<?xml version="1.0" encoding="latin1" ?> <!DOCTYPE erlref SYSTEM "erlref.dtd"> <erlref> <header> <copyright> <year>1996</year><year>2009</year> <holder>Ericsson AB. All Rights Reserved.</holder> </copyright> <legalnotice> The contents of this file are subject to the Erlang Public License, Version 1.1, (the "License"); you may not use this file except in compliance with the License. You should have received a copy of the Erlang Public License along with this software. If not, it can be retrieved online at http://www.erlang.org/. Software distributed under the License is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License for the specific language governing rights and limitations under the License. </legalnotice> <title>gen_server</title> <prepared></prepared> <docno></docno> <date></date> <rev></rev> </header> <module>gen_server</module> <modulesummary>Generic Server Behaviour</modulesummary> <description> <p>A behaviour module for implementing the server of a client-server relation. A generic server process (gen_server) implemented using this module will have a standard set of interface functions and include functionality for tracing and error reporting. It will also fit into an OTP supervision tree. Refer to <seealso marker="doc/design_principles:gen_server_concepts">OTP Design Principles</seealso> for more information.</p> <p>A gen_server assumes all specific parts to be located in a callback module exporting a pre-defined set of functions. The relationship between the behaviour functions and the callback functions can be illustrated as follows:</p> <pre> gen_server module Callback module ----------------- --------------- gen_server:start_link -----> Module:init/1 gen_server:call gen_server:multi_call -----> Module:handle_call/3 gen_server:cast gen_server:abcast -----> Module:handle_cast/2 - -----> Module:handle_info/2 - -----> Module:terminate/2 - -----> Module:code_change/3 </pre> <p>If a callback function fails or returns a bad value, the gen_server will terminate.</p> <p>A gen_server handles system messages as documented in <seealso marker="sys">sys(3)</seealso>. The <c>sys</c> module can be used for debugging a gen_server.</p> <p>Note that a gen_server does not trap exit signals automatically, this must be explicitly initiated in the callback module.</p> <p>Unless otherwise stated, all functions in this module fail if the specified gen_server does not exist or if bad arguments are given.</p> <p>The gen_server process can go into hibernation (see <seealso marker="erts:erlang#erlang:hibernate/3">erlang(3)</seealso>) if a callback function specifies <c>'hibernate'</c> instead of a timeout value. This might be useful if the server is expected to be idle for a long time. However this feature should be used with care as hibernation implies at least two garbage collections (when hibernating and shortly after waking up) and is not something you'd want to do between each call to a busy server.</p> </description> <funcs> <func> <name>start_link(Module, Args, Options) -> Result</name> <name>start_link(ServerName, Module, Args, Options) -> Result</name> <fsummary>Create a gen_server process in a supervision tree.</fsummary> <type> <v>ServerName = {local,Name} | {global,GlobalName}</v> <v> Name = atom()</v> <v> GlobalName = term()</v> <v>Module = atom()</v> <v>Args = term()</v> <v>Options = [Option]</v> <v> Option = {debug,Dbgs} | {timeout,Time} | {spawn_opt,SOpts}</v> <v> Dbgs = [Dbg]</v> <v> Dbg = trace | log | statistics | {log_to_file,FileName} | {install,{Func,FuncState}}</v> <v> SOpts = [term()]</v> <v>Result = {ok,Pid} | ignore | {error,Error}</v> <v> Pid = pid()</v> <v> Error = {already_started,Pid} | term()</v> </type> <desc> <p>Creates a gen_server process as part of a supervision tree. The function should be called, directly or indirectly, by the supervisor. It will, among other things, ensure that the gen_server is linked to the supervisor.</p> <p>The gen_server process calls <c>Module:init/1</c> to initialize. To ensure a synchronized start-up procedure, <c>start_link/3,4</c> does not return until <c>Module:init/1</c> has returned.</p> <p>If <c>ServerName={local,Name}</c> the gen_server is registered locally as <c>Name</c> using <c>register/2</c>. If <c>ServerName={global,GlobalName}</c> the gen_server is registered globally as <c>GlobalName</c> using <c>global:register_name/2</c>. If no name is provided, the gen_server is not registered.</p> <p><c>Module</c> is the name of the callback module.</p> <p><c>Args</c> is an arbitrary term which is passed as the argument to <c>Module:init/1</c>.</p> <p>If the option <c>{timeout,Time}</c> is present, the gen_server is allowed to spend <c>Time</c> milliseconds initializing or it will be terminated and the start function will return <c>{error,timeout}</c>.</p> <p>If the option <c>{debug,Dbgs}</c> is present, the corresponding <c>sys</c> function will be called for each item in <c>Dbgs</c>. See <seealso marker="sys">sys(3)</seealso>.</p> <p>If the option <c>{spawn_opt,SOpts}</c> is present, <c>SOpts</c> will be passed as option list to the <c>spawn_opt</c> BIF which is used to spawn the gen_server. See <seealso marker="erts:erlang#spawn_opt/2">erlang(3)</seealso>.</p> <note> <p>Using the spawn option <c>monitor</c> is currently not allowed, but will cause the function to fail with reason <c>badarg</c>.</p> </note> <p>If the gen_server is successfully created and initialized the function returns <c>{ok,Pid}</c>, where <c>Pid</c> is the pid of the gen_server. If there already exists a process with the specified <c>ServerName</c> the function returns <c>{error,{already_started,Pid}}</c>, where <c>Pid</c> is the pid of that process.</p> <p>If <c>Module:init/1</c> fails with <c>Reason</c>, the function returns <c>{error,Reason}</c>. If <c>Module:init/1</c> returns <c>{stop,Reason}</c> or <c>ignore</c>, the process is terminated and the function returns <c>{error,Reason}</c> or <c>ignore</c>, respectively.</p> </desc> </func> <func> <name>start(Module, Args, Options) -> Result</name> <name>start(ServerName, Module, Args, Options) -> Result</name> <fsummary>Create a stand-alone gen_server process.</fsummary> <type> <v>ServerName = {local,Name} | {global,GlobalName}</v> <v> Name = atom()</v> <v> GlobalName = term()</v> <v>Module = atom()</v> <v>Args = term()</v> <v>Options = [Option]</v> <v> Option = {debug,Dbgs} | {timeout,Time} | {spawn_opt,SOpts}</v> <v> Dbgs = [Dbg]</v> <v> Dbg = trace | log | statistics | {log_to_file,FileName} | {install,{Func,FuncState}}</v> <v> SOpts = [term()]</v> <v>Result = {ok,Pid} | ignore | {error,Error}</v> <v> Pid = pid()</v> <v> Error = {already_started,Pid} | term()</v> </type> <desc> <p>Creates a stand-alone gen_server process, i.e. a gen_server which is not part of a supervision tree and thus has no supervisor.</p> <p>See <seealso marker="#start_link/3">start_link/3,4</seealso> for a description of arguments and return values.</p> </desc> </func> <func> <name>call(ServerRef, Request) -> Reply</name> <name>call(ServerRef, Request, Timeout) -> Reply</name> <fsummary>Make a synchronous call to a generic server.</fsummary> <type> <v>ServerRef = Name | {Name,Node} | {global,GlobalName} | pid()</v> <v> Node = atom()</v> <v> GlobalName = term()</v> <v>Request = term()</v> <v>Timeout = int()>0 | infinity</v> <v>Reply = term()</v> </type> <desc> <p>Makes a synchronous call to the gen_server <c>ServerRef</c> by sending a request and waiting until a reply arrives or a timeout occurs. The gen_server will call <c>Module:handle_call/3</c> to handle the request.</p> <p><c>ServerRef</c> can be:</p> <list type="bulleted"> <item>the pid,</item> <item><c>Name</c>, if the gen_server is locally registered,</item> <item><c>{Name,Node}</c>, if the gen_server is locally registered at another node, or</item> <item><c>{global,GlobalName}</c>, if the gen_server is globally registered.</item> </list> <p><c>Request</c> is an arbitrary term which is passed as one of the arguments to <c>Module:handle_call/3</c>.</p> <p><c>Timeout</c> is an integer greater than zero which specifies how many milliseconds to wait for a reply, or the atom <c>infinity</c> to wait indefinitely. Default value is 5000. If no reply is received within the specified time, the function call fails. If the caller catches the failure and continues running, and the server is just late with the reply, it may arrive at any time later into the caller's message queue. The caller must in this case be prepared for this and discard any such garbage messages that are two element tuples with a reference as the first element.</p> <p>The return value <c>Reply</c> is defined in the return value of <c>Module:handle_call/3</c>.</p> <p>The call may fail for several reasons, including timeout and the called gen_server dying before or during the call.</p> <p>The ancient behaviour of sometimes consuming the server exit message if the server died during the call while linked to the client has been removed in OTP R12B/Erlang 5.6.</p> </desc> </func> <func> <name>multi_call(Name, Request) -> Result</name> <name>multi_call(Nodes, Name, Request) -> Result</name> <name>multi_call(Nodes, Name, Request, Timeout) -> Result</name> <fsummary>Make a synchronous call to several generic servers.</fsummary> <type> <v>Nodes = [Node]</v> <v> Node = atom()</v> <v>Name = atom()</v> <v>Request = term()</v> <v>Timeout = int()>=0 | infinity</v> <v>Result = {Replies,BadNodes}</v> <v> Replies = [{Node,Reply}]</v> <v> Reply = term()</v> <v>BadNodes = [Node]</v> </type> <desc> <p>Makes a synchronous call to all gen_servers locally registered as <c>Name</c> at the specified nodes by first sending a request to every node and then waiting for the replies. The gen_servers will call <c>Module:handle_call/3</c> to handle the request.</p> <p>The function returns a tuple <c>{Replies,BadNodes}</c> where <c>Replies</c> is a list of <c>{Node,Reply}</c> and <c>BadNodes</c> is a list of node that either did not exist, or where the gen_server <c>Name</c> did not exist or did not reply.</p> <p><c>Nodes</c> is a list of node names to which the request should be sent. Default value is the list of all known nodes <c>[node()|nodes()]</c>.</p> <p><c>Name</c> is the locally registered name of each gen_server.</p> <p><c>Request</c> is an arbitrary term which is passed as one of the arguments to <c>Module:handle_call/3</c>.</p> <p><c>Timeout</c> is an integer greater than zero which specifies how many milliseconds to wait for each reply, or the atom <c>infinity</c> to wait indefinitely. Default value is <c>infinity</c>. If no reply is received from a node within the specified time, the node is added to <c>BadNodes</c>.</p> <p>When a reply <c>Reply</c> is received from the gen_server at a node <c>Node</c>, <c>{Node,Reply}</c> is added to <c>Replies</c>. <c>Reply</c> is defined in the return value of <c>Module:handle_call/3</c>.</p> <warning> <p>If one of the nodes is not capable of process monitors, for example C or Java nodes, and the gen_server is not started when the requests are sent, but starts within 2 seconds, this function waits the whole <c>Timeout</c>, which may be infinity.</p> <p>This problem does not exist if all nodes are Erlang nodes.</p> </warning> <p>To avoid that late answers (after the timeout) pollutes the caller's message queue, a middleman process is used to do the actual calls. Late answers will then be discarded when they arrive to a terminated process.</p> </desc> </func> <func> <name>cast(ServerRef, Request) -> ok</name> <fsummary>Send an asynchronous request to a generic server.</fsummary> <type> <v>ServerRef = Name | {Name,Node} | {global,GlobalName} | pid()</v> <v> Node = atom()</v> <v> GlobalName = term()</v> <v>Request = term()</v> </type> <desc> <p>Sends an asynchronous request to the gen_server <c>ServerRef</c> and returns <c>ok</c> immediately, ignoring if the destination node or gen_server does not exist. The gen_server will call <c>Module:handle_cast/2</c> to handle the request.</p> <p>See <seealso marker="#call/2">call/2,3</seealso> for a description of <c>ServerRef</c>.</p> <p><c>Request</c> is an arbitrary term which is passed as one of the arguments to <c>Module:handle_cast/2</c>.</p> </desc> </func> <func> <name>abcast(Name, Request) -> abcast</name> <name>abcast(Nodes, Name, Request) -> abcast</name> <fsummary>Send an asynchronous request to several generic servers.</fsummary> <type> <v>Nodes = [Node]</v> <v> Node = atom()</v> <v>Name = atom()</v> <v>Request = term()</v> </type> <desc> <p>Sends an asynchronous request to the gen_servers locally registered as <c>Name</c> at the specified nodes. The function returns immediately and ignores nodes that do not exist, or where the gen_server <c>Name</c> does not exist. The gen_servers will call <c>Module:handle_cast/2</c> to handle the request.</p> <p>See <seealso marker="#multi_call/2">multi_call/2,3,4</seealso> for a description of the arguments.</p> </desc> </func> <func> <name>reply(Client, Reply) -> Result</name> <fsummary>Send a reply to a client.</fsummary> <type> <v>Client - see below</v> <v>Reply = term()</v> <v>Result = term()</v> </type> <desc> <p>This function can be used by a gen_server to explicitly send a reply to a client that called <c>call/2,3</c> or <c>multi_call/2,3,4</c>, when the reply cannot be defined in the return value of <c>Module:handle_call/3</c>.</p> <p><c>Client</c> must be the <c>From</c> argument provided to the callback function. <c>Reply</c> is an arbitrary term, which will be given back to the client as the return value of <c>call/2,3</c> or <c>multi_call/2,3,4</c>.</p> <p>The return value <c>Result</c> is not further defined, and should always be ignored.</p> </desc> </func> <func> <name>enter_loop(Module, Options, State)</name> <name>enter_loop(Module, Options, State, ServerName)</name> <name>enter_loop(Module, Options, State, Timeout)</name> <name>enter_loop(Module, Options, State, ServerName, Timeout)</name> <fsummary>Enter the gen_server receive loop</fsummary> <type> <v>Module = atom()</v> <v>Options = [Option]</v> <v> Option = {debug,Dbgs}</v> <v> Dbgs = [Dbg]</v> <v> Dbg = trace | log | statistics</v> <v> | {log_to_file,FileName} | {install,{Func,FuncState}}</v> <v>State = term()</v> <v>ServerName = {local,Name} | {global,GlobalName}</v> <v> Name = atom()</v> <v> GlobalName = term()</v> <v>Timeout = int() | infinity</v> </type> <desc> <p>Makes an existing process into a gen_server. Does not return, instead the calling process will enter the gen_server receive loop and become a gen_server process. The process <em>must</em> have been started using one of the start functions in <c>proc_lib</c>, see <seealso marker="proc_lib">proc_lib(3)</seealso>. The user is responsible for any initialization of the process, including registering a name for it.</p> <p>This function is useful when a more complex initialization procedure is needed than the gen_server behaviour provides.</p> <p><c>Module</c>, <c>Options</c> and <c>ServerName</c> have the same meanings as when calling <seealso marker="#start_link/3">gen_server:start[_link]/3,4</seealso>. However, if <c>ServerName</c> is specified, the process must have been registered accordingly <em>before</em> this function is called.</p> <p><c>State</c> and <c>Timeout</c> have the same meanings as in the return value of <seealso marker="#Moduleinit">Module:init/1</seealso>. Also, the callback module <c>Module</c> does not need to export an <c>init/1</c> function. </p> <p>Failure: If the calling process was not started by a <c>proc_lib</c> start function, or if it is not registered according to <c>ServerName</c>.</p> </desc> </func> </funcs> <section> <title>CALLBACK FUNCTIONS</title> <p>The following functions should be exported from a <c>gen_server</c> callback module.</p> </section> <funcs> <func> <name>Module:init(Args) -> Result</name> <fsummary>Initialize process and internal state.</fsummary> <type> <v>Args = term()</v> <v>Result = {ok,State} | {ok,State,Timeout} | {ok,State,hibernate}</v> <v> | {stop,Reason} | ignore</v> <v> State = term()</v> <v> Timeout = int()>=0 | infinity</v> <v> Reason = term()</v> </type> <desc> <marker id="Moduleinit"></marker> <p>Whenever a gen_server is started using <seealso marker="#start/3">gen_server:start/3,4</seealso> or <seealso marker="#start_link/3">gen_server:start_link/3,4</seealso>, this function is called by the new process to initialize.</p> <p><c>Args</c> is the <c>Args</c> argument provided to the start function.</p> <p>If the initialization is successful, the function should return <c>{ok,State}</c>, <c>{ok,State,Timeout}</c> or <c>{ok,State,hibernate}</c>, where <c>State</c> is the internal state of the gen_server.</p> <p>If an integer timeout value is provided, a timeout will occur unless a request or a message is received within <c>Timeout</c> milliseconds. A timeout is represented by the atom <c>timeout</c> which should be handled by the <c>handle_info/2</c> callback function. The atom <c>infinity</c> can be used to wait indefinitely, this is the default value.</p> <p>If <c>hibernate</c> is specified instead of a timeout value, the process will go into hibernation when waiting for the next message to arrive (by calling <seealso marker="proc_lib#hibernate/3">proc_lib:hibernate/3</seealso>).</p> <p>If something goes wrong during the initialization the function should return <c>{stop,Reason}</c> where <c>Reason</c> is any term, or <c>ignore</c>.</p> </desc> </func> <func> <name>Module:handle_call(Request, From, State) -> Result</name> <fsummary>Handle a synchronous request.</fsummary> <type> <v>Request = term()</v> <v>From = {pid(),Tag}</v> <v>State = term()</v> <v>Result = {reply,Reply,NewState} | {reply,Reply,NewState,Timeout}</v> <v> | {reply,Reply,NewState,hibernate}</v> <v> | {noreply,NewState} | {noreply,NewState,Timeout}</v> <v> | {noreply,NewState,hibernate}</v> <v> | {stop,Reason,Reply,NewState} | {stop,Reason,NewState}</v> <v> Reply = term()</v> <v> NewState = term()</v> <v> Timeout = int()>=0 | infinity</v> <v> Reason = term()</v> </type> <desc> <p>Whenever a gen_server receives a request sent using <seealso marker="#call/2">gen_server:call/2,3</seealso> or <seealso marker="#multi_call/2">gen_server:multi_call/2,3,4</seealso>, this function is called to handle the request.</p> <p><c>Request</c> is the <c>Request</c> argument provided to <c>call</c> or <c>multi_call</c>.</p> <p><c>From</c> is a tuple <c>{Pid,Tag}</c> where <c>Pid</c> is the pid of the client and <c>Tag</c> is a unique tag.</p> <p><c>State</c> is the internal state of the gen_server.</p> <p>If the function returns <c>{reply,Reply,NewState}</c>, <c>{reply,Reply,NewState,Timeout}</c> or <c>{reply,Reply,NewState,hibernate}</c>, <c>Reply</c> will be given back to <c>From</c> as the return value of <c>call/2,3</c> or included in the return value of <c>multi_call/2,3,4</c>. The gen_server then continues executing with the possibly updated internal state <c>NewState</c>. See <c>Module:init/1</c> for a description of <c>Timeout</c> and <c>hibernate</c>.</p> <p>If the functions returns <c>{noreply,NewState}</c>, <c>{noreply,NewState,Timeout}</c> or <c>{noreply,NewState,hibernate}</c>, the gen_server will continue executing with <c>NewState</c>. Any reply to <c>From</c> must be given explicitly using <seealso marker="#reply/2">gen_server:reply/2</seealso>.</p> <p>If the function returns <c>{stop,Reason,Reply,NewState}</c>, <c>Reply</c> will be given back to <c>From</c>. If the function returns <c>{stop,Reason,NewState}</c>, any reply to <c>From</c> must be given explicitly using <c>gen_server:reply/2</c>. The gen_server will then call <c>Module:terminate(Reason,NewState)</c> and terminate.</p> </desc> </func> <func> <name>Module:handle_cast(Request, State) -> Result</name> <fsummary>Handle an asynchronous request.</fsummary> <type> <v>Request = term()</v> <v>State = term()</v> <v>Result = {noreply,NewState} | {noreply,NewState,Timeout}</v> <v> | {noreply,NewState,hibernate}</v> <v> | {stop,Reason,NewState}</v> <v> NewState = term()</v> <v> Timeout = int()>=0 | infinity</v> <v> Reason = term()</v> </type> <desc> <p>Whenever a gen_server receives a request sent using <seealso marker="#cast/2">gen_server:cast/2</seealso> or <seealso marker="#abcast/2">gen_server:abcast/2,3</seealso>, this function is called to handle the request.</p> <p>See <c>Module:handle_call/3</c> for a description of the arguments and possible return values.</p> </desc> </func> <func> <name>Module:handle_info(Info, State) -> Result</name> <fsummary>Handle an incoming message.</fsummary> <type> <v>Info = timeout | term()</v> <v>State = term()</v> <v>Result = {noreply,NewState} | {noreply,NewState,Timeout} </v> <v> | {noreply,NewState,hibernate}</v> <v> | {stop,Reason,NewState}</v> <v> NewState = term()</v> <v> Timeout = int()>=0 | infinity</v> <v> Reason = normal | term()</v> </type> <desc> <p>This function is called by a gen_server when a timeout occurs or when it receives any other message than a synchronous or asynchronous request (or a system message).</p> <p><c>Info</c> is either the atom <c>timeout</c>, if a timeout has occurred, or the received message.</p> <p>See <c>Module:handle_call/3</c> for a description of the other arguments and possible return values.</p> </desc> </func> <func> <name>Module:terminate(Reason, State)</name> <fsummary>Clean up before termination.</fsummary> <type> <v>Reason = normal | shutdown | {shutdown,term()} | term()</v> <v>State = term()</v> </type> <desc> <p>This function is called by a gen_server when it is about to terminate. It should be the opposite of <c>Module:init/1</c> and do any necessary cleaning up. When it returns, the gen_server terminates with <c>Reason</c>. The return value is ignored.</p> <p><c>Reason</c> is a term denoting the stop reason and <c>State</c> is the internal state of the gen_server.</p> <p><c>Reason</c> depends on why the gen_server is terminating. If it is because another callback function has returned a stop tuple <c>{stop,..}</c>, <c>Reason</c> will have the value specified in that tuple. If it is due to a failure, <c>Reason</c> is the error reason.</p> <p>If the gen_server is part of a supervision tree and is ordered by its supervisor to terminate, this function will be called with <c>Reason=shutdown</c> if the following conditions apply:</p> <list type="bulleted"> <item>the gen_server has been set to trap exit signals, and</item> <item>the shutdown strategy as defined in the supervisor's child specification is an integer timeout value, not <c>brutal_kill</c>.</item> </list> <p>Even if the gen_server is <em>not</em> part of a supervision tree, this function will be called if it receives an <c>'EXIT'</c> message from its parent. <c>Reason</c> will be the same as in the <c>'EXIT'</c> message.</p> <p>Otherwise, the gen_server will be immediately terminated.</p> <p>Note that for any other reason than <c>normal</c>, <c>shutdown</c>, or <c>{shutdown,Term}</c> the gen_server is assumed to terminate due to an error and an error report is issued using <seealso marker="kernel:error_logger#format/2">error_logger:format/2</seealso>.</p> </desc> </func> <func> <name>Module:code_change(OldVsn, State, Extra) -> {ok, NewState}</name> <fsummary>Update the internal state during upgrade/downgrade.</fsummary> <type> <v>OldVsn = Vsn | {down, Vsn}</v> <v> Vsn = term()</v> <v>State = NewState = term()</v> <v>Extra = term()</v> </type> <desc> <p>This function is called by a gen_server when it should update its internal state during a release upgrade/downgrade, i.e. when the instruction <c>{update,Module,Change,...}</c> where <c>Change={advanced,Extra}</c> is given in the <c>appup</c> file. See <seealso marker="doc/design_principles:release_handling#instr">OTP Design Principles</seealso> for more information.</p> <p>In the case of an upgrade, <c>OldVsn</c> is <c>Vsn</c>, and in the case of a downgrade, <c>OldVsn</c> is <c>{down,Vsn}</c>. <c>Vsn</c> is defined by the <c>vsn</c> attribute(s) of the old version of the callback module <c>Module</c>. If no such attribute is defined, the version is the checksum of the BEAM file.</p> <p><c>State</c> is the internal state of the gen_server.</p> <p><c>Extra</c> is passed as-is from the <c>{advanced,Extra}</c> part of the update instruction.</p> <p>The function should return the updated internal state.</p> </desc> </func> </funcs> <section> <title>SEE ALSO</title> <p><seealso marker="gen_event">gen_event(3)</seealso>, <seealso marker="gen_fsm">gen_fsm(3)</seealso>, <seealso marker="supervisor">supervisor(3)</seealso>, <seealso marker="proc_lib">proc_lib(3)</seealso>, <seealso marker="sys">sys(3)</seealso></p> </section> </erlref>