View Source Phoenix.Endpoint behaviour (Phoenix v1.7.18)

Defines a Phoenix endpoint.

The endpoint is the boundary where all requests to your web application start. It is also the interface your application provides to the underlying web servers.

Overall, an endpoint has three responsibilities:

  • to provide a wrapper for starting and stopping the endpoint as part of a supervision tree

  • to define an initial plug pipeline for requests to pass through

  • to host web specific configuration for your application

Endpoints

An endpoint is simply a module defined with the help of Phoenix.Endpoint. If you have used the mix phx.new generator, an endpoint was automatically generated as part of your application:

defmodule YourAppWeb.Endpoint do
  use Phoenix.Endpoint, otp_app: :your_app

  # plug ...
  # plug ...

  plug YourApp.Router
end

Endpoints must be explicitly started as part of your application supervision tree. Endpoints are added by default to the supervision tree in generated applications. Endpoints can be added to the supervision tree as follows:

children = [
  YourAppWeb.Endpoint
]

Endpoint configuration

All endpoints are configured in your application environment. For example:

config :your_app, YourAppWeb.Endpoint,
  secret_key_base: "kjoy3o1zeidquwy1398juxzldjlksahdk3"

Endpoint configuration is split into two categories. Compile-time configuration means the configuration is read during compilation and changing it at runtime has no effect. The compile-time configuration is mostly related to error handling.

Runtime configuration, instead, is accessed during or after your application is started and can be read through the config/2 function:

YourAppWeb.Endpoint.config(:port)
YourAppWeb.Endpoint.config(:some_config, :default_value)

Compile-time configuration

Compile-time configuration may be set on config/dev.exs, config/prod.exs and so on, but has no effect on config/runtime.exs:

  • :code_reloader - when true, enables code reloading functionality. For the list of code reloader configuration options see Phoenix.CodeReloader.reload/1. Keep in mind code reloading is based on the file-system, therefore it is not possible to run two instances of the same app at the same time with code reloading in development, as they will race each other and only one will effectively recompile the files. In such cases, tweak your config files so code reloading is enabled in only one of the apps or set the MIX_BUILD environment variable to give them distinct build directories

  • :debug_errors - when true, uses Plug.Debugger functionality for debugging failures in the application. Recommended to be set to true only in development as it allows listing of the application source code during debugging. Defaults to false

  • :force_ssl - ensures no data is ever sent via HTTP, always redirecting to HTTPS. It expects a list of options which are forwarded to Plug.SSL. By default it sets the "strict-transport-security" header in HTTPS requests, forcing browsers to always use HTTPS. If an unsafe request (HTTP) is sent, it redirects to the HTTPS version using the :host specified in the :url configuration. To dynamically redirect to the host of the current request, set :host in the :force_ssl configuration to nil

Runtime configuration

The configuration below may be set on config/dev.exs, config/prod.exs and so on, as well as on config/runtime.exs. Typically, if you need to configure them with system environment variables, you set them in config/runtime.exs. These options may also be set when starting the endpoint in your supervision tree, such as {MyApp.Endpoint, options}.

  • :adapter - which webserver adapter to use for serving web requests. See the "Adapter configuration" section below

  • :cache_static_manifest - a path to a json manifest file that contains static files and their digested version. This is typically set to "priv/static/cache_manifest.json" which is the file automatically generated by mix phx.digest. It can be either: a string containing a file system path or a tuple containing the application name and the path within that application.

  • :cache_static_manifest_latest - a map of the static files pointing to their digest version. This is automatically loaded from cache_static_manifest on boot. However, if you have your own static handling mechanism, you may want to set this value explicitly. This is used by projects such as LiveView to detect if the client is running on the latest version of all assets.

  • :cache_manifest_skip_vsn - when true, skips the appended query string "?vsn=d" when generating paths to static assets. This query string is used by Plug.Static to set long expiry dates, therefore, you should set this option to true only if you are not using Plug.Static to serve assets, for example, if you are using a CDN. If you are setting this option, you should also consider passing --no-vsn to mix phx.digest. Defaults to false.

  • :check_origin - configure the default :check_origin setting for transports. See socket/3 for options. Defaults to true.

  • :secret_key_base - a secret key used as a base to generate secrets for encrypting and signing data. For example, cookies and tokens are signed by default, but they may also be encrypted if desired. Defaults to nil as it must be set per application

  • :server - when true, starts the web server when the endpoint supervision tree starts. Defaults to false. The mix phx.server task automatically sets this to true

  • :url - configuration for generating URLs throughout the app. Accepts the :host, :scheme, :path and :port options. All keys except :path can be changed at runtime. Defaults to:

    [host: "localhost", path: "/"]

    The :port option requires either an integer or string. The :host option requires a string.

    The :scheme option accepts "http" and "https" values. Default value is inferred from top level :http or :https option. It is useful when hosting Phoenix behind a load balancer or reverse proxy and terminating SSL there.

    The :path option can be used to override root path. Useful when hosting Phoenix behind a reverse proxy with URL rewrite rules

  • :static_url - configuration for generating URLs for static files. It will fallback to url if no option is provided. Accepts the same options as url

  • :watchers - a set of watchers to run alongside your server. It expects a list of tuples containing the executable and its arguments. Watchers are guaranteed to run in the application directory, but only when the server is enabled (unless :force_watchers configuration is set to true). For example, the watcher below will run the "watch" mode of the webpack build tool when the server starts. You can configure it to whatever build tool or command you want:

    [
      node: [
        "node_modules/webpack/bin/webpack.js",
        "--mode",
        "development",
        "--watch",
        "--watch-options-stdin"
      ]
    ]

    The :cd and :env options can be given at the end of the list to customize the watcher:

    [node: [..., cd: "assets", env: [{"TAILWIND_MODE", "watch"}]]]

    A watcher can also be a module-function-args tuple that will be invoked accordingly:

    [another: {Mod, :fun, [arg1, arg2]}]
  • :force_watchers - when true, forces your watchers to start even when the :server option is set to false.

  • :live_reload - configuration for the live reload option. Configuration requires a :patterns option which should be a list of file patterns to watch. When these files change, it will trigger a reload.

    live_reload: [
      url: "ws://localhost:4000",
      patterns: [
        ~r"priv/static/(?!uploads/).*(js|css|png|jpeg|jpg|gif|svg)$",
        ~r"lib/app_web/(live|views)/.*(ex)$",
        ~r"lib/app_web/templates/.*(eex)$"
      ]
    ]
  • :pubsub_server - the name of the pubsub server to use in channels and via the Endpoint broadcast functions. The PubSub server is typically started in your supervision tree.

  • :render_errors - responsible for rendering templates whenever there is a failure in the application. For example, if the application crashes with a 500 error during a HTML request, render("500.html", assigns) will be called in the view given to :render_errors. A :formats list can be provided to specify a module per format to handle error rendering. Example:

    [formats: [html: MyApp.ErrorHTML], layout: false, log: :debug]
  • :log_access_url - log the access url once the server boots

Note that you can also store your own configurations in the Phoenix.Endpoint. For example, Phoenix LiveView expects its own configuration under the :live_view key. In such cases, you should consult the documentation of the respective projects.

Adapter configuration

Phoenix allows you to choose which webserver adapter to use. Newly generated applications created via the phx.new Mix task use the Bandit webserver via the Bandit.PhoenixAdapter adapter. If not otherwise specified via the adapter option Phoenix will fall back to the Phoenix.Endpoint.Cowboy2Adapter for backwards compatibility with applications generated prior to Phoenix 1.7.8.

Both adapters can be configured in a similar manner using the following two top-level options:

  • :http - the configuration for the HTTP server. It accepts all options as defined by either Bandit or Plug.Cowboy depending on your choice of adapter. Defaults to false

  • :https - the configuration for the HTTPS server. It accepts all options as defined by either Bandit or Plug.Cowboy depending on your choice of adapter. Defaults to false

In addition, the connection draining can be configured for the Cowboy webserver via the following top-level option (this is not required for Bandit as it has connection draining built-in):

  • :drainer - a drainer process waits for any on-going request to finish during application shutdown. It accepts the :shutdown and :check_interval options as defined by Plug.Cowboy.Drainer. Note the draining does not terminate any existing connection, it simply waits for them to finish. Socket connections run their own drainer before this one is invoked. That's because sockets are stateful and can be gracefully notified, which allows us to stagger them over a longer period of time. See the documentation for socket/3 for more information

Endpoint API

In the previous section, we have used the config/2 function that is automatically generated in your endpoint. Here's a list of all the functions that are automatically defined in your endpoint:

Summary

Callbacks

Broadcasts a msg as event in the given topic to all nodes.

Broadcasts a msg as event in the given topic to all nodes.

Broadcasts a msg from the given from as event in the given topic to all nodes.

Broadcasts a msg from the given from as event in the given topic to all nodes.

Access the endpoint configuration given by key.

Reload the endpoint configuration on application upgrades.

Returns the host from the :url configuration.

Broadcasts a msg as event in the given topic within the current node.

Broadcasts a msg from the given from as event in the given topic within the current node.

Generates the path information when routing to this endpoint.

Returns the script name from the :url configuration.

Returns the address and port that the server is running on

Starts the endpoint supervision tree.

Generates an integrity hash to a static file in priv/static.

Generates a two item tuple containing the static_path and static_integrity.

Generates a route to a static file in priv/static.

Generates the static URL without any path information.

Generates the endpoint base URL, but as a URI struct.

Subscribes the caller to the given topic.

Unsubscribes the caller from the given topic.

Generates the endpoint base URL without any path information.

Functions

Checks if Endpoint's web server has been configured to start.

Defines a websocket/longpoll mount-point for a socket.

Types

event()

@type event() :: String.t()

msg()

@type msg() :: map() | {:binary, binary()}

topic()

@type topic() :: String.t()

Callbacks

broadcast(topic, event, msg)

@callback broadcast(topic(), event(), msg()) :: :ok | {:error, term()}

Broadcasts a msg as event in the given topic to all nodes.

broadcast!(topic, event, msg)

@callback broadcast!(topic(), event(), msg()) :: :ok

Broadcasts a msg as event in the given topic to all nodes.

Raises in case of failures.

broadcast_from(from, topic, event, msg)

@callback broadcast_from(from :: pid(), topic(), event(), msg()) :: :ok | {:error, term()}

Broadcasts a msg from the given from as event in the given topic to all nodes.

broadcast_from!(from, topic, event, msg)

@callback broadcast_from!(from :: pid(), topic(), event(), msg()) :: :ok

Broadcasts a msg from the given from as event in the given topic to all nodes.

Raises in case of failures.

config key, default

@callback config(key :: atom(), default :: term()) :: term()

Access the endpoint configuration given by key.

config_change(changed, removed)

@callback config_change(changed :: term(), removed :: term()) :: term()

Reload the endpoint configuration on application upgrades.

host()

@callback host() :: String.t()

Returns the host from the :url configuration.

local_broadcast(topic, event, msg)

@callback local_broadcast(topic(), event(), msg()) :: :ok

Broadcasts a msg as event in the given topic within the current node.

local_broadcast_from(from, topic, event, msg)

@callback local_broadcast_from(from :: pid(), topic(), event(), msg()) :: :ok

Broadcasts a msg from the given from as event in the given topic within the current node.

path(path)

@callback path(path :: String.t()) :: String.t()

Generates the path information when routing to this endpoint.

script_name()

@callback script_name() :: [String.t()]

Returns the script name from the :url configuration.

server_info(scheme)

@callback server_info(Plug.Conn.scheme()) ::
  {:ok,
   {:inet.ip_address(), :inet.port_number()} | :inet.returned_non_ip_address()}
  | {:error, term()}

Returns the address and port that the server is running on

start_link(keyword)

@callback start_link(keyword()) :: Supervisor.on_start()

Starts the endpoint supervision tree.

Starts endpoint's configuration cache and possibly the servers for handling requests.

static_integrity(path)

@callback static_integrity(path :: String.t()) :: String.t() | nil

Generates an integrity hash to a static file in priv/static.

static_lookup(path)

@callback static_lookup(path :: String.t()) ::
  {String.t(), String.t()} | {String.t(), nil}

Generates a two item tuple containing the static_path and static_integrity.

static_path(path)

@callback static_path(path :: String.t()) :: String.t()

Generates a route to a static file in priv/static.

static_url()

@callback static_url() :: String.t()

Generates the static URL without any path information.

struct_url()

@callback struct_url() :: URI.t()

Generates the endpoint base URL, but as a URI struct.

subscribe(topic, opts)

@callback subscribe(topic(), opts :: Keyword.t()) :: :ok | {:error, term()}

Subscribes the caller to the given topic.

See Phoenix.PubSub.subscribe/3 for options.

unsubscribe(topic)

@callback unsubscribe(topic()) :: :ok | {:error, term()}

Unsubscribes the caller from the given topic.

url()

@callback url() :: String.t()

Generates the endpoint base URL without any path information.

Functions

server?(otp_app, endpoint)

Checks if Endpoint's web server has been configured to start.

  • otp_app - The OTP app running the endpoint, for example :my_app
  • endpoint - The endpoint module, for example MyAppWeb.Endpoint

Examples

iex> Phoenix.Endpoint.server?(:my_app, MyAppWeb.Endpoint)
true

socket(path, module, opts \\ [])

(macro)

Defines a websocket/longpoll mount-point for a socket.

It expects a path, a socket module, and a set of options. The socket module is typically defined with Phoenix.Socket.

Both websocket and longpolling connections are supported out of the box.

Options

  • :websocket - controls the websocket configuration. Defaults to true. May be false or a keyword list of options. See "Common configuration" and "WebSocket configuration" for the whole list

  • :longpoll - controls the longpoll configuration. Defaults to false. May be true or a keyword list of options. See "Common configuration" and "Longpoll configuration" for the whole list

  • :drainer - a keyword list or a custom MFA function returning a keyword list, for example:

    {MyAppWeb.Socket, :drainer_configuration, []}

    configuring how to drain sockets on application shutdown. The goal is to notify all channels (and LiveViews) clients to reconnect. The supported options are:

    • :batch_size - How many clients to notify at once in a given batch. Defaults to 10000.
    • :batch_interval - The amount of time in milliseconds given for a batch to terminate. Defaults to 2000ms.
    • :shutdown - The maximum amount of time in milliseconds allowed to drain all batches. Defaults to 30000ms.

    For example, if you have 150k connections, the default values will split them into 15 batches of 10k connections. Each batch takes 2000ms before the next batch starts. In this case, we will do everything right under the maximum shutdown time of 30000ms. Therefore, as you increase the number of connections, remember to adjust the shutdown accordingly. Finally, after the socket drainer runs, the lower level HTTP/HTTPS connection drainer will still run, and apply to all connections. Set it to false to disable draining.

You can also pass the options below on use Phoenix.Socket. The values specified here override the value in use Phoenix.Socket.

Examples

socket "/ws", MyApp.UserSocket

socket "/ws/admin", MyApp.AdminUserSocket,
  longpoll: true,
  websocket: [compress: true]

Path params

It is possible to include variables in the path, these will be available in the params that are passed to the socket.

socket "/ws/:user_id", MyApp.UserSocket,
  websocket: [path: "/project/:project_id"]

Common configuration

The configuration below can be given to both :websocket and :longpoll keys:

  • :path - the path to use for the transport. Will default to the transport name ("/websocket" or "/longpoll")

  • :serializer - a list of serializers for messages. See Phoenix.Socket for more information

  • :transport_log - if the transport layer itself should log and, if so, the level

  • :check_origin - if the transport should check the origin of requests when the origin header is present. May be true, false, a list of hosts that are allowed, or a function provided as MFA tuple. Defaults to :check_origin setting at endpoint configuration.

    If true, the header is checked against :host in YourAppWeb.Endpoint.config(:url)[:host].

    If false and you do not validate the session in your socket, your app is vulnerable to Cross-Site WebSocket Hijacking (CSWSH) attacks. Only use in development, when the host is truly unknown or when serving clients that do not send the origin header, such as mobile apps.

    You can also specify a list of explicitly allowed origins. Wildcards are supported.

    check_origin: [
      "https://example.com",
      "//another.com:888",
      "//*.other.com"
    ]

    Or to accept any origin matching the request connection's host, port, and scheme:

    check_origin: :conn

    Or a custom MFA function:

    check_origin: {MyAppWeb.Auth, :my_check_origin?, []}

    The MFA is invoked with the request %URI{} as the first argument, followed by arguments in the MFA list, and must return a boolean.

  • :code_reloader - enable or disable the code reloader. Defaults to your endpoint configuration

  • :connect_info - a list of keys that represent data to be copied from the transport to be made available in the user socket connect/3 callback. See the "Connect info" subsection for valid keys

Connect info

The valid keys are:

  • :peer_data - the result of Plug.Conn.get_peer_data/1

  • :trace_context_headers - a list of all trace context headers. Supported headers are defined by the W3C Trace Context Specification. These headers are necessary for libraries such as OpenTelemetry to extract trace propagation information to know this request is part of a larger trace in progress.

  • :x_headers - all request headers that have an "x-" prefix

  • :uri - a %URI{} with information from the conn

  • :user_agent - the value of the "user-agent" request header

  • {:session, session_config} - the session information from Plug.Conn. The session_config is typically an exact copy of the arguments given to Plug.Session. In order to validate the session, the "_csrf_token" must be given as request parameter when connecting the socket with the value of URI.encode_www_form(Plug.CSRFProtection.get_csrf_token()). The CSRF token request parameter can be modified via the :csrf_token_key option.

    Additionally, session_config may be a MFA, such as {MyAppWeb.Auth, :get_session_config, []}, to allow loading config in runtime.

Arbitrary keywords may also appear following the above valid keys, which is useful for passing custom connection information to the socket.

For example:

  socket "/socket", AppWeb.UserSocket,
      websocket: [
        connect_info: [:peer_data, :trace_context_headers, :x_headers, :uri, session: [store: :cookie]]
      ]

With arbitrary keywords:

  socket "/socket", AppWeb.UserSocket,
      websocket: [
        connect_info: [:uri, custom_value: "abcdef"]
      ]

Where are my headers?

Phoenix only gives you limited access to the connection headers for security reasons. WebSockets are cross-domain, which means that, when a user "John Doe" visits a malicious website, the malicious website can open up a WebSocket connection to your application, and the browser will gladly submit John Doe's authentication/cookie information. If you were to accept this information as is, the malicious website would have full control of a WebSocket connection to your application, authenticated on John Doe's behalf.

To safe-guard your application, Phoenix limits and validates the connection information your socket can access. This means your application is safe from these attacks, but you can't access cookies and other headers in your socket. You may access the session stored in the connection via the :connect_info option, provided you also pass a csrf token when connecting over WebSocket.

Websocket configuration

The following configuration applies only to :websocket.

  • :timeout - the timeout for keeping websocket connections open after it last received data, defaults to 60_000ms

  • :max_frame_size - the maximum allowed frame size in bytes, defaults to "infinity"

  • :fullsweep_after - the maximum number of garbage collections before forcing a fullsweep for the socket process. You can set it to 0 to force more frequent cleanups of your websocket transport processes. Setting this option requires Erlang/OTP 24

  • :compress - whether to enable per message compression on all data frames, defaults to false

  • :subprotocols - a list of supported websocket subprotocols. Used for handshake Sec-WebSocket-Protocol response header, defaults to nil.

    For example:

    subprotocols: ["sip", "mqtt"]
  • :error_handler - custom error handler for connection errors. If Phoenix.Socket.connect/3 returns an {:error, reason} tuple, the error handler will be called with the error reason. For WebSockets, the error handler must be a MFA tuple that receives a Plug.Conn, the error reason, and returns a Plug.Conn with a response. For example:

    socket "/socket", MySocket,
        websocket: [
          error_handler: {MySocket, :handle_error, []}
        ]

    and a {:error, :rate_limit} return may be handled on MySocket as:

    def handle_error(conn, :rate_limit), do: Plug.Conn.send_resp(conn, 429, "Too many requests")

Longpoll configuration

The following configuration applies only to :longpoll:

  • :window_ms - how long the client can wait for new messages in its poll request in milliseconds (ms). Defaults to 10_000.

  • :pubsub_timeout_ms - how long a request can wait for the pubsub layer to respond in milliseconds (ms). Defaults to 2000.

  • :crypto - options for verifying and signing the token, accepted by Phoenix.Token. By default tokens are valid for 2 weeks