1 | =head1 NAME |
1 | =head1 NAME |
2 | |
2 | |
3 | AnyEvent::MP - multi-processing/message-passing framework |
3 | AnyEvent::MP - erlang-style multi-processing/message-passing framework |
4 | |
4 | |
5 | =head1 SYNOPSIS |
5 | =head1 SYNOPSIS |
6 | |
6 | |
7 | use AnyEvent::MP; |
7 | use AnyEvent::MP; |
8 | |
8 | |
… | |
… | |
30 | rcv $port, pong => sub { warn "pong received\n" }; |
30 | rcv $port, pong => sub { warn "pong received\n" }; |
31 | |
31 | |
32 | # create a port on another node |
32 | # create a port on another node |
33 | my $port = spawn $node, $initfunc, @initdata; |
33 | my $port = spawn $node, $initfunc, @initdata; |
34 | |
34 | |
|
|
35 | # destroy a port again |
|
|
36 | kil $port; # "normal" kill |
|
|
37 | kil $port, my_error => "everything is broken"; # error kill |
|
|
38 | |
35 | # monitoring |
39 | # monitoring |
36 | mon $port, $cb->(@msg) # callback is invoked on death |
40 | mon $localport, $cb->(@msg) # callback is invoked on death |
37 | mon $port, $otherport # kill otherport on abnormal death |
41 | mon $localport, $otherport # kill otherport on abnormal death |
38 | mon $port, $otherport, @msg # send message on death |
42 | mon $localport, $otherport, @msg # send message on death |
|
|
43 | |
|
|
44 | # temporarily execute code in port context |
|
|
45 | peval $port, sub { die "kill the port!" }; |
|
|
46 | |
|
|
47 | # execute callbacks in $SELF port context |
|
|
48 | my $timer = AE::timer 1, 0, psub { |
|
|
49 | die "kill the port, delayed"; |
|
|
50 | }; |
39 | |
51 | |
40 | =head1 CURRENT STATUS |
52 | =head1 CURRENT STATUS |
41 | |
53 | |
42 | bin/aemp - stable. |
54 | bin/aemp - stable. |
43 | AnyEvent::MP - stable API, should work. |
55 | AnyEvent::MP - stable API, should work. |
44 | AnyEvent::MP::Intro - epxlains most concepts. |
56 | AnyEvent::MP::Intro - explains most concepts. |
45 | AnyEvent::MP::Kernel - mostly stable. |
57 | AnyEvent::MP::Kernel - mostly stable API. |
46 | AnyEvent::MP::Global - stable API, protocol not yet final. |
58 | AnyEvent::MP::Global - stable API. |
47 | |
|
|
48 | stay tuned. |
|
|
49 | |
59 | |
50 | =head1 DESCRIPTION |
60 | =head1 DESCRIPTION |
51 | |
61 | |
52 | This module (-family) implements a simple message passing framework. |
62 | This module (-family) implements a simple message passing framework. |
53 | |
63 | |
… | |
… | |
61 | |
71 | |
62 | =over 4 |
72 | =over 4 |
63 | |
73 | |
64 | =item port |
74 | =item port |
65 | |
75 | |
66 | A port is something you can send messages to (with the C<snd> function). |
76 | Not to be confused with a TCP port, a "port" is something you can send |
|
|
77 | messages to (with the C<snd> function). |
67 | |
78 | |
68 | Ports allow you to register C<rcv> handlers that can match all or just |
79 | Ports allow you to register C<rcv> handlers that can match all or just |
69 | some messages. Messages send to ports will not be queued, regardless of |
80 | some messages. Messages send to ports will not be queued, regardless of |
70 | anything was listening for them or not. |
81 | anything was listening for them or not. |
71 | |
82 | |
|
|
83 | Ports are represented by (printable) strings called "port IDs". |
|
|
84 | |
72 | =item port ID - C<nodeid#portname> |
85 | =item port ID - C<nodeid#portname> |
73 | |
86 | |
74 | A port ID is the concatenation of a node ID, a hash-mark (C<#>) as |
87 | A port ID is the concatenation of a node ID, a hash-mark (C<#>) as |
75 | separator, and a port name (a printable string of unspecified format). |
88 | separator, and a port name (a printable string of unspecified format). |
76 | |
89 | |
… | |
… | |
80 | which enables nodes to manage each other remotely, and to create new |
93 | which enables nodes to manage each other remotely, and to create new |
81 | ports. |
94 | ports. |
82 | |
95 | |
83 | Nodes are either public (have one or more listening ports) or private |
96 | Nodes are either public (have one or more listening ports) or private |
84 | (no listening ports). Private nodes cannot talk to other private nodes |
97 | (no listening ports). Private nodes cannot talk to other private nodes |
85 | currently. |
98 | currently, but all nodes can talk to public nodes. |
86 | |
99 | |
|
|
100 | Nodes is represented by (printable) strings called "node IDs". |
|
|
101 | |
87 | =item node ID - C<[a-za-Z0-9_\-.:]+> |
102 | =item node ID - C<[A-Za-z0-9_\-.:]*> |
88 | |
103 | |
89 | A node ID is a string that uniquely identifies the node within a |
104 | A node ID is a string that uniquely identifies the node within a |
90 | network. Depending on the configuration used, node IDs can look like a |
105 | network. Depending on the configuration used, node IDs can look like a |
91 | hostname, a hostname and a port, or a random string. AnyEvent::MP itself |
106 | hostname, a hostname and a port, or a random string. AnyEvent::MP itself |
92 | doesn't interpret node IDs in any way. |
107 | doesn't interpret node IDs in any way except to uniquely identify a node. |
93 | |
108 | |
94 | =item binds - C<ip:port> |
109 | =item binds - C<ip:port> |
95 | |
110 | |
96 | Nodes can only talk to each other by creating some kind of connection to |
111 | Nodes can only talk to each other by creating some kind of connection to |
97 | each other. To do this, nodes should listen on one or more local transport |
112 | each other. To do this, nodes should listen on one or more local transport |
|
|
113 | endpoints - binds. |
|
|
114 | |
98 | endpoints - binds. Currently, only standard C<ip:port> specifications can |
115 | Currently, only standard C<ip:port> specifications can be used, which |
99 | be used, which specify TCP ports to listen on. |
116 | specify TCP ports to listen on. So a bind is basically just a tcp socket |
|
|
117 | in listening mode thta accepts conenctions form other nodes. |
100 | |
118 | |
|
|
119 | =item seed nodes |
|
|
120 | |
|
|
121 | When a node starts, it knows nothing about the network it is in - it |
|
|
122 | needs to connect to at least one other node that is already in the |
|
|
123 | network. These other nodes are called "seed nodes". |
|
|
124 | |
|
|
125 | Seed nodes themselves are not special - they are seed nodes only because |
|
|
126 | some other node I<uses> them as such, but any node can be used as seed |
|
|
127 | node for other nodes, and eahc node cna use a different set of seed nodes. |
|
|
128 | |
|
|
129 | In addition to discovering the network, seed nodes are also used to |
|
|
130 | maintain the network - all nodes using the same seed node form are part of |
|
|
131 | the same network. If a network is split into multiple subnets because e.g. |
|
|
132 | the network link between the parts goes down, then using the same seed |
|
|
133 | nodes for all nodes ensures that eventually the subnets get merged again. |
|
|
134 | |
|
|
135 | Seed nodes are expected to be long-running, and at least one seed node |
|
|
136 | should always be available. They should also be relatively responsive - a |
|
|
137 | seed node that blocks for long periods will slow down everybody else. |
|
|
138 | |
|
|
139 | For small networks, it's best if every node uses the same set of seed |
|
|
140 | nodes. For large networks, it can be useful to specify "regional" seed |
|
|
141 | nodes for most nodes in an area, and use all seed nodes as seed nodes for |
|
|
142 | each other. What's important is that all seed nodes connections form a |
|
|
143 | complete graph, so that the network cannot split into separate subnets |
|
|
144 | forever. |
|
|
145 | |
|
|
146 | Seed nodes are represented by seed IDs. |
|
|
147 | |
101 | =item seeds - C<host:port> |
148 | =item seed IDs - C<host:port> |
102 | |
149 | |
103 | When a node starts, it knows nothing about the network. To teach the node |
150 | Seed IDs are transport endpoint(s) (usually a hostname/IP address and a |
104 | about the network it first has to contact some other node within the |
151 | TCP port) of nodes that should be used as seed nodes. |
105 | network. This node is called a seed. |
|
|
106 | |
152 | |
107 | Seeds are transport endpoint(s) of as many nodes as one wants. Those nodes |
153 | =item global nodes |
108 | are expected to be long-running, and at least one of those should always |
|
|
109 | be available. When nodes run out of connections (e.g. due to a network |
|
|
110 | error), they try to re-establish connections to some seednodes again to |
|
|
111 | join the network. |
|
|
112 | |
154 | |
113 | Apart from being sued for seeding, seednodes are not special in any way - |
155 | An AEMP network needs a discovery service - nodes need to know how to |
114 | every public node can be a seednode. |
156 | connect to other nodes they only know by name. In addition, AEMP offers a |
|
|
157 | distributed "group database", which maps group names to a list of strings |
|
|
158 | - for example, to register worker ports. |
|
|
159 | |
|
|
160 | A network needs at least one global node to work, and allows every node to |
|
|
161 | be a global node. |
|
|
162 | |
|
|
163 | Any node that loads the L<AnyEvent::MP::Global> module becomes a global |
|
|
164 | node and tries to keep connections to all other nodes. So while it can |
|
|
165 | make sense to make every node "global" in small networks, it usually makes |
|
|
166 | sense to only make seed nodes into global nodes in large networks (nodes |
|
|
167 | keep connections to seed nodes and global nodes, so makign them the same |
|
|
168 | reduces overhead). |
115 | |
169 | |
116 | =back |
170 | =back |
117 | |
171 | |
118 | =head1 VARIABLES/FUNCTIONS |
172 | =head1 VARIABLES/FUNCTIONS |
119 | |
173 | |
… | |
… | |
131 | |
185 | |
132 | use AE (); |
186 | use AE (); |
133 | |
187 | |
134 | use base "Exporter"; |
188 | use base "Exporter"; |
135 | |
189 | |
136 | our $VERSION = $AnyEvent::MP::Kernel::VERSION; |
190 | our $VERSION = '1.30'; |
137 | |
191 | |
138 | our @EXPORT = qw( |
192 | our @EXPORT = qw( |
139 | NODE $NODE *SELF node_of after |
193 | NODE $NODE *SELF node_of after |
140 | configure |
194 | configure |
141 | snd rcv mon mon_guard kil reg psub spawn |
195 | snd rcv mon mon_guard kil psub peval spawn cal |
142 | port |
196 | port |
143 | ); |
197 | ); |
144 | |
198 | |
145 | our $SELF; |
199 | our $SELF; |
146 | |
200 | |
… | |
… | |
158 | |
212 | |
159 | =item $nodeid = node_of $port |
213 | =item $nodeid = node_of $port |
160 | |
214 | |
161 | Extracts and returns the node ID from a port ID or a node ID. |
215 | Extracts and returns the node ID from a port ID or a node ID. |
162 | |
216 | |
|
|
217 | =item configure $profile, key => value... |
|
|
218 | |
163 | =item configure key => value... |
219 | =item configure key => value... |
164 | |
220 | |
165 | Before a node can talk to other nodes on the network (i.e. enter |
221 | Before a node can talk to other nodes on the network (i.e. enter |
166 | "distributed mode") it has to configure itself - the minimum a node needs |
222 | "distributed mode") it has to configure itself - the minimum a node needs |
167 | to know is its own name, and optionally it should know the addresses of |
223 | to know is its own name, and optionally it should know the addresses of |
168 | some other nodes in the network to discover other nodes. |
224 | some other nodes in the network to discover other nodes. |
169 | |
225 | |
|
|
226 | The key/value pairs are basically the same ones as documented for the |
|
|
227 | F<aemp> command line utility (sans the set/del prefix). |
|
|
228 | |
170 | This function configures a node - it must be called exactly once (or |
229 | This function configures a node - it must be called exactly once (or |
171 | never) before calling other AnyEvent::MP functions. |
230 | never) before calling other AnyEvent::MP functions. |
172 | |
231 | |
173 | =over 4 |
232 | =over 4 |
174 | |
233 | |
175 | =item step 1, gathering configuration from profiles |
234 | =item step 1, gathering configuration from profiles |
176 | |
235 | |
177 | The function first looks up a profile in the aemp configuration (see the |
236 | The function first looks up a profile in the aemp configuration (see the |
178 | L<aemp> commandline utility). The profile name can be specified via the |
237 | L<aemp> commandline utility). The profile name can be specified via the |
179 | named C<profile> parameter. If it is missing, then the nodename (F<uname |
238 | named C<profile> parameter or can simply be the first parameter). If it is |
180 | -n>) will be used as profile name. |
239 | missing, then the nodename (F<uname -n>) will be used as profile name. |
181 | |
240 | |
182 | The profile data is then gathered as follows: |
241 | The profile data is then gathered as follows: |
183 | |
242 | |
184 | First, all remaining key => value pairs (all of which are conviniently |
243 | First, all remaining key => value pairs (all of which are conveniently |
185 | undocumented at the moment) will be interpreted as configuration |
244 | undocumented at the moment) will be interpreted as configuration |
186 | data. Then they will be overwritten by any values specified in the global |
245 | data. Then they will be overwritten by any values specified in the global |
187 | default configuration (see the F<aemp> utility), then the chain of |
246 | default configuration (see the F<aemp> utility), then the chain of |
188 | profiles chosen by the profile name (and any C<parent> attributes). |
247 | profiles chosen by the profile name (and any C<parent> attributes). |
189 | |
248 | |
… | |
… | |
207 | used, meaning the node will bind on a dynamically-assigned port on every |
266 | used, meaning the node will bind on a dynamically-assigned port on every |
208 | local IP address it finds. |
267 | local IP address it finds. |
209 | |
268 | |
210 | =item step 3, connect to seed nodes |
269 | =item step 3, connect to seed nodes |
211 | |
270 | |
212 | As the last step, the seeds list from the profile is passed to the |
271 | As the last step, the seed ID list from the profile is passed to the |
213 | L<AnyEvent::MP::Global> module, which will then use it to keep |
272 | L<AnyEvent::MP::Global> module, which will then use it to keep |
214 | connectivity with at least one node at any point in time. |
273 | connectivity with at least one node at any point in time. |
215 | |
274 | |
216 | =back |
275 | =back |
217 | |
276 | |
218 | Example: become a distributed node using the locla node name as profile. |
277 | Example: become a distributed node using the local node name as profile. |
219 | This should be the most common form of invocation for "daemon"-type nodes. |
278 | This should be the most common form of invocation for "daemon"-type nodes. |
220 | |
279 | |
221 | configure |
280 | configure |
222 | |
281 | |
223 | Example: become an anonymous node. This form is often used for commandline |
282 | Example: become an anonymous node. This form is often used for commandline |
… | |
… | |
357 | msg1 => sub { ... }, |
416 | msg1 => sub { ... }, |
358 | ... |
417 | ... |
359 | ; |
418 | ; |
360 | |
419 | |
361 | Example: temporarily register a rcv callback for a tag matching some port |
420 | Example: temporarily register a rcv callback for a tag matching some port |
362 | (e.g. for a rpc reply) and unregister it after a message was received. |
421 | (e.g. for an rpc reply) and unregister it after a message was received. |
363 | |
422 | |
364 | rcv $port, $otherport => sub { |
423 | rcv $port, $otherport => sub { |
365 | my @reply = @_; |
424 | my @reply = @_; |
366 | |
425 | |
367 | rcv $SELF, $otherport; |
426 | rcv $SELF, $otherport; |
… | |
… | |
380 | if (ref $_[0]) { |
439 | if (ref $_[0]) { |
381 | if (my $self = $PORT_DATA{$portid}) { |
440 | if (my $self = $PORT_DATA{$portid}) { |
382 | "AnyEvent::MP::Port" eq ref $self |
441 | "AnyEvent::MP::Port" eq ref $self |
383 | or Carp::croak "$port: rcv can only be called on message matching ports, caught"; |
442 | or Carp::croak "$port: rcv can only be called on message matching ports, caught"; |
384 | |
443 | |
385 | $self->[2] = shift; |
444 | $self->[0] = shift; |
386 | } else { |
445 | } else { |
387 | my $cb = shift; |
446 | my $cb = shift; |
388 | $PORT{$portid} = sub { |
447 | $PORT{$portid} = sub { |
389 | local $SELF = $port; |
448 | local $SELF = $port; |
390 | eval { &$cb }; _self_die if $@; |
449 | eval { &$cb }; _self_die if $@; |
391 | }; |
450 | }; |
392 | } |
451 | } |
393 | } elsif (defined $_[0]) { |
452 | } elsif (defined $_[0]) { |
394 | my $self = $PORT_DATA{$portid} ||= do { |
453 | my $self = $PORT_DATA{$portid} ||= do { |
395 | my $self = bless [$PORT{$port} || sub { }, { }, $port], "AnyEvent::MP::Port"; |
454 | my $self = bless [$PORT{$portid} || sub { }, { }, $port], "AnyEvent::MP::Port"; |
396 | |
455 | |
397 | $PORT{$portid} = sub { |
456 | $PORT{$portid} = sub { |
398 | local $SELF = $port; |
457 | local $SELF = $port; |
399 | |
458 | |
400 | if (my $cb = $self->[1]{$_[0]}) { |
459 | if (my $cb = $self->[1]{$_[0]}) { |
… | |
… | |
422 | } |
481 | } |
423 | |
482 | |
424 | $port |
483 | $port |
425 | } |
484 | } |
426 | |
485 | |
|
|
486 | =item peval $port, $coderef[, @args] |
|
|
487 | |
|
|
488 | Evaluates the given C<$codref> within the contetx of C<$port>, that is, |
|
|
489 | when the code throews an exception the C<$port> will be killed. |
|
|
490 | |
|
|
491 | Any remaining args will be passed to the callback. Any return values will |
|
|
492 | be returned to the caller. |
|
|
493 | |
|
|
494 | This is useful when you temporarily want to execute code in the context of |
|
|
495 | a port. |
|
|
496 | |
|
|
497 | Example: create a port and run some initialisation code in it's context. |
|
|
498 | |
|
|
499 | my $port = port { ... }; |
|
|
500 | |
|
|
501 | peval $port, sub { |
|
|
502 | init |
|
|
503 | or die "unable to init"; |
|
|
504 | }; |
|
|
505 | |
|
|
506 | =cut |
|
|
507 | |
|
|
508 | sub peval($$) { |
|
|
509 | local $SELF = shift; |
|
|
510 | my $cb = shift; |
|
|
511 | |
|
|
512 | if (wantarray) { |
|
|
513 | my @res = eval { &$cb }; |
|
|
514 | _self_die if $@; |
|
|
515 | @res |
|
|
516 | } else { |
|
|
517 | my $res = eval { &$cb }; |
|
|
518 | _self_die if $@; |
|
|
519 | $res |
|
|
520 | } |
|
|
521 | } |
|
|
522 | |
427 | =item $closure = psub { BLOCK } |
523 | =item $closure = psub { BLOCK } |
428 | |
524 | |
429 | Remembers C<$SELF> and creates a closure out of the BLOCK. When the |
525 | Remembers C<$SELF> and creates a closure out of the BLOCK. When the |
430 | closure is executed, sets up the environment in the same way as in C<rcv> |
526 | closure is executed, sets up the environment in the same way as in C<rcv> |
431 | callbacks, i.e. runtime errors will cause the port to get C<kil>ed. |
527 | callbacks, i.e. runtime errors will cause the port to get C<kil>ed. |
|
|
528 | |
|
|
529 | The effect is basically as if it returned C<< sub { peval $SELF, sub { |
|
|
530 | BLOCK }, @_ } >>. |
432 | |
531 | |
433 | This is useful when you register callbacks from C<rcv> callbacks: |
532 | This is useful when you register callbacks from C<rcv> callbacks: |
434 | |
533 | |
435 | rcv delayed_reply => sub { |
534 | rcv delayed_reply => sub { |
436 | my ($delay, @reply) = @_; |
535 | my ($delay, @reply) = @_; |
… | |
… | |
472 | |
571 | |
473 | Monitor the given port and do something when the port is killed or |
572 | Monitor the given port and do something when the port is killed or |
474 | messages to it were lost, and optionally return a guard that can be used |
573 | messages to it were lost, and optionally return a guard that can be used |
475 | to stop monitoring again. |
574 | to stop monitoring again. |
476 | |
575 | |
|
|
576 | In the first form (callback), the callback is simply called with any |
|
|
577 | number of C<@reason> elements (no @reason means that the port was deleted |
|
|
578 | "normally"). Note also that I<< the callback B<must> never die >>, so use |
|
|
579 | C<eval> if unsure. |
|
|
580 | |
|
|
581 | In the second form (another port given), the other port (C<$rcvport>) |
|
|
582 | will be C<kil>'ed with C<@reason>, if a @reason was specified, i.e. on |
|
|
583 | "normal" kils nothing happens, while under all other conditions, the other |
|
|
584 | port is killed with the same reason. |
|
|
585 | |
|
|
586 | The third form (kill self) is the same as the second form, except that |
|
|
587 | C<$rvport> defaults to C<$SELF>. |
|
|
588 | |
|
|
589 | In the last form (message), a message of the form C<@msg, @reason> will be |
|
|
590 | C<snd>. |
|
|
591 | |
|
|
592 | Monitoring-actions are one-shot: once messages are lost (and a monitoring |
|
|
593 | alert was raised), they are removed and will not trigger again. |
|
|
594 | |
|
|
595 | As a rule of thumb, monitoring requests should always monitor a port from |
|
|
596 | a local port (or callback). The reason is that kill messages might get |
|
|
597 | lost, just like any other message. Another less obvious reason is that |
|
|
598 | even monitoring requests can get lost (for example, when the connection |
|
|
599 | to the other node goes down permanently). When monitoring a port locally |
|
|
600 | these problems do not exist. |
|
|
601 | |
477 | C<mon> effectively guarantees that, in the absence of hardware failures, |
602 | C<mon> effectively guarantees that, in the absence of hardware failures, |
478 | after starting the monitor, either all messages sent to the port will |
603 | after starting the monitor, either all messages sent to the port will |
479 | arrive, or the monitoring action will be invoked after possible message |
604 | arrive, or the monitoring action will be invoked after possible message |
480 | loss has been detected. No messages will be lost "in between" (after |
605 | loss has been detected. No messages will be lost "in between" (after |
481 | the first lost message no further messages will be received by the |
606 | the first lost message no further messages will be received by the |
482 | port). After the monitoring action was invoked, further messages might get |
607 | port). After the monitoring action was invoked, further messages might get |
483 | delivered again. |
608 | delivered again. |
484 | |
609 | |
485 | Note that monitoring-actions are one-shot: once messages are lost (and a |
610 | Inter-host-connection timeouts and monitoring depend on the transport |
486 | monitoring alert was raised), they are removed and will not trigger again. |
611 | used. The only transport currently implemented is TCP, and AnyEvent::MP |
|
|
612 | relies on TCP to detect node-downs (this can take 10-15 minutes on a |
|
|
613 | non-idle connection, and usually around two hours for idle connections). |
487 | |
614 | |
488 | In the first form (callback), the callback is simply called with any |
615 | This means that monitoring is good for program errors and cleaning up |
489 | number of C<@reason> elements (no @reason means that the port was deleted |
616 | stuff eventually, but they are no replacement for a timeout when you need |
490 | "normally"). Note also that I<< the callback B<must> never die >>, so use |
617 | to ensure some maximum latency. |
491 | C<eval> if unsure. |
|
|
492 | |
|
|
493 | In the second form (another port given), the other port (C<$rcvport>) |
|
|
494 | will be C<kil>'ed with C<@reason>, iff a @reason was specified, i.e. on |
|
|
495 | "normal" kils nothing happens, while under all other conditions, the other |
|
|
496 | port is killed with the same reason. |
|
|
497 | |
|
|
498 | The third form (kill self) is the same as the second form, except that |
|
|
499 | C<$rvport> defaults to C<$SELF>. |
|
|
500 | |
|
|
501 | In the last form (message), a message of the form C<@msg, @reason> will be |
|
|
502 | C<snd>. |
|
|
503 | |
|
|
504 | As a rule of thumb, monitoring requests should always monitor a port from |
|
|
505 | a local port (or callback). The reason is that kill messages might get |
|
|
506 | lost, just like any other message. Another less obvious reason is that |
|
|
507 | even monitoring requests can get lost (for exmaple, when the connection |
|
|
508 | to the other node goes down permanently). When monitoring a port locally |
|
|
509 | these problems do not exist. |
|
|
510 | |
618 | |
511 | Example: call a given callback when C<$port> is killed. |
619 | Example: call a given callback when C<$port> is killed. |
512 | |
620 | |
513 | mon $port, sub { warn "port died because of <@_>\n" }; |
621 | mon $port, sub { warn "port died because of <@_>\n" }; |
514 | |
622 | |
… | |
… | |
542 | } |
650 | } |
543 | |
651 | |
544 | $node->monitor ($port, $cb); |
652 | $node->monitor ($port, $cb); |
545 | |
653 | |
546 | defined wantarray |
654 | defined wantarray |
547 | and AnyEvent::Util::guard { $node->unmonitor ($port, $cb) } |
655 | and ($cb += 0, AnyEvent::Util::guard { $node->unmonitor ($port, $cb) }) |
548 | } |
656 | } |
549 | |
657 | |
550 | =item $guard = mon_guard $port, $ref, $ref... |
658 | =item $guard = mon_guard $port, $ref, $ref... |
551 | |
659 | |
552 | Monitors the given C<$port> and keeps the passed references. When the port |
660 | Monitors the given C<$port> and keeps the passed references. When the port |
… | |
… | |
575 | |
683 | |
576 | =item kil $port[, @reason] |
684 | =item kil $port[, @reason] |
577 | |
685 | |
578 | Kill the specified port with the given C<@reason>. |
686 | Kill the specified port with the given C<@reason>. |
579 | |
687 | |
580 | If no C<@reason> is specified, then the port is killed "normally" (ports |
688 | If no C<@reason> is specified, then the port is killed "normally" - |
581 | monitoring other ports will not necessarily die because a port dies |
689 | monitor callback will be invoked, but the kil will not cause linked ports |
582 | "normally"). |
690 | (C<mon $mport, $lport> form) to get killed. |
583 | |
691 | |
584 | Otherwise, linked ports get killed with the same reason (second form of |
692 | If a C<@reason> is specified, then linked ports (C<mon $mport, $lport> |
585 | C<mon>, see above). |
693 | form) get killed with the same reason. |
586 | |
694 | |
587 | Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks |
695 | Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks |
588 | will be reported as reason C<< die => $@ >>. |
696 | will be reported as reason C<< die => $@ >>. |
589 | |
697 | |
590 | Transport/communication errors are reported as C<< transport_error => |
698 | Transport/communication errors are reported as C<< transport_error => |
… | |
… | |
609 | the package, then the package above the package and so on (e.g. |
717 | the package, then the package above the package and so on (e.g. |
610 | C<MyApp::Chat::Server>, C<MyApp::Chat>, C<MyApp>) until the function |
718 | C<MyApp::Chat::Server>, C<MyApp::Chat>, C<MyApp>) until the function |
611 | exists or it runs out of package names. |
719 | exists or it runs out of package names. |
612 | |
720 | |
613 | The init function is then called with the newly-created port as context |
721 | The init function is then called with the newly-created port as context |
614 | object (C<$SELF>) and the C<@initdata> values as arguments. |
722 | object (C<$SELF>) and the C<@initdata> values as arguments. It I<must> |
|
|
723 | call one of the C<rcv> functions to set callbacks on C<$SELF>, otherwise |
|
|
724 | the port might not get created. |
615 | |
725 | |
616 | A common idiom is to pass a local port, immediately monitor the spawned |
726 | A common idiom is to pass a local port, immediately monitor the spawned |
617 | port, and in the remote init function, immediately monitor the passed |
727 | port, and in the remote init function, immediately monitor the passed |
618 | local port. This two-way monitoring ensures that both ports get cleaned up |
728 | local port. This two-way monitoring ensures that both ports get cleaned up |
619 | when there is a problem. |
729 | when there is a problem. |
620 | |
730 | |
|
|
731 | C<spawn> guarantees that the C<$initfunc> has no visible effects on the |
|
|
732 | caller before C<spawn> returns (by delaying invocation when spawn is |
|
|
733 | called for the local node). |
|
|
734 | |
621 | Example: spawn a chat server port on C<$othernode>. |
735 | Example: spawn a chat server port on C<$othernode>. |
622 | |
736 | |
623 | # this node, executed from within a port context: |
737 | # this node, executed from within a port context: |
624 | my $server = spawn $othernode, "MyApp::Chat::Server::connect", $SELF; |
738 | my $server = spawn $othernode, "MyApp::Chat::Server::connect", $SELF; |
625 | mon $server; |
739 | mon $server; |
… | |
… | |
639 | |
753 | |
640 | sub _spawn { |
754 | sub _spawn { |
641 | my $port = shift; |
755 | my $port = shift; |
642 | my $init = shift; |
756 | my $init = shift; |
643 | |
757 | |
|
|
758 | # rcv will create the actual port |
644 | local $SELF = "$NODE#$port"; |
759 | local $SELF = "$NODE#$port"; |
645 | eval { |
760 | eval { |
646 | &{ load_func $init } |
761 | &{ load_func $init } |
647 | }; |
762 | }; |
648 | _self_die if $@; |
763 | _self_die if $@; |
… | |
… | |
683 | ? $action[0]() |
798 | ? $action[0]() |
684 | : snd @action; |
799 | : snd @action; |
685 | }; |
800 | }; |
686 | } |
801 | } |
687 | |
802 | |
|
|
803 | =item cal $port, @msg, $callback[, $timeout] |
|
|
804 | |
|
|
805 | A simple form of RPC - sends a message to the given C<$port> with the |
|
|
806 | given contents (C<@msg>), but adds a reply port to the message. |
|
|
807 | |
|
|
808 | The reply port is created temporarily just for the purpose of receiving |
|
|
809 | the reply, and will be C<kil>ed when no longer needed. |
|
|
810 | |
|
|
811 | A reply message sent to the port is passed to the C<$callback> as-is. |
|
|
812 | |
|
|
813 | If an optional time-out (in seconds) is given and it is not C<undef>, |
|
|
814 | then the callback will be called without any arguments after the time-out |
|
|
815 | elapsed and the port is C<kil>ed. |
|
|
816 | |
|
|
817 | If no time-out is given (or it is C<undef>), then the local port will |
|
|
818 | monitor the remote port instead, so it eventually gets cleaned-up. |
|
|
819 | |
|
|
820 | Currently this function returns the temporary port, but this "feature" |
|
|
821 | might go in future versions unless you can make a convincing case that |
|
|
822 | this is indeed useful for something. |
|
|
823 | |
|
|
824 | =cut |
|
|
825 | |
|
|
826 | sub cal(@) { |
|
|
827 | my $timeout = ref $_[-1] ? undef : pop; |
|
|
828 | my $cb = pop; |
|
|
829 | |
|
|
830 | my $port = port { |
|
|
831 | undef $timeout; |
|
|
832 | kil $SELF; |
|
|
833 | &$cb; |
|
|
834 | }; |
|
|
835 | |
|
|
836 | if (defined $timeout) { |
|
|
837 | $timeout = AE::timer $timeout, 0, sub { |
|
|
838 | undef $timeout; |
|
|
839 | kil $port; |
|
|
840 | $cb->(); |
|
|
841 | }; |
|
|
842 | } else { |
|
|
843 | mon $_[0], sub { |
|
|
844 | kil $port; |
|
|
845 | $cb->(); |
|
|
846 | }; |
|
|
847 | } |
|
|
848 | |
|
|
849 | push @_, $port; |
|
|
850 | &snd; |
|
|
851 | |
|
|
852 | $port |
|
|
853 | } |
|
|
854 | |
688 | =back |
855 | =back |
689 | |
856 | |
690 | =head1 AnyEvent::MP vs. Distributed Erlang |
857 | =head1 AnyEvent::MP vs. Distributed Erlang |
691 | |
858 | |
692 | AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node |
859 | AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node |
693 | == aemp node, Erlang process == aemp port), so many of the documents and |
860 | == aemp node, Erlang process == aemp port), so many of the documents and |
694 | programming techniques employed by Erlang apply to AnyEvent::MP. Here is a |
861 | programming techniques employed by Erlang apply to AnyEvent::MP. Here is a |
695 | sample: |
862 | sample: |
696 | |
863 | |
697 | http://www.Erlang.se/doc/programming_rules.shtml |
864 | http://www.erlang.se/doc/programming_rules.shtml |
698 | http://Erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4 |
865 | http://erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4 |
699 | http://Erlang.org/download/Erlang-book-part1.pdf # chapters 5 and 6 |
866 | http://erlang.org/download/erlang-book-part1.pdf # chapters 5 and 6 |
700 | http://Erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5 |
867 | http://erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5 |
701 | |
868 | |
702 | Despite the similarities, there are also some important differences: |
869 | Despite the similarities, there are also some important differences: |
703 | |
870 | |
704 | =over 4 |
871 | =over 4 |
705 | |
872 | |
706 | =item * Node IDs are arbitrary strings in AEMP. |
873 | =item * Node IDs are arbitrary strings in AEMP. |
707 | |
874 | |
708 | Erlang relies on special naming and DNS to work everywhere in the same |
875 | Erlang relies on special naming and DNS to work everywhere in the same |
709 | way. AEMP relies on each node somehow knowing its own address(es) (e.g. by |
876 | way. AEMP relies on each node somehow knowing its own address(es) (e.g. by |
710 | configuraiton or DNS), but will otherwise discover other odes itself. |
877 | configuration or DNS), and possibly the addresses of some seed nodes, but |
|
|
878 | will otherwise discover other nodes (and their IDs) itself. |
711 | |
879 | |
712 | =item * Erlang has a "remote ports are like local ports" philosophy, AEMP |
880 | =item * Erlang has a "remote ports are like local ports" philosophy, AEMP |
713 | uses "local ports are like remote ports". |
881 | uses "local ports are like remote ports". |
714 | |
882 | |
715 | The failure modes for local ports are quite different (runtime errors |
883 | The failure modes for local ports are quite different (runtime errors |
… | |
… | |
724 | ports being the special case/exception, where transport errors cannot |
892 | ports being the special case/exception, where transport errors cannot |
725 | occur. |
893 | occur. |
726 | |
894 | |
727 | =item * Erlang uses processes and a mailbox, AEMP does not queue. |
895 | =item * Erlang uses processes and a mailbox, AEMP does not queue. |
728 | |
896 | |
729 | Erlang uses processes that selectively receive messages, and therefore |
897 | Erlang uses processes that selectively receive messages out of order, and |
730 | needs a queue. AEMP is event based, queuing messages would serve no |
898 | therefore needs a queue. AEMP is event based, queuing messages would serve |
731 | useful purpose. For the same reason the pattern-matching abilities of |
899 | no useful purpose. For the same reason the pattern-matching abilities |
732 | AnyEvent::MP are more limited, as there is little need to be able to |
900 | of AnyEvent::MP are more limited, as there is little need to be able to |
733 | filter messages without dequeing them. |
901 | filter messages without dequeuing them. |
734 | |
902 | |
735 | (But see L<Coro::MP> for a more Erlang-like process model on top of AEMP). |
903 | This is not a philosophical difference, but simply stems from AnyEvent::MP |
|
|
904 | being event-based, while Erlang is process-based. |
|
|
905 | |
|
|
906 | You cna have a look at L<Coro::MP> for a more Erlang-like process model on |
|
|
907 | top of AEMP and Coro threads. |
736 | |
908 | |
737 | =item * Erlang sends are synchronous, AEMP sends are asynchronous. |
909 | =item * Erlang sends are synchronous, AEMP sends are asynchronous. |
738 | |
910 | |
739 | Sending messages in Erlang is synchronous and blocks the process (and |
911 | Sending messages in Erlang is synchronous and blocks the process until |
|
|
912 | a conenction has been established and the message sent (and so does not |
740 | so does not need a queue that can overflow). AEMP sends are immediate, |
913 | need a queue that can overflow). AEMP sends return immediately, connection |
741 | connection establishment is handled in the background. |
914 | establishment is handled in the background. |
742 | |
915 | |
743 | =item * Erlang suffers from silent message loss, AEMP does not. |
916 | =item * Erlang suffers from silent message loss, AEMP does not. |
744 | |
917 | |
745 | Erlang makes few guarantees on messages delivery - messages can get lost |
918 | Erlang implements few guarantees on messages delivery - messages can get |
746 | without any of the processes realising it (i.e. you send messages a, b, |
919 | lost without any of the processes realising it (i.e. you send messages a, |
747 | and c, and the other side only receives messages a and c). |
920 | b, and c, and the other side only receives messages a and c). |
748 | |
921 | |
749 | AEMP guarantees correct ordering, and the guarantee that after one message |
922 | AEMP guarantees (modulo hardware errors) correct ordering, and the |
750 | is lost, all following ones sent to the same port are lost as well, until |
923 | guarantee that after one message is lost, all following ones sent to the |
751 | monitoring raises an error, so there are no silent "holes" in the message |
924 | same port are lost as well, until monitoring raises an error, so there are |
752 | sequence. |
925 | no silent "holes" in the message sequence. |
|
|
926 | |
|
|
927 | If you want your software to be very reliable, you have to cope with |
|
|
928 | corrupted and even out-of-order messages in both Erlang and AEMP. AEMP |
|
|
929 | simply tries to work better in common error cases, such as when a network |
|
|
930 | link goes down. |
753 | |
931 | |
754 | =item * Erlang can send messages to the wrong port, AEMP does not. |
932 | =item * Erlang can send messages to the wrong port, AEMP does not. |
755 | |
933 | |
756 | In Erlang it is quite likely that a node that restarts reuses a process ID |
934 | In Erlang it is quite likely that a node that restarts reuses an Erlang |
757 | known to other nodes for a completely different process, causing messages |
935 | process ID known to other nodes for a completely different process, |
758 | destined for that process to end up in an unrelated process. |
936 | causing messages destined for that process to end up in an unrelated |
|
|
937 | process. |
759 | |
938 | |
760 | AEMP never reuses port IDs, so old messages or old port IDs floating |
939 | AEMP does not reuse port IDs, so old messages or old port IDs floating |
761 | around in the network will not be sent to an unrelated port. |
940 | around in the network will not be sent to an unrelated port. |
762 | |
941 | |
763 | =item * Erlang uses unprotected connections, AEMP uses secure |
942 | =item * Erlang uses unprotected connections, AEMP uses secure |
764 | authentication and can use TLS. |
943 | authentication and can use TLS. |
765 | |
944 | |
… | |
… | |
768 | |
947 | |
769 | =item * The AEMP protocol is optimised for both text-based and binary |
948 | =item * The AEMP protocol is optimised for both text-based and binary |
770 | communications. |
949 | communications. |
771 | |
950 | |
772 | The AEMP protocol, unlike the Erlang protocol, supports both programming |
951 | The AEMP protocol, unlike the Erlang protocol, supports both programming |
773 | language independent text-only protocols (good for debugging) and binary, |
952 | language independent text-only protocols (good for debugging), and binary, |
774 | language-specific serialisers (e.g. Storable). By default, unless TLS is |
953 | language-specific serialisers (e.g. Storable). By default, unless TLS is |
775 | used, the protocol is actually completely text-based. |
954 | used, the protocol is actually completely text-based. |
776 | |
955 | |
777 | It has also been carefully designed to be implementable in other languages |
956 | It has also been carefully designed to be implementable in other languages |
778 | with a minimum of work while gracefully degrading functionality to make the |
957 | with a minimum of work while gracefully degrading functionality to make the |
779 | protocol simple. |
958 | protocol simple. |
780 | |
959 | |
781 | =item * AEMP has more flexible monitoring options than Erlang. |
960 | =item * AEMP has more flexible monitoring options than Erlang. |
782 | |
961 | |
783 | In Erlang, you can chose to receive I<all> exit signals as messages |
962 | In Erlang, you can chose to receive I<all> exit signals as messages or |
784 | or I<none>, there is no in-between, so monitoring single processes is |
963 | I<none>, there is no in-between, so monitoring single Erlang processes is |
785 | difficult to implement. Monitoring in AEMP is more flexible than in |
964 | difficult to implement. |
786 | Erlang, as one can choose between automatic kill, exit message or callback |
965 | |
787 | on a per-process basis. |
966 | Monitoring in AEMP is more flexible than in Erlang, as one can choose |
|
|
967 | between automatic kill, exit message or callback on a per-port basis. |
788 | |
968 | |
789 | =item * Erlang tries to hide remote/local connections, AEMP does not. |
969 | =item * Erlang tries to hide remote/local connections, AEMP does not. |
790 | |
970 | |
791 | Monitoring in Erlang is not an indicator of process death/crashes, in the |
971 | Monitoring in Erlang is not an indicator of process death/crashes, in the |
792 | same way as linking is (except linking is unreliable in Erlang). |
972 | same way as linking is (except linking is unreliable in Erlang). |
… | |
… | |
814 | overhead, as well as having to keep a proxy object everywhere. |
994 | overhead, as well as having to keep a proxy object everywhere. |
815 | |
995 | |
816 | Strings can easily be printed, easily serialised etc. and need no special |
996 | Strings can easily be printed, easily serialised etc. and need no special |
817 | procedures to be "valid". |
997 | procedures to be "valid". |
818 | |
998 | |
819 | And as a result, a miniport consists of a single closure stored in a |
999 | And as a result, a port with just a default receiver consists of a single |
820 | global hash - it can't become much cheaper. |
1000 | code reference stored in a global hash - it can't become much cheaper. |
821 | |
1001 | |
822 | =item Why favour JSON, why not a real serialising format such as Storable? |
1002 | =item Why favour JSON, why not a real serialising format such as Storable? |
823 | |
1003 | |
824 | In fact, any AnyEvent::MP node will happily accept Storable as framing |
1004 | In fact, any AnyEvent::MP node will happily accept Storable as framing |
825 | format, but currently there is no way to make a node use Storable by |
1005 | format, but currently there is no way to make a node use Storable by |
… | |
… | |
841 | |
1021 | |
842 | L<AnyEvent::MP::Intro> - a gentle introduction. |
1022 | L<AnyEvent::MP::Intro> - a gentle introduction. |
843 | |
1023 | |
844 | L<AnyEvent::MP::Kernel> - more, lower-level, stuff. |
1024 | L<AnyEvent::MP::Kernel> - more, lower-level, stuff. |
845 | |
1025 | |
846 | L<AnyEvent::MP::Global> - network maintainance and port groups, to find |
1026 | L<AnyEvent::MP::Global> - network maintenance and port groups, to find |
847 | your applications. |
1027 | your applications. |
|
|
1028 | |
|
|
1029 | L<AnyEvent::MP::DataConn> - establish data connections between nodes. |
|
|
1030 | |
|
|
1031 | L<AnyEvent::MP::LogCatcher> - simple service to display log messages from |
|
|
1032 | all nodes. |
848 | |
1033 | |
849 | L<AnyEvent>. |
1034 | L<AnyEvent>. |
850 | |
1035 | |
851 | =head1 AUTHOR |
1036 | =head1 AUTHOR |
852 | |
1037 | |