<?xml version="1.0"?>

<!--
  Copyright (C) Nginx, Inc.
  -->

<!DOCTYPE module SYSTEM "../../../../dtd/module.dtd">

<module name="Module ngx_stream_upstream_module"
        link="/en/docs/stream/ngx_stream_upstream_module.html"
        lang="en"
        rev="39">

<section id="summary">

<para>
The <literal>ngx_stream_upstream_module</literal> module (1.9.0)
is used to define groups of servers that can be referenced
by the <link doc="ngx_stream_proxy_module.xml" id="proxy_pass"/>
directive.
</para>

</section>


<section id="example" name="Example Configuration">

<para>
<example>
upstream <emphasis>backend</emphasis> {
    hash $remote_addr consistent;

    server backend1.example.com:12345  weight=5;
    server backend2.example.com:12345;
    server unix:/tmp/backend3;

    server backup1.example.com:12345   backup;
    server backup2.example.com:12345   backup;
}

server {
    listen 12346;
    proxy_pass <emphasis>backend</emphasis>;
}
</example>
</para>

<para>
Dynamically configurable group with
periodic <link doc="ngx_stream_upstream_hc_module.xml">health checks</link> is
available as part of our
<commercial_version>commercial subscription</commercial_version>:
<example>
resolver 10.0.0.1;

upstream <emphasis>dynamic</emphasis> {
    zone upstream_dynamic 64k;

    server backend1.example.com:12345 weight=5;
    server backend2.example.com:12345 fail_timeout=5s slow_start=30s;
    server 192.0.2.1:12345            max_fails=3;
    server backend3.example.com:12345 resolve;
    server backend4.example.com       service=http resolve;

    server backup1.example.com:12345  backup;
    server backup2.example.com:12345  backup;
}

server {
    listen 12346;
    proxy_pass <emphasis>dynamic</emphasis>;
    health_check;
}
</example>
</para>

</section>


<section id="directives" name="Directives">

<directive name="upstream">
<syntax block="yes"><value>name</value></syntax>
<default/>
<context>stream</context>

<para>
Defines a group of servers.
Servers can listen on different ports.
In addition, servers listening on TCP and UNIX-domain sockets
can be mixed.
</para>

<para>
Example:
<example>
upstream backend {
    server backend1.example.com:12345 weight=5;
    server 127.0.0.1:12345            max_fails=3 fail_timeout=30s;
    server unix:/tmp/backend2;
    server backend3.example.com:12345 resolve;

    server backup1.example.com:12345  backup;
}
</example>
</para>

<para>
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 <literal>backend1.example.com:12345</literal>
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.
</para>

</directive>


<directive name="server">
<syntax><value>address</value> [<value>parameters</value>]</syntax>
<default/>
<context>upstream</context>

<para>
Defines the <value>address</value> and other <value>parameters</value>
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 “<literal>unix:</literal>” prefix.
A domain name that resolves to several IP addresses defines
multiple servers at once.
</para>

<para>
The following parameters can be defined:
<list type="tag">

<tag-name id="weight">
<literal>weight</literal>=<value>number</value>
</tag-name>
<tag-desc>
sets the weight of the server, by default, 1.
</tag-desc>

<tag-name id="max_conns">
<literal>max_conns</literal>=<value>number</value>
</tag-name>
<tag-desc>
limits the maximum <value>number</value> of simultaneous
connections to the proxied server (1.11.5).
Default value is zero, meaning there is no limit.
If the server group does not reside in the <link id="zone">shared memory</link>,
the limitation works per each worker process.
<note>
Prior to version 1.11.5, this parameter was available as part of our
<commercial_version>commercial subscription</commercial_version>.
</note>
</tag-desc>

<tag-name id="max_fails">
<literal>max_fails</literal>=<value>number</value>
</tag-name>
<tag-desc>
sets the number of unsuccessful attempts to communicate with the server
that should happen in the duration set by the <literal>fail_timeout</literal>
parameter to consider the server unavailable for a duration also set by the
<literal>fail_timeout</literal> 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.
</tag-desc>

<tag-name id="fail_timeout">
<literal>fail_timeout</literal>=<value>time</value>
</tag-name>
<tag-desc>
sets
<list type="bullet">

<listitem>
the time during which the specified number of unsuccessful attempts to
communicate with the server should happen to consider the server unavailable;
</listitem>

<listitem>
and the period of time the server will be considered unavailable.
</listitem>

</list>
By default, the parameter is set to 10 seconds.
</tag-desc>

<tag-name id="backup">
<literal>backup</literal>
</tag-name>
<tag-desc>
marks the server as a backup server.
Connections to the backup server will be passed
when the primary servers are unavailable.
<note>
The parameter cannot be used along with the
<link id="hash"/> and <link id="random"/> load balancing methods.
</note>
</tag-desc>

<tag-name id="down">
<literal>down</literal>
</tag-name>
<tag-desc>
marks the server as permanently unavailable.
</tag-desc>

</list>
</para>

<para>
Additionally,
the following parameters are available as part of our
<commercial_version>commercial subscription</commercial_version>:
<list type="tag">

<tag-name id="resolve">
<literal>resolve</literal>
</tag-name>
<tag-desc>
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.
The server group must reside in the <link id="zone">shared memory</link>.
<para>
In order for this parameter to work,
the <literal>resolver</literal> directive
must be specified in the
<link doc="ngx_stream_core_module.xml" id="resolver">stream</link> block
or in the corresponding <link id="resolver">upstream</link> block.
</para>
</tag-desc>

<tag-name id="service">
<literal>service</literal>=<value>name</value>
</tag-name>
<tag-desc>
enables resolving of DNS
<link url="https://datatracker.ietf.org/doc/html/rfc2782">SRV</link>
records and sets the service <value>name</value> (1.9.13).
In order for this parameter to work, it is necessary to specify
the <link id="resolve"/> parameter for the server
and specify a hostname without a port number.
<para>
If the service name does not contain a dot (“<literal>.</literal>”), then
the <link url="https://datatracker.ietf.org/doc/html/rfc2782">RFC</link>-compliant name
is constructed
and the TCP protocol is added to the service prefix.
For example, to look up the
<literal>_http._tcp.backend.example.com</literal> SRV record,
it is necessary to specify the directive:
<example>
server backend.example.com service=http resolve;
</example>
If the service name contains one or more dots, then the name is constructed
by joining the service prefix and the server name.
For example, to look up the <literal>_http._tcp.backend.example.com</literal>
and <literal>server1.backend.example.com</literal> SRV records,
it is necessary to specify the directives:
<example>
server backend.example.com service=_http._tcp resolve;
server example.com service=server1.backend resolve;
</example>
</para>

<para>
Highest-priority SRV records
(records with the same lowest-number priority value)
are resolved as primary servers,
the rest of SRV records are resolved as backup servers.
If the <link id="backup"/> parameter is specified for the server,
high-priority SRV records are resolved as backup servers,
the rest of SRV records are ignored.
</para>
</tag-desc>

<tag-name id="slow_start">
<literal>slow_start</literal>=<value>time</value>
</tag-name>
<tag-desc>
sets the <value>time</value> during which the server will recover its weight
from zero to a nominal value, when unhealthy server becomes
<link doc="ngx_stream_upstream_hc_module.xml" id="health_check">healthy</link>,
or when the server becomes available after a period of time
it was considered <link id="fail_timeout">unavailable</link>.
Default value is zero, i.e. slow start is disabled.
<note>
The parameter cannot be used along with the
<link id="hash"/> and <link id="random"/> load balancing methods.
</note>
</tag-desc>

</list>
</para>

<para>
<note>
If there is only a single server in a group, <literal>max_fails</literal>,
<literal>fail_timeout</literal> and <literal>slow_start</literal> parameters
are ignored, and such a server will never be considered unavailable.
</note>
</para>

</directive>


<directive name="zone">
<syntax><value>name</value> [<value>size</value>]</syntax>
<default/>
<context>upstream</context>

<para>
Defines the <value>name</value> and <value>size</value> of the shared
memory zone that keeps the group’s configuration and run-time state that are
shared between worker processes.
Several groups may share the same zone.
In this case, it is enough to specify the <value>size</value> only once.
</para>

<para>
Additionally,
as part of our <commercial_version>commercial subscription</commercial_version>,
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 the
<link doc="../http/ngx_http_api_module.xml">API</link> module (1.13.3).
<note>
Prior to version 1.13.3,
the configuration was accessible only via a special location
handled by
<link doc="../http/ngx_http_upstream_conf_module.xml" id="upstream_conf"/>.
</note>
</para>

</directive>


<directive name="state">
<syntax><value>file</value></syntax>
<default/>
<context>upstream</context>
<appeared-in>1.9.7</appeared-in>

<para>
Specifies a <value>file</value> that keeps the state
of the dynamically configurable group.
</para>

<para>
Examples:
<example>
state /var/lib/nginx/state/servers.conf; # path for Linux
state /var/db/nginx/state/servers.conf;  # path for FreeBSD
</example>
</para>

<para>
The state is currently limited to the list of servers with their parameters.
The file is read when parsing the configuration and is updated each time
the upstream configuration is
<link doc="../http/ngx_http_api_module.xml" id="stream_upstreams_stream_upstream_name_servers_">changed</link>.
Changing the file content directly should be avoided.
The directive cannot be used
along with the <link id="server"/> directive.
</para>

<para>
<note>
Changes made during
<link doc="../control.xml" id="reconfiguration">configuration reload</link>
or <link doc="../control.xml" id="upgrade">binary upgrade</link>
can be lost.
</note>
</para>

<para>
<note>
This directive is available as part of our
<commercial_version>commercial subscription</commercial_version>.
</note>
</para>

</directive>


<directive name="hash">
<syntax><value>key</value> [<literal>consistent</literal>]</syntax>
<default/>
<context>upstream</context>

<para>
Specifies a load balancing method for a server group
where the client-server mapping is based on the hashed <value>key</value> value.
The <value>key</value> can contain text, variables,
and their combinations (1.11.2).
Usage example:
<example>
hash $remote_addr;
</example>
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
<link url="https://metacpan.org/pod/Cache::Memcached">Cache::Memcached</link>
Perl library.
</para>

<para>
If the <literal>consistent</literal> parameter is specified,
the <link url="https://www.metabrew.com/article/libketama-consistent-hashing-algo-memcached-clients">ketama</link>
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
<link url="https://metacpan.org/pod/Cache::Memcached::Fast">Cache::Memcached::Fast</link>
Perl library with the <value>ketama_points</value> parameter set to 160.
</para>

</directive>


<directive name="least_conn">
<syntax/>
<default/>
<context>upstream</context>

<para>
Specifies that a 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.
</para>

</directive>


<directive name="least_time">
<syntax>
    <literal>connect</literal> |
    <literal>first_byte</literal> |
    <literal>last_byte</literal>
    [<literal>inflight</literal>]</syntax>
<default/>
<context>upstream</context>

<para>
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.
</para>

<para>
If the <literal>connect</literal> parameter is specified,
time to
<link id="var_upstream_connect_time">connect</link>
to the upstream server is used.
If the <literal>first_byte</literal> parameter is specified,
time to receive the
<link id="var_upstream_first_byte_time">first byte</link> of data is used.
If the <literal>last_byte</literal> is specified,
time to receive the
<link id="var_upstream_session_time">last byte</link> of data is used.
If the <literal>inflight</literal> parameter is specified (1.11.6),
incomplete connections are also taken into account.
<note>
Prior to version 1.11.6,
incomplete connections were taken into account by default.
</note>
</para>

<para>
<note>
This directive is available as part of our
<commercial_version>commercial subscription</commercial_version>.
</note>
</para>

</directive>


<directive name="random">
<syntax>[<literal>two</literal> [<value>method</value>]]</syntax>
<default/>
<context>upstream</context>
<appeared-in>1.15.1</appeared-in>

<para>
Specifies that a group should use a load balancing method where a connection
is passed to a randomly selected server, taking into account weights
of servers.
</para>

<para>
The optional <literal>two</literal> parameter
instructs nginx to randomly select
<link url="https://homes.cs.washington.edu/~karlin/papers/balls.pdf">two</link>
servers and then choose a server
using the specified <literal>method</literal>.
The default method is <literal>least_conn</literal>
which passes a connection to a server
with the least number of active connections.
</para>

<para id="random_least_time">
The <literal>least_time</literal> method passes a connection to a server
with the least average time and least number of active connections.
If <literal>least_time=connect</literal> parameter is specified,
time to
<link id="var_upstream_connect_time">connect</link>
to the upstream server is used.
If <literal>least_time=first_byte</literal> parameter is specified,
time to receive the
<link id="var_upstream_first_byte_time">first byte</link> of data is used.
If <literal>least_time=last_byte</literal> is specified,
time to receive the
<link id="var_upstream_session_time">last byte</link> of data is used.
<note>
The <literal>least_time</literal> method is available as a part of our
<commercial_version>commercial subscription</commercial_version>.
</note>
</para>

</directive>


<directive name="resolver">
<syntax>
    <value>address</value> ...
    [<literal>valid</literal>=<value>time</value>]
    [<literal>ipv4</literal>=<literal>on</literal>|<literal>off</literal>]
    [<literal>ipv6</literal>=<literal>on</literal>|<literal>off</literal>]
    [<literal>status_zone</literal>=<value>zone</value>]</syntax>
<default/>
<context>upstream</context>
<appeared-in>1.17.5</appeared-in>

<para>
Configures name servers used to resolve names of upstream servers
into addresses, for example:
<example>
resolver 127.0.0.1 [::1]:5353;
</example>
The address can be specified as a domain name or IP address,
with an optional port.
If port is not specified, the port 53 is used.
Name servers are queried in a round-robin fashion.
</para>

<para id="resolver_ipv6">
By default, nginx will look up both IPv4 and IPv6 addresses while resolving.
If looking up of IPv4 or IPv6 addresses is not desired,
the <literal>ipv4=off</literal> (1.23.1) or
the <literal>ipv6=off</literal> parameter can be specified.
</para>

<para id="resolver_valid">
By default, nginx caches answers using the TTL value of a response.
The optional <literal>valid</literal> parameter allows overriding it:
<example>
resolver 127.0.0.1 [::1]:5353 valid=30s;
</example>
<note>
To prevent DNS spoofing, it is recommended
configuring DNS servers in a properly secured trusted local network.
</note>
</para>

<para id="resolver_status_zone">
The optional <literal>status_zone</literal> parameter
enables
<link doc="../http/ngx_http_api_module.xml" id="resolvers_">collection</link>
of DNS server statistics of requests and responses
in the specified <value>zone</value>.
</para>

<para>
<note>
This directive is available as part of our
<commercial_version>commercial subscription</commercial_version>.
</note>
</para>

</directive>


<directive name="resolver_timeout">
<syntax><value>time</value></syntax>
<default>30s</default>
<context>upstream</context>
<appeared-in>1.17.5</appeared-in>

<para>
Sets a timeout for name resolution, for example:
<example>
resolver_timeout 5s;
</example>
</para>

<para>
<note>
This directive is available as part of our
<commercial_version>commercial subscription</commercial_version>.
</note>
</para>

</directive>

</section>


<section id="variables" name="Embedded Variables">

<para>
The <literal>ngx_stream_upstream_module</literal> module
supports the following embedded variables:
<list type="tag">

<tag-name id="var_upstream_addr"><var>$upstream_addr</var></tag-name>
<tag-desc>
keeps the IP address and port,
or the path to the UNIX-domain socket of the upstream server (1.11.4).
If several servers were contacted during proxying,
their addresses are separated by commas, e.g.
“<literal>192.168.1.1:12345, 192.168.1.2:12345, unix:/tmp/sock</literal>”.
If a server cannot be selected,
the variable keeps the name of the server group.
</tag-desc>

<tag-name id="var_upstream_bytes_received"><var>$upstream_bytes_received</var></tag-name>
<tag-desc>
number of bytes received from an upstream server (1.11.4).
Values from several connections
are separated by commas like addresses in the
<link id="var_upstream_addr">$upstream_addr</link> variable.
</tag-desc>

<tag-name id="var_upstream_bytes_sent"><var>$upstream_bytes_sent</var></tag-name>
<tag-desc>
number of bytes sent to an upstream server (1.11.4).
Values from several connections
are separated by commas like addresses in the
<link id="var_upstream_addr">$upstream_addr</link> variable.
</tag-desc>

<tag-name id="var_upstream_connect_time"><var>$upstream_connect_time</var></tag-name>
<tag-desc>
time to connect to the upstream server (1.11.4);
the time is kept in seconds with millisecond resolution.
Times of several connections
are separated by commas like addresses in the
<link id="var_upstream_addr">$upstream_addr</link> variable.
</tag-desc>

<tag-name id="var_upstream_first_byte_time"><var>$upstream_first_byte_time</var></tag-name>
<tag-desc>
time to receive the first byte of data (1.11.4);
the time is kept in seconds with millisecond resolution.
Times of several connections
are separated by commas like addresses in the
<link id="var_upstream_addr">$upstream_addr</link> variable.
</tag-desc>

<tag-name id="var_upstream_session_time"><var>$upstream_session_time</var></tag-name>
<tag-desc>
session duration in seconds with millisecond resolution (1.11.4).
Times of several connections
are separated by commas like addresses in the
<link id="var_upstream_addr">$upstream_addr</link> variable.
</tag-desc>

</list>
</para>

</section>

</module>