Upstream#
The module is used to define groups of servers that can be referenced by the proxy_pass directive.
Configuration Example#
upstream backend {
hash $remote_addr consistent;
zone backend 1m;
server backend1.example.com:1935 weight=5;
server unix:/tmp/backend3;
server backend3.example.com service=_example._tcp resolve;
server backup1.example.com:1935 backup;
server backup2.example.com:1935 backup;
}
resolver 127.0.0.53 status_zone=resolver;
server {
listen 1936;
proxy_pass backend;
}
Directives#
upstream#
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:1935 weight=5;
server 127.0.0.1:1935 max_fails=3 fail_timeout=30s;
server unix:/tmp/backend2;
server backend3.example.com:1935 resolve;
server backup1.example.com:1935 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#
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:
|
sets the weight of the server |
|
limits the maximum number of simultaneous active connections to the proxied server. |
max_fails=
number — sets the number of unsuccessful attempts
to communicate with the server
that should happen in the duration set by fail_timeout
to consider the server unavailable;
it is then retried after the same duration.
Here, an unsuccessful attempt is an error or timeout while establishing a connection with the server.
Note
If a server
in an upstream resolves into multiple peers,
its max_fails
setting applies to each peer individually.
If an upstream contains only one peer
after all its server
directives are resolved,
the max_fails
setting has no effect and will be ignored.
|
the default number of unsuccessful attempts |
|
disables the accounting of attempts |
fail_timeout=
time — sets the period of time during which a number of
unsuccessful attempts to communicate with the server (max_fails) should happen to consider the server unavailable. The server
then becomes unavailable for the same amount of time before it is retried.
By default, this is set to 10 seconds.
Note
If a server
in an upstream resolves into multiple peers,
its fail_timeout
setting applies to each peer individually.
If an upstream contains only one peer
after all its server
directives are resolved,
the fail_timeout
setting has no effect and will be ignored.
|
marks the server as a backup server. It will be passed requests when the primary servers are unavailable. |
|
marks the server as permanently unavailable. |
|
sets the server to draining; this means
it receives only requests from the sessions
that were bound earlier with sticky.
Otherwise it behaves similarly to |
Caution
The backup
parameter cannot be used along with the hash and random load balancing methods.
The down
and drain
options are mutually exclusive.
Added in version 1.3.0.
|
Enables monitoring changes to the list of IP addresses that corresponds to a domain name, updating it without a configuration reload. The group should be stored in a shared memory zone; also, you need to define a resolver. |
|
Enables resolving DNS SRV records and sets the service name. For this parameter to work, specify the resolve server parameter, providing a hostname without a port number. If there are no dots in the service name,
the name is formed according to the RFC standard:
the service name is prefixed with Angie resolves the SRV records by combining the normalized service name and the hostname and obtaining the list of servers for the combination via DNS, along with their priorities and weights.
|
This example will look up the _http._tcp.backend.example.com
record:
server backend.example.com service=http resolve;
Added in version 1.4.0.
|
sets the time to recover the If the value is set and the server is again considered available and healthy as defined by max_fails and upstream_probe (PRO), the server will steadily recover its designated weight within the allocated timeframe. If the value isn’t set, the server in a similar situation will recover its designated weight immediately. |
Note
If there’s only one server
in an upstream,
slow_start
has no effect and will be ignored.
state (PRO)#
Added in version 1.4.0: PRO
Specifies the file where the upstream’s server list is persisted.
When installing from
our packages,
a designated
/var/lib/angie/state/
(/var/db/angie/state/
on FreeBSD)
directory with appropriate permissions
is created to store these files,
so you will only need to add the file’s basename in the configuration:
upstream backend {
zone backend 1m;
state /var/lib/angie/state/<FILE NAME>;
}
The format of this server list is similar to server
. The contents of
the file change whenever there is any modification to servers in the
/config/stream/upstreams/ section
via the configuration API.
The file is read at Angie start or configuration reload.
Caution
For the state
directive to be used in an upstream
block,
the block should have no server
directives;
instead, it must have a shared memory zone (zone).
zone#
Defines the name and size 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 size only once.
feedback (PRO)#
Added in version 1.7.0: PRO
|
|
Default |
— |
upstream |
Enables a feedback-based load balancing mechanism for the upstream
.
It adjusts the load balancing decisions dynamically,
multiplying each peer’s weight by its average feedback value
that is affected by the value of a variable over time
and is subject to an optional condition.
The following parameters are accepted:
|
The variable from which the feedback value is taken. It should represent a performance or health metric, and is intended to be supplied by the peer. The value is assessed at each response from the peer
and factored into the rolling average
according to |
|
If set, the feedback value is interpreted inversely, meaning lower values indicate better performance. |
|
The factor by which the feedback value is weighted
when calculating the average.
Valid values are integers between 0 and 99.
By default — The average feedback is calculated using the exponential moving average formula. The larger is the factor, the less is the average affected by new values;
if the factor is set to |
|
Specifies a condition variable
that controls how connections are included in the calculation.
The average is updated with the feedback value
only if the condition variable
isn’t Note By default, traffic from probes
isn’t included in the calculation;
combining the $upstream_probe variable
with |
Example:
upstream backend {
zone backend 1m;
feedback $feedback_value factor=80 account=$condition_value;
server backend1.example.com:1935 weight=1;
server backend2.example.com:1935 weight=2;
}
map $protocol $feedback_value {
"TCP" 100;
"UDP" 75;
default 10;
}
map $upstream_probe $condition_value {
"high_priority" "1";
"low_priority" "0";
default "1";
}
This categorizes servers into different feedback levels
based on specific protocols used for different sessions,
and also adds a condition mapped from $upstream_probe
to account only for the high_priority
probe
or regular client sessions.
hash#
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 (1.11.2). Usage example:
hash $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#
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.
least_time (PRO)#
|
|
Default |
— |
upstream |
Sets the load balancing method for a group where the probability of forwarding a connection to an active server is inversely proportional to the average time it takes to respond; the smaller the response time, the more conections the server will receive.
|
The directive accounts for the average time to establish the connection. |
|
The directive uses the average time to receive the first byte of the response. |
|
The directive uses the average time to receive the entire response. |
Added in version 1.7.0: PRO
|
Serves the same purpose as response_time_factor (PRO) and overrides it if set. |
|
Specifies a condition variable
that controls which connections should be included in the calculation.
The average is updated
only if the condition variable for the connection
isn’t Note By default, probes
aren’t included in the calculation;
combining the $upstream_probe variable
with |
The respective moving averages, adjusted for factor
and account
,
are also presented as connect_time
, first_byte_time
, and
last_byte_time
in the health
object of the server among the
stream upstream metrics in the API.
random#
Specifies that a group should use a load balancing method where a request is passed to a randomly selected server, taking into account weights of servers.
The optional two
parameter instructs Angie to randomly select two servers and then choose a server using the specified method. The default method is least_conn which passes a request to a server with the least number of active connections.
response_time_factor (PRO)#
Sets the smoothing factor for the least_time (PRO) load balancing method, using the previous value when calculating the average response time according to the formula of the exponential weighted moving average.
The larger the specified number, the less new values influence the average;
if 90
is specified, 90% of the previous value will be taken,
and only 10% of the new value. Acceptable values range from 0 to 99 inclusive.
The respective moving averages are presented as connect_time
(time to
establish the connection), first_byte_time
(time to receive the first
byte of the response), and last_byte_time
(time to receive the complete
response) in the health
object of the server among the stream
upstream metrics in the API.
Note
Only successful responses are considered in the calculation; what constitutes an unsuccessful response is determined by the proxy_next_upstream directives.
sticky#
Added in version 1.6.0: Angie
Added in version 1.6.0: Angie PRO
|
|
Default |
— |
upstream |
Configures the binding of client sessions to proxied servers
in the mode specified by the first parameter;
to drain requests from servers
that have sticky
defined,
use the drain
option in the server block.
Attention
The sticky
directive must be used after all directives
that set the load balancing method;
otherwise, it won’t work.
This mode uses predefined route identifiers that can be embedded in any connection properties Angie can access. It is less flexible because it relies on predefined values but can suit better if such identifiers are already in place.
Here, when a connection is established with the proxied server, it can assign a route to the client and return its identifier in a manner that they both are aware of. The value of the sid parameter of the server directive must be used as the route identifier. Note that the parameter is additionally hashed if the sticky_secret directive is set.
Subsequent connections from clients that wish to use this route must contain the identifier issued by the server in a way that ensures it ends up in Angie variables.
The directive lists specific variables used for routing. To select the server where the incoming connection is routed, the first non-empty variable is used; it is then compared with the sid parameter of the server directive. If selecting a server fails or the chosen server can’t accept the connection, another server is selected according to the configured balancing method.
Here,
Angie looks for the identifier in a custom $route
variable,
which is mapped from $ssl_preread_server_name
(note that ssl_preread must be enabled):
stream {
map $ssl_preread_server_name $route {
a.example.com a;
b.example.com b;
default "";
}
upstream backend {
server 127.0.0.1:8081 sid=a;
server 127.0.0.1:8082 sid=b;
sticky route $route;
}
server {
listen 127.0.0.1:8080;
ssl_preread on;
proxy_pass backend;
}
}
This mode uses a dynamically generated key to associate a client with a particular proxied server; it’s more flexible because it assigns servers on the go, stores sessions in a shared memory zone, and supports different ways of passing session identifiers.
Here, a session is created
based on the connection properties from the proxied server.
The create
and lookup
parameters list variables
indicating how new sessions are created
and existing sessions are looked up.
Both parameters can occur multiple times.
The session identifier is the value of the first non-empty variable
specified with create
;
for example, this could be the
name of the proxied server.
Sessions are stored in a shared memory zone;
its name and size are set by the zone
parameter.
If a session has been inactive for the time set by timeout
,
it is deleted.
The default is 1 hour.
Subsequent connections from clients that wish to use the session
must contain its identifier,
ensuring that it ends up in a non-empty variable
specified with lookup
;
its value will then be matched against sessions in shared memory.
If selecting a server fails
or the chosen server can’t accept the connection,
another server is selected
according to the configured balancing method.
The connect
parameter allows creating a session
immediately after the connection to the proxied server was established.
Without it, a session is created only after processing the connection.
In the example, Angie creates and looks up sessions, using the $rdp_cookie variable:
stream {
upstream backend {
server 127.0.0.1:3390 sid=a;
server 127.0.0.1:3391 sid=b;
sticky learn lookup=$rdp_cookie create=$rdp_cookie zone=sessions:1m;
}
server {
listen 127.0.0.1:3389;
ssl_preread on;
proxy_pass backend;
}
}
sticky_strict#
Added in version 1.6.0: Angie
Added in version 1.6.0: Angie PRO
When enabled, causes Angie to return a connection error to the client if the desired server is unavailable, rather than using any other available server as it would when no servers in the upstream are available.
sticky_secret#
Added in version 1.6.0: Angie
Added in version 1.6.0: Angie PRO
Adds the string as the salt value to the MD5 hashing function
for the sticky directive in the route
mode.
The string may contain variables, for example, $remote_addr:
upstream backend {
server 127.0.0.1:8081 sid=a;
server 127.0.0.1:8082 sid=b;
sticky route $route;
sticky_secret my_secret.$remote_addr;
}
Salt is appended to the value being hashed; to verify the hashing mechanism independently:
$ echo -n "<VALUE><SALT>" | md5sum
Built-in Variables#
The stream_upstream
module supports the following built-in 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 commas, e.g. :
192.168.1.1:1935, 192.168.1.2:1935, unix:/tmp/sock
If a server cannot be selected, the variable keeps the name of the server group.
$upstream_bytes_received
#
number of bytes received from an upstream server. Values from several connections are separated by commas like addresses in the $upstream_addr variable.
$upstream_bytes_sent
#
number of bytes sent to an upstream server. Values from several connections are separated by commas like addresses in the $upstream_addr variable.
$upstream_connect_time
#
time to connect to the upstream server; the time is kept in seconds with millisecond resolution. Times of several connections are separated by commas like addresses in the $upstream_addr variable.
$upstream_first_byte_time
#
time to receive the first byte of data; the time is kept in seconds with millisecond resolution. Times of several connections are separated by commas like addresses in the $upstream_addr variable.
$upstream_session_time
name#
session duration in seconds with millisecond resolution. Times of several connections are separated by commas like addresses in the $upstream_addr variable.
$upstream_sticky_status
#
Status of sticky connections.
|
Connection routed to upstream without sticky enabled. |
|
Connection without sticky information. |
|
Connection with sticky information routed to the desired backend. |
|
Connection with sticky information routed to the backend selected by the load balancing algorithm. |
Values from multiple connections are separated by commas and colons, similar to addresses in the $upstream_addr variable.