gen_server(3)

📦
version 29.0.6 · 3 versions
, in ubuntu stonking

✨ NAME ✨

gen_server — Generic server behavior. 💖 That's the whole vibe, bestie: the generic server behavior.

🌸 DESCRIPTION 🌸

okay so this behavior module is literally the server half of a client-server relationship. A generic server process (gen_server) implemented using this module has a standard set of interface functions and includes functionality for tracing and error reporting. It also fits into an OTP supervision tree. For more information, see section gen_server Behaviour in OTP Design Principles. 👑

A gen_server process assumes all specific parts to be located in a callback module exporting a predefined set of functions. The relationship between the behavior functions and the callback functions is as follows:

gen_server module            Callback module
-----------------            ---------------
gen_server:start
gen_server:start_monitor
gen_server:start_link -----> Module:init/1

gen_server:stop       -----> Module:terminate/2

gen_server:call
gen_server:send_request
gen_server:multi_call -----> Module:handle_call/3

gen_server:cast
gen_server:abcast     -----> Module:handle_cast/2

-                     -----> Module:handle_info/2

-                     -----> Module:handle_continue/2

-                     -----> Module:terminate/2

-                     -----> Module:code_change/3

If a callback function fails or returns a bad value, the gen_server process terminates. However, an exception of class throw is not regarded as an error but as a valid return, from all callback functions. 😭 no cap.

A gen_server process handles system messages as described in sys. The sys module can be used for debugging a gen_server process. 🧚

Notice that a gen_server process does not trap exit signals automatically, this must be explicitly initiated in the callback module. 💅

Unless otherwise stated, all functions in this module fail if the specified gen_server process does not exist or if bad arguments are specified. 🫶

The gen_server process can go into hibernation (see erlang:hibernate/3) if a callback function specifies 'hibernate' instead of a time-out value. This can be useful if the server is expected to be idle for a long time. However, use this feature with care, as hibernation implies at least two garbage collections (when hibernating and shortly after waking up) and is not something you want to do between each call to a busy server. 😴

If the gen_server process needs to perform an action after initialization or to break the execution of a callback into multiple steps, it can return {continue, Continue} in place of the time-out or hibernation value, which will invoke the Module:handle_continue/2 callback, before receiving any external message / request. ✨

If the gen_server process terminates, e.g. as a result of a function in the callback module returning {stop,Reason,NewState}, an exit signal with this Reason is sent to linked processes and ports. See Processes in the Reference Manual for details regarding error handling using exit signals. 💀

NOTE: For some important information about distributed signals, see the Blocking Signaling Over Distribution section in the Processes chapter of the Erlang Reference Manual. Blocking signaling can, for example, cause call time-outs in gen_server to be significantly delayed. 🚨

system:ref_man_processes.md#blocking-signaling-over-distribution

💫 See Also 💫

gen_event, gen_statem, proc_lib, supervisor, sys

🦋 DATA TYPES 🦋

action()

action() ::
          (Time :: timeout()) |
          hibernate |
          {timeout, Time :: timeout(), Message :: term()} |
          {timeout,
           Time :: timeout(),
           Message :: term(),
           Options :: timeout_option() | [timeout_option()]} |
          {hibernate, Time :: timeout(), Message :: term()} |
          {hibernate,
           Time :: timeout(),
           Message :: term(),
           Options :: timeout_option() | [timeout_option()]} |
          {continue, Continue :: term()}.

Callback and server loop action. 👑

Returned by the callbacks Module:init/1, Module:handle_call/3, Module:handle_cast/2, Module:handle_info/2, and Module:handle_continue/2. It is also the last argument to enter_loop/4,5.

action/0 is one of:

  • Time :: timeout() — If the action is an integer Time, a time-out occurs unless a request or a message is received within that many milliseconds. A time-out is represented by the atom timeout as the Info argument to be handled by the Module:handle_info/2 callback function. The action infinity can be used to wait indefinitely, which is the default when there is no action/0 specified.

For Time = 0, if there is a request or message waiting to be received, it interrupts (cancels) the time-out.

NOTE: A system message restarts the time-out, which is a known and unfortunate flaw in its implementation. This also applies to stray (cancelled) timer messages from the {timeout|hibernate, ...} time-outs described below, so it is recommended to not use them in combination with this legacy time-out type. 💀
  • {timeout, Time, Message} / {timeout, Time, Message, Options} — Like Time above, but the delivered Info argument is specified by Message, and the time-out is not affected by system messages.

Options can be used to trigger the time-out at an absolute point in time instead. See timeout_option/0 and erlang:start_timer/4 for details.

A relative time-out with Time = 0 is immediately delivered, before any request or message is received, including system messages.

A time-out with Time = infinity will never be delivered so the action is ignored; no time-out is started. 🫠

  • {hibernate, Time, Message} / {hibernate, Time, Message, Options} — A combination of the action hibernate below, and {timeout, Time, Message} or {timeout, Time, Message, Options} above.

The process goes into hibernation while waiting for the next request or message to arrive, or for the time-out to expire. 😴

  • hibernate — The process goes into hibernation (by calling erlang:hibernate/0), waiting for the next request or message to arrive.
  • {continue, Continue} — The process will immediately execute the Module:handle_continue/2 callback function, with Continue as the first argument. 🏃‍♀️

enter_loop_opt()

enter_loop_opt() ::
          {hibernate_after, HibernateAfterTimeout :: timeout()} |
          {debug, Dbgs :: [sys:debug_option()]}.

Server start options for the start or enter_loop functions.

Options that can be used when starting a gen_server server through enter_loop/3-5 or the start functions such as start_link/3,4.

  • {hibernate_after, HibernateAfterTimeout} — Specifies that the gen_server process awaits any message for HibernateAfterTimeout milliseconds and if no message is received, the process goes into hibernation automatically (by calling proc_lib:hibernate/3). 💤
  • {debug, Dbgs} — For every entry in Dbgs, the corresponding function in sys is called. 🔍

format_status()

format_status() ::
          #{state => term(),
            message => term(),
            reason => term(),
            log => [sys:system_event()]}.

A map that describes the gen_server status. 🗺️

The keys are:

  • state — The internal state of the gen_server process.
  • message — The message that caused the server to terminate.
  • reason — The reason that caused the server to terminate.
  • log — The sys log of the server.

New associations may be added to the status map without prior notice. 👀

from()

from() :: {Client :: pid(), Tag :: reply_tag()}.

A call's reply destination. 📬

Destination, given to the gen_server as the first argument to the callback function Module:handle_call/3, to be used by the when replying through reply/2 (instead of through the callback function's return value), to the process Client that has called the gen_server using call/2,3. Tag is a term that is unique for this call/request instance.

-opaque reply_tag()

A handle that associates a reply to the corresponding request. 🎀

-opaque request_id()

An opaque request identifier. See send_request/2 for details. 🪪

-opaque request_id_collection()

An opaque collection of request identifiers (request_id/0).

Each request identifier can be associated with a label chosen by the user. For more information see reqids_new/0. 📚

response_timeout()

response_timeout() :: timeout() | {abs, integer()}.

Response time-out for an asynchronous call. ⏳

Used to set a time limit on how long to wait for a response using either receive_response/2, receive_response/3, wait_response/2, or wait_response/3. The time unit used is millisecond.

Currently valid values:

  • 0..4294967295 — Time-out relative to current time in milliseconds.
  • infinity — Infinite time-out. That is, the operation will never time out.
  • {abs, Timeout} — An absolute Erlang monotonic time time-out in milliseconds. That is, the operation will time out when erlang:monotonic_time(millisecond) returns a value larger than or equal to Timeout. Timeout is not allowed to identify a time further into the future than 4294967295 milliseconds. Specifying the time-out using an absolute value is especially handy when you have a deadline for responses corresponding to a complete collection of requests (request_id_collection/0), since you do not have to recalculate the relative time until the deadline over and over again. 🔥

server_name()

server_name() ::
          {local, LocalName :: atom()} |
          {global, GlobalName :: term()} |
          {via, RegMod :: module(), ViaName :: term()}.

Server name specification: local, global, or via registered. 🌐

To be used when starting a gen_server. See functions start/3,4, start_link/3,4, start_monitor/3,4, enter_loop/3,4,5, and the type server_ref/0.

  • {local, LocalName} — Register the gen_server locally as LocalName using register/2.
  • {global, GlobalName} — registers the gen_server process id globally as GlobalName using global:register_name/2. 🌍
  • {via, RegMod, ViaName} — registers the gen_server process with the registry represented by RegMod. The RegMod callback needs to export the functions register_name/2, unregister_name/1, whereis_name/1, and send/2, and they gotta behave like the corresponding functions in global. So {via, global, GlobalName} is a valid reference that's equivalent to {global, GlobalName}. ✨

Next up, server_ref() — it's giving "how do I point at a server?" energy. Here's the type:

server_ref() ::
          pid() |
          (LocalName :: atom()) |
          {Name :: atom(), Node :: atom()} |
          {global, GlobalName :: term()} |
          {via, RegMod :: module(), ViaName :: term()}.

Server specification: pid/0 or registered server_name/0.

Use this when you're addressing a gen_server. Peep call/2,3, cast/2, send_request/2, check_response/2, wait_response/2, stop/2,3 and the type server_name/0.

It can be:

  • pid/0 — the gen_server's process identifier. 👑
  • LocalName — the gen_server is locally registered as LocalName with register/2. 💖
  • {Name,Node} — the gen_server is locally registered on another node. 🦋
  • {global, GlobalName} — the gen_server is globally registered in global. 🌈
  • {via, RegMod, ViaName} — the gen_server is registered in an alternative process registry. See the same term described for server_name/0. ✨

start_mon_ret() — the return value from the start_monitor/3,4 functions.

start_mon_ret() ::
          {ok, {Pid :: pid(), MonRef :: reference()}} |
          ignore |
          {error, Reason :: term()}.

It's the same as start_ret/0 except when it successful it gives you both the process identifier Pid and a monitor/2,3 MonRef. 💅

start_opt() — server start options for the start functions.

start_opt() ::
          {timeout, Timeout :: timeout()} |
          {spawn_opt, SpawnOptions :: [proc_lib:start_spawn_option()]} |
          enter_loop_opt().

Options that can be used when starting a gen_server server through, for example, start_link/3,4. ✨

  • {timeout, Timeout} — how many milliseconds the gen_server process is allowed to spend initializing, or else it gets terminated and the start function returns {error, timeout}. ⏰
  • {spawn_opt, SpawnOptions} — the SpawnOptions option list is passed to the function used to spawn the gen_server; see proc_lib:start_spawn_option/0. 🍼
NOTE: using spawn option monitor is not allowed — it causes a badarg failure. 💀
  • enter_loop_opt/0 — see the type enter_loop_opt/0 below for more start options that are also allowed by enter_loop/3,4,5. 🌀

start_ret() — return value from the start/3,4 and start_link/3,4 functions.

start_ret() ::
          {ok, Pid :: pid()} | ignore | {error, Reason :: term()}.

Here's the breakdown:

  • {ok, Pid} — the gen_server process was successfully created and initialized, with the process identifier Pid. 🎉
  • {error, {already_started, OtherPid}} — a process with the specified ServerName exists already with the process identifier OtherPid. This function failed to start a gen_server. It exited with reason normal before calling Module:init/1. 💀
  • {error, timeout} — the gen_server process failed to initialize since Module:init/1 did not return within the start time-out. The gen_server process was killed with exit_signal(_, kill). ⏰
  • ignore — the gen_server process failed to initialize since Module:init/1 returned ignore. 🙈
  • {error,Reason} — the gen_server process failed to initialize since Module:init/1 returned {stop,Reason}, {error,Reason}, or it failed with reason Reason. 😭

See Module:init/1 about the exit reason for the gen_server process when it fails to initialize.

timeout_option() :: {abs, Abs :: boolean()}. — time-out timer start option, to select absolute time of expiry.

If Abs is true an absolute timer is started, and if it is false a relative, which is the default. See erlang:start_timer/4 for details. ⏳

✨ FUNCTIONS ✨

abcast(Name :: atom(), Request :: term()) -> abcast.

Cast a request to multiple nodes.

Equivalent to abcast(Nodes, Name, Request) where Nodes is all nodes connected to the calling node, including the calling node itself. 📢

abcast(Nodes :: [node()], Name :: atom(), Request :: term()) -> abcast.

Cast a request to multiple nodes.

Sends an asynchronous request to the gen_server processes locally registered as Name at the specified nodes. The function returns immediately and ignores nodes that do not exist, or where the gen_server Name does not exist. The gen_server processes call Module:handle_cast/2 to handle the request.

For a description of the arguments, see multi_call/2,3,4. 🌐

call(ServerRef :: server_ref(), Request :: term(), Timeout :: timeout()) -> Reply :: term().

Call a server: send request and wait for response.

Makes a synchronous call to the ServerRef of the gen_server process by sending a request and waiting until a reply arrives or a time-out occurs. The gen_server process calls Module:handle_call/3 to handle the request.

See also ServerRef's type server_ref/0.

Request is any term that is passed as the first argument to Module:handle_call/3.

Timeout is an integer that specifies how many milliseconds to wait for a reply, or the atom infinity to wait indefinitely. If no reply is received within the specified time, this function exits the calling process with an exit term containing Reason = timeout as described below. 💅

NOTE: Before OTP 24, if the caller uses (try...)catch to avoid process exit, and the server happens to just be late with the reply, it may arrive to the process message queue any time later. The calling process must therefore after catching a time-out exit be prepared to receive garbage message(s) of the form {reference(), _} and deal with them appropriately (discard them) so they do not clog the process message queue, or gets mistaken for other messages. Starting with OTP 24, gen_server:call uses process aliases, so late replies will not be received. 👀

The return value Reply is passed from the return value of Module:handle_call/3.

This call may exit the calling process with an exit term on the form {Reason, Location} where Location = {gen_server, call, ArgList} and Reason can be (at least) one of:

  • timeout — the call was aborted after waiting Timeout milliseconds for a reply, as described above. ⏰
  • noproc — the ServerRef refers to a server by name (it is not a pid/0) and looking up the server process failed, or the pid/0 was already terminated. 💀
  • {nodedown,Node} — the ServerRef refers to a server on the remote node Node and the connection to that node failed. 🌐
  • calling_self — a call to self/0 would hang indefinitely. 🌀
  • shutdown — the server was stopped during the call by its supervisor. See also stop/3. 🛑
  • normal {shutdown,Term} — the server stopped during the call by returning {stop,Reason,_} from one of its callbacks without replying to this call. See also stop/3. 🛑
  • _OtherTerm — the server process exited during the call, with reason Reason. Either by returning {stop,Reason,_} from one of its callbacks (without replying to this call), by raising an exception, or due to getting an exit signal it did not trap. 💀

cast(ServerRef :: server_ref(), Request :: term()) -> ok.

Cast a request to a server.

Sends an asynchronous request to the gen_server ServerRef and returns ok immediately, ignoring if the destination node or gen_server process does not exist.

The gen_server process calls Module:handle_cast(Request, _) to handle the request. 📨

check_response(Msg, ReqId) -> Result

check_response(Msg, ReqId) -> Result
                        when
                            Msg :: term(),
                            ReqId :: request_id(),
                            Response ::
                                {reply, Reply :: term()} |
                                {error,
                                 {Reason :: term(), server_ref()}},
                            Result :: Response | no_reply.

Check if a received message is a request response.

Checks if Msg is a response corresponding to the request identifier ReqId. The request must have been made by send_request/2, and by the same process calling this function.

If Msg is a reply to the handle ReqId the result of the request is returned in Reply. Otherwise this function returns no_reply and no cleanup is done, and thus the function shall be invoked repeatedly until the response is returned.

The return value Reply is passed from the return value of Module:handle_call/3.

If the gen_statem server process has died when this function is called, that is; Msg reports the server's death, this function returns an error return with the exit Reason. 💅

check_response(Msg, ReqIdCollection, Delete) -> Result

check_response(Msg, ReqIdCollection, Delete) -> Result
                        when
                            Msg :: term(),
                            ReqIdCollection :: request_id_collection(),
                            Delete :: boolean(),
                            Response ::
                                {reply, Reply :: term()} |
                                {error,
                                 {Reason :: term(), server_ref()}},
                            Result ::
                                {Response,
                                 Label :: term(),
                                 NewReqIdCollection ::
                                     request_id_collection()} |
                                no_request | no_reply.

Check if a received message is a request response in a collection. 💅

Okay so bestie, we were saying 👇

It checks if Msg is literally a response that matches up with a request identifier chillin' in ReqIdCollection. Important tho: every single request id in that collection has to be from requests made with send_request/2 or send_request/4, and made by the process that's calling this function. Like, it's their collection, no borrowing 💅

The Label in the response equals the Label that's tied to the request identifier the response is for. And when does a request id get a Label? Either when you store the request id in a collection, or when you send it with send_request/4. That's the vibe.

Now, compared to check_response/2, this one wraps everything in a cute little 3-tuple: {Response, Label, NewReqIdCollection}. Real quick, bestie:

  • Response — literally the same value check_response/2 would've given you
  • Label — the label associated with that specific request identifier
  • NewReqIdCollection — a possibly modified request identifier collection 👑

If ReqIdCollection is empty? You get no_request back. Periodt.

If Msg doesn't match any of the request identifiers in ReqIdCollection? Then it's no_reply, bestie. Nothing matched, we move on 💀

If Delete is true, the association with Label gets deleted from ReqIdCollection in the resulting NewReqIdCollection. If Delete is false, NewReqIdCollection will equal ReqIdCollection. But lowkey — deleting an association is not free, and a collection that still contains already-handled requests can still totally be used by subsequent calls to check_response/3, receive_response/3, and wait_response/3. So it's not like they go stale or anything 🫶

However — and this is the tea ☕ — without deleting handled associations, those calls above can't detect when there are no more outstanding requests to handle. So you'll have to track that yourself somehow, instead of relying on no_request being returned. And heads up: if you pass in a collection that only has associations of already handled or abandoned requests? This function will always return no_reply. Every time. No cap.

✨ enter_loop/3,4,5 ✨

enter_loop(Module :: module(),
                 Options :: [enter_loop_opt()],
                 State :: term(),
                 ServerName :: server_name() | pid()) ->
                    no_return();
                (Module :: module(),
                 Options :: [enter_loop_opt()],
                 State :: term(),
                 Action :: action()) ->
                    no_return().

Make the calling process become a gen_server process. Glow-up arc, honestly ✨

With the ServerName argument, it's equivalent to enter_loop(Module, Options, State, ServerName, infinity).

With the Action argument, it's equivalent to enter_loop(Module, Options, State, self(), Action).

enter_loop(Module :: module(),
                 Options :: [enter_loop_opt()],
                 State :: term(),
                 ServerName :: server_name() | pid(),
                 Action :: action()) ->
                    no_return().

Make the calling process become a gen_server process. Same energy, now with extra args 💖

This one does not return. Instead, the calling process enters the gen_server process receive loop and becomes a gen_server process. And listen — the process must have been started using one of the start functions in proc_lib. You, bestie, are responsible for any initialization of the process, including registering a name for it. That's on you 🫵

This function is your girl when you need a more complex initialization procedure than the one Module:init/1 (that callback) provides. Like when init needs to do a whole side quest first 🌸

Module, Options, and ServerName mean the same things as when you call start[_link|_monitor]/3,4 — or ServerName can be self/0 for an anonymous server, which is the same as calling an enter_loop/3,4 function without a ServerName argument. However — if ServerName is specified (and it's not self/0), the process has to have been registered accordingly before this function is called. No takebacks, no doing it after 🙈

State means the same as in the return value of Module:init/1, which doesn't get called when enter_loop/3,4,5 is used. And btw, to adhere to the gen_server Behaviour, such a callback function still needs to be defined, and it might as well be the one you used when starting the gen_server process through proc_lib, and then be the one that calls enter_loop/3,4,5. But if such a Module:init/1 function — in error cases, for example — cannot call enter_loop/3,4,5, it should return a value that follows the type specification for Module:init/1, such as ignore. Although ngl, that value will be lost when returning to the spawning function 😭

Action is described by the action/0 type.

This function fails if the calling process wasn't started by a proc_lib start function, or if it's not registered according to ServerName. Vibe check failed, sorry bestie 💅

🌈 multi_call/2 🌈

multi_call(Name :: atom(), Request :: term()) ->
                    {Replies :: [{Node :: node(), Reply :: term()}],
                     BadNodes :: [node()]}.

Call servers on multiple nodes in parallel. Multitasking queen behavior 👑

Equivalent to multi_call(Nodes, Name, Request) where Nodes is all nodes connected to the calling node, including the calling node itself. Yes bestie, including yourself 🪞

🌈 multi_call/4 (and yes, multi_call/3 iykyk) 🌈

multi_call(Nodes :: [node()],
                 Name :: atom(),
                 Request :: term(),
                 Timeout :: timeout()) ->
                    {Replies :: [{Node :: node(), Reply :: term()}],
                     BadNodes :: [node()]}.

Call servers on multiple nodes in parallel. Same slay, now with parameters 🎀

It makes a synchronous call to all gen_server processes locally registered as Name at the specified nodes — first sending the request to the nodes, then waiting for the replies. The gen_server processes on those nodes call Module:handle_call/3 to handle the request. Very demure, very mindful 🧚

The function returns a tuple {Replies, BadNodes} where Replies is a list of {Node, Reply} tuples, and BadNodes is a list of nodes that either didn't exist, where Name wasn't a registered gen_server, or where it just didn't reply. Ghosted 💀

  • Nodes — a list of node names to which the request is to be sent
  • Name — the locally registered name for each gen_server process
  • Request — any term that gets passed as the first argument to Module:handle_call/3
  • Timeout — an integer specifying how many milliseconds to wait for all replies, or the atom infinity to wait indefinitely. If no reply comes from a node within the specified time, that node gets added to BadNodes ⏰

When a reply Reply is received from the gen_server process at a node Node, {Node,Reply} is added to Replies. And Reply is passed from the return value of Module:handle_call/3. That's the whole loop 💖

WARNING: If one of the nodes can't process monitors — for example, C or Java nodes — and the gen_server process isn't started when the requests are sent, but starts within 2 seconds, this function waits the whole Timeout, which may be infinity. Scary. This problem doesn't exist if all nodes are Erlang nodes, so like... keep it in the family bestie 🫶

And to prevent late answers (the ones that show up after the time-out) from polluting the message queue of the caller, a middleman process is used to do the calls. Late answers then get discarded when they arrive to the terminated middleman process. Middleman took one for the team, iconic 😭

🦋 receive_response/2 🦋

receive_response(ReqId, Timeout) -> Result
                          when
                              ReqId :: request_id(),
                              Timeout :: response_timeout(),
                              Response ::
                                  {reply, Reply :: term()} |
                                  {error,
                                   {Reason :: term(), server_ref()}},
                              Result :: Response | timeout.

Receive a request response. We're waiting, bestie 📬

This receives a response corresponding to the request identifier ReqId. The request must have been made by send_request/2, and it must have been made by the same process calling this function. No outsourcing 🙅‍♀️

Timeout specifies how long to wait for a response. If no response arrives within the specified time, this function returns timeout. Assuming the server executes on a node supporting aliases (introduced in OTP 24), the request will also be abandoned. That is, no response will be received after a time-out. Otherwise, a stray response might show up at a later time. Awkward 😳

The return value Reply is passed from the return value of Module:handle_call/3.

The function returns an error if the gen_server died before a reply was sent. RIP 🪦

The difference between receive_response/2 and wait_response/2? receive_response/2 abandons the request at time-out so that a potential future response is ignored, while wait_response/2 does not. One of them is a ghoster, one of them is a waiter fr fr 💅

🦋 receive_response/3 🦋

receive_response(ReqIdCollection, Timeout, Delete) -> Result
                          when
                              ReqIdCollection :: request_id_collection(),
                              Timeout :: response_timeout(),
                              Delete :: boolean(),
                              Response ::
                                  {reply, Reply :: term()} |
                                  {error,
                                   {Reason :: term(), server_ref()}},
                              Result ::
                                  {Response,
                                   Label :: term(),
                                   NewReqIdCollection ::
                                       request_id_collection()} |
                                  no_request | timeout.

Receive a request response in a collection. Group chat edition ✨

This receives a response in ReqIdCollection. All request identifiers of ReqIdCollection must correspond to requests that have been made using send_request/2 or send_request/4, and all requests must have been made by the process calling this function. Everybody has to be on the same account, okay? 🫶

The Label in the response is the Label associated with the request identifier that the response corresponds to. And the Label of a request identifier gets associated when you add the request id to a collection, or when you send the request using send_request/4. Same rules as before, no cap 👑

So compared to receive_response/2, this one hands you back the result or exception tied to one specific request identifier, all wrapped up in a little 3-tuple: {Response, Label, NewReqIdCollection}. Break it down bestie: Response is literally the exact value receive_response/2 would have produced, Label is the value associated with that specific request identifier, and NewReqIdCollection is the request identifier collection — possibly modified, so like, keep an eye on that one. 🙈

If ReqIdCollection is empty? You get no_request. Nothing. Crickets. 💀

Timeout is how long you're willing to wait for a response — patience is a virtue and you have some of it. If no response shows up within the specified time, the function returns timeout. And assuming the server executes on a node that supports aliases (that's the OTP 24 glow-up ✨), every request identified by ReqIdCollection also gets abandoned, meaning no responses will be received after a time-out. If your node doesn't do aliases though, stray responses might still arrive at a later time. Surprise delivery, but make it cursed. 🎀

The difference between receive_response/3 and wait_response/3 (because I know you were gonna ask) is that receive_response/3 abandons the requests at time-out so that potential future responses are ignored, while wait_response/3 does not. One of them ghosts. The other one leaves the door open. 💅

Delete is the switch:

  • true → the association with Label is deleted from ReqIdCollection in the resulting NewReqIdCollection.
  • false → NewReqIdCollection will equal ReqIdCollection, untouched, exactly as you handed it over.

Note that deleting an association is not for free (nothing ever is 😭) and that a collection containing already handled requests can still be used by subsequent calls to receive_response/3, check_response/3, and wait_response/3.

However — and this is the part everyone sleeps on — without deleting handled associations, the calls above will not be able to detect when there are no more outstanding requests to handle. So you'll have to keep track of that yourself, some other way, instead of relying on a no_request return. Also: if you pass a collection that only contains associations of already handled or abandoned requests to this function, it will always block until Timeout expires and then return timeout. That's it, that's the vibe. Periodt.

✨ reply/2 ✨

reply(Client :: from(), Reply :: term()) -> ok.

Send a reply to a client. The classic. 💌

This function can be used by a gen_server process to explicitly send a reply to a client that called call/2,3 or multi_call/2,3,4, for those situations where the reply simply cannot be passed in the return value of Module:handle_call/3.

Client must be the From argument that was provided to the handle_call/3 callback function. Reply is any term that gets passed back to the client as the return value of call/2,3 or multi_call/2,3,4. 💅

🌈 reqids_add/3 🌈

reqids_add(ReqId :: request_id(),
                Label :: term(),
                ReqIdCollection :: request_id_collection()) ->
                   NewReqIdCollection :: request_id_collection().

Store a request identifier in a collection. 🗂️

Stores ReqId and associates a Label with the request identifier by adding this information to ReqIdCollection, then returns the resulting request identifier collection.

🧚 reqids_new/0 🧚

reqids_new() -> NewReqIdCollection :: request_id_collection().

Create an empty request identifier collection.

Returns a new, empty request identifier collection. A request identifier collection can be utilized to handle multiple outstanding requests, which is very organized of you. 🫶

Request identifiers of requests made by send_request/2 can be stored in a collection using reqids_add/3. Such a collection of request identifiers can later be used in order to get one response corresponding to a request in the collection, by passing the collection as an argument to receive_response/3, wait_response/3, or check_response/3.

And reqids_size/1 can be used to determine the number of request identifiers in a collection. Handy! 👑

⭐ reqids_size/1 ⭐

reqids_size(ReqIdCollection :: request_id_collection()) ->
                    non_neg_integer().

Returns the number of request identifiers in ReqIdCollection. Count 'em. 🔢

🎀 reqids_to_list/1 🎀

reqids_to_list(ReqIdCollection :: request_id_collection()) ->
                       [{ReqId :: request_id(), Label :: term()}].

Convert a request identifier collection to a list.

Returns a list of {ReqId, Label} tuples which corresponds to all request identifiers with their associated labels in ReqIdCollection. So tidy, so cute, so useful. 🎀

💖 send_request/2 💖

send_request(ServerRef :: server_ref(), Request :: term()) ->
                     ReqId :: request_id().

Send an asynchronous call request. Fire now, deal with the response later. 🔥

Sends Request to the gen_server process identified by ServerRef and returns a request identifier ReqId.

That ReqId return value shall later be used with receive_response/2, wait_response/2, or check_response/2 to fetch the actual result of the request. Besides passing the request identifier directly to those functions, it can also be stored in a request identifier collection using reqids_add/3. Such a collection of request identifiers can later be used in order to get one response corresponding to a request in the collection, by passing the collection as an argument to receive_response/3, wait_response/3, or check_response/3. And if you're about to store the request identifier in a collection anyway, you may want to consider using send_request/4 instead. Lowkey more efficient. 👀

The call

gen_server:receive_response(gen_server:send_request(ServerRef, Request), Timeout)

can be seen as equivalent to gen_server:call(ServerRef, Request, Timeout), ignoring the error handling.

The gen_server process calls Module:handle_call/3 to handle the request.

See the type server_ref/0 for the possible values for ServerRef.

Request is any term that is passed as the first argument to Module:handle_call/3.

🦋 send_request/4 🦋

send_request(ServerRef :: server_ref(),
                  Request :: term(),
                  Label :: term(),
                  ReqIdCollection :: request_id_collection()) ->
                     NewReqIdCollection :: request_id_collection().

Send an asynchronous call request and drop it into a request identifier collection. Multitasking icon behavior. 💅

Sends Request to the gen_server process identified by ServerRef. The Label will be associated with the request identifier of the operation and added to the returned request identifier collection NewReqIdCollection. That collection can later be used in order to get one response corresponding to a request in the collection, by passing the collection as an argument to receive_response/3, wait_response/3, or check_response/3.

The same as calling reqids_add(send_request(ServerRef, Request), Label, ReqIdCollection), but slightly more efficient. Efficiency is self-care, bestie. ✨

👑 start/3 👑

start(Module :: module(),
           Args :: term(),
           Options :: [start_opt()]) ->
              start_ret().

Start a server — neither linked nor registered. Free spirit energy. 🌸

Equivalent to start/4 except that the gen_server process is not registered with any name service.

🌸 start/4 🌸

start(ServerName :: server_name(),
           Module :: module(),
           Args :: term(),
           Options :: [start_opt()]) ->
              start_ret().

Start a server — registered, but not linked.

Creates a standalone gen_server process, that is, a gen_server process that is not part of a supervision tree and thus has no supervisor. Brave little toaster. 🫶

Other than that, see start_link/4.

start_link(Module :: module(),
                Args :: term(),
                Options :: [start_opt()]) ->
                   start_ret().

Start a server — linked, but not registered.

Equivalent to start_link/4 except that the gen_server process is not registered with any name service.

start_link(ServerName :: server_name(),
                Module :: module(),
                Args :: term(),
                Options :: [start_opt()]) ->
                   start_ret().

Start a server — linked and registered. The full package, the main character, the whole moment. 💖

Creates a gen_server process as part of a supervision tree. This function is to be called, directly or indirectly, by the supervisor. For example, it ensures that the gen_server process is spawned as linked to the caller (the supervisor).

The gen_server process calls Module:init/1 to initialize. To ensure a synchronized startup procedure, start_link/3,4 does not return until Module:init/1 has returned or failed.

ServerName specifies with what name and how to register the server name. See type server_name/0 for different name registrations.

Module is the name of the callback module.

Args is any term that is passed as the argument to Module:init/1.

See type start_opt/0 for Options for starting the gen_server process.

See type start_ret/0 for a description of this function's return values.

If start_link/3,4 returns ignore or {error, _}, the started gen_server process has terminated. If an 'EXIT' message was delivered to the calling process (due to the process link), that message has been consumed.

WARNING: Before OTP 26.0, if the started gen_server process returned e.g. {stop, Reason} from Module:init/1, this function could return {error, Reason} before the started gen_server process had terminated, so starting again might fail because VM resources such as the registered name was not yet unregistered. An 'EXIT' message could arrive later to the process calling this function. But if the started gen_server process instead failed during Module:init/1, a process link {'EXIT', Pid, Reason} message caused this function to return {error, Reason}, so the 'EXIT' message had been consumed and the started gen_server process had terminated. Since it was impossible to tell the difference between these two cases from start_link/3,4's return value, this inconsistency was cleaned up in OTP 26.0. 😭

The difference between returning {stop, _} and {error, _} from Module:init/1 is that {error, _} results in a graceful ("silent") termination, since the gen_server process exits with reason normal.

🌈 start_monitor/3 🌈

start_monitor(Module :: module(),
                   Args :: term(),
                   Options :: [start_opt()]) ->
                      start_mon_ret().

Start a server — monitored, but neither linked nor registered.

...and yeah, it's basically start_monitor/4's whole deal, except the gen_server process is not registered with any name service. No name, no registration, just vibes. 🦋

start_monitor/4

start_monitor(ServerName :: server_name(),
              Module :: module(),
              Args :: term(),
              Options :: [start_opt()]) ->
                 start_mon_ret().

Starts a server that's monitored and registered, but not linked. Bestie, read that again — monitored, yes; linked, absolutely not. 💅

It creates a standalone gen_server process, meaning a gen_server process that is not part of a supervision tree (so it has literally no supervisor looking out for it 😭) and atomically sets up a monitor to the newly created server. Atomic. The monitor is there from the jump, no awkward gap.

Other than that, go look at start_link/3,4. But heads up: the return value for a successful start differs, because it hands you back a monitor reference. See type start_mon_ret/0.

If the start is not successful, the caller will be blocked until the monitor's 'DOWN' message has been received and removed from the message queue. Not "probably blocked", not "eventually" — blocked until that message is plucked out of the queue. 🫶

stop/3

stop(ServerRef :: server_ref(),
     Reason :: term(),
     Timeout :: timeout()) ->
        ok.

Stop a server. Simple as. ✋

Orders the generic server specified by ServerRef to exit with the specified Reason and waits for it to terminate. The gen_server process calls Module:terminate/2 before exiting — so it gets its little goodbye moment. 🎀

The function returns ok if the server terminates with the expected reason. Any other reason than normal, shutdown, or {shutdown,Term} causes an error report to be issued using logger. An exit signal with the same reason is sent to linked processes and ports. So if you were messy with the reason, everyone linked is hearing about it. 💀

Timeout is an integer that specifies how many milliseconds to wait for the server to terminate, or the atom infinity to wait indefinitely. If the server has not terminated within the specified time, the call exits the calling process with reason timeout.

If the process does not exist, the call exits the calling process with reason noproc, or with reason {nodedown,Node} if the connection fails to the remote Node where the server runs.

so real. noproc = there's nobody home. {nodedown,Node} = the whole node ghosted you. neither one is a you problem, but you do get exited either way. 🙈

wait_response/2

wait_response(ReqId, WaitTime) -> Result
                       when
                           ReqId :: request_id(),
                           WaitTime :: response_timeout(),
                           Response ::
                               {reply, Reply :: term()} |
                               {error, {Reason :: term(),
server_ref()}},
                           Result :: Response | timeout.

Wait for a request response. ⏳

Wait for the response to the request identifier ReqId. The request must have been made by send_request/2, and it must have been made by the same process calling this function. Same process. Not your friend, not your supervisor, not your cousin — same process, bestie. 👑

WaitTime specifies how long to wait for a reply. If no reply is received within the specified time, the function returns timeout and no cleanup is done. Which is actually great news, because it means the function can be invoked repeatedly until a reply is returned. Obsessed with a retry loop that doesn't burn the request. ✨

The return value Reply is passed from the return value of Module:handle_call/3.

The function returns an error if the gen_server died before a reply was sent. 😭

The difference between receive_response/2 and wait_response/2 is that receive_response/2 abandons the request at time-out so that a potential future response is ignored, while wait_response/2 does not. So receive_response/2 is the "okay we're done here" one, and wait_response/2 keeps the door open. Two completely different energies. 🌸

wait_response/3

wait_response(ReqIdCollection, WaitTime, Delete) -> Result
                       when
                           ReqIdCollection :: request_id_collection(),
                           WaitTime :: response_timeout(),
                           Delete :: boolean(),
                           Response ::
                               {reply, Reply :: term()} |
                               {error, {Reason :: term(),
server_ref()}},
                           Result ::
                               {Response,
                                Label :: term(),
                                NewReqIdCollection ::
                                    request_id_collection()} |
                               no_request | timeout.

Wait for any request response in a collection. Not one specific request — whichever one comes home first. 🎯

Wait for a response in a ReqIdCollection. All request identifiers of ReqIdCollection must correspond to requests that have been made using send_request/2 or send_request/4, and all requests must have been made by the process calling this function. Again: same process. Periodt.

The Label in the response is the Label associated with the request identifier that the response corresponds to. The Label of a request identifier is associated when adding the request id to a collection, or when sending the request using send_request/4.

Compared to wait_response/2, the returned result or exception associated with a specific request identifier will be wrapped in a 3-tuple {Response, Label, NewReqIdCollection}. Response is the value that would have been produced by wait_response/2, Label is the value associated with the specific request identifier and NewReqIdCollection is a possibly modified request identifier collection. So it's the same tea, just steeped in extra context. 🍓

If ReqIdCollection is empty, no_request will be returned.

If no response is received before WaitTime has expired, timeout is returned. It is valid to continue waiting for a response as many times as needed up until a response has been received and completed by check_response(), receive_response(), or wait_response().

The difference between receive_response/3 and wait_response/3 is that receive_response/3 abandons requests at time-out so that potential future responses are ignored, while wait_response/3 does not.

If Delete is true, the association with Label has been deleted from ReqIdCollection in the resulting NewReqIdCollection. If Delete is false, NewReqIdCollection will equal ReqIdCollection. Note that deleting an association is not for free and that a collection containing already handled requests can still be used by subsequent calls to wait_response/3, check_response/3, and receive_response/3.

ok here's the catch, and it's a real one 😭 — without deleting handled associations, the above calls will not be able to detect when there are no more outstanding requests to handle, so you will have to keep track of this some other way than relying on a no_request return. Note that if you pass a collection only containing associations of already handled or abandoned requests to this function, it will always block until WaitTime expires and then return timeout. Like, always. No shortcuts. 💀

✨ CALLBACKS ✨

code_change(OldVsn, State, Extra)

Update the server state after a code change. Glow-up, but make it internal. 🌈

This function is called by a gen_server process when it is to update its internal state during a release upgrade/downgrade, that is, when the instruction {update, Module, Change, ...} is specified in the appup file.

For more information, see section Release Handling Instructions in OTP Design Principles.

For an upgrade, OldVsn is Vsn, and for a downgrade, OldVsn is {down,Vsn}. Vsn is defined by the vsn attribute(s) of the old version of the callback module Module. If no such attribute is defined, the version is the checksum of the Beam file.

State is the internal state of the gen_server process.

Extra is passed "as is" from the {advanced,Extra} part of the update instruction.

If successful, the function must return the updated internal state.

If the function returns {error,Reason}, the ongoing upgrade fails and rolls back to the old release. No partial glow-ups allowed. 🚫

NOTE: If a release upgrade/downgrade with Change = {advanced, Extra} specified in the .appup file is made when Module:code_change/3 is not implemented, the callback call will crash with an undef error reason. fr fr, it will just crash.

format_status(Status)

Format/limit the status value. ✨

This function is called by a gen_server process in in order to format/limit the server state for debugging and logging purposes.

It is called in the following situations:

  • sys:get_status/1,2 is invoked to get the gen_server status.
  • The gen_server process terminates abnormally and logs an error.

This callback is used to limit the status of the process returned by sys:get_status/1,2 or sent to logger.

The callback gets a map Status describing the current status and shall return a map NewStatus with the same keys, but it may transform some values.

Two possible use cases for this callback is to remove sensitive information from the state to prevent it from being printed in log files, or to compact large irrelevant status items that would only clutter the logs. Basically: protect the secrets 🫣 and clean up the noise. 🧚

Example:

format_status(Status) ->
  maps:map(
    fun(state,State) ->
            maps:remove(private_key, State);
       (message,{password, _Pass}) ->
            {password, removed};
       (_,Value) ->
            Value
    end, Status).
NOTE: This callback is optional, so callback modules need not export it. The gen_server module provides a default implementation of this function that returns the callback module state. If this callback is exported but fails, to hide possibly sensitive data, the default function will instead return the fact that Module:format_status/1 has crashed. The safety net catches it and tells on it. 🤭

format_status(Opt, StatusData)

Deprecated: the callback gen_server:format_status(_,_) is deprecated; use format_status/1 instead. It's giving legacy. 👵 Use the new one, bestie.

Format/limit the status value. ✨

This function is called by a gen_server process in in order to format/limit the server state for debugging and logging purposes.

It is called in the following situations:

  • One of sys:get_status/1,2 is invoked to get the gen_server status. Opt is set to the atom normal.
  • The gen_server process terminates abnormally and logs an error. Opt is set to the atom terminate.

so yeah, all that stuff above is why a callback module that wants to tweak how its status looks exports its own Module:format_status/2 — and that's the one whose return value gets used for sys:get_status/1,2 and for how the status shows up in termination error logs. 💖

PDict is the current value of the process dictionary of the gen_server process. (yes it has its own little diary, we don't judge 🫶)

State is the internal state of the gen_server process.

And the whole point of the function is that it returns Status — a term that changes the details of the current state and status of the gen_server process. There are no restrictions on the form Status can take, bestie, go feral. But for the sys:get_status/1,2 case (so when Opt is normal), the recommended form for the Status value is [{data, [{"State", Term}]}], where Term provides relevant details of the gen_server state. Following this recommendation is not required — but it makes the callback module status consistent with the rest of the sys:get_status/1,2 return value, and consistency is the vibe ✨

One use for this function is to return compact alternative state representations, so that large state terms don't get printed in log files. Lowkey iconic of it honestly.

NOTE: This callback is optional, so callback modules need not export it. The gen_server module provides a default implementation of this function that returns the callback module state.

🦋 CALLBACKS (the rest, bestie) 🦋

handle_call(Request, From, State) — handle a call 📞

Whenever a gen_server process receives a request sent using call/2,3, multi_call/2,3,4, or send_request/2,4, this function is called to handle the request.

State is the internal state of the gen_server process, and NewState a possibly updated one.

Request is passed from the same argument provided to call or multi_call.

The return value Result is interpreted like this, so pay attention 👀:

  • {reply,Reply,NewState}

{reply,Reply,NewState,_}

The Reply value is sent back to the client request and there becomes its return value.

The gen_server process continues executing with the possibly updated internal state NewState.

  • {noreply,NewState}

{noreply,NewState,_}

The gen_server process continues executing with the possibly updated internal state NewState.

A reply to the client request has to be created by calling reply(From, Reply), either in this or in a later callback. no skipping it, the client is waiting 💀

  • {reply,_,_,Action}

{noreply,_,Action}

Action is described by the action/0 type.

  • {stop,Reason,NewState}

{stop,Reason,Reply,NewState}

The gen_server process will call Module:terminate(Reason,NewState), and then terminate.

{stop,_,Reply,_} will create a reply to the client request just as {reply,Reply,...} while {stop,_,_} will not, so just as for {noreply,NewState,...} a reply has to be created by calling reply(From, Reply) before returning {stop,_,_}.

handle_cast(Request, State) — handle a cast message 📨

Whenever a gen_server process receives a request sent using cast/2 or abcast/2,3, this function is called to handle the request.

For a description of the arguments and possible return values, see Module:handle_call/3. (same energy, no reply tho, that's the whole point 🎀)

handle_continue(Info, State) — handle a callback continuation 🏃♀️

This function is called by a gen_server process whenever a previous callback returns one of the tuples containing {continue, Continue}. The call is invoked immediately after the previous callback, which makes it useful for performing work after initialization, or for splitting the work in a callback into multiple steps while updating the process state along the way. the way that it just. keeps. going. 🔥

For a description of the other arguments and possible return values, see Module:handle_call/3.

NOTE: This callback is optional, so callback modules need to export it only if they return one of the tuples containing {continue,Continue} from another callback. If such a {continue,_} tuple is used and the callback is not implemented, the process will exit with undef error. she will not survive that, fr fr.

handle_info(Info, State) — handle an info message (regular process message) 📬

This function is called by a gen_server process when a time-out occurs or when it receives any other message than a synchronous or asynchronous request (or a system message).

Info is either the atom timeout, if a time-out has occurred, or the received message.

For a description of the other arguments and possible return values, see Module:handle_call/3.

NOTE: This callback is optional, so callback modules need not export it. The gen_server module provides a default implementation of this function that logs about the unexpected Info message, drops it and returns {noreply, State}. like, noted, ignored, moving on 💅

init(Args) — initialize the server 🌱✨

Whenever a gen_server process is started using start/3,4, start_monitor/3,4, or start_link/3,4, this function is called by the new process to initialize the server.

Args is the Args argument provided to the start function.

The return value Result is interpreted as follows:

  • {ok,State}

{ok,State,_}

Initialization was succesful and State is the internal state of the gen_server process.

  • {ok,_,Action} — Action is described by the action/0 type.
  • {stop,Reason}

Initialization failed. The gen_server process exits with reason Reason. 💀

  • {error,Reason} since OTP 26.0

Initialization failed. The gen_server process exits with reason normal.

See function start_link/3,4's return value start_ret/0 in these different cases.

terminate(Reason, State) — handle server termination 🥀

This function is called by a gen_server process when it is about to terminate.

It is to be the opposite of Module:init/1 and do any necessary cleaning up. When it returns, the gen_server process terminates with Reason. The return value is ignored. (she said what she said, the return value does not matter here 😌)

Reason is a term denoting the stop reason and State is the internal state of the gen_server process.

Reason depends on why the gen_server process is terminating. If it is because another callback function has returned a stop tuple {stop,..}, Reason has the value specified in that tuple. If it is because of a failure, Reason is the error reason.

If the gen_server process is part of a supervision tree and is ordered by its supervisor to terminate, this function is called with Reason=shutdown if the following conditions apply:

  • The gen_server process has been set to trap exit signals.
  • The shutdown strategy as defined in the child specification of the supervisor is an integer time-out value, not brutal_kill.

Even if the gen_server process is not part of a supervision tree, this function is called if it receives an 'EXIT' message from its parent. Reason is the same as in the 'EXIT' message.

If the gen_server process does not trap exits, the gen_server process terminates immediately. no cleanup arc, just gone 🙈

Notice that for any other reason than normal, shutdown, or {shutdown,Term}, see stop/3, the gen_server process is assumed to terminate because of an error, and an error report is issued using logger.

When the gen_server process exits, an exit signal with the same reason is sent to linked processes and ports. chain reaction, main character energy, everybody's involved 👑

NOTE: This callback is optional, so callback modules need not export it. The gen_server module provides a default implementation with no cleanup.

and that's a wrap on gen_server(3) bestie 🎀✨ that was a lot of callbacks and honestly you ate every single one of them. go forth and build your little processes, they'll behave, probably 💖

this whole thing is brought to you by Ericsson AB, living in stdlib 8.0.4, and it's the gen_server(3) page — mwah, kiss the manual, iykyk 💅🫶

📦 where this page comes from