number
sets the weight of the server, by default, 1.
max_fails=number
sets the number of unsuccessful attempts to communicate with the
server that should happen in the duration set by the fail_timeout
parameter to consider the server unavailable for a duration also set by
the fail_timeout parameter. By default, the number of unsuccessful
attempts is set to 1. The zero value disables the accounting of
attempts. What is considered an unsuccessful attempt is defined by
the proxy next upstream, fastcgi next upstream, uwsgi next upstream,
scgi next upstream, and memcached next upstream directives.
fail_timeout=time
sets
the time during which the specified number of unsuccessful attempts
to communicate with the server should happen to consider the server
unavailable;
and the period of time the server will be considered unavailable.
By default, the parameter is set to 10 seconds.
backup
marks the server as a backup server. It will be passed requests when the
primary servers are unavailable.
down
marks the server as permanently unavailable.

Additionally, the following parameters are available as part of our

commercial subscription:

max_conns=number
limits the maximum number of simultaneous connections to the proxied
server (1.5.9). Default value is zero, meaning there is no limit.
resolve
monitors changes of the IP addresses that correspond to a domain name
of the server, and automatically modifies the upstream configuration
without the need of restarting nginx (1.5.12).
In order for this parameter to work, the resolver directive must be
specified in the http block. Example:

http {
resolver 10.0.0.1;

upstream u {
zone ...;
...
server example . com resolve ;
}
}

Nginx, Inc. p.230 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

route=string
sets the server route name.
slow_start=time
sets the time during which the server will recover its weight from zero
to a nominal value, when unhealthy server becomes healthy, or when
the server becomes available after a period of time it was considered
unavailable. Default value is zero, i.e. slow start is disabled.

If there is only a single server in a group, max_fails, fail_timeout

and slow_start parameters are ignored, and such a server will never be
considered unavailable.

zone
Syntax: zone name size;
Default
Context: upstream

Defines the name and size of the shared memory zone that keeps the groups
configuration and run-time state that are shared between worker processes.
Such groups allow changing the group membership or modifying the settings
of a particular server without the need of restarting nginx. The configuration
is accessible via a special location handled by upstream conf.

This directive is available as part of our commercial subscription.

hash
Syntax: hash key [consistent];
Default
Context: upstream
This directive appeared in version 1.7.2.

Specifies a load balancing method for a server group where the client-server
mapping is based on the hashed key value. The key can contain text, variables,
and their combinations. Note that adding or removing a server from the group
may result in remapping most of the keys to different servers. The method is
compatible with the Cache::Memcached Perl library.
If the consistent parameter is specified the ketama consistent hashing
method will be used instead. The method ensures that only a few keys will be
remapped to different servers when a server is added to or removed from the
group. This helps to achieve a higher cache hit ratio for caching servers. The
method is compatible with the Cache::Memcached::Fast Perl library with the
ketama points parameter set to 160.

Nginx, Inc. p.231 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

ip hash
Syntax: ip_hash;
Default
Context: upstream

Specifies that a group should use a load balancing method where requests
are distributed between servers based on client IP addresses. The first three
octets of the client IPv4 address, or the entire IPv6 address, are used as a
hashing key. The method ensures that requests from the same client will
always be passed to the same server except when this server is unavailable. In
the latter case client requests will be passed to another server. Most probably,
it will always be the same server as well.

IPv6 addresses are supported starting from versions 1.3.2 and 1.2.2.

If one of the servers needs to be temporarily removed, it should be marked

with the down parameter in order to preserve the current hashing of client IP
addresses.
Example:

upstream backend {
ip_hash ;

server backend1 . example . com ;

server backend2 . example . com ;
server backend3 . example . com down ;
server backend4 . example . com ;
}

Until versions 1.3.1 and 1.2.2, it was not possible to specify a weight for
servers using the ip_hash load balancing method.

keepalive
Syntax: keepalive connections;
Default
Context: upstream
This directive appeared in version 1.1.4.

Activates the cache for connections to upstream servers.

The connections parameter sets the maximum number of idle keepalive
connections to upstream servers that are preserved in the cache of each worker
process. When this number is exceeded, the least recently used connections
are closed.

It should be particularly noted that the keepalive directive does not limit
the total number of connections to upstream servers that an nginx worker
process can open. The connections parameter should be set to a number
small enough to let upstream servers process new incoming connections as
well.

Nginx, Inc. p.232 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

Example configuration of memcached upstream with keepalive connections:

upstream m e m c a c h e d _ b a c k e n d {
server 12 7 .0 . 0. 1: 1 12 1 1;
server 10 .0. 0.2: 1121 1;

keepalive 32;
}

server {
...

location / memcached / {
set $memcached_key $uri ;
memcached_pass m e m c a c h e d _ b a c k e n d ;
}

For HTTP, the proxy http version directive should be set to 1.1 and the
Connection header field should be cleared:

upstream http_backend {
server 12 7.0 .0.1 :808 0;

keepalive 16;
}

server {
...

location / http / {
proxy_pass http :// http_backend ;
p r o x y _ h t t p _ v e r s i o n 1.1;
p ro x y_ s et _ he ad e r Connection "";
...
}
}

Alternatively, HTTP/1.0 persistent connections can be used by passing

the Connection: Keep-Alive header field to an upstream server, though
this method is not recommended.

For FastCGI servers, it is required to set fastcgi keep conn for keepalive
connections to work:

upstream fast cgi_ back end {

server 12 7.0 .0.1 :900 0;

keepalive 8;
}

server {
...

location / fastcgi / {
fastcgi_pass fast cgi_ bac kend ;
f a s t c g i _ k e e p _ c o n n on ;
...
}
}

Nginx, Inc. p.233 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

When using load balancer methods other than the default round-robin
method, it is necessary to activate them before the keepalive directive.

SCGI and uwsgi protocols do not have a notion of keepalive connections.

least conn
Syntax: least_conn;
Default
Context: upstream
This directive appeared in versions 1.3.1 and 1.2.2.

Specifies that a group should use a load balancing method where a request
is passed to the server with the least number of active connections, taking into
account weights of servers. If there are several such servers, they are tried in
turn using a weighted round-robin balancing method.

least time
Syntax: least_time header | last_byte;
Default
Context: upstream
This directive appeared in version 1.7.10.

Specifies that a group should use a load balancing method where a request
is passed to the server with the least average response time and least number of
active connections, taking into account weights of servers. If there are several
such servers, they are tried in turn using a weighted round-robin balancing
method.
If the header parameter is specified, time to receive the response header is
used. If the last_byte parameter is specified, time to receive the full response
is used.

This directive is available as part of our commercial subscription.

health check
Syntax: health_check [parameters];
Default
Context: location

Enables periodic health checks of the servers in a group referenced in the

surrounding location.
The following optional parameters are supported:

interval=time
sets the interval between two consecutive health checks, by default, 5
seconds;

Nginx, Inc. p.234 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

fails=number
sets the number of consecutive failed health checks of a particular server
after which this server will be considered unhealthy, by default, 1;
passes=number
sets the number of consecutive passed health checks of a particular server
after which the server will be considered healthy, by default, 1;
uri=uri
defines the URI used in health check requests, by default, /;
match=name
specifies the match block configuring the tests that a response should
pass in order for a health check to pass; by default, the response should
have status code 2xx or 3xx.

For example,

location / {
proxy_pass http :// backend ;
health_check ;
}

will send/requests to each server in the backend group every five seconds.
If any communication error or timeout occurs, or a proxied server responds
with the status code other than 2xx or 3xx, the health check will fail, and
the server will be considered unhealthy. Client requests are not passed to
unhealthy servers.
Health checks can be configured to test the status code of a response,
presence of certain header fields and their values, and the body contents. Tests
are configured separately using the match directive and referenced in the match
parameter. For example:

http {
server {
...
location / {
proxy_pass http :// backend ;
health_check match = welcome ;
}
}

match welcome {
status 200;
header Content - Type = text / html ;
body ~ " Welcome to nginx !";
}
}

This configuration tells that for a health check to pass, the response
to a health check request should succeed, have status 200, content type
text/html, and contain Welcome to nginx! in the body.
The server group must reside in the shared memory.
If several health checks are defined for the same group of servers, a single
failure of any check will make the corresponding server be considered unhealthy.

Nginx, Inc. p.235 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

Please note that most of the variables will have empty values when used
with health checks.

This directive is available as part of our commercial subscription.

match
Syntax: match name { . . . }
Default
Context: http

Defines the named test set used to verify responses to health check requests.
The following items can be tested in a response:

status 200;
status is 200
status ! 500;
status is not 500
status 200 204;
status is 200 or 204
status ! 301 302;
status is neither 301 nor 302
status 200-399;
status is in the range from 200 to 399
status ! 400-599;
status is not in the range from 400 to 599
status 301-303 307;
status is either 301, 302, 303, or 307

header Content-Type = text/html;

header contains Content-Type with value text/html
header Content-Type != text/html;
header contains Content-Type with value other than text/html
header Connection ~ close;
header contains Connection with value matching regular expression
close
header Connection !~ close;
header contains Connection with value not matching regular expression
close
header Host;
header contains Host
header ! X-Accel-Redirect;
header lacks X-Accel-Redirect

body ~ "Welcome to nginx!";

body matches regular expression Welcome to nginx!

Nginx, Inc. p.236 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

body !~ "Welcome to nginx!";

body does not match regular expression Welcome to nginx!

If several tests are specified, the response matches only if it matches all
tests.

Only the first 256k of the response body are examined.

Examples:

# status is 200 , content type is " text / html " ,

# and body contains " Welcome to nginx !"
match welcome {
status 200;
header Content - Type = text / html ;
body ~ " Welcome to nginx !";
}

# status is not one of 301 , 302 , 303 , or 307 , and header does not have "
Refresh :"
match not_redirect {
status ! 301 -303 307;
header ! Refresh ;
}

# status ok and not in maintenance mode

match server_ok {
status 200 -399;
body !~ " maintenance mode ";
}

This directive is available as part of our commercial subscription.

queue
Syntax: queue number [timeout=time];
Default
Context: upstream
This directive appeared in version 1.5.12.

If an upstream server cannot be selected immediately while processing a

request, and there are the servers in the group that have reached the max -
conns limit, the request will be placed into the queue. The directive specifies
the maximum number of requests that can be in the queue at the same time. If
the queue is filled up, or the server to pass the request to cannot been selected
within the time period specified in the timeout parameter, an error will be
returned to the client.
The default value of the timeout parameter is 60 seconds.

Nginx, Inc. p.237 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

This directive is available as part of our commercial subscription.

sticky
Syntax: sticky cookie name [expires=time] [domain=domain] [httponly]
[secure] [path=path];
Syntax: sticky route $variable . . . ;
Syntax: sticky learn create=$variable lookup=$variable zone=name:size
[timeout=time];
Default
Context: upstream
This directive appeared in version 1.5.7.

Enables session affinity, which causes requests from the same client to be
passed to the same server in a group of servers. Three methods are available:
cookie
When the cookie method is used, information about the designated
server is passed in an HTTP cookie generated by nginx:

upstream backend {
server backend1 . example . com ;
server backend2 . example . com ;

sticky cookie srv_id expires =1 h domain =. example . com path =/;

}

A request that comes from a client not yet bound to a particular server
is passed to the server selected by the configured balancing method.
Further requests with this cookie will be passed to the designated server.
If the designated server cannot process a request, the new server is
selected as if the client has not been bound yet.
The first parameter sets the name of the cookie to be set or inspected.
Additional parameters may be as follows:
expires=time
Sets the time for which a browser should keep the cookie.
The special value max will cause the cookie to expire on
31 Dec 2037 23:55:55 GMT. If the parameter is not specified,
it will cause the cookie to expire at the end of a browser session.
domain=domain
Defines the domain for which the cookie is set.
httponly
Adds the HttpOnly attribute to the cookie (1.7.11).
secure
Adds the Secure attribute to the cookie (1.7.11).
path=path
Defines the path for which the cookie is set.
If any parameters are omitted, the corresponding cookie fields are not
set.

Nginx, Inc. p.238 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

route
When the route method is used, proxied server assigns client a route
on receipt of the first request. All subsequent requests from this client
will carry routing information in a cookie or URI. This information is
compared with the route parameter of the server directive to identify
the server to which the request should be proxied. If the designated server
cannot process a request, the new server is selected by the configured
balancing method as if there is no routing information in the request.
The parameters of the route method specify variables that may contain
routing information. The first non-empty variable is used to find the
matching server.
Example:

map $ c o o k i e _ j s e s s i o n i d $route_cookie {
~.+\.(? P < route >\ w +) $ $route ;
}

map $request_uri $route_uri {

~ jsessionid =.+\.(? P < route >\ w +) $ $route ;
}

upstream backend {
server backend1 . example . com route = a ;
server backend2 . example . com route = b ;

sticky route $route_cookie $route_uri ;

}

Here, the route is taken from the JSESSIONID cookie if present in a

request. Otherwise, the route from the URI is used.
learn
When the learn method (1.7.1) is used, nginx analyzes upstream server
responses and learns server-initiated sessions usually passed in an HTTP
cookie.

upstream backend {
server backend1 . example . com :8080;
server backend2 . example . com :8081;

sticky learn
create = $ u p s t r e a m _ c o o k i e _ s e s s i o n i d
lookup = $ c o o k i e _ s e s s i o n i d
zone = cl ient _se ssio ns :1 m ;
}

In the example, the upstream server creates a session by setting the

cookie SESSIONID in the response. Further requests with this cookie
will be passed to the same server. If the server cannot process the request,
the new server is selected as if the client has not been bound yet.
The parameters create and lookup specify variables that indicate how
new sessions are created and existing sessions are searched, respectively.
Both parameters may be specified more than once, in which case the first
non-empty variable is used.
Sessions are stored in a shared memory zone, whose name and size are

Nginx, Inc. p.239 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

configured by the zone parameter. One megabyte zone can store about
8000 sessions on the 64-bit platform. The sessions that are not accessed
during the time specified by the timeout parameter get removed from
the zone. By default, timeout is set to 10 minutes.

This directive is available as part of our commercial subscription.

sticky cookie insert

Syntax: sticky_cookie_insert name [expires=time] [domain=domain]
[path=path];
Default
Context: upstream

This directive is obsolete since version 1.5.7. An equivalent sticky directive

with a new syntax should be used instead:

sticky cookie name [expires=time] [domain=domain] [path=path];

2.45.4 Embedded Variables

The ngx_http_upstream_module module supports the following embedded
variables:

$upstream addr
keeps the IP address and port, or the path to the UNIX-domain
socket of the upstream server. If several servers were contacted
during request processing, their addresses are separated by com-
mas, e.g. 192.168.1.1:80, 192.168.1.2:80, unix:/tmp/sock.
If an internal redirect from one server group to another hap-
pens, initiated by X-Accel-Redirect or error page, then the
server addresses from different groups are separated by colons, e.g.
192.168.1.1:80, 192.168.1.2:80, unix:/tmp/sock : 192.168.10.1:80, 192.16
$upstream cache status
keeps the status of accessing a response cache (0.8.3). The status
can be either MISS, BYPASS, EXPIRED, STALE, UPDATING,
REVALIDATED, or HIT.
$upstream cookie name
cookie with the specified name sent by the upstream server in the
Set-Cookie response header field (1.7.1). Only the cookies from the
response of the last server are saved.
$upstream header time
keeps time spent on receiving the response header from the upstream
server (1.7.10); the time is kept in seconds with millisecond resolution.
Times of several responses are separated by commas and colons like
addresses in the $upstream addr variable.

Nginx, Inc. p.240 of 330

CHAPTER 2. HTTP SERVER MODULES 2.45. MODULE NGX HTTP UPSTREAM MODULE

$upstream http name

keep server response header fields. For example, the Server response
header field is available through the $upstream http server variable. The
rules of converting header field names to variable names are the same
as for the variables that start with the $http prefix. Only the header
fields from the response of the last server are saved.
$upstream response length
keeps the length of the response obtained from the upstream server
(0.7.27); the length is kept in bytes. Lengths of several responses are
separated by commas and colons like addresses in the $upstream addr
variable.
$upstream response time
keeps time spent on receiving the response from the upstream server; the
time is kept in seconds with millisecond resolution. Times of several
responses are separated by commas and colons like addresses in the
$upstream addr variable.
$upstream status
keeps status code of the response obtained from the upstream server.
Status codes of several responses are separated by commas and colons
like addresses in the $upstream addr variable.

Nginx, Inc. p.241 of 330

CHAPTER 2. HTTP SERVER MODULES2.46. MODULE NGX HTTP UPSTREAM CONF MODULE

2.46 Module ngx http upstream conf module

2.46.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 242
2.46.2 Example Configuration . . . . . . . . . . . . . . . . . . . 242
2.46.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 242
upstream conf . . . . . . . . . . . . . . . . . . . . . . . . 242

2.46.1 Summary
The ngx_http_upstream_conf_module module allows configuring up-
stream server groups on-the-fly via a simple HTTP interface without the need
of restarting nginx. The http or stream server group must reside in the shared
memory.

2.46.2 Example Configuration

upstream backend {
zone up s tr e am _ ba c ke n d 64 k ;

...
}

server {
location / upstream_conf {
upstream_conf ;
allow 127.0.0.1;
deny all ;
}
}

2.46.3 Directives
upstream conf
Syntax: upstream_conf;
Default
Context: location

Turns on the HTTP interface of upstream configuration in the surrounding

location. Access to this location should be limited.
Configuration commands can be used to:

view the group configuration;

view, modify, or remove a server;

add a new server.

Since addresses in a group are not required to be unique, specific servers

in a group are referenced by their IDs. IDs are assigned automatically and
shown when adding a new server or viewing the group configuration.

Nginx, Inc. p.242 of 330

CHAPTER 2. HTTP SERVER MODULES2.46. MODULE NGX HTTP UPSTREAM CONF MODULE

A configuration command consists of parameters passed as request

arguments, for example:

http ://127.0.0.1/ upstream_conf ? upstream = backend

The following parameters are supported:

stream=
Selects a stream upstream server group. Without this parameter, selects
an http upstream server group.
upstream=name
Selects a group to work with. This parameter is mandatory.
id=number
Selects a server for viewing, modifying, or removing.
remove=
Removes a server from the group.
add=
Adds a new server to the group.
backup=
Required to add a backup server.

Before version 1.7.2, backup= was also required to view, modify, or

remove existing backup servers.

server=address
Same as the address parameter of the http or stream upstream server.
When adding a server, it is possible to specify it as a domain name.
In this case, changes of the IP addresses that correspond to a domain
name will be monitored and automatically applied to the upstream
configuration without the need of restarting nginx (1.7.2). This requires
the resolver directive in the http or stream block. See also the
resolve parameter of the http or stream upstream server.
weight=number
Same as the weight parameter of the http or stream upstream server.
max_conns=number
Same as the max_conns parameter of the http or stream upstream
server.
max_fails=number
Same as the max_fails parameter of the http or stream upstream
server.
fail_timeout=time
Same as the fail_timeout parameter of the http or stream upstream
server.
slow_start=time
Same as the slow_start parameter of the http or stream upstream
server.

Nginx, Inc. p.243 of 330

CHAPTER 2. HTTP SERVER MODULES2.46. MODULE NGX HTTP UPSTREAM CONF MODULE

down=
Same as the down parameter of the http or stream upstream server.
drain=
Puts the http upstream server in the draining mode (1.7.5). In this
mode, only requests of the bound to the server will be proxied to it.
up=
The opposite of the down parameter of the http or stream upstream
server.
route=string
Same as the route parameter of the http upstream server.

The first three parameters select an object. This can be either the whole
http or stream upstream server group, or a specific server. Without other
parameters, the configuration of the selected group or server is shown.
For example, to view the configuration of the whole group, send:

http ://127.0.0.1/ upstream_conf ? upstream = backend

To view the configuration of a specific server, also specify its ID:

http ://127.0.0.1/ upstream_conf ? upstream = backend & id =42

To add a new server, specify its address in the server= parameter.

Without other parameters specified, a server will be added with other
parameters set to their default values (see the http or stream server
directive).
For example, to add a new primary server, send:

http ://127.0.0.1/ upstream_conf ? add =& upstream = backend & server =12 7.0. 0.1: 808 0

To add a new backup server, send:

http ://127.0.0.1/ upstream_conf ? add =& upstream = backend & backup =& server
=1 27.0 .0.1 :808 0

To add a new primary server, set its parameters to non-default values and
mark it as down, send:

http ://127.0.0.1/ upstream_conf ? add =& upstream = backend & server

= 12 7 .0 . 0. 1 :8 0 80 & weight =2& down =

To remove a server, specify its ID:

http ://127.0.0.1/ upstream_conf ? remove =& upstream = backend & id =42

To mark an existing server as down, send:

http ://127.0.0.1/ upstream_conf ? upstream = backend & id =42& down =

Nginx, Inc. p.244 of 330

CHAPTER 2. HTTP SERVER MODULES2.46. MODULE NGX HTTP UPSTREAM CONF MODULE

To modify the address of an existing server, send:

http ://127.0.0.1/ upstream_conf ? upstream = backend & id =42& server

=1 92.0 .2.3 :812 3

To modify other parameters of an existing server, send:

http ://127.0.0.1/ upstream_conf ? upstream = backend & id =42& max_fails =3& weight =4

The above examples are for an http upstream server group. Similar
examples for a stream upstream server group require the stream= parameter.

Nginx, Inc. p.245 of 330

CHAPTER 2. HTTP SERVER MODULES 2.47. MODULE NGX HTTP USERID MODULE

2.47 Module ngx http userid module

2.47.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 246
2.47.2 Example Configuration . . . . . . . . . . . . . . . . . . . 246
2.47.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 246
userid . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
userid domain . . . . . . . . . . . . . . . . . . . . . . . . 247
userid expires . . . . . . . . . . . . . . . . . . . . . . . . 247
userid mark . . . . . . . . . . . . . . . . . . . . . . . . . 247
userid name . . . . . . . . . . . . . . . . . . . . . . . . . 247
userid p3p . . . . . . . . . . . . . . . . . . . . . . . . . . 248
userid path . . . . . . . . . . . . . . . . . . . . . . . . . 248
userid service . . . . . . . . . . . . . . . . . . . . . . . . 248
2.47.4 Embedded Variables . . . . . . . . . . . . . . . . . . . . 248

2.47.1 Summary
The ngx_http_userid_module module sets cookies suitable for client
identification. Received and set cookies can be logged using the embedded
variables $uid got and $uid set. This module is compatible with the mod uid
module for Apache.

2.47.2 Example Configuration

userid on ;
userid_name uid ;
userid_domain example . com ;
userid_path /;
userid_expires 365 d ;
userid_p3p policyref ="/ w3c / p3p . xml " , CP =" CUR ADM OUR NOR STA NID " ;

2.47.3 Directives
userid
Syntax: userid on | v1 | log | off;
Default off
Context: http, server, location

Enables or disables setting cookies and logging the received cookies:

on
enables the setting of version 2 cookies and logging of the received
cookies;
v1
enables the setting of version 1 cookies and logging of the received
cookies;

Nginx, Inc. p.246 of 330

CHAPTER 2. HTTP SERVER MODULES 2.47. MODULE NGX HTTP USERID MODULE

log
disables the setting of cookies, but enables logging of the received cookies;
off
disables the setting of cookies and logging of the received cookies.

userid domain
Syntax: userid_domain name | none;
Default none
Context: http, server, location

Defines a domain for which the cookie is set. The none parameter disables
setting of a domain for the cookie.

userid expires
Syntax: userid_expires time | max | off;
Default off
Context: http, server, location

Sets a time during which a browser should keep the cookie. The parameter
max will cause the cookie to expire on 31 Dec 2037 23:55:55 GMT. The
parameter off will cause the cookie to expire at the end of a browser session.

userid mark
Syntax: userid_mark letter | digit | = | off;
Default off
Context: http, server, location

If the parameter is not off, enables the cookie marking mechanism and sets
the character used as a mark. This mechanism is used to add or change userid -
p3p and/or a cookie expiration time while preserving the client identifier. A
mark can be any letter of the English alphabet (case-sensitive), digit, or the
= character.
If the mark is set, it is compared with the first padding symbol in the
base64 representation of the client identifier passed in a cookie. If they do not
match, the cookie is resent with the specified mark, expiration time, and P3P
header.

userid name
Syntax: userid_name name;
Default uid
Context: http, server, location

Sets the cookie name.

Nginx, Inc. p.247 of 330

CHAPTER 2. HTTP SERVER MODULES 2.47. MODULE NGX HTTP USERID MODULE

userid p3p
Syntax: userid_p3p string | none;
Default none
Context: http, server, location

Sets a value for the P3P header field that will be sent along with the cookie.
If the directive is set to the special value none, the P3P header will not be sent
in a response.

userid path
Syntax: userid_path path;
Default /
Context: http, server, location

Defines a path for which the cookie is set.

userid service
Syntax: userid_service number;
Default IP address of the server
Context: http, server, location

If identifiers are issued by multiple servers (services), each service should be

assigned its own number to ensure that client identifiers are unique. For version
1 cookies, the default value is zero. For version 2 cookies, the default value is
the number composed from the last four octets of the servers IP address.

2.47.4 Embedded Variables

The ngx_http_userid_module module supports the following embedded
variables:

$uid got
The cookie name and received client identifier.
$uid reset
If the variable is set to a non-empty string that is not 0, the client
identifiers are reset. The special value log additionally leads to the
output of messages about the reset identifiers to the error log.
$uid set
The cookie name and sent client identifier.

Nginx, Inc. p.248 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

2.48 Module ngx http uwsgi module

2.48.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 250
2.48.2 Example Configuration . . . . . . . . . . . . . . . . . . . 250
2.48.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 250
uwsgi bind . . . . . . . . . . . . . . . . . . . . . . . . . . 250
uwsgi buffer size . . . . . . . . . . . . . . . . . . . . . . 250
uwsgi buffering . . . . . . . . . . . . . . . . . . . . . . . 251
uwsgi buffers . . . . . . . . . . . . . . . . . . . . . . . . 251
uwsgi busy buffers size . . . . . . . . . . . . . . . . . . . 251
uwsgi cache . . . . . . . . . . . . . . . . . . . . . . . . . 252
uwsgi cache bypass . . . . . . . . . . . . . . . . . . . . . 252
uwsgi cache key . . . . . . . . . . . . . . . . . . . . . . . 252
uwsgi cache lock . . . . . . . . . . . . . . . . . . . . . . 252
uwsgi cache lock age . . . . . . . . . . . . . . . . . . . . 253
uwsgi cache lock timeout . . . . . . . . . . . . . . . . . 253
uwsgi cache methods . . . . . . . . . . . . . . . . . . . . 253
uwsgi cache min uses . . . . . . . . . . . . . . . . . . . . 253
uwsgi cache path . . . . . . . . . . . . . . . . . . . . . . 253
uwsgi cache purge . . . . . . . . . . . . . . . . . . . . . 254
uwsgi cache revalidate . . . . . . . . . . . . . . . . . . . 255
uwsgi cache use stale . . . . . . . . . . . . . . . . . . . . 255
uwsgi cache valid . . . . . . . . . . . . . . . . . . . . . . 256
uwsgi connect timeout . . . . . . . . . . . . . . . . . . . 256
uwsgi force ranges . . . . . . . . . . . . . . . . . . . . . 257
uwsgi hide header . . . . . . . . . . . . . . . . . . . . . . 257
uwsgi ignore client abort . . . . . . . . . . . . . . . . . . 257
uwsgi ignore headers . . . . . . . . . . . . . . . . . . . . 257
uwsgi intercept errors . . . . . . . . . . . . . . . . . . . 258
uwsgi limit rate . . . . . . . . . . . . . . . . . . . . . . . 258
uwsgi max temp file size . . . . . . . . . . . . . . . . . . 258
uwsgi modifier1 . . . . . . . . . . . . . . . . . . . . . . . 259
uwsgi modifier2 . . . . . . . . . . . . . . . . . . . . . . . 259
uwsgi next upstream . . . . . . . . . . . . . . . . . . . . 259
uwsgi next upstream timeout . . . . . . . . . . . . . . . 260
uwsgi next upstream tries . . . . . . . . . . . . . . . . . 260
uwsgi no cache . . . . . . . . . . . . . . . . . . . . . . . 260
uwsgi param . . . . . . . . . . . . . . . . . . . . . . . . 260
uwsgi pass . . . . . . . . . . . . . . . . . . . . . . . . . . 261
uwsgi pass header . . . . . . . . . . . . . . . . . . . . . 261
uwsgi pass request body . . . . . . . . . . . . . . . . . . 262
uwsgi pass request headers . . . . . . . . . . . . . . . . 262
uwsgi read timeout . . . . . . . . . . . . . . . . . . . . . 262
uwsgi request buffering . . . . . . . . . . . . . . . . . . . 262
uwsgi send timeout . . . . . . . . . . . . . . . . . . . . . 262
uwsgi ssl certificate . . . . . . . . . . . . . . . . . . . . . 263

Nginx, Inc. p.249 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

uwsgi ssl certificate key . . . . . . . . . . . . . . . . . . 263

uwsgi ssl ciphers . . . . . . . . . . . . . . . . . . . . . . 263
uwsgi ssl crl . . . . . . . . . . . . . . . . . . . . . . . . . 263
uwsgi ssl name . . . . . . . . . . . . . . . . . . . . . . . 263
uwsgi ssl password file . . . . . . . . . . . . . . . . . . . 264
uwsgi ssl protocols . . . . . . . . . . . . . . . . . . . . . 264
uwsgi ssl server name . . . . . . . . . . . . . . . . . . . 264
uwsgi ssl session reuse . . . . . . . . . . . . . . . . . . . 264
uwsgi ssl trusted certificate . . . . . . . . . . . . . . . . 265
uwsgi ssl verify . . . . . . . . . . . . . . . . . . . . . . . 265
uwsgi ssl verify depth . . . . . . . . . . . . . . . . . . . 265
uwsgi store . . . . . . . . . . . . . . . . . . . . . . . . . 265
uwsgi store access . . . . . . . . . . . . . . . . . . . . . 266
uwsgi temp file write size . . . . . . . . . . . . . . . . . 266
uwsgi temp path . . . . . . . . . . . . . . . . . . . . . . 266

2.48.1 Summary
The ngx_http_uwsgi_module module allows passing requests to a uwsgi
server.

2.48.2 Example Configuration

location / {
include uwsgi_params ;
uwsgi_pass localhost :9000;
}

2.48.3 Directives
uwsgi bind
Syntax: uwsgi_bind address | off;
Default
Context: http, server, location

Makes outgoing connections to a uwsgi server originate from the specified

local IP address. Parameter value can contain variables (1.3.12). The special
value off (1.3.12) cancels the effect of the uwsgi_bind directive inherited
from the previous configuration level, which allows the system to auto-assign
the local IP address.

uwsgi buffer size

Syntax: uwsgi_buffer_size size;
Default 4k|8k
Context: http, server, location

Nginx, Inc. p.250 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

Sets the size of the buffer used for reading the first part of the response
received from the uwsgi server. This part usually contains a small response
header. By default, the buffer size is equal to the size of one buffer set by the
uwsgi buffers directive. It can be made smaller, however.

uwsgi buffering
Syntax: uwsgi_buffering on | off;
Default on
Context: http, server, location

Enables or disables buffering of responses from the uwsgi server.

When buffering is enabled, nginx receives a response from the uwsgi server
as soon as possible, saving it into the buffers set by the uwsgi buffer size and
uwsgi buffers directives. If the whole response does not fit into memory, a part
of it can be saved to a temporary file on the disk. Writing to temporary files
is controlled by the uwsgi max temp file size and uwsgi temp file write size
directives.
When buffering is disabled, the response is passed to a client synchronously,
immediately as it is received. nginx will not try to read the whole response
from the uwsgi server. The maximum size of the data that nginx can receive
from the server at a time is set by the uwsgi buffer size directive.
Buffering can also be enabled or disabled by passing yes or no in the
X-Accel-Buffering response header field. This capability can be disabled
using the uwsgi ignore headers directive.

uwsgi buffers
Syntax: uwsgi_buffers number size;
Default 8 4k|8k
Context: http, server, location

Sets the number and size of the buffers used for reading a response from
the uwsgi server, for a single connection. By default, the buffer size is equal to
one memory page. This is either 4K or 8K, depending on a platform.

uwsgi busy buffers size

Syntax: uwsgi_busy_buffers_size size;
Default 8k|16k
Context: http, server, location

When buffering of responses from the uwsgi server is enabled, limits the
total size of buffers that can be busy sending a response to the client while the
response is not yet fully read. In the meantime, the rest of the buffers can be
used for reading the response and, if needed, buffering part of the response to
a temporary file. By default, size is limited by the size of two buffers set by
the uwsgi buffer size and uwsgi buffers directives.

Nginx, Inc. p.251 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

uwsgi cache
Syntax: uwsgi_cache zone | off;
Default off
Context: http, server, location

Defines a shared memory zone used for caching. The same zone can be
used in several places. Parameter value can contain variables (1.7.9). The off
parameter disables caching inherited from the previous configuration level.

uwsgi cache bypass

Syntax: uwsgi_cache_bypass string . . . ;
Default
Context: http, server, location

Defines conditions under which the response will not be taken from a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be taken from the cache:

u w s g i _ c a c h e _ b y p a s s $ coo kie_ noca che $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

u w s g i _ c a c h e _ b y p a s s $http_pragma $http_authorization ;

Can be used along with the uwsgi no cache directive.

uwsgi cache key

Syntax: uwsgi_cache_key string;
Default
Context: http, server, location

Defines a key for caching, for example

uw sgi_ cach e_ke y localhost :9000 $request_uri ;

uwsgi cache lock

Syntax: uwsgi_cache_lock on | off;
Default off
Context: http, server, location
This directive appeared in version 1.1.12.

When enabled, only one request at a time will be allowed to populate a new
cache element identified according to the uwsgi cache key directive by passing
a request to a uwsgi server. Other requests of the same cache element will
either wait for a response to appear in the cache or the cache lock for this
element to be released, up to the time set by the uwsgi cache lock timeout
directive.

Nginx, Inc. p.252 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

uwsgi cache lock age

Syntax: uwsgi_cache_lock_age time;
Default 5s
Context: http, server, location
This directive appeared in version 1.7.8.

If the last request passed to the uwsgi server for populating a new cache
element has not completed for the specified time, one more request may be
passed to the uwsgi server.

uwsgi cache lock timeout

Syntax: uwsgi_cache_lock_timeout time;
Default 5s
Context: http, server, location
This directive appeared in version 1.1.12.

Sets a timeout for uwsgi cache lock. When the time expires, the request
will be passed to the uwsgi server, however, the response will not be cached.

Before 1.7.8, the response could be cached.

uwsgi cache methods

Syntax: uwsgi_cache_methods GET | HEAD | POST . . . ;
Default GET HEAD
Context: http, server, location

If the client request method is listed in this directive then the response will
be cached. GET and HEAD methods are always added to the list, though it is
recommended to specify them explicitly. See also the uwsgi no cache directive.

uwsgi cache min uses

Syntax: uwsgi_cache_min_uses number;
Default 1
Context: http, server, location

Sets the number of requests after which the response will be cached.

uwsgi cache path

Syntax: uwsgi_cache_path path [levels=levels] [use_temp_path=on|off]
keys_zone=name:size [inactive=time] [max_size=size]
[loader_files=number] [loader_sleep=time]
[loader_threshold=time];
Default
Context: http

Nginx, Inc. p.253 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

Sets the path and other parameters of a cache. Cache data are stored in
files. The file name in a cache is a result of applying the MD5 function to
the cache key. The levels parameter defines hierarchy levels of a cache. For
example, in the following configuration

u ws g i_ c ac h e_ p at h / data / nginx / cache levels =1:2 keys_zone = one :10 m ;

file names in a cache will look like this:

/ data / nginx / cache /c /29 / b 7 f 5 4 b 2 d f 7 7 7 3 7 2 2 d 3 8 2 f 4 8 0 9 d 6 5 0 29c

A cached response is first written to a temporary file, and then the file is
renamed. Starting from version 0.8.9, temporary files and the cache can be put
on different file systems. However, be aware that in this case a file is copied
across two file systems instead of the cheap renaming operation. It is thus
recommended that for any given location both cache and a directory holding
temporary files are put on the same file system. A directory for temporary
files is set based on the use_temp_path parameter (1.7.10). If this parameter
is omitted or set to the value on, the directory set by the uwsgi temp path
directive for the given location will be used. If the value is set to off, temporary
files will be put directly in the cache directory.
In addition, all active keys and information about data are stored in a
shared memory zone, whose name and size are configured by the keys_zone
parameter. One megabyte zone can store about 8 thousand keys.
Cached data that are not accessed during the time specified by the
inactive parameter get removed from the cache regardless of their freshness.
By default, inactive is set to 10 minutes.
The special cache manager process monitors the maximum cache size set
by the max_size parameter. When this size is exceeded, it removes the least
recently used data.
A minute after the start the special cache loader process is activated. It
loads information about previously cached data stored on file system into a
cache zone. The loading is done in iterations. During one iteration no more
than loader_files items are loaded (by default, 100). Besides, the duration of
one iteration is limited by the loader_threshold parameter (by default, 200
milliseconds). Between iterations, a pause configured by the loader_sleep
parameter (by default, 50 milliseconds) is made.

uwsgi cache purge

Syntax: uwsgi_cache_purgestring . . . ;
Default
Context: http, server, location
This directive appeared in version 1.5.7.

Defines conditions under which the request will be considered a cache purge
request. If at least one value of the string parameters is not empty and
is not equal to 0 then the cache entry with a corresponding cache key is

Nginx, Inc. p.254 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

removed. The result of successful operation is indicated by returning the 204

No Content response.
If the cache key of a purge request ends with an asterisk (*), all cache
entries matching the wildcard key will be removed from the cache.
Example configuration:

u ws g i_ c ac h e_ p at h / data / nginx / cache keys_zone = cache_zone :10 m ;

map $re ques t_m etho d $purge_method {

PURGE 1;
default 0;
}

server {
...
location / {
uwsgi_pass backend ;
uwsgi_cache cache_zone ;
uws gi_ cach e_ke y $uri ;
uwsgi_cache_purge $purge_method ;
}
}

This functionality is available as part of our commercial subscription.

uwsgi cache revalidate

Syntax: uwsgi_cache_revalidate on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.7.

Enables revalidation of expired cache items using conditional requests with

the If-Modified-Since and If-None-Match header fields.

uwsgi cache use stale

Syntax: uwsgi_cache_use_stale error | timeout | invalid_header | updating
| http_500 | http_503 | http_403 | http_404 | off . . . ;
Default off
Context: http, server, location

Determines in which cases a stale cached response can be used when an

error occurs during communication with the uwsgi server. The directives
parameters match the parameters of the uwsgi next upstream directive.
Additionally, the updating parameter permits using a stale cached response
if it is currently being updated. This allows minimizing the number of accesses
to uwsgi servers when updating cached data.
To minimize the number of accesses to uwsgi servers when populating a
new cache element, the uwsgi cache lock directive can be used.

Nginx, Inc. p.255 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

uwsgi cache valid

Syntax: uwsgi_cache_valid [code . . . ] time;
Default
Context: http, server, location

Sets caching time for different response codes. For example, the following
directives

u w s g i _ c a c h e _ v a l i d 200 302 10 m ;
u w s g i _ c a c h e _ v a l i d 404 1m;

set 10 minutes of caching for responses with codes 200 and 302 and 1 minute
for responses with code 404.
If only caching time is specified

uwsgi_cache_valid 5m;

then only 200, 301, and 302 responses are cached.

In addition, the any parameter can be specified to cache any responses:

u w s g i _ c a c h e _ v a l i d 200 302 10 m ;
u w s g i _ c a c h e _ v a l i d 301 1h;
u w s g i _ c a c h e _ v a l i d any 1m;

Parameters of caching can also be set directly in the response header. This
has higher priority than setting of caching time using the directive.
The X-Accel-Expires header field sets caching time of a response in
seconds. The zero value disables caching for a response. If the value
starts with the @ prefix, it sets an absolute time in seconds since Epoch,
up to which the response may be cached.

If the header does not include the X-Accel-Expires field, parameters of

caching may be set in the header fields Expires or Cache-Control.

If the header includes the Set-Cookie field, such a response will not be
cached.

If the header includes the Vary field with the special value *, such a
response will not be cached (1.7.7). If the header includes the Vary field
with another value, such a response will be cached taking into account
the corresponding request header fields (1.7.7).
Processing of one or more of these response header fields can be disabled using
the uwsgi ignore headers directive.

uwsgi connect timeout

Syntax: uwsgi_connect_timeout time;
Default 60s
Context: http, server, location

Nginx, Inc. p.256 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

Defines a timeout for establishing a connection with a uwsgi server. It

should be noted that this timeout cannot usually exceed 75 seconds.

uwsgi force ranges

Syntax: uwsgi_force_ranges on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.7.

Enables byte-range support for both cached and uncached responses from
the uwsgi server regardless of the Accept-Ranges field in these responses.

uwsgi hide header

Syntax: uwsgi_hide_header field;
Default
Context: http, server, location

By default, nginx does not pass the header fields Status and X-Accel-...
from the response of a uwsgi server to a client. The uwsgi_hide_header
directive sets additional fields that will not be passed. If, on the contrary, the
passing of fields needs to be permitted, the uwsgi pass header directive can be
used.

uwsgi ignore client abort

Syntax: uwsgi_ignore_client_abort on | off;
Default off
Context: http, server, location

Determines whether the connection with a uwsgi server should be closed

when a client closes the connection without waiting for a response.

uwsgi ignore headers

Syntax: uwsgi_ignore_headers field . . . ;
Default
Context: http, server, location

Disables processing of certain response header fields from the uwsgi server.
The following fields can be ignored: X-Accel-Redirect, X-Accel-Expires,
X-Accel-Limit-Rate (1.1.6), X-Accel-Buffering (1.1.6), X-Accel-Charset
(1.1.6), Expires, Cache-Control, Set-Cookie (0.8.44), and Vary (1.7.7).
If not disabled, processing of these header fields has the following effect:

X-Accel-Expires, Expires, Cache-Control, Set-Cookie, and Vary set

the parameters of response caching;

X-Accel-Redirect performs an internal redirect to the specified URI;

Nginx, Inc. p.257 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

X-Accel-Limit-Rate sets the rate limit for transmission of a response

to a client;

X-Accel-Buffering enables or disables buffering of a response;

X-Accel-Charset sets the desired charset of a response.

uwsgi intercept errors

Syntax: uwsgi_intercept_errors on | off;
Default off
Context: http, server, location

Determines whether a uwsgi server responses with codes greater than or

equal to 300 should be passed to a client or be redirected to nginx for processing
with the error page directive.

uwsgi limit rate

Syntax: uwsgi_limit_rate rate;
Default 0
Context: http, server, location
This directive appeared in version 1.7.7.

Limits the speed of reading the response from the uwsgi server. The rate is
specified in bytes per second. The zero value disables rate limiting. The limit
is set per a request, and so if nginx simultaneously opens two connections to
the uwsgi server, the overall rate will be twice as much as the specified limit.
The limitation works only if buffering of responses from the uwsgi server is
enabled.

uwsgi max temp file size

Syntax: uwsgi_max_temp_file_size size;
Default 1024m
Context: http, server, location

When buffering of responses from the uwsgi server is enabled, and the whole
response does not fit into the buffers set by the uwsgi buffer size and uwsgi -
buffers directives, a part of the response can be saved to a temporary file.
This directive sets the maximum size of the temporary file. The size of data
written to the temporary file at a time is set by the uwsgi temp file write size
directive.
The zero value disables buffering of responses to temporary files.

This restriction does not apply to responses that will be cached or stored
on disk.

Nginx, Inc. p.258 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

uwsgi modifier1
Syntax: uwsgi_modifier1 number;
Default 0
Context: http, server, location

Sets the value of the modifier1 field in the uwsgi packet header.

uwsgi modifier2
Syntax: uwsgi_modifier2 number;
Default 0
Context: http, server, location

Sets the value of the modifier2 field in the uwsgi packet header.

uwsgi next upstream

Syntax: uwsgi_next_upstream error | timeout | invalid_header | http_500 |
http_503 | http_403 | http_404 | off . . . ;
Default error timeout
Context: http, server, location

Specifies in which cases a request should be passed to the next server:

error
an error occurred while establishing a connection with the server, passing
a request to it, or reading the response header;
timeout
a timeout has occurred while establishing a connection with the server,
passing a request to it, or reading the response header;
invalid_header
a server returned an empty or invalid response;
http_500
a server returned a response with the code 500;
http_503
a server returned a response with the code 503;
http_403
a server returned a response with the code 403;
http_404
a server returned a response with the code 404;
off
disables passing a request to the next server.

One should bear in mind that passing a request to the next server is only
possible if nothing has been sent to a client yet. That is, if an error or timeout
occurs in the middle of the transferring of a response, fixing this is impossible.
The directive also defines what is considered an unsuccessful attempt
of communication with a server. The cases of error, timeout and

Nginx, Inc. p.259 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

invalid_header are always considered unsuccessful attempts, even if they

are not specified in the directive. The cases of http_500 and http_503 are
considered unsuccessful attempts only if they are specified in the directive. The
cases of http_403 and http_404 are never considered unsuccessful attempts.
Passing a request to the next server can be limited by the number of tries
and by time.

uwsgi next upstream timeout

Syntax: uwsgi_next_upstream_timeout time;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the time allowed to pass a request to the next server. The 0 value
turns off this limitation.

uwsgi next upstream tries

Syntax: uwsgi_next_upstream_tries number;
Default 0
Context: http, server, location
This directive appeared in version 1.7.5.

Limits the number of possible tries for passing a request to the next server.
The 0 value turns off this limitation.

uwsgi no cache
Syntax: uwsgi_no_cache string . . . ;
Default
Context: http, server, location

Defines conditions under which the response will not be saved to a cache.
If at least one value of the string parameters is not empty and is not equal to
0 then the response will not be saved:

uwsgi_no_cache $ cook ie_n oca che $ a r g _ n o c a c h e $ a r g _ c o m m e n t ;

uwsgi_no_cache $http_pragma $http_authorization ;

Can be used along with the uwsgi cache bypass directive.

uwsgi param
Syntax: uwsgi_param parameter value [if_not_empty];
Default
Context: http, server, location

Sets a parameter that should be passed to the uwsgi server. The value can
contain text, variables, and their combination. These directives are inherited

Nginx, Inc. p.260 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

from the previous level if and only if there are no uwsgi_param directives
defined on the current level.
Standard CGI environment variables should be provided as uwsgi headers,
see the uwsgi_params file provided in the distribution:

location / {
include uwsgi_params ;
...
}

If a directive is specified with if_not_empty (1.1.11) then such a parameter

will not be passed to the server until its value is not empty:

uwsgi_param HTTPS $https if_not_empty ;

uwsgi pass
Syntax: uwsgi_pass [protocol://]address;
Default
Context: location, if in location

Sets the protocol and address of a uwsgi server. As a protocol, uwsgi or

suwsgi (secured uwsgi, uwsgi over SSL) can be specified. The address can
be specified as a domain name or IP address, and an optional port:

uwsgi_pass localhost :9000;

uwsgi_pass uwsgi :// localhost :9000;
uwsgi_pass suwsgi ://[2001: db8 ::1]:9090;

or as a UNIX-domain socket path:

uwsgi_pass unix :/ tmp / uwsgi . socket ;

If a domain name resolves to several addresses, all of them will be used

in a round-robin fashion. In addition, an address can be specified as a server
group.

Secured uwsgi protocol is supported since version 1.5.8.

uwsgi pass header

Syntax: uwsgi_pass_header field;
Default
Context: http, server, location

Permits passing otherwise disabled header fields from a uwsgi server to a

client.

Nginx, Inc. p.261 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

uwsgi pass request body

Syntax: uwsgi_pass_request_body on | off;
Default on
Context: http, server, location

Indicates whether the original request body is passed to the uwsgi server.
See also the uwsgi pass request headers directive.

uwsgi pass request headers

Syntax: uwsgi_pass_request_headers on | off;
Default on
Context: http, server, location

Indicates whether the header fields of the original request are passed to the
uwsgi server. See also the uwsgi pass request body directive.

uwsgi read timeout

Syntax: uwsgi_read_timeout time;
Default 60s
Context: http, server, location

Defines a timeout for reading a response from the uwsgi server. The timeout
is set only between two successive read operations, not for the transmission of
the whole response. If the uwsgi server does not transmit anything within this
time, the connection is closed.

uwsgi request buffering

Syntax: uwsgi_request_buffering on | off;
Default on
Context: http, server, location
This directive appeared in version 1.7.11.

Enables or disables buffering of a client request body.

When buffering is enabled, the entire request body is read from the client
before sending the request to a uwsgi server.
When buffering is disabled, the request body is sent to the uwsgi server
immediately as it is received. In this case, the request cannot be passed to the
next server if nginx already started sending the request body.
When HTTP/1.1 chunked transfer encoding is used to send the original
request body, the request body will be buffered regardless of the directive
value.

uwsgi send timeout

Syntax: uwsgi_send_timeout time;
Default 60s
Context: http, server, location

Nginx, Inc. p.262 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

Sets a timeout for transmitting a request to the uwsgi server. The timeout
is set only between two successive write operations, not for the transmission
of the whole request. If the uwsgi server does not receive anything within this
time, the connection is closed.

uwsgi ssl certificate

Syntax: uwsgi_ssl_certificate file;
Default
Context: http, server, location
This directive appeared in version 1.7.8.

Specifies a file with the certificate in the PEM format used for
authentication to a secured uwsgi server.

uwsgi ssl certificate key

Syntax: uwsgi_ssl_certificate_key file;
Default
Context: http, server, location
This directive appeared in version 1.7.8.

Specifies a file with the secret key in the PEM format used for
authentication to a secured uwsgi server.

uwsgi ssl ciphers

Syntax: uwsgi_ssl_ciphers ciphers;
Default DEFAULT
Context: http, server, location
This directive appeared in version 1.5.8.

Specifies the enabled ciphers for requests to a secured uwsgi server. The
ciphers are specified in the format understood by the OpenSSL library.
The full list can be viewed using the openssl ciphers command.

uwsgi ssl crl

Syntax: uwsgi_ssl_crl file;
Default
Context: http, server, location
This directive appeared in version 1.7.0.

Specifies a file with revoked certificates (CRL) in the PEM format used to
verify the certificate of the secured uwsgi server.

uwsgi ssl name

Syntax: uwsgi_ssl_name name;
Default host from uwsgi_pass
Context: http, server, location

Nginx, Inc. p.263 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

This directive appeared in version 1.7.0.

Allows overriding the server name used to verify the certificate of the
secured uwsgi server and to be passed through SNI when establishing a
connection with the secured uwsgi server.
By default, the host part from uwsgi pass is used.

uwsgi ssl password file

Syntax: uwsgi_ssl_password_file file;
Default
Context: http, server, location
This directive appeared in version 1.7.8.

Specifies a file with passphrases for secret keys where each passphrase is
specified on a separate line. Passphrases are tried in turn when loading the
key.

uwsgi ssl protocols

Syntax: uwsgi_ssl_protocols [SSLv2] [SSLv3] [TLSv1] [TLSv1.1] [TLSv1.2];
Default SSLv3 TLSv1 TLSv1.1 TLSv1.2
Context: http, server, location
This directive appeared in version 1.5.8.

Enables the specified protocols for requests to a secured uwsgi server.

uwsgi ssl server name

Syntax: uwsgi_ssl_server_name on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.0.

Enables or disables passing of the server name through TLS Server Name
Indication extension (SNI, RFC 6066) when establishing a connection with the
secured uwsgi server.

uwsgi ssl session reuse

Syntax: uwsgi_ssl_session_reuse on | off;
Default on
Context: http, server, location
This directive appeared in version 1.5.8.

Determines whether SSL sessions can be reused when working with a se-
cured uwsgi server. If the errorsSSL3_GET_FINISHED:digest check failed
appear in the logs, try disabling session reuse.

Nginx, Inc. p.264 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

uwsgi ssl trusted certificate

Syntax: uwsgi_ssl_trusted_certificate file;
Default
Context: http, server, location
This directive appeared in version 1.7.0.

Specifies a file with trusted CA certificates in the PEM format used to

verify the certificate of the secured uwsgi server.

uwsgi ssl verify

Syntax: uwsgi_ssl_verify on | off;
Default off
Context: http, server, location
This directive appeared in version 1.7.0.

Enables or disables verification of the secured uwsgi server certificate.

uwsgi ssl verify depth

Syntax: uwsgi_ssl_verify_depth number;
Default 1
Context: http, server, location
This directive appeared in version 1.7.0.

Sets the verification depth in the secured uwsgi server certificates chain.

uwsgi store
Syntax: uwsgi_store on | off | string;
Default off
Context: http, server, location

Enables saving of files to a disk. The on parameter saves files with paths
corresponding to the directives alias or root. The off parameter disables saving
of files. In addition, the file name can be set explicitly using the string with
variables:

uwsgi_store / data / w w w$ or i gi n al _ ur i ;

The modification time of files is set according to the received

Last-Modified response header field. The response is first written to a
temporary file, and then the file is renamed. Starting from version 0.8.9,
temporary files and the persistent store can be put on different file systems.
However, be aware that in this case a file is copied across two file systems
instead of the cheap renaming operation. It is thus recommended that for any
given location both saved files and a directory holding temporary files, set by
the uwsgi temp path directive, are put on the same file system.
This directive can be used to create local copies of static unchangeable files,
e.g.:

Nginx, Inc. p.265 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

location / images / {
root / data / www ;
error_page 404 = / fetch$uri ;
}

location / fetch / {
internal ;

uwsgi_pass backend :9000;

...

uwsgi_store on ;
u w s g i _ s t o r e _ a c c e s s user : rw group : rw all : r ;
uw sgi_ temp _pat h / data / temp ;

alias / data / www /;

}

uwsgi store access

Syntax: uwsgi_store_access users:permissions . . . ;
Default user:rw
Context: http, server, location

Sets access permissions for newly created files and directories, e.g.:

u w s g i _ s t o r e _ a c c e s s user : rw group : rw all : r ;

If any group or all access permissions are specified then user permissions
may be omitted:

u w s g i _ s t o r e _ a c c e s s group : rw all : r ;

uwsgi temp file write size

Syntax: uwsgi_temp_file_write_size size;
Default 8k|16k
Context: http, server, location

Limits the size of data written to a temporary file at a time, when buffering
of responses from the uwsgi server to temporary files is enabled. By default,
size is limited by two buffers set by the uwsgi buffer size and uwsgi buffers
directives. The maximum size of a temporary file is set by the uwsgi max -
temp file size directive.

uwsgi temp path

Syntax: uwsgi_temp_path path [level1 [level2 [level3]]];
Default uwsgi_temp
Context: http, server, location

Nginx, Inc. p.266 of 330

CHAPTER 2. HTTP SERVER MODULES 2.48. MODULE NGX HTTP UWSGI MODULE

Defines a directory for storing temporary files with data received from uwsgi
servers. Up to three-level subdirectory hierarchy can be used underneath the
specified directory. For example, in the following configuration

uw sgi_ temp _pat h / spool / nginx / uwsgi_temp 1 2;

a temporary file might look like this:

/ spool / nginx / uwsgi_temp /7 /45 /00000123457

See also the use_temp_path parameter of the uwsgi cache path directive.

Nginx, Inc. p.267 of 330

CHAPTER 2. HTTP SERVER MODULES 2.49. MODULE NGX HTTP XSLT MODULE

2.49 Module ngx http xslt module

2.49.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 268
2.49.2 Example Configuration . . . . . . . . . . . . . . . . . . . 268
2.49.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 268
xml entities . . . . . . . . . . . . . . . . . . . . . . . . . 268
xslt last modified . . . . . . . . . . . . . . . . . . . . . . 269
xslt param . . . . . . . . . . . . . . . . . . . . . . . . . . 269
xslt string param . . . . . . . . . . . . . . . . . . . . . . 269
xslt stylesheet . . . . . . . . . . . . . . . . . . . . . . . . 269
xslt types . . . . . . . . . . . . . . . . . . . . . . . . . . 270

2.49.1 Summary
The ngx_http_xslt_module (0.7.8+) is a filter that transforms XML
responses using one or more XSLT stylesheets.
This module is not built by default, it should be enabled with the
--with-http_xslt_module configuration parameter.

This module requires the libxml2 and libxslt libraries.

2.49.2 Example Configuration

location / {
xml_entities / site / dtd / entities . dtd ;
xs lt_s tyle shee t / site / xslt / one . xslt param = value ;
xs lt_s tyle shee t / site / xslt / two . xslt ;
}

2.49.3 Directives
xml entities
Syntax: xml_entities path;
Default
Context: http, server, location

Specifies the DTD file that declares character entities. This file is compiled
at the configuration stage. For technical reasons, the module is unable to
use the external subset declared in the processed XML, so it is ignored and a
specially defined file is used instead. This file should not describe the XML
structure. It is enough to declare just the required character entities, for
example:

<! ENTITY nbsp "&# xa0 ;" >

Nginx, Inc. p.268 of 330

CHAPTER 2. HTTP SERVER MODULES 2.49. MODULE NGX HTTP XSLT MODULE

xslt last modified

Syntax: xslt_last_modified on | off;
Default off
Context: http, server, location
This directive appeared in version 1.5.1.

Allows preserving the Last-Modified header field from the original

response during XSLT transformations to facilitate response caching.
By default, the header field is removed as contents of the response are
modified during transformations and may contain dynamically generated
elements or parts that are changed independently of the original response.

xslt param
Syntax: xslt_param parameter value;
Default
Context: http, server, location
This directive appeared in version 1.1.18.

Defines the parameters for XSLT stylesheets. The value is treated as an

XPath expression. The value can contain variables. To pass a string value to
a stylesheet, the xslt string param directive can be used.
There could be several xslt_param directives. These directives are
inherited from the previous level if and only if there are no xslt_param and
xslt string param directives defined on the current level.

xslt string param

Syntax: xslt_string_param parameter value;
Default
Context: http, server, location
This directive appeared in version 1.1.18.

Defines the string parameters for XSLT stylesheets. XPath expressions in

the value are not interpreted. The value can contain variables.
There could be several xslt_string_param directives. These directives
are inherited from the previous level if and only if there are no xslt param and
xslt_string_param directives defined on the current level.

xslt stylesheet
Syntax: xslt_stylesheet stylesheet [parameter=value . . . ];
Default
Context: location

Defines the XSLT stylesheet and its optional parameters. A stylesheet is

compiled at the configuration stage.
Parameters can either be specified separately, or grouped in a single line
using the : delimiter. If a parameter includes the : character, it should be

Nginx, Inc. p.269 of 330

CHAPTER 2. HTTP SERVER MODULES 2.49. MODULE NGX HTTP XSLT MODULE

escaped as %3A. Also, libxslt requires to enclose parameters that contain

non-alphanumeric characters into single or double quotes, for example:

param1 = http %3 A // www . example . com : param2 = value2

The parameters description can contain variables, for example, the whole
line of parameters can be taken from a single variable:

location / {
xs lt_s tyle shee t / site / xslt / one . xslt
$ ar g _x s lt _ pa ra m s
param1 = $value1 : param2 = value2
param3 = value3 ;
}

It is possible to specify several stylesheets. They will be applied sequentially

in the specified order.

xslt types
Syntax: xslt_types mime-type . . . ;
Default text/xml
Context: http, server, location

Enables transformations in responses with the specified MIME types in

addition to text/xml. The special value * matches any MIME type
(0.8.29). If the transformation result is an HTML response, its MIME type is
changed to text/html.

Nginx, Inc. p.270 of 330

Chapter 3

Stream proxy modules

3.1 Module ngx stream core module

3.1.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 271
3.1.2 Example Configuration . . . . . . . . . . . . . . . . . . . 271
3.1.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 272
listen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
resolver . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
resolver timeout . . . . . . . . . . . . . . . . . . . . . . . 273
server . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
stream . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274

3.1.1 Summary

This module is available as part of our commercial subscription.

3.1.2 Example Configuration

w or k er _ pr o ce s se s auto ;

error_log / var / log / nginx / error . log info ;

stream {
upstream backend {
hash $remote_addr consistent ;

server backend1 . example . com :12345 weight =5;

server 12 7.0. 0.1: 123 45 max_fails =3 fail_timeout =30 s ;
server unix :/ tmp / backend3 ;
}

server {
listen 12345;
proxy_connect_timeout 1s;
proxy_timeout 3 s ;
proxy_pass backend ;
}

server {
listen [::1]:12345;
proxy_pass unix :/ tmp / stream . socket ;

271
CHAPTER 3. STREAM PROXY MODULES 3.1. MODULE NGX STREAM CORE MODULE

}
}

3.1.3 Directives
listen
Syntax: listen address:port [ssl] [bind] [ipv6only=on|off]
[so_keepalive=on|off|[keepidle]:[keepintvl]:[keepcnt]];
Default
Context: server

Sets the address and port for the socket on which the server will accept
connections. It is possible to specify just the port. The address can also be a
hostname, for example:

listen 12 7 .0 .0 . 1: 1 23 4 5;
listen *:12345;
listen 12345; # same as *:12345
listen localhost :12345;

IPv6 addresses are specified in square brackets:

listen [::1]:12345;
listen [::]:12345;

UNIX-domain sockets are specified with the unix: prefix:

listen unix :/ var / run / nginx . sock ;

The ssl parameter (1.7.10) allows specifying that all connections accepted
on this port should work in SSL mode.
The listen directive can have several additional parameters specific to
socket-related system calls.

bind
this parameter instructs to make a separate bind call for a given
address:port pair. The fact is that if there are several listen directives
with the same port but different addresses, and one of the listen
directives listens on all addresses for the given port (*:port), nginx will
bind only to *:port. It should be noted that the getsockname system
call will be made in this case to determine the address that accepted the
connection. If the ipv6only or so_keepalive parameters are used then
for a given address:port pair a separate bind call will always be made.
ipv6only=on|off
this parameter determines (via the IPV6_V6ONLY socket option) whether
an IPv6 socket listening on a wildcard address [::] will accept only
IPv6 connections or both IPv6 and IPv4 connections. This parameter is
turned on by default. It can only be set once on start.

Nginx, Inc. p.272 of 330

CHAPTER 3. STREAM PROXY MODULES 3.1. MODULE NGX STREAM CORE MODULE

so_keepalive=on|off|[keepidle]:[keepintvl]:[keepcnt]
this parameter configures the TCP keepalive behavior for the listening
socket. If this parameter is omitted then the operating systems settings
will be in effect for the socket. If it is set to the value on, the
SO_KEEPALIVE option is turned on for the socket. If it is set to the
value off, the SO_KEEPALIVE option is turned off for the socket. Some
operating systems support setting of TCP keepalive parameters on a per-
socket basis using the TCP_KEEPIDLE, TCP_KEEPINTVL, and TCP_KEEPCNT
socket options. On such systems (currently, Linux 2.4+, NetBSD 5+,
and FreeBSD 9.0-STABLE), they can be configured using the keepidle,
keepintvl, and keepcnt parameters. One or two parameters may be
omitted, in which case the system default setting for the corresponding
socket option will be in effect. For example,
so_keepalive =30 m ::10

will set the idle timeout (TCP_KEEPIDLE) to 30 minutes, leave the probe
interval (TCP_KEEPINTVL) at its system default, and set the probes count
(TCP_KEEPCNT) to 10 probes.
Different servers must listen on different address:port pairs.

resolver
Syntax: resolver address . . . [valid=time] [ipv6=on|off];
Default
Context: stream, server
This directive appeared in version 1.7.10.

Configures name servers used to resolve names of upstream servers into

addresses, for example:

resolver 127.0.0.1 [::1]:5353;

An address can be specified as a domain name or IP address, and an

optional port. If port is not specified, the port 53 is used. Name servers
are queried in a round-robin fashion.
By default, nginx will look up both IPv4 and IPv6 addresses while resolving.
If looking up of IPv6 addresses is not desired, the ipv6=off parameter can be
specified.
By default, nginx caches answers using the TTL value of a response. The
optional valid parameter allows overriding it:

resolver 127.0.0.1 [::1]:5353 valid =30 s ;

resolver timeout
Syntax: resolver_timeout time;
Default 30s
Context: stream, server

Nginx, Inc. p.273 of 330

CHAPTER 3. STREAM PROXY MODULES 3.1. MODULE NGX STREAM CORE MODULE

This directive appeared in version 1.7.10.

Sets a timeout for name resolution, for example:

r es o lv e r_ t im e ou t 5 s ;

server
Syntax: server { . . . }
Default
Context: stream

Sets the configuration for a server.

stream
Syntax: stream { . . . }
Default
Context: main

Provides the configuration file context in which the stream server directives
are specified.

Nginx, Inc. p.274 of 330

CHAPTER 3. STREAM PROXY MODULES 3.2. MODULE NGX STREAM PROXY MODULE

3.2 Module ngx stream proxy module

3.2.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 275
3.2.2 Example Configuration . . . . . . . . . . . . . . . . . . . 275
3.2.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 276
proxy connect timeout . . . . . . . . . . . . . . . . . . . 276
proxy downstream buffer . . . . . . . . . . . . . . . . . . 276
proxy next upstream . . . . . . . . . . . . . . . . . . . . 276
proxy next upstream timeout . . . . . . . . . . . . . . . 276
proxy next upstream tries . . . . . . . . . . . . . . . . . 276
proxy pass . . . . . . . . . . . . . . . . . . . . . . . . . . 277
proxy ssl . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
proxy ssl certificate . . . . . . . . . . . . . . . . . . . . . 277
proxy ssl certificate key . . . . . . . . . . . . . . . . . . 277
proxy ssl ciphers . . . . . . . . . . . . . . . . . . . . . . 278
proxy ssl crl . . . . . . . . . . . . . . . . . . . . . . . . . 278
proxy ssl name . . . . . . . . . . . . . . . . . . . . . . . 278
proxy ssl password file . . . . . . . . . . . . . . . . . . . 278
proxy ssl server name . . . . . . . . . . . . . . . . . . . 278
proxy ssl session reuse . . . . . . . . . . . . . . . . . . . 279
proxy ssl protocols . . . . . . . . . . . . . . . . . . . . . 279
proxy ssl trusted certificate . . . . . . . . . . . . . . . . 279
proxy ssl verify . . . . . . . . . . . . . . . . . . . . . . . 279
proxy ssl verify depth . . . . . . . . . . . . . . . . . . . 279
proxy timeout . . . . . . . . . . . . . . . . . . . . . . . . 280
proxy upstream buffer . . . . . . . . . . . . . . . . . . . 280

3.2.1 Summary
The ngx_stream_proxy_module module (1.7.7) allows passing connections
to another server over TCP and UNIX-domain sockets.

This module is available as part of our commercial subscription.

3.2.2 Example Configuration

server {
listen 12 7 .0 . 0. 1: 1 23 4 5;
proxy_pass 1 27.0 .0.1 :808 0;
}

server {
listen 12345;
proxy_connect_timeout 1s;
proxy_timeout 3 s ;
proxy_pass example . com :12345;
}

server {

Nginx, Inc. p.275 of 330

CHAPTER 3. STREAM PROXY MODULES 3.2. MODULE NGX STREAM PROXY MODULE

listen [::1]:12345;
proxy_pass unix :/ tmp / stream . socket ;
}

3.2.3 Directives
proxy connect timeout
Syntax: proxy_connect_timeout time;
Default 60s
Context: stream, server

Defines a timeout for establishing a connection with a proxied server.

proxy downstream buffer

Syntax: proxy_downstream_buffer size;
Default 16k
Context: stream, server

Sets the size of the buffer used for reading data from the client.

proxy next upstream

Syntax: proxy_next_upstream on | off;
Default on
Context: stream, server
This directive appeared in version 1.7.10.

When a connection to the proxied server cannot be established, determines

whether a client connection will be passed to the next server.
Passing a connection to the next server can be limited by the number of
tries and by time.

proxy next upstream timeout

Syntax: proxy_next_upstream_timeout time;
Default 0
Context: stream, server
This directive appeared in version 1.7.10.

Limits the time allowed to pass a connection to the next server. The 0
value turns off this limitation.

proxy next upstream tries

Syntax: proxy_next_upstream_tries number;
Default 0
Context: stream, server
This directive appeared in version 1.7.10.

Nginx, Inc. p.276 of 330

CHAPTER 3. STREAM PROXY MODULES 3.2. MODULE NGX STREAM PROXY MODULE

Limits the number of possible tries for passing a connection to the next
server. The 0 value turns off this limitation.

proxy pass
Syntax: proxy_pass address;
Default
Context: server

Sets the address of a proxied server. The address can be specified as a

domain name or IP address, and an obligatory port:

proxy_pass localhost :12345;

or as a UNIX-domain socket path:

proxy_pass unix :/ tmp / stream . socket ;

If a domain name resolves to several addresses, all of them will be used

in a round-robin fashion. In addition, an address can be specified as a server
group.

proxy ssl
Syntax: proxy_ssl on | off;
Default off
Context: stream, server
This directive appeared in version 1.7.10.

Enables the SSL/TLS protocol for connections to a proxied server.

proxy ssl certificate

Syntax: proxy_ssl_certificate file;
Default
Context: stream, server
This directive appeared in version 1.7.10.

Specifies a file with the certificate in the PEM format used for
authentication to a proxied server.

proxy ssl certificate key

Syntax: proxy_ssl_certificate_key file;
Default
Context: stream, server
This directive appeared in version 1.7.10.

Specifies a file with the secret key in the PEM format used for
authentication to a proxied server.

Nginx, Inc. p.277 of 330

CHAPTER 3. STREAM PROXY MODULES 3.2. MODULE NGX STREAM PROXY MODULE

proxy ssl ciphers

Syntax: proxy_ssl_ciphers ciphers;
Default DEFAULT
Context: stream, server
This directive appeared in version 1.7.10.

Specifies the enabled ciphers for connections to a proxied server. The

ciphers are specified in the format understood by the OpenSSL library.
The full list can be viewed using the openssl ciphers command.

proxy ssl crl

Syntax: proxy_ssl_crl file;
Default
Context: stream, server
This directive appeared in version 1.7.10.

Specifies a file with revoked certificates (CRL) in the PEM format used to
verify the certificate of the proxied server.

proxy ssl name

Syntax: proxy_ssl_name name;
Default host from proxy_pass
Context: stream, server
This directive appeared in version 1.7.10.

Allows to override the server name used to verify the certificate of the
proxied server and to be passed through SNI when establishing a connection
with the proxied server.
By default, the host part of the proxy pass address is used.

proxy ssl password file

Syntax: proxy_ssl_password_file file;
Default
Context: stream, server
This directive appeared in version 1.7.10.

Specifies a file with passphrases for secret keys where each passphrase is
specified on a separate line. Passphrases are tried in turn when loading the
key.

proxy ssl server name

Syntax: proxy_ssl_server_name on | off;
Default off
Context: stream, server
This directive appeared in version 1.7.10.

Nginx, Inc. p.278 of 330

CHAPTER 3. STREAM PROXY MODULES 3.2. MODULE NGX STREAM PROXY MODULE

Enables or disables passing of the server name through TLS Server Name
Indication extension (SNI, RFC 6066) when establishing a connection with the
proxied server.

proxy ssl session reuse

Syntax: proxy_ssl_session_reuse on | off;
Default on
Context: stream, server
This directive appeared in version 1.7.10.

Determines whether SSL sessions can be reused when working with the
proxied server. If the errors SSL3_GET_FINISHED:digest check failed
appear in the logs, try disabling session reuse.

proxy ssl protocols

Syntax: proxy_ssl_protocols [SSLv2] [SSLv3] [TLSv1] [TLSv1.1] [TLSv1.2];
Default SSLv3 TLSv1 TLSv1.1 TLSv1.2
Context: stream, server
This directive appeared in version 1.7.10.

Enables the specified protocols for connections to a proxied server.

proxy ssl trusted certificate

Syntax: proxy_ssl_trusted_certificate file;
Default
Context: stream, server
This directive appeared in version 1.7.10.

Specifies a file with trusted CA certificates in the PEM format used to

verify the certificate of the proxied server.

proxy ssl verify

Syntax: proxy_ssl_verify on | off;
Default off
Context: stream, server
This directive appeared in version 1.7.10.

Enables or disables verification of the proxied server certificate.

proxy ssl verify depth

Syntax: proxy_ssl_verify_depth number;
Default 1
Context: stream, server
This directive appeared in version 1.7.10.

Sets the verification depth in the proxied server certificates chain.

Nginx, Inc. p.279 of 330

CHAPTER 3. STREAM PROXY MODULES 3.2. MODULE NGX STREAM PROXY MODULE

proxy timeout
Syntax: proxy_timeout timeout;
Default 10m
Context: stream, server

Defines a timeout used after the proxying to the backend had started.

proxy upstream buffer

Syntax: proxy_upstream_buffer size;
Default 16k
Context: stream, server

Sets the size of the buffer used for reading data from the upstream server.

Nginx, Inc. p.280 of 330

CHAPTER 3. STREAM PROXY MODULES 3.3. MODULE NGX STREAM SSL MODULE

3.3 Module ngx stream ssl module

3.3.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 281
3.3.2 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 281
ssl certificate . . . . . . . . . . . . . . . . . . . . . . . . 281
ssl certificate key . . . . . . . . . . . . . . . . . . . . . . 281
ssl ciphers . . . . . . . . . . . . . . . . . . . . . . . . . . 282
ssl dhparam . . . . . . . . . . . . . . . . . . . . . . . . . 282
ssl ecdh curve . . . . . . . . . . . . . . . . . . . . . . . . 282
ssl handshake timeout . . . . . . . . . . . . . . . . . . . 282
ssl password file . . . . . . . . . . . . . . . . . . . . . . . 282
ssl prefer server ciphers . . . . . . . . . . . . . . . . . . 283
ssl protocols . . . . . . . . . . . . . . . . . . . . . . . . . 283
ssl session cache . . . . . . . . . . . . . . . . . . . . . . 283
ssl session ticket key . . . . . . . . . . . . . . . . . . . . 284
ssl session tickets . . . . . . . . . . . . . . . . . . . . . . 284
ssl session timeout . . . . . . . . . . . . . . . . . . . . . 284

3.3.1 Summary
The ngx_stream_ssl_module module (1.7.10) provides the necessary
support for a stream proxy server to work with the SSL/TLS protocol.

This module is available as part of our commercial subscription.

3.3.2 Directives
ssl certificate
Syntax: ssl_certificate file;
Default
Context: stream, server

Specifies a file with the certificate in the PEM format for the given server. If
intermediate certificates should be specified in addition to a primary certificate,
they should be specified in the same file in the following order: the primary
certificate comes first, then the intermediate certificates. A secret key in the
PEM format may be placed in the same file.

ssl certificate key

Syntax: ssl_certificate_key file;
Default
Context: stream, server

Specifies a file with the secret key in the PEM format for the given server.

Nginx, Inc. p.281 of 330

CHAPTER 3. STREAM PROXY MODULES 3.3. MODULE NGX STREAM SSL MODULE

ssl ciphers
Syntax: ssl_ciphers ciphers;
Default HIGH:!aNULL:!MD5
Context: stream, server

Specifies the enabled ciphers. The ciphers are specified in the format
understood by the OpenSSL library, for example:

ssl_ciphers ALL :! aNULL :! EXPORT56 : RC4 + RSA :+ HIGH :+ MEDIUM :+ LOW :+ SSLv2 :+ EXP ;

The full list can be viewed using the openssl ciphers command.

ssl dhparam
Syntax: ssl_dhparam file;
Default
Context: stream, server

Specifies a file with DH parameters for EDH ciphers.

ssl ecdh curve

Syntax: ssl_ecdh_curve curve;
Default prime256v1
Context: stream, server

Specifies a curve for ECDHE ciphers.

ssl handshake timeout

Syntax: ssl_handshake_timeout time;
Default 60s
Context: stream, server

Specifies a timeout for the SSL handshake to complete.

ssl password file

Syntax: ssl_password_file file;
Default
Context: stream, server

Specifies a file with passphrases for secret keys where each passphrase is
specified on a separate line. Passphrases are tried in turn when loading the
key.
Example:

stream {
s s l _ p a s s w o r d _ f i l e / etc / keys / global . pass ;
...

server {

Nginx, Inc. p.282 of 330

CHAPTER 3. STREAM PROXY MODULES 3.3. MODULE NGX STREAM SSL MODULE

listen 1 2 7. 0 .0 . 1: 1 23 4 5;
s s l _ c e r t i f i c a t e _ k e y / etc / keys / first . key ;
}

server {
listen 1 2 7. 0 .0 . 1: 1 23 4 6;

# named pipe can also be used instead of a file

s s l _ p a s s w o r d _ f i l e / etc / keys / fifo ;
s s l _ c e r t i f i c a t e _ k e y / etc / keys / second . key ;
}
}

ssl prefer server ciphers

Syntax: ssl_prefer_server_ciphers on | off;
Default off
Context: stream, server

Specifies that server ciphers should be preferred over client ciphers when
the SSLv3 and TLS protocols are used.

ssl protocols
Syntax: ssl_protocols [SSLv2] [SSLv3] [TLSv1] [TLSv1.1] [TLSv1.2];
Default SSLv3 TLSv1 TLSv1.1 TLSv1.2
Context: stream, server

Enables the specified protocols. The TLSv1.1 and TLSv1.2 parameters

work only when the OpenSSL library of version 1.0.1 or higher is used.

ssl session cache

Syntax: ssl_session_cache off | none | [builtin[:size]] [shared:name:size];
Default none
Context: stream, server

Sets the types and sizes of caches that store session parameters. A cache
can be of any of the following types:

off
the use of a session cache is strictly prohibited: nginx explicitly tells a
client that sessions may not be reused.
none
the use of a session cache is gently disallowed: nginx tells a client that
sessions may be reused, but does not actually store session parameters
in the cache.
builtin
a cache built in OpenSSL; used by one worker process only. The cache
size is specified in sessions. If size is not given, it is equal to 20480
sessions. Use of the built-in cache can cause memory fragmentation.

Nginx, Inc. p.283 of 330

CHAPTER 3. STREAM PROXY MODULES 3.3. MODULE NGX STREAM SSL MODULE

shared
a cache shared between all worker processes. The cache size is specified
in bytes; one megabyte can store about 4000 sessions. Each shared cache
should have an arbitrary name. A cache with the same name can be used
in several servers.

Both cache types can be used simultaneously, for example:

s s l _ s e s s i o n _ c a c h e builtin :1000 shared : SSL :10 m ;

but using only shared cache without the built-in cache should be more
efficient.

ssl session ticket key

Syntax: ssl_session_ticket_key file;
Default
Context: stream, server

Sets a file with the secret key used to encrypt and decrypt TLS session
tickets. The directive is necessary if the same key has to be shared between
multiple servers. By default, a randomly generated key is used.
If several keys are specified, only the first key is used to encrypt TLS session
tickets. This allows configuring key rotation, for example:

s s l _ s e s s i o n _ t i c k e t _ k e y current . key ;
s s l _ s e s s i o n _ t i c k e t _ k e y previous . key ;

The file must contain 48 bytes of random data and can be created using
the following command:

openssl rand 48 > ticket . key

ssl session tickets

Syntax: ssl_session_tickets on | off;
Default on
Context: stream, server

Enables or disables session resumption through TLS session tickets.

ssl session timeout

Syntax: ssl_session_timeout time;
Default 5m
Context: stream, server

Specifies a time during which a client may reuse the session parameters
stored in a cache.

Nginx, Inc. p.284 of 330

CHAPTER 3. STREAM PROXY MODULES 3.4. MODULE NGX STREAM UPSTREAM MODULE

3.4 Module ngx stream upstream module

3.4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 285
3.4.2 Example Configuration . . . . . . . . . . . . . . . . . . . 285
3.4.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 285
upstream . . . . . . . . . . . . . . . . . . . . . . . . . . 285
server . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
hash . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
least conn . . . . . . . . . . . . . . . . . . . . . . . . . . 288
least time . . . . . . . . . . . . . . . . . . . . . . . . . . 288
health check . . . . . . . . . . . . . . . . . . . . . . . . . 289
health check timeout . . . . . . . . . . . . . . . . . . . . 289
match . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290

3.4.1 Summary
The ngx_stream_upstream_module module (1.7.7) is used to define groups
of servers that can be referenced by the proxy pass directive.

This module is available as part of our commercial subscription.

3.4.2 Example Configuration

resolver 10.0.0.1;

upstream backend {
zone up s tr e am _ ba c ke n d 64 k ;

hash $remote_addr consistent ;

server backend1 . example . com :12345 weight =5;

server 12 7.0 .0.1 :123 45 max_fails =3 fail_timeout =30 s ;
server unix :/ tmp / backend2 ;
server backend3 . example . com :12345 resolve ;
}

server {
listen 12346;
proxy_pass backend ;
health_check ;
}

3.4.3 Directives
upstream
Syntax: upstream name { . . . }
Default
Context: stream

Nginx, Inc. p.285 of 330

CHAPTER 3. STREAM PROXY MODULES 3.4. MODULE NGX STREAM UPSTREAM MODULE

Defines a group of servers. Servers can listen on different ports. In addition,

servers listening on TCP and UNIX-domain sockets can be mixed.
Example:

upstream backend {
server backend1 . example . com :12345 weight =5;
server 12 7.0 .0.1 :123 45 max_fails =3 fail_timeout =30 s ;
server unix :/ tmp / backend2 ;
server backend3 . example . com :12345 resolve ;

server backup1 . example . com :12345 backup ;

}

By default, connections are distributed between the servers using a

weighted round-robin balancing method. In the above example, each
7 connections will be distributed as follows: 5 connections go to
backend1.example.com:12345 and one connection to each of the second and
third servers. If an error occurs during communication with a server, the
connection will be passed to the next server, and so on until all of the
functioning servers will be tried. If communication with all servers fails, the
connection will be closed.

server
Syntax: server address [parameters];
Default
Context: upstream

Defines the address and other parameters of a server. The address can be
specified as a domain name or IP address with an obligatory port, or as a
UNIX-domain socket path specified after the unix: prefix. A domain name
that resolves to several IP addresses defines multiple servers at once.
The following parameters can be defined:
weight=number
sets the weight of the server, by default, 1.
max_fails=number
sets the number of unsuccessful attempts to communicate with the
server that should happen in the duration set by the fail_timeout
parameter to consider the server unavailable for a duration also set by
the fail_timeout parameter. By default, the number of unsuccessful
attempts is set to 1. The zero value disables the accounting of attempts.
Here, an unsuccessful attempt is an error or timeout while establishing
a connection with the server.
fail_timeout=time
sets
the time during which the specified number of unsuccessful attempts
to communicate with the server should happen to consider the server
unavailable;
and the period of time the server will be considered unavailable.

Nginx, Inc. p.286 of 330

CHAPTER 3. STREAM PROXY MODULES 3.4. MODULE NGX STREAM UPSTREAM MODULE

By default, the parameter is set to 10 seconds.

backup
marks the server as a backup server. Connections to the backup server
will be passed when the primary servers are unavailable.
down
marks the server as permanently unavailable.
max_conns=number
limits the maximum number of simultaneous connections to the proxied
server. Default value is zero, meaning there is no limit.
resolve
monitors changes of the IP addresses that correspond to a domain name
of the server, and automatically modifies the upstream configuration
without the need of restarting nginx (1.7.10).
In order for this parameter to work, the resolver directive must be
specified in the stream block. Example:

stream {
resolver 10.0.0.1;

upstream u {
zone ...;
...
server example . com :12345 resolve ;
}
}

slow_start=time
sets the time during which the server will recover its weight from zero to
a nominal value, or when the server becomes available after a period of
time it was considered unavailable. Default value is zero, i.e. slow start
is disabled.

If there is only a single server in a group, max_fails, fail_timeout

and slow_start parameters are ignored, and such a server will never be
considered unavailable.

zone
Syntax: zone name size;
Default
Context: upstream
This directive appeared in version 1.7.10.

Defines the name and size of the shared memory zone that keeps the groups
configuration and run-time state that are shared between worker processes.
Such groups allow changing the group membership or modifying the settings
of a particular server without the need of restarting nginx. The configuration
is accessible via a special location handled by upstream conf.

Nginx, Inc. p.287 of 330

CHAPTER 3. STREAM PROXY MODULES 3.4. MODULE NGX STREAM UPSTREAM MODULE

hash
Syntax: hash key [consistent];
Default
Context: upstream

Specifies a load balancing method for a server group where client-server

mapping is based on the hashed key value. Currently, the only supported
value for the key is the client remote address specified as $remote_addr. Note
that adding or removing a server from the group may result in remapping
most of the keys to different servers. The method is compatible with the
Cache::Memcached Perl library.
If the consistent parameter is specified, the ketama consistent hashing
method will be used instead. The method ensures that only a few keys will be
remapped to different servers when a server is added to or removed from the
group. This helps to achieve a higher cache hit ratio for caching servers. The
method is compatible with the Cache::Memcached::Fast Perl library with the
ketama points parameter set to 160.

least conn
Syntax: least_conn;
Default
Context: upstream

Specifies that a server group should use a load balancing method where a
connection is passed to the server with the least number of active connections,
taking into account weights of servers. If there are several such servers, they
are tried in turn using a weighted round-robin balancing method.

least time
Syntax: least_time connect | first_byte | last_byte;
Default
Context: upstream
This directive appeared in version 1.7.11.

Specifies that a group should use a load balancing method where a

connection is passed to the server with the least average time and least number
of active connections, taking into account weights of servers. If there are several
such servers, they are tried in turn using a weighted round-robin balancing
method.
If the connect parameter is specified, time to connect to the upstream
server is used. If the first_byte parameter is specified, time to receive the
first byte of data is used. If the last_byte is specified, time to receive the last
byte of data is used.

Nginx, Inc. p.288 of 330

CHAPTER 3. STREAM PROXY MODULES 3.4. MODULE NGX STREAM UPSTREAM MODULE

health check
Syntax: health_check [parameters];
Default
Context: server
This directive appeared in version 1.7.10.

Enables periodic health checks of the servers in a group.

The following optional parameters are supported:

interval=time
sets the interval between two consecutive health checks, by default, 5
seconds;
fails=number
sets the number of consecutive failed health checks of a particular server
after which this server will be considered unhealthy, by default, 1;
passes=number
sets the number of consecutive passed health checks of a particular server
after which the server will be considered healthy, by default, 1;
match=name
specifies the match block configuring the tests that a successful
connection should pass in order for a health check to pass; by default,
only the ability to connect to the server is checked.

For example,

server {
proxy_pass backend ;
health_check ;
}

will check the ability to connect to each server in the backend group every
five seconds. When a connection to the server cannot be established, the health
check will fail, and the server will be considered unhealthy. Client connections
are not passed to unhealthy servers.
Health checks can also be configured to test data obtained from the server.
Tests are configured separately using the match directive and referenced in the
match parameter.
The server group must reside in the shared memory.
If several health checks are defined for the same group of servers, a single
failure of any check will make the corresponding server be considered unhealthy.

health check timeout

Syntax: health_check_timeout timeout;
Default 5s
Context: stream, server
This directive appeared in version 1.7.10.

Overrides the proxy timeout value for health checks.

Nginx, Inc. p.289 of 330

CHAPTER 3. STREAM PROXY MODULES 3.4. MODULE NGX STREAM UPSTREAM MODULE

match
Syntax: match name { . . . }
Default
Context: stream
This directive appeared in version 1.7.10.

Defines the named test set used to verify server responses to health checks.
The following parameters can be configured:

send string;
sends a string to the server;
expect ~ regexp;
a regular expression that the data obtained from the server should match.
The regular expression is specified with the preceding ~* modifier
(for case-insensitive matching), or the ~ modifier (for case-sensitive
matching).

Health check is passed if:

the connection was successfully established;

the string from the send parameter, if specified, was sent;

the data obtained from the server matched the regular expression from
the expect parameter, if specified;

the time elapsed does not exceed the value specified in the health check -
timeout directive.

Example:

upstream backend {
zone up s tr e am _ ba ck e nd 10 m ;
server 1 27 . 0. 0 .1 : 12 3 45 ;
}

match http {
send " GET / HTTP /1.0\ r \ nHost : localhost \ r \ n \ r \ n ";
expect ~ "200 OK ";
}

server {
listen 12346;
proxy_pass backend ;
health_check match = http ;
}

Only the first proxy upstream buffer bytes of data obtained from the
server are examined.

Nginx, Inc. p.290 of 330

Chapter 4

Mail server modules

4.1 Module ngx mail core module

4.1.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 291
4.1.2 Example Configuration . . . . . . . . . . . . . . . . . . . 291
4.1.3 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 292
listen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
mail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
resolver . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
resolver timeout . . . . . . . . . . . . . . . . . . . . . . . 294
server . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
server name . . . . . . . . . . . . . . . . . . . . . . . . . 294
so keepalive . . . . . . . . . . . . . . . . . . . . . . . . . 295
timeout . . . . . . . . . . . . . . . . . . . . . . . . . . . 295

4.1.1 Summary
This module is not built by default, it should be enabled with the
--with-mail configuration parameter.

4.1.2 Example Configuration

w or k er _ pr o ce s se s 1;

error_log / var / log / nginx / error . log info ;

mail {
server_name mail . example . com ;
auth_http localhost :9000/ cgi - bin / nginxauth . cgi ;

i m a p _ c a p a b i l i t i e s IMAP4rev1 UIDPLUS IDLE LITERAL + QUOTA ;

pop3_auth plain apop cram - md5 ;

p o p 3 _ c a p a b i l i t i e s LAST TOP USER PIPELINING UIDL ;

smtp_auth login plain cram - md5 ;

s m t p _ c a p a b i l i t i e s " SIZE 10485760" E N H A N C E D S T A T U S C O D E S 8 BITMIME DSN ;
xclient off ;

291
CHAPTER 4. MAIL SERVER MODULES 4.1. MODULE NGX MAIL CORE MODULE

server {
listen 25;
protocol smtp ;
}
server {
listen 110;
protocol pop3 ;
p r o x y _ p a s s _ e r r o r _ m e s s a g e on ;
}
server {
listen 143;
protocol imap ;
}
server {
listen 587;
protocol smtp ;
}
}

4.1.3 Directives
listen
Syntax: listen address:port [bind];
Default
Context: server

Sets the address and port for the socket on which the server will accept
requests. It is possible to specify just the port. The address can also be a
hostname, for example:

listen 127.0.0.1:110;
listen *:110;
listen 110; # same as *:110
listen localhost :110;

IPv6 addresses (0.7.58) are specified in square brackets:

listen [::1]:110;
listen [::]:110;

UNIX-domain sockets (1.3.5) are specified with the unix: prefix:

listen unix :/ var / run / nginx . sock ;

The optional bind parameter instructs to make a separate bind call for a
given address:port pair. The fact is that if there are several listen directives
with the same port but different addresses, and one of the listen directives
listens on all addresses for the given port (*:port), nginx will bind only to
*:port. It should be noted that the getsockname system call will be made in
this case to determine the address that accepted the connection.
Different servers must listen on different address:port pairs.

Nginx, Inc. p.292 of 330

CHAPTER 4. MAIL SERVER MODULES 4.1. MODULE NGX MAIL CORE MODULE

mail
Syntax: mail { . . . }
Default
Context: main

Provides the configuration file context in which the mail server directives
are specified.

protocol
Syntax: protocol imap | pop3 | smtp;
Default
Context: server

Sets the protocol for a proxied server. Supported protocols are IMAP,
POP3, and SMTP.
If the directive is not set, the protocol can be detected automatically based
on the well-known port specified in the listen directive:

imap: 143, 993

pop3: 110, 995

smtp: 25, 587, 465

Unnecessary protocols can be disabled using the configuration param-

eters --without-mail_imap_module, --without-mail_pop3_module, and
--without-mail_smtp_module.

resolver
Syntax: resolver address . . . [valid=time];
Syntax: resolver off;
Default off
Context: mail, server

Configures name servers used to find the clients hostname to pass it to the
authentication server, and in the XCLIENT command when proxying SMTP.
For example:

resolver 127.0.0.1 [::1]:5353;

An address can be specified as a domain name or IP address, and an

optional port (1.3.1, 1.2.2). If port is not specified, the port 53 is used. Name
servers are queried in a round-robin fashion.

Before version 1.1.7, only a single name server could be configured.

Specifying name servers using IPv6 addresses is supported starting from
versions 1.3.1 and 1.2.2.

Nginx, Inc. p.293 of 330

CHAPTER 4. MAIL SERVER MODULES 4.1. MODULE NGX MAIL CORE MODULE

By default, nginx caches answers using the TTL value of a response. An

optional valid parameter allows overriding it:

resolver 127.0.0.1 [::1]:5353 valid =30 s ;

Before version 1.1.9, tuning of caching time was not possible, and nginx
always cached answers for the duration of 5 minutes.

The special value off disables resolving.

resolver timeout
Syntax: resolver_timeout time;
Default 30s
Context: mail, server

Sets a timeout for DNS operations, for example:

r es o lv e r_ t im e ou t 5 s ;

server
Syntax: server { . . . }
Default
Context: mail

Sets the configuration for a server.

server name
Syntax: server_name name;
Default hostname
Context: mail, server

Sets the server name that is used:

in the initial POP3/SMTP server greeting;

in the salt during the SASL CRAM-MD5 authentication;

in the EHLO command when connecting to the SMTP backend, if the

passing of the XCLIENT command is enabled.

If the directive is not specified, the machines hostname is used.

Nginx, Inc. p.294 of 330

CHAPTER 4. MAIL SERVER MODULES 4.1. MODULE NGX MAIL CORE MODULE

so keepalive
Syntax: so_keepalive on | off;
Default off
Context: mail, server

Indicates if the TCP keepalive mode should be enabled on the clients

connection (SO_KEEPALIVE socket parameter) when connecting to a proxied
server.

timeout
Syntax: timeout time;
Default 60s
Context: mail, server

Sets the timeout that is used before proxying to the backend starts.

Nginx, Inc. p.295 of 330

CHAPTER 4. MAIL SERVER MODULES 4.2. MODULE NGX MAIL AUTH HTTP MODULE

4.2 Module ngx mail auth http module

4.2.1 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 296
auth http . . . . . . . . . . . . . . . . . . . . . . . . . . 296
auth http header . . . . . . . . . . . . . . . . . . . . . . 296
auth http pass client cert . . . . . . . . . . . . . . . . . 296
auth http timeout . . . . . . . . . . . . . . . . . . . . . 296
4.2.2 Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . 297

4.2.1 Directives
auth http
Syntax: auth_http URL;
Default
Context: mail, server

Sets the URL of the HTTP authentication server. The protocol is described
below.

auth http header

Syntax: auth_http_header header value;
Default
Context: mail, server

Appends the specified header to requests sent to the authentication server.

This header can be used as the shared secret to verify that the request comes
from nginx. For example:

a ut h _h t tp _ he a de r X - Auth - Key " secret_string ";

auth http pass client cert

Syntax: auth_http_pass_client_cert on | off;
Default off
Context: mail, server
This directive appeared in version 1.7.11.

Appends the Auth-SSL-Cert header with the client certificate in the PEM
format (urlencoded) to requests sent to the authentication server.

auth http timeout

Syntax: auth_http_timeout time;
Default 60s
Context: mail, server

Sets the timeout for communication with the authentication server.

Nginx, Inc. p.296 of 330

CHAPTER 4. MAIL SERVER MODULES 4.2. MODULE NGX MAIL AUTH HTTP MODULE

4.2.2 Protocol
The HTTP protocol is used to communicate with the authentication server.
The data in the response body is ignored, the information is passed only in
the headers.
Examples of requests and responses:
Request:

GET / auth HTTP /1.0

Host : localhost
Auth - Method : plain # plain / apop / cram - md5
Auth - User : user
Auth - Pass : password
Auth - Protocol : imap # imap / pop3 / smtp
Auth - Login - Attempt : 1
Client - IP : 192.0.2.42
Client - Host : client . example . org

Good response:

HTTP /1.0 200 OK

Auth - Status : OK
Auth - Server : 198.51.100.1
Auth - Port : 143

Bad response:

HTTP /1.0 200 OK

Auth - Status : Invalid login or password
Auth - Wait : 3

If there is no Auth-Wait header, an error will be returned and the

connection will be closed. The current implementation allocates memory for
each authentication attempt. The memory is freed only at the end of a session.
Therefore, the number of invalid authentication attempts in a single session
must be limited the server must respond without the Auth-Wait header after
10-20 attempts (the attempt number is passed in the Auth-Login-Attempt
header).
When the APOP or CRAM-MD5 are used, request-response will look as
follows:

GET / auth HTTP /1.0

Host : localhost
Auth - Method : apop
Auth - User : user
Auth - Salt : <238188073.1163692009 @mail . example . com >
Auth - Pass : auth_response
Auth - Protocol : imap
Auth - Login - Attempt : 1
Client - IP : 192.0.2.42
Client - Host : client . example . org

Good response:

HTTP /1.0 200 OK

Auth - Status : OK
Auth - Server : 198.51.100.1

Nginx, Inc. p.297 of 330

CHAPTER 4. MAIL SERVER MODULES 4.2. MODULE NGX MAIL AUTH HTTP MODULE

Auth - Port : 143

Auth - Pass : plain - text - pass

If the Auth-User header exists in the response, it overrides the username

used to authenticate with the backend.
For the SMTP, the response additionally takes into account the
Auth-Error-Code header if exists, it is used as a response code in case
of an error. Otherwise, the 535 5.7.0 code will be added to the Auth-Status
header.
For example, if the following response is received from the authentication
server:

HTTP /1.0 200 OK

Auth - Status : Temporary server problem , try again later
Auth - Error - Code : 451 4.3.0
Auth - Wait : 3

then the SMTP client will receive an error

451 4.3.0 Temporary server problem , try again later

If proxying SMTP does not require authentication, the request will look as
follows:

GET / auth HTTP /1.0

Host : localhost
Auth - Method : none
Auth - User :
Auth - Pass :
Auth - Protocol : smtp
Auth - Login - Attempt : 1
Client - IP : 192.0.2.42
Client - Host : client . example . org
Auth - SMTP - Helo : client . example . org
Auth - SMTP - From : MAIL FROM : <>
Auth - SMTP - To : RCPT TO : < p ostm aste r@ma il . example . com >

For the SSL/TLS client connection (1.7.11), the Auth-SSL header is added,
and Auth-SSL-Verify will contain the result of client certificate verification,
if enabled: SUCCESS, FAILED, and NONE if a certificate was not present.
When the client certificate was present, its details are passed in the following
request headers: Auth-SSL-Subject, Auth-SSL-Issuer, Auth-SSL-Serial,
and Auth-SSL-Fingerprint. If auth http pass client cert is enabled, the
certificate itself is passed in the Auth-SSL-Cert header. The request will look
as follows:

GET / auth HTTP /1.0

Host : localhost
Auth - Method : plain
Auth - User : user
Auth - Pass : password
Auth - Protocol : imap
Auth - Login - Attempt : 1
Client - IP : 192.0.2.42
Auth - SSL : on
Auth - SSL - Verify : SUCCESS

Nginx, Inc. p.298 of 330

CHAPTER 4. MAIL SERVER MODULES 4.2. MODULE NGX MAIL AUTH HTTP MODULE

Auth - SSL - Subject : / CN = example . com

Auth - SSL - Issuer : / CN = example . com
Auth - SSL - Serial : C0 7 AD 5 6B 8 46 B 5B F F
Auth - SSL - Fingerprint : 29 d 6 a 8 0 a 1 2 3 d 1 3 3 5 5 e d 1 6 b 4 b 0 4 6 0 5 e 2 9 c b 5 5 a 5 a d

Nginx, Inc. p.299 of 330

CHAPTER 4. MAIL SERVER MODULES 4.3. MODULE NGX MAIL PROXY MODULE

4.3 Module ngx mail proxy module

4.3.1 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 300
proxy buffer . . . . . . . . . . . . . . . . . . . . . . . . . 300
proxy pass error message . . . . . . . . . . . . . . . . . 300
proxy timeout . . . . . . . . . . . . . . . . . . . . . . . . 300
xclient . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300

4.3.1 Directives
proxy buffer
Syntax: proxy_buffer size;
Default 4k|8k
Context: mail, server

Sets the size of the buffer used for proxying. By default, the buffer size is
equal to one memory page. Depending on a platform, it is either 4K or 8K.

proxy pass error message

Syntax: proxy_pass_error_message on | off;
Default off
Context: mail, server

Indicates whether to pass the error message obtained during the

authentication on the backend to the client.
Usually, if the authentication in nginx is a success, the backend cannot
return an error. If it nevertheless returns an error, it means some internal
error has occurred. In such case the backend message can contain information
that should not be shown to the client. However, responding with an error
for the correct password is a normal behavior for some POP3 servers. For
example, CommuniGatePro informs a user about mailbox overflow or other
events by periodically outputting the authentication error. The directive
should be enabled in this case.

proxy timeout
Syntax: proxy_timeout timeout;
Default 24h
Context: mail, server

Defines a timeout used after the proxying to the backend had started.

xclient
Syntax: xclient on | off;
Default on
Context: mail, server

Nginx, Inc. p.300 of 330

CHAPTER 4. MAIL SERVER MODULES 4.3. MODULE NGX MAIL PROXY MODULE

Enables or disables the passing of the XCLIENT command with client

parameters when connecting to the SMTP backend.
With XCLIENT, the MTA is able to write client information to the log and
apply various limitations based on this data.
If XCLIENT is enabled then nginx passes the following commands when
connecting to the backend:

EHLO with the server name

XCLIENT

EHLO or HELO, as passed by the client

If the name found by the client IP address points to the same address, it
is passed in the NAME parameter of the XCLIENT command. If the name could
not be found, points to a different address, or resolver is not specified, the
[UNAVAILABLE] is passed in the NAME parameter. If an error has occurred in
the process of resolving, the [TEMPUNAVAIL] value is used.
If XCLIENT is disabled then nginx passes the EHLO command with the server
name when connecting to the backend if the client has passed EHLO, or HELO
with the server name, otherwise.

Nginx, Inc. p.301 of 330

CHAPTER 4. MAIL SERVER MODULES 4.4. MODULE NGX MAIL SSL MODULE

4.4 Module ngx mail ssl module

4.4.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . 302
4.4.2 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 302
ssl . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
ssl certificate . . . . . . . . . . . . . . . . . . . . . . . . 303
ssl certificate key . . . . . . . . . . . . . . . . . . . . . . 303
ssl ciphers . . . . . . . . . . . . . . . . . . . . . . . . . . 303
ssl client certificate . . . . . . . . . . . . . . . . . . . . . 303
ssl crl . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
ssl dhparam . . . . . . . . . . . . . . . . . . . . . . . . . 304
ssl ecdh curve . . . . . . . . . . . . . . . . . . . . . . . . 304
ssl password file . . . . . . . . . . . . . . . . . . . . . . . 304
ssl prefer server ciphers . . . . . . . . . . . . . . . . . . 305
ssl protocols . . . . . . . . . . . . . . . . . . . . . . . . . 305
ssl session cache . . . . . . . . . . . . . . . . . . . . . . 305
ssl session ticket key . . . . . . . . . . . . . . . . . . . . 306
ssl session tickets . . . . . . . . . . . . . . . . . . . . . . 306
ssl session timeout . . . . . . . . . . . . . . . . . . . . . 306
ssl trusted certificate . . . . . . . . . . . . . . . . . . . . 306
ssl verify client . . . . . . . . . . . . . . . . . . . . . . . 307
ssl verify depth . . . . . . . . . . . . . . . . . . . . . . . 307
starttls . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307

4.4.1 Summary
The ngx_mail_ssl_module module provides the necessary support for a
mail proxy server to work with the SSL/TLS protocol.
This module is not built by default, it should be enabled with the
--with-mail_ssl_module configuration parameter.

This module requires the OpenSSL library.

4.4.2 Directives
ssl
Syntax: ssl on | off;
Default off
Context: mail, server

Enables the SSL/TLS protocol for the given server.

Nginx, Inc. p.302 of 330

CHAPTER 4. MAIL SERVER MODULES 4.4. MODULE NGX MAIL SSL MODULE

ssl certificate
Syntax: ssl_certificate file;
Default
Context: mail, server

Specifies a file with the certificate in the PEM format for the given server. If
intermediate certificates should be specified in addition to a primary certificate,
they should be specified in the same file in the following order: the primary
certificate comes first, then the intermediate certificates. A secret key in the
PEM format may be placed in the same file.

ssl certificate key

Syntax: ssl_certificate_key file;
Default
Context: mail, server

Specifies a file with the secret key in the PEM format for the given server.

ssl ciphers
Syntax: ssl_ciphers ciphers;
Default HIGH:!aNULL:!MD5
Context: mail, server

Specifies the enabled ciphers. The ciphers are specified in the format
understood by the OpenSSL library, for example:

ssl_ciphers ALL :! aNULL :! EXPORT56 : RC4 + RSA :+ HIGH :+ MEDIUM :+ LOW :+ SSLv2 :+ EXP ;

The full list can be viewed using the openssl ciphers command.

The previous versions of nginx used different ciphers by default.

ssl client certificate

Syntax: ssl_client_certificate file;
Default
Context: mail, server
This directive appeared in version 1.7.11.

Specifies a file with trusted CA certificates in the PEM format used to

verify client certificates.
The list of certificates will be sent to clients. If this is not desired, the
ssl trusted certificate directive can be used.

ssl crl
Syntax: ssl_crl file;
Default
Context: mail, server

Nginx, Inc. p.303 of 330

CHAPTER 4. MAIL SERVER MODULES 4.4. MODULE NGX MAIL SSL MODULE

This directive appeared in version 1.7.11.

Specifies a file with revoked certificates (CRL) in the PEM format used to
verify client certificates.

ssl dhparam
Syntax: ssl_dhparam file;
Default
Context: mail, server
This directive appeared in version 0.7.2.

Specifies a file with DH parameters for EDH ciphers.

ssl ecdh curve

Syntax: ssl_ecdh_curve curve;
Default prime256v1
Context: mail, server
This directive appeared in versions 1.1.0 and 1.0.6.

Specifies a curve for ECDHE ciphers.

ssl password file

Syntax: ssl_password_file file;
Default
Context: mail, server
This directive appeared in version 1.7.3.

Specifies a file with passphrases for secret keys where each passphrase is
specified on a separate line. Passphrases are tried in turn when loading the
key.
Example:

mail {
s s l _ p a s s w o r d _ f i l e / etc / keys / global . pass ;
...

server {
server_name mail1 . example . com ;
s s l _ c e r t i f i c a t e _ k e y / etc / keys / first . key ;
}

server {
server_name mail2 . example . com ;

# named pipe can also be used instead of a file

s s l _ p a s s w o r d _ f i l e / etc / keys / fifo ;
s s l _ c e r t i f i c a t e _ k e y / etc / keys / second . key ;
}
}

Nginx, Inc. p.304 of 330

CHAPTER 4. MAIL SERVER MODULES 4.4. MODULE NGX MAIL SSL MODULE

ssl prefer server ciphers

Syntax: ssl_prefer_server_ciphers on | off;
Default off
Context: mail, server

Specifies that server ciphers should be preferred over client ciphers when
the SSLv3 and TLS protocols are used.

ssl protocols
Syntax: ssl_protocols [SSLv2] [SSLv3] [TLSv1] [TLSv1.1] [TLSv1.2];
Default SSLv3 TLSv1 TLSv1.1 TLSv1.2
Context: mail, server

Enables the specified protocols. The TLSv1.1 and TLSv1.2 parameters

work only when the OpenSSL library of version 1.0.1 or higher is used.

The TLSv1.1 and TLSv1.2 parameters are supported starting from

versions 1.1.13 and 1.0.12 so when the OpenSSL version 1.0.1 or higher is
used on older nginx versions, these protocols work, but cannot be disabled.

ssl session cache

Syntax: ssl_session_cache off | none | [builtin[:size]] [shared:name:size];
Default none
Context: mail, server

Sets the types and sizes of caches that store session parameters. A cache
can be of any of the following types:

off
the use of a session cache is strictly prohibited: nginx explicitly tells a
client that sessions may not be reused.
none
the use of a session cache is gently disallowed: nginx tells a client that
sessions may be reused, but does not actually store session parameters
in the cache.
builtin
a cache built in OpenSSL; used by one worker process only. The cache
size is specified in sessions. If size is not given, it is equal to 20480
sessions. Use of the built-in cache can cause memory fragmentation.
shared
a cache shared between all worker processes. The cache size is specified
in bytes; one megabyte can store about 4000 sessions. Each shared cache
should have an arbitrary name. A cache with the same name can be used
in several servers.

Both cache types can be used simultaneously, for example:

Nginx, Inc. p.305 of 330

CHAPTER 4. MAIL SERVER MODULES 4.4. MODULE NGX MAIL SSL MODULE

s s l _ s e s s i o n _ c a c h e builtin :1000 shared : SSL :10 m ;

but using only shared cache without the built-in cache should be more
efficient.

ssl session ticket key

Syntax: ssl_session_ticket_key file;
Default
Context: mail, server
This directive appeared in version 1.5.7.

Sets a file with the secret key used to encrypt and decrypt TLS session
tickets. The directive is necessary if the same key has to be shared between
multiple servers. By default, a randomly generated key is used.
If several keys are specified, only the first key is used to encrypt TLS session
tickets. This allows configuring key rotation, for example:

s s l _ s e s s i o n _ t i c k e t _ k e y current . key ;
s s l _ s e s s i o n _ t i c k e t _ k e y previous . key ;

The file must contain 48 bytes of random data and can be created using
the following command:

openssl rand 48 > ticket . key

ssl session tickets

Syntax: ssl_session_tickets on | off;
Default on
Context: mail, server
This directive appeared in version 1.5.9.

Enables or disables session resumption through TLS session tickets.

ssl session timeout

Syntax: ssl_session_timeout time;
Default 5m
Context: mail, server

Specifies a time during which a client may reuse the session parameters
stored in a cache.

ssl trusted certificate

Syntax: ssl_trusted_certificate file;
Default
Context: mail, server

Nginx, Inc. p.306 of 330

CHAPTER 4. MAIL SERVER MODULES 4.4. MODULE NGX MAIL SSL MODULE

This directive appeared in version 1.7.11.

Specifies a file with trusted CA certificates in the PEM format used to

verify client certificates.
In contrast to the certificate set by ssl client certificate, the list of these
certificates will not be sent to clients.

ssl verify client

Syntax: ssl_verify_client on | off | optional | optional_no_ca;
Default off
Context: mail, server
This directive appeared in version 1.7.11.

Enables verification of client certificates. The verification result is passed

in the Auth-SSL-Verify header of the authentication request.
The optional parameter requests the client certificate and verifies it if the
certificate is present.
The optional_no_ca parameter requests the client certificate but does not
require it to be signed by a trusted CA certificate. This is intended for the use
in cases when a service that is external to nginx performs the actual certificate
verification. The contents of the certificate is accessible through requests sent
to the authentication server.

ssl verify depth

Syntax: ssl_verify_depth number;
Default 1
Context: mail, server
This directive appeared in version 1.7.11.

Sets the verification depth in the client certificates chain.

starttls
Syntax: starttls on | off | only;
Default off
Context: mail, server

on
allow usage of the STLS command for the POP3 and the STARTTLS
command for the IMAP;
off
deny usage of the STLS and STARTTLS commands;
only
require preliminary TLS transition.

Nginx, Inc. p.307 of 330

CHAPTER 4. MAIL SERVER MODULES 4.5. MODULE NGX MAIL IMAP MODULE

4.5 Module ngx mail imap module

4.5.1 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 308
imap auth . . . . . . . . . . . . . . . . . . . . . . . . . . 308
imap capabilities . . . . . . . . . . . . . . . . . . . . . . 308
imap client buffer . . . . . . . . . . . . . . . . . . . . . . 308

4.5.1 Directives
imap auth
Syntax: imap_auth method . . . ;
Default plain
Context: mail, server

Sets permitted methods of authentication for IMAP clients. Supported

methods are:
login
AUTH=LOGIN
plain
AUTH=PLAIN
cram-md5
AUTH=CRAM-MD5. In order for this method to work, the password
must be stored unencrypted.

imap capabilities
Syntax: imap_capabilities extension . . . ;
Default IMAP4 IMAP4rev1 UIDPLUS
Context: mail, server

Sets the IMAP protocol extensions list that is passed to the client in
response to the CAPABILITY command. The authentication methods specified
in the imap auth and STARTTLS directives are automatically added to this
list if the starttls directive is enabled.
It makes sense to specify the extensions supported by the IMAP backends to
which the clients are proxied (if these extensions are related to commands used
after the authentication, when nginx transparently proxies a client connection
to the backend).
The current list of standardized extensions is published at www.iana.org.

imap client buffer

Syntax: imap_client_buffer size;
Default 4k|8k
Context: mail, server

Sets the IMAP commands read buffer size. By default, the buffer size is
equal to one memory page. This is either 4K or 8K, depending on a platform.

Nginx, Inc. p.308 of 330

CHAPTER 4. MAIL SERVER MODULES 4.6. MODULE NGX MAIL POP3 MODULE

4.6 Module ngx mail pop3 module

4.6.1 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 309
pop3 auth . . . . . . . . . . . . . . . . . . . . . . . . . . 309
pop3 capabilities . . . . . . . . . . . . . . . . . . . . . . 309

4.6.1 Directives
pop3 auth
Syntax: pop3_auth method . . . ;
Default plain
Context: mail, server

Sets permitted methods of authentication for POP3 clients. Supported

methods are:

plain
USER/PASS, AUTH PLAIN, AUTH LOGIN. It is not possible to disable
these methods.
apop
APOP. In order for this method to work, the password must be stored
unencrypted.
cram-md5
AUTH CRAM-MD5. In order for this method to work, the password
must be stored unencrypted.

pop3 capabilities
Syntax: pop3_capabilities extension . . . ;
Default TOP USER UIDL
Context: mail, server

Sets the POP3 protocol extensions list that is passed to the client in
response to the CAPA command.
The authentication methods specified in the pop3 auth and (SASL
extension) and STLS directives, are automatically added to this list if the
starttls directive is enabled.
It makes sense to specify the extensions supported by the POP3 backends
to which the clients are proxied (if these extensions are related to commands
used after the authentication, when nginx transparently proxies the client
connection to the backend).
The current list of standardized extensions is published at www.iana.org.

Nginx, Inc. p.309 of 330

CHAPTER 4. MAIL SERVER MODULES 4.7. MODULE NGX MAIL SMTP MODULE

4.7 Module ngx mail smtp module

4.7.1 Directives . . . . . . . . . . . . . . . . . . . . . . . . . . 310
smtp auth . . . . . . . . . . . . . . . . . . . . . . . . . . 310
smtp capabilities . . . . . . . . . . . . . . . . . . . . . . 310

4.7.1 Directives
smtp auth
Syntax: smtp_auth method . . . ;
Default login plain
Context: mail, server

Sets permitted methods of SASL authentication for SMTP clients.

Supported methods are:

login
AUTH LOGIN
plain
AUTH PLAIN
cram-md5
AUTH CRAM-MD5. In order for this method to work, the password
must be stored unencrypted.
none
Authentication is not required.

smtp capabilities
Syntax: smtp_capabilities extension . . . ;
Default
Context: mail, server

Sets the SMTP protocol extensions list that is passed to the client in
response to the EHLO command. Authentication methods specified in the
smtp auth directive are automatically added to this list.
It makes sense to specify the extensions supported by the MTA to which
the clients are proxied (if these extensions are related to commands used after
the authentication, when nginx transparently proxies the client connection to
the backend).
The current list of standardized extensions is published at www.iana.org.

Nginx, Inc. p.310 of 330

Chapter 5

Miscellaneous

5.1 High Availability support for NGINX

Plus
5.1.1 High Availability support . . . . . . . . . . . . . . . . . 311
5.1.2 Configuring HA setup . . . . . . . . . . . . . . . . . . . 312
5.1.3 Check scripts . . . . . . . . . . . . . . . . . . . . . . . . 313
5.1.4 Checking the status of HA setup . . . . . . . . . . . . . 313
5.1.5 Forcing state change . . . . . . . . . . . . . . . . . . . . 314
5.1.6 Adding more virtual IP addresses . . . . . . . . . . . . . 314
5.1.7 Troubleshooting keepalived and VRRP . . . . . . . . . . 315
5.1.8 Miscellaneous . . . . . . . . . . . . . . . . . . . . . . . . 316

5.1.1 High Availability support

NGINX-HA-Keepalived is a solution for fast and easy configuration of

NGINX Plus in an active-passive high-availability (HA) setup. It is based
on keepalived.
The keepalived project provides a keepalive facility for Linux servers, an
implementation of the VRRP protocol to manage virtual routers (virtual IP
addresses), and a health check facility to determine if a service (web server,
PHP back end, database server, etc.) is up and operational. If a service on
a node fails a configurable number of health checks, keepalived reassigns the
virtual IP address of the node to a secondary node.
The VRRP protocol ensures that one of participating nodes is master. The
backup node listens for VRRP advertisement packets from the master node.
If it does not receive an advertisement packet for a period longer than three
times the configured advertisement interval, the backup node takes over as
master and assigns the configured virtual IP addresses to itself.

311
CHAPTER 5. MISCELLANEOUS 5.1. HIGH AVAILABILITY SUPPORT FOR NGINX PLUS

5.1.2 Configuring HA setup

Run the nginx-ha-setup script (available in the nginx-ha-keepalived
package, must be installed separately) on both nodes as the root user.
The script configures a high-availability NGINX Plus environment with an
active-passive pair of nodes acting as master and backup. It prompts for the
following data:

IP address of the local and remote nodes (one of which will be configured
as a master, the other one as a backup.

One free IP address to be used as the cluster endpoints (floating) virtual

IP address.

The configuration of the keepalived daemon is recorded in a text file, /etc

/keepalived/keepalived.conf. The configuration blocks in the file control
notification settings, the virtual IP addresses to manage, and the health checks
to use to test the services that rely on virtual IP addresses. Following is the
configuration created by the nginx-ha-setup script on a CentOS 7 machine:

vrrp_script c h k _ n g i n x _ s e r v i c e {
script "/ usr / libexec / keepalived / nginx - ha - check "
interval 3
weight 50
}

vrrp_instance VI_1 {
interface eth0
state BACKUP
priority 101
v i r t u a l _ r o u t e r _ i d 51
advert_int 1
unicast_src_ip 1 92.1 68.1 00.1 00
unicast_peer {
192 .16 8.10 0.10 1
}
authentication {
auth_type PASS
auth_pass f 8 f 0 e 5 1 1 4 c b e 0 3 1 a 3 e 1 e 6 2 2 d a f 1 8 f 8 2 a
}
virtual_ipaddress {
192 .16 8.10 0.15 0
}
track_script {
chk_nginx_service
}
notify "/ usr / libexec / keepalived / nginx - ha - notify "
}

The configuration shown above is self-explanatory, but a few items are

worth noting:

Each node in the HA setup needs its own copy of the configuration
file, with values for the priority, unicast_src_ip, and unicast_peer
directives that are appropriate to the nodes status (master or backup).

The priority directive controls which host becomes the master, as

explained in the next section.

Nginx, Inc. p.312 of 330

CHAPTER 5. MISCELLANEOUS 5.1. HIGH AVAILABILITY SUPPORT FOR NGINX PLUS

The notify directive names the notification script included in the

distribution, which can be used to generate syslog messages (or other
notifications) when a state transition or fault occurs.

The value 51 for the virtual_router_id directive in the

vrrp_instance VI_1 block is a sample value.

If you have multiple pairs of keepalived instances (or other VRRP

instances) running in your local network, create a vrrp_instance block
for each one, with a unique name (like VI_1 in the sample) and
virtual_router_id number.

5.1.3 Check scripts

There is no fencing mechanism in keepalived. If the two nodes in a pair are
not aware of each other, each assumes it is the master and assigns the virtual
IP address to itself. To prevent this situation, the chk_nginx_service script
is executed regularly to check its exit code and adjust the nodes priority as
necessary. Code 0 indicates correct operation, and code 1 (or any nonzero
code) indicates an error.
In the default configuration of the chk_nginx_service script, the weight
directive is set to 50, which means that when the check script succeeds:

The priority of the first node (which has a base priority of 101) is set to
151.

The priority of the second node (which has a base priority of 100) is set
to 150.

The first node has higher priority (151 in this case) and becomes master.
Use the interval directive to specify how often the check script executes, in
seconds (it is set to 3 in the default configuration). Note that the check also
fails when the timeout is reached (by default, the timeout is the same as the
check interval).
Use the rise and fall directives to specify how many times the script
must succeed or fail before action is taken (they are not set in the default
configuration).
The default script provided with the nginx-ha-keepalived package checks
if nginx is up. We recommend creating additional scripts as appropriate for
your local setup.

5.1.4 Checking the status of HA setup

To see which node is currently the master for a given virtual IP address,
run the ip addr show command for the interface on which the vrrp instance
is defined (in the following commands, interface eth0):

Nginx, Inc. p.313 of 330

CHAPTER 5. MISCELLANEOUS 5.1. HIGH AVAILABILITY SUPPORT FOR NGINX PLUS

centos7 -1 # ip addr show eth0

2: eth0 : < BROADCAST , MULTICAST , UP , LOWER_UP > mtu 1500 qdisc pfifo_fast state
UP qlen 1000
link / ether 52:54:00:33: a5 : a5 brd ff : ff : ff : ff : ff : ff
inet 1 9 2 . 1 6 8 . 1 0 0 . 1 0 0 / 2 4 brd 192 .168 .122 .25 5 scope global dynamic eth0
valid_lft 3071 sec preferred_lft 3071 sec
inet 1 9 2 . 1 6 8 . 1 0 0 . 1 5 0 / 3 2 scope global eth0
valid_lft forever preferred_lft forever

centos7 -2 # ip addr show eth0

2: eth0 : < BROADCAST , MULTICAST , UP , LOWER_UP > mtu 1500 qdisc pfifo_fast state
UP qlen 1000
link / ether 52:54:00:33: a5 :87 brd ff : ff : ff : ff : ff : ff
inet 1 9 2 . 1 6 8 . 1 0 0 . 1 0 1 / 2 4 brd 192 .168 .122 .25 5 scope global eth0
valid_lft forever preferred_lft forever

In this output, the defined virtual IP address (192.168.100.150) is currently

assigned to the host with real IP address of 192.168.100.100.
When a hosts HA state changes, nginx-ha-keepalived writes it to the
/var/run/nginx-ha-keepalived.state file:

centos7 -1 # cat / var / run / nginx - ha - keepalived . state

STATE = MASTER

centos7 -2 # cat / var / run / nginx - ha - keepalived . state

STATE = BACKUP

5.1.5 Forcing state change

To force the master node to switch to backup state, run the following
command on it:

# service keepalived stop

As it shuts down, keepalived sends a VRRP packet with priority 0 to the

backup node, which causes the backup node to take over the virtual IP address.

5.1.6 Adding more virtual IP addresses

The configuration created by nginx-ha-setup is very basic, and makes
a single IP address highly available. To make more than one IP address
highly available, add each new IP address to the virtual_ipaddress block
in the /etc/keepalived/keepalived.conf configuration file. Then run
the service keepalived reload command on both nodes to reload the
keepalived service:

virtual_ipaddress {
192 .16 8.10 0.15 0
192 .16 8.10 0.20 0
1234:5678:9 abc : def ::1/64
}

Nginx, Inc. p.314 of 330

CHAPTER 5. MISCELLANEOUS 5.1. HIGH AVAILABILITY SUPPORT FOR NGINX PLUS

As indicated in this example, keepalived can be utilized in dual-stack

IPv4/IPv6 environments to fail over both IPv4 and IPv6 addresses.
The syntax in the virtual ipaddress block replicates the syntax of the ip
utility.

5.1.7 Troubleshooting keepalived and VRRP

The keepalived daemon logs to syslog. On CentOS, RHEL, and SLES-based
systems, the output is typically written to /var/log/messages, whereas on
Ubuntu and Debian-based systems it is written to /var/log/syslog. Log
entries record events such as startup of the keepalived daemon and state
transitions. Here are a few sample entries that show the keepalived daemon
starting up, and the node transitioning a VRRP instance to the master state:

Feb 27 14:42:04 centos7 -1 systemd : Starting LVS and VRRP High Availability
Monitor ...
Feb 27 14:42:04 centos7 -1 Keepalived [19242]: Starting Keepalived v1 .2.15
(02/26 ,2015)
Feb 27 14:42:04 centos7 -1 Keepalived [19243]: Starting VRRP child process ,
pid =19244
Feb 27 14:42:04 centos7 -1 Keep aliv ed_ vrrp [19244]: Registering Kernel
netlink reflector
Feb 27 14:42:04 centos7 -1 Keep aliv ed_ vrrp [19244]: Registering Kernel
netlink command channel
Feb 27 14:42:04 centos7 -1 Keep aliv ed_ vrrp [19244]: Registering gratuitous
ARP shared channel
Feb 27 14:42:05 centos7 -1 systemd : Started LVS and VRRP High Availability
Monitor .
Feb 27 14:42:05 centos7 -1 Keep aliv ed_ vrrp [19244]: Opening file / etc /
keepalived / keepalived . conf .
Feb 27 14:42:05 centos7 -1 Keep aliv ed_ vrrp [19244]: Truncating auth_pass to
8 characters
Feb 27 14:42:05 centos7 -1 Keep aliv ed_ vrrp [19244]: Configuration is using :
64631 Bytes
Feb 27 14:42:05 centos7 -1 Keep aliv ed_ vrrp [19244]: Using LinkWatch kernel
netlink reflector ...
Feb 27 14:42:05 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
Entering BACKUP STATE
Feb 27 14:42:05 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP sockpool : [ ifindex
(2) , proto (112) , unicast (1) , fd (14 ,15) ]
Feb 27 14:42:05 centos7 -1 nginx - ha - keepalived : Transition to state BACKUP
on VRRP instance VI_1 .
Feb 27 14:42:05 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Script (
c h k _ n g i n x _ s e r v i c e ) succeeded
Feb 27 14:42:06 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
forcing a new MASTER election
Feb 27 14:42:06 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
forcing a new MASTER election
Feb 27 14:42:07 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
Transition to MASTER STATE
Feb 27 14:42:08 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
Entering MASTER STATE
Feb 27 14:42:08 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
setting protocol VIPs .
Feb 27 14:42:08 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
Sending gratuitous ARPs on eth0 for 19 2.16 8.10 0.15 0
Feb 27 14:42:08 centos7 -1 nginx - ha - keepalived : Transition to state MASTER
on VRRP instance VI_1 .
Feb 27 14:42:13 centos7 -1 Keep aliv ed_ vrrp [19244]: VRRP_Instance ( VI_1 )
Sending gratuitous ARPs on eth0 for 19 2.16 8.10 0.15 0

Nginx, Inc. p.315 of 330

CHAPTER 5. MISCELLANEOUS 5.1. HIGH AVAILABILITY SUPPORT FOR NGINX PLUS

If the system log does not explain the source of a problem, run the tcpdump
command with the following parameters to display the VRRP advertisements
that are sent on the local network:

# tcpdump - vvv - ni eth0 proto vrrp

If you have multiple VRRP instances on the local network and want to
filter the traffic for select hosts, include the host parameter to specify the IP
address that is defined in the unicast_peer block, as in the following example:

centos7 -1 # tcpdump - vvv - ni eth0 proto vrrp and host 192 .168 .100 .101
tcpdump : listening on eth0 , link - type EN10MB ( Ethernet ) , capture size
65535 bytes
14 :48: 27.1 8810 0 IP ( tos 0 xc0 , ttl 255 , id 382 , offset 0 , flags [ none ] ,
proto VRRP (112) , length 40)
19 2.16 8.10 0.10 0 > 1 9 2. 1 68 . 10 0 .1 01 : vrrp 1 92.1 68. 100. 100 >
1 92 . 16 8 .1 0 0. 10 1 : VRRPv2 , Advertisement , vrid 51 , prio 151 ,
authtype simple , intvl 1s , length 20 , addrs : 192 .16 8.10 0.15 0 auth
" f8f0e511 "

Several fields in the output are useful for debugging:

authtype - the type of authentication in use (authentication directive)

vrid - the virtual router ID (virtual_router_id directive)

prio - the nodes priority (priority directive)

intvl - the frequency at which advertisements are sent (advert_int

directive)

auth - the authentication token sent (auth_pass directive)

5.1.8 Miscellaneous
Note that NGINX configuration files on both nodes must define the services
that are being made highly available. Keeping the configuration files in sync
is outside the scope of the provided clustering software.
The nginx-ha-keepalived package comes with numerous configuration
examples, in the /usr/share/doc/nginx-ha-keepalived/ directory. They
show how to configure numerous aspects of an HA setup.

Nginx, Inc. p.316 of 330

Appendix A

Changelog for NGINX Plus

This appendix contains the most important changes that may apply to both NGINX Plus
and nginx/OSS. Full changelog for nginx/OSS is available in the packages and by the
following link: http://nginx.org/en/CHANGES

NGINX Plus r6 (1.7.11), released Apr 14, 2015

TCP proxy enhancements (health checks, dynamic reconfiguration, SSL
support, logging, status counters).
New least time load balancing method.
Unbuffered upload support (proxy request buffering and friends).
Proxy SSL authentication support for http and uwsgi.
Proxy cache enhancements (variables support in proxy cache, use_temp_path
parameter in proxy cache path).
Client SSL certificates support in mail proxy.
Autoindex module enhancement (the autoindex format directive).
New status dashboard.
Lua module updated to version 0.9.16rc1 (nginx-plus-lua,
nginx-plus-extras).
Passenger module updated to version 4.0.59 (nginx-plus-extras).
set-misc module updated to version 0.28 (nginx-plus-extras).
NGINX Plus r5 (1.7.7), released Dec 1, 2014
New TCP proxying and load balancing mode (the stream module).
Sticky session timeout now applies from the most recent request in the session.
Upstream draining can be used to remove an upstream server without
interrupting any user sessions (the drain command of the upstream conf
dynamic configuration interface).
Improved control over request retries in the event of failure, based on number
of tries and time. Also available for fastcgi, uwsgi, scgi and memcached
modules.
Caching: the Vary response header is correctly handled (multiple variants of
the same resource can be cached). Note that the on-disk cache format has
changed, so cached content will be invalidated after the upgrade.
Caching: improved support for byte-range requests.
Ability to control upstream bandwidth with the proxy limit rate directive.
Lua module updated to version 0.9.13 (nginx-plus-lua,
nginx-plus-extras).
Passenger module updated to version 4.0.53 (nginx-plus-extras).

317
APPENDIX A. CHANGELOG FOR NGINX PLUS

NGINX Plus r4 (1.7.3), released Jul 22, 2014

MP4 module now supports the end query argument which sets the end point
of playback.
Added the ability to verify backend SSL certificates.
Added support for SNI while working with SSL backends.
Added conditional logging for requests (the if parameter of the access log
directive).
New load balancing method based on user-defined keys with optional
consistency.
Cache revalidation now uses If-None-Match header if possible.
Passphrases for SSL private keys can now be stored in an external file.
Introduced a new session affinity mechanism (sticky learn) based on
server-initiated sessions.
Added the ability to retrieve a subset of the extended status data.
Lua module updated to version 0.9.10 (nginx-plus-lua,
nginx-plus-extras).
Passenger module updated to version 4.0.45 (nginx-plus-extras).
NGINX Plus r3 (1.5.12), released Apr 2, 2014
SPDY protocol updated to version 3.1. SPDY/2 is no longer supported.
Added PROXY protocol support (the proxy_protocol parameter of the
listen directive).
IPv6 support added to resolver.
DNS names in upstream groups are periodically re-resolved (the resolve
parameter of the server directive).
Introduced limiting connections to upstream servers (the max_conns
parameter) with optional support for connections queue.
NGINX Plus r2 (1.5.7), released Dec 12, 2013
Enhanced sticky routing support.
Additional status metrics for virtual hosts and cache zones.
Cache purge support (also available for FastCGI).
Added support for cache revalidation.
New module: ngx http auth request module (authorization based on the
result of a subrequest).
NGINX Plus r1 (1.5.3), released Aug 12, 2013
Enhanced status monitoring.
Load balancing: slow start feature.
Added syslog support for both error log and access log.
Support for Apple HTTP Live Streaming.
NGINX Plus 1.5.0-2, released May 27, 2013
Added support for active healthchecks.
NGINX Plus 1.5.0, released May 7, 2013
Security: fixed CVE-2013-2028.
NGINX Plus 1.3.16, released Apr 19, 2013
Added SPDY support.

Nginx, Inc. p.318 of 330

APPENDIX A. CHANGELOG FOR NGINX PLUS

NGINX Plus 1.3.13, released Feb 22, 2013

Added sticky sessions support.
Added support for proxying WebSocket connections.
NGINX Plus 1.3.11, released Jan 18, 2013
Added base module ngx http gunzip module.
New extra module: ngx http f4f module (Adobe HDS Dynamic Streaming).
New extra module: ngx http session log module (aggregated session logging).
NGINX Plus 1.3.9-2, released Dec 20, 2012
License information updated.
End-User License Agreement added to the package.
NGINX Plus 1.3.9, released Nov 27, 2012
Added dynamic upstream management feature.
PDF documentation bundled into package.
NGINX Plus 1.3.7, released Oct 18, 2012
Initial release of NGINX Plus package.

Nginx, Inc. p.319 of 330

Appendix B

Legal Notices

At the release moment of this document, there are three versions of NGINX Plus package
in distribution:
NGINX Plus (package name is nginx-plus)
NGINX Plus Lua (package name is nginx-plus-lua)
NGINX Plus Extras (package name is nginx-plus-extras)
These distributions contain a different set of various open source software components
described below.

Open source components included in NGINX Plus, NGINX Plus Lua and NGINX Plus
Extras are:
nginx/OSS, distributed under 2-clause BSD license.

Copyright 2002-2015 Igor Sysoev

Copyright 2011-2015 Nginx, Inc.
All rights reserved.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

1. Redistributions of source code must retain the above copyright notice, this list
of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.

THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS

AS IS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT
NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY
AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE FOR ANY
DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

MurmurHash algorithm, distributed under MIT license.

Copyright Austin Appleby

Permission is hereby granted, free of charge, to any person obtaining a copy of this
software and associated documentation files (the Software), to deal in the Software

320
APPENDIX B. LEGAL NOTICES

without restriction, including without limitation the rights to use, copy, modify,
merge, publish, distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to the following
conditions:
The above copyright notice and this permission notice shall be included in all copies
or substantial portions of the Software.
THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY
KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS
OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Ractive.js JavaScript library, distributed under MIT license.

Copyright 2012-14 Rich Harris and contributors

Permission is hereby granted, free of charge, to any person obtaining a copy of this
software and associated documentation files (the Software), to deal in the Software
without restriction, including without limitation the rights to use, copy, modify,
merge, publish, distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to the following
conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY
KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS
OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

opentip JavaScript tooltip framework, distributed under MIT license.

Copyright 2009-2012, Matias Meno

Permission is hereby granted, free of charge, to any person obtaining a copy of this
software and associated documentation files (the Software), to deal in the Software
without restriction, including without limitation the rights to use, copy, modify,
merge, publish, distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to the following
conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY
KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS
OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Open source components included in NGINX Plus Lua and NGINX Plus Extras are:
Nginx Development Kit (NDK) module, distributed under BSD license.
Copyright Marcus Clyne

Nginx, Inc. p.321 of 330

APPENDIX B. LEGAL NOTICES

lua-nginx-module, distributed under 2-clause BSD license.

Copyright 2009-2015, by Xiaozhe Wang (chaoslawful)

Copyright 2009-2015, by Yichun agentzh Zhang ()
All rights reserved.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list
of conditions and the following disclaimer.
Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND

CONTRIBUTORS AS IS AND ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.

Open source components included in NGINX Plus Extras are:

headers-more-nginx-module, distributed under 2-clause BSD license.

Copyright 2009-2014, Yichun agentzh Zhang ()

Copyright 2010-2013, Bernd Dorn
This module is licensed under the terms of the BSD license.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list
of conditions and the following disclaimer.
Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND

CONTRIBUTORS AS IS AND ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.

Nginx, Inc. p.322 of 330

APPENDIX B. LEGAL NOTICES

set-misc-nginx-module, distributed under 2-clause BSD license.

Copyright 2009-2014, Yichun agentzh Zhang ()

This module is licensed under the terms of the BSD license.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list
of conditions and the following disclaimer.
Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND

CONTRIBUTORS AS IS AND ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.

Phusion Passenger module for nginx (open source)

Copyright 2010-2013 Phusion

Permission is hereby granted, free of charge, to any person obtaining a copy of this
software and associated documentation files (the Software), to deal in the Software
without restriction, including without limitation the rights to use, copy, modify,
merge, publish, distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to the following
conditions:

The above copyright notice and this permission notice shall be included in all copies
or substantial portions of the Software.

THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY

KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS
OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Open source components included in NGINX Plus Extras are:
nginx-rtmp-module, distributed under 2-clause BSD license.

Copyright 2012-2014, Roman Arutyunyan

All rights reserved.
Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

Redistributions of source code must retain the above copyright notice, this list
of conditions and the following disclaimer.
Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.

Nginx, Inc. p.323 of 330

APPENDIX B. LEGAL NOTICES

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND

CONTRIBUTORS AS IS AND ANY EXPRESS OR IMPLIED WARRANTIES,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.

Nginx, Inc. p.324 of 330

Index

accept mutex, 7 create full put path, 71

accept mutex delay, 7
access log, 122 daemon, 7
add after body, 57 dav access, 70
add before body, 57 dav methods, 71
add header, 106 debug connection, 8
addition types, 57 debug points, 8
aio, 21 default type, 25
alias, 22 deny, 55
allow, 55 directio, 25
ancient browser, 66 directio alignment, 26
ancient browser value, 66 disable symlinks, 26
auth basic, 59 empty gif, 73
auth basic user file, 59 env, 9
auth http, 296 error log, 8
auth http header, 296 error page, 27
auth http pass client cert, 296 etag, 28
auth http timeout, 296 events, 9
auth request, 61 expires, 106
auth request set, 62
autoindex, 63 f4f, 74
autoindex exact size, 63 f4f buffer size, 74
autoindex format, 63 fastcgi bind, 76
autoindex localtime, 64 fastcgi buffer size, 76
fastcgi buffering, 77
break, 172
fastcgi buffers, 77
charset, 67 fastcgi busy buffers size, 77
charset map, 68 fastcgi cache, 78
charset types, 69 fastcgi cache bypass, 78
chunked transfer encoding, 23 fastcgi cache key, 78
client body buffer size, 23 fastcgi cache lock, 78
client body in file only, 23 fastcgi cache lock age, 79
client body in single buffer, 23 fastcgi cache lock timeout, 79
client body temp path, 24 fastcgi cache methods, 79
client body timeout, 24 fastcgi cache min uses, 79
client header buffer size, 24 fastcgi cache path, 79
client header timeout, 24 fastcgi cache purge, 80
client max body size, 25 fastcgi cache revalidate, 81
connection pool size, 25 fastcgi cache use stale, 81

325
INDEX INDEX

fastcgi cache valid, 82 gzip static, 105

fastcgi catch stderr, 82 gzip types, 103
fastcgi connect timeout, 83 gzip vary, 104
fastcgi force ranges, 83
fastcgi hide header, 83 hash, 231, 288
fastcgi ignore client abort, 83 health check, 234, 289
fastcgi ignore headers, 84 health check timeout, 289
fastcgi index, 84 hls, 109
fastcgi intercept errors, 84 hls buffers, 109
fastcgi keep conn, 85 hls forward args, 109
fastcgi limit rate, 85 hls fragment, 110
fastcgi max temp file size, 85 hls mp4 buffer size, 110
fastcgi next upstream, 85 hls mp4 max buffer size, 110
fastcgi next upstream timeout, 86 http, 28
fastcgi next upstream tries, 86 if, 173
fastcgi no cache, 87 if modified since, 28
fastcgi param, 87 ignore invalid headers, 28
fastcgi pass, 88 image filter, 112
fastcgi pass header, 88 image filter buffer, 113
fastcgi pass request body, 88 image filter interlace, 113
fastcgi pass request headers, 89 image filter jpeg quality, 114
fastcgi read timeout, 88 image filter sharpen, 114
fastcgi request buffering, 89 image filter transparency, 114
fastcgi send lowat, 89 imap auth, 308
fastcgi send timeout, 89 imap capabilities, 308
fastcgi split path info, 89 imap client buffer, 308
fastcgi store, 90 include, 10
fastcgi store access, 91 index, 115
fastcgi temp file write size, 91 internal, 29
fastcgi temp path, 91 ip hash, 232
flv, 93
keepalive, 232
geo, 94 keepalive disable, 29
geoip city, 98 keepalive requests, 30
geoip country, 97 keepalive timeout, 30
geoip org, 99
geoip proxy, 99 large client header buffers, 30
geoip proxy recursive, 99 least conn, 234, 288
gunzip, 100 least time, 234, 288
gunzip buffers, 100 limit conn, 116
gzip, 101 limit conn log level, 117
gzip buffers, 101 limit conn status, 117
gzip comp level, 102 limit conn zone, 117
gzip disable, 102 limit except, 31
gzip http version, 102 limit rate, 31
gzip min length, 102 limit rate after, 31
gzip proxied, 103 limit req, 119

Nginx, Inc. p.326 of 330

INDEX INDEX

limit req log level, 120 open file cache errors, 39

limit req status, 120 open file cache min uses, 39
limit req zone, 120 open file cache valid, 40
limit zone, 118 open log file cache, 125
lingering close, 32 optimize server names, 40
lingering time, 32 output buffers, 40
lingering timeout, 32 override charset, 69
listen, 33, 272, 292
location, 35 pcre jit, 11
lock file, 10 perl, 137
log format, 124 perl modules, 138
log not found, 37 perl require, 138
log subrequest, 37 perl set, 138
pid, 11
mail, 293 pop3 auth, 309
map, 126 pop3 capabilities, 309
map hash bucket size, 128 port in redirect, 40
map hash max size, 128 postpone output, 40
master process, 10 protocol, 293
match, 236, 290 proxy bind, 143
max ranges, 37 proxy buffer, 300
memcached bind, 129 proxy buffer size, 144
memcached buffer size, 130 proxy buffering, 144
memcached connect timeout, 130 proxy buffers, 144
memcached force ranges, 130 proxy busy buffers size, 145
memcached gzip flag, 130 proxy cache, 145
memcached next upstream, 130 proxy cache bypass, 145
memcached next upstream timeout, proxy cache key, 145
131 proxy cache lock, 146
memcached next upstream tries, 131 proxy cache lock age, 146
memcached pass, 131 proxy cache lock timeout, 146
memcached read timeout, 132 proxy cache methods, 146
memcached send timeout, 132 proxy cache min uses, 147
merge slashes, 38 proxy cache path, 147
min delete depth, 71 proxy cache purge, 148
modern browser, 66 proxy cache revalidate, 148
modern browser value, 66 proxy cache use stale, 149
mp4, 134 proxy cache valid, 149
mp4 buffer size, 134 proxy connect timeout, 150, 276
mp4 limit rate, 135 proxy cookie domain, 150
mp4 limit rate after, 135 proxy cookie path, 151
mp4 max buffer size, 134 proxy downstream buffer, 276
msie padding, 38 proxy force ranges, 152
msie refresh, 38 proxy headers hash bucket size, 152
multi accept, 10 proxy headers hash max size, 152
proxy hide header, 152
open file cache, 39 proxy http version, 152

Nginx, Inc. p.327 of 330

INDEX INDEX

proxy ignore client abort, 153 real ip header, 168

proxy ignore headers, 153 real ip recursive, 169
proxy intercept errors, 153 recursive error pages, 41
proxy limit rate, 153 referer hash bucket size, 170
proxy max temp file size, 154 referer hash max size, 170
proxy method, 154 request pool size, 41
proxy next upstream, 154, 276 reset timedout connection, 41
proxy next upstream timeout, 155, resolver, 41, 273, 293
276 resolver timeout, 42, 273, 294
proxy next upstream tries, 155, 276 return, 174
proxy no cache, 156 rewrite, 174
proxy pass, 156, 277 rewrite log, 175
proxy pass error message, 300 root, 42
proxy pass header, 157
proxy pass request body, 158 satisfy, 43
proxy pass request headers, 158 satisfy any, 43
proxy read timeout, 158 scgi bind, 179
proxy redirect, 158 scgi buffer size, 179
proxy request buffering, 160 scgi buffering, 179
proxy send lowat, 160 scgi buffers, 180
proxy send timeout, 160 scgi busy buffers size, 180
proxy set body, 161 scgi cache, 180
proxy set header, 161 scgi cache bypass, 180
proxy ssl, 277 scgi cache key, 181
proxy ssl certificate, 162, 277 scgi cache lock, 181
proxy ssl certificate key, 162, 277 scgi cache lock age, 181
proxy ssl ciphers, 162, 278 scgi cache lock timeout, 181
proxy ssl crl, 162, 278 scgi cache methods, 182
proxy ssl name, 162, 278 scgi cache min uses, 182
proxy ssl password file, 163, 278 scgi cache path, 182
proxy ssl protocols, 163, 279 scgi cache purge, 183
proxy ssl server name, 163, 278 scgi cache revalidate, 184
proxy ssl session reuse, 163, 279 scgi cache use stale, 184
proxy ssl trusted certificate, 163, 279 scgi cache valid, 184
proxy ssl verify, 164, 279 scgi connect timeout, 185
proxy ssl verify depth, 164, 279 scgi force ranges, 185
proxy store, 164 scgi hide header, 185
proxy store access, 165 scgi ignore client abort, 186
proxy temp file write size, 165 scgi ignore headers, 186
proxy temp path, 166 scgi intercept errors, 186
proxy timeout, 280, 300 scgi limit rate, 187
proxy upstream buffer, 280 scgi max temp file size, 187
scgi next upstream, 187
queue, 237 scgi next upstream timeout, 188
scgi next upstream tries, 188
random index, 167 scgi no cache, 188
read ahead, 41 scgi param, 189

Nginx, Inc. p.328 of 330

INDEX INDEX

scgi pass, 189 ssl ciphers, 208, 282, 303

scgi pass header, 190 ssl client certificate, 209, 303
scgi pass request body, 190 ssl crl, 209, 303
scgi pass request headers, 190 ssl dhparam, 209, 282, 304
scgi read timeout, 190 ssl ecdh curve, 209, 282, 304
scgi request buffering, 190 ssl engine, 11
scgi send timeout, 191 ssl handshake timeout, 282
scgi store, 191 ssl password file, 209, 282, 304
scgi store access, 192 ssl prefer server ciphers, 210, 283, 305
scgi temp file write size, 192 ssl protocols, 210, 283, 305
scgi temp path, 192 ssl session cache, 210, 283, 305
secure link, 193 ssl session ticket key, 211, 284, 306
secure link md5, 194 ssl session tickets, 211, 284, 306
secure link secret, 194 ssl session timeout, 212, 284, 306
send lowat, 43 ssl stapling, 212
send timeout, 44 ssl stapling file, 212
sendfile, 44 ssl stapling responder, 212
sendfile max chunk, 44 ssl stapling verify, 213
server, 45, 229, 274, 286, 294 ssl trusted certificate, 213, 306
server name, 45, 294 ssl verify client, 213, 307
server name in redirect, 47 ssl verify depth, 213, 307
server names hash bucket size, 47 starttls, 307
server names hash max size, 47 status, 217
server tokens, 47 status format, 217
session log, 197 status zone, 218
session log format, 196 sticky, 238
session log zone, 196 sticky cookie insert, 240
set, 176 stream, 274
set real ip from, 168 stub status, 224
smtp auth, 310 sub filter, 226
smtp capabilities, 310 sub filter last modified, 226
so keepalive, 295 sub filter once, 227
source charset, 69 sub filter types, 227
spdy chunk size, 199
spdy headers comp, 199 tcp nodelay, 47
split clients, 200 tcp nopush, 48
ssi, 201 thread pool, 11
ssi last modified, 201 timeout, 295
ssi min file chunk, 202 timer resolution, 12
ssi silent errors, 202 try files, 48
ssi types, 202 types, 50
ssi value length, 202 types hash bucket size, 50
ssl, 207, 302 types hash max size, 51
ssl buffer size, 207 underscores in headers, 51
ssl certificate, 208, 281, 303 uninitialized variable warn, 176
ssl certificate key, 208, 281, 303 upstream, 229, 285

Nginx, Inc. p.329 of 330

INDEX INDEX

upstream conf, 242 uwsgi pass request body, 262

use, 12 uwsgi pass request headers, 262
user, 12 uwsgi read timeout, 262
userid, 246 uwsgi request buffering, 262
userid domain, 247 uwsgi send timeout, 262
userid expires, 247 uwsgi ssl certificate, 263
userid mark, 247 uwsgi ssl certificate key, 263
userid name, 247 uwsgi ssl ciphers, 263
userid p3p, 248 uwsgi ssl crl, 263
userid path, 248 uwsgi ssl name, 263
userid service, 248 uwsgi ssl password file, 264
uwsgi bind, 250 uwsgi ssl protocols, 264
uwsgi buffer size, 250 uwsgi ssl server name, 264
uwsgi buffering, 251 uwsgi ssl session reuse, 264
uwsgi buffers, 251 uwsgi ssl trusted certificate, 265
uwsgi busy buffers size, 251 uwsgi ssl verify, 265
uwsgi cache, 252 uwsgi ssl verify depth, 265
uwsgi cache bypass, 252 uwsgi store, 265
uwsgi cache key, 252 uwsgi store access, 266
uwsgi cache lock, 252 uwsgi temp file write size, 266
uwsgi cache lock age, 253 uwsgi temp path, 266
uwsgi cache lock timeout, 253
uwsgi cache methods, 253 valid referers, 171
uwsgi cache min uses, 253 variables hash bucket size, 51
uwsgi cache path, 253 variables hash max size, 51
uwsgi cache purge, 254 worker aio requests, 12
uwsgi cache revalidate, 255 worker connections, 13
uwsgi cache use stale, 255 worker cpu affinity, 13
uwsgi cache valid, 256 worker priority, 13
uwsgi connect timeout, 256 worker processes, 14
uwsgi force ranges, 257 worker rlimit core, 14
uwsgi hide header, 257 worker rlimit nofile, 14
uwsgi ignore client abort, 257 worker rlimit sigpending, 14
uwsgi ignore headers, 257 working directory, 15
uwsgi intercept errors, 258
uwsgi limit rate, 258 xclient, 300
uwsgi max temp file size, 258 xml entities, 268
uwsgi modifier1, 259 xslt last modified, 269
uwsgi modifier2, 259 xslt param, 269
uwsgi next upstream, 259 xslt string param, 269
uwsgi next upstream timeout, 260 xslt stylesheet, 269
uwsgi next upstream tries, 260 xslt types, 270
uwsgi no cache, 260
uwsgi param, 260 zone, 231, 287
uwsgi pass, 261
uwsgi pass header, 261

Nginx, Inc. p.330 of 330