1 |
root |
1.1 |
=head1 NAME |
2 |
|
|
|
3 |
root |
1.89 |
AnyEvent::MP - erlang-style multi-processing/message-passing framework |
4 |
root |
1.1 |
|
5 |
|
|
=head1 SYNOPSIS |
6 |
|
|
|
7 |
|
|
use AnyEvent::MP; |
8 |
|
|
|
9 |
root |
1.75 |
$NODE # contains this node's node ID |
10 |
|
|
NODE # returns this node's node ID |
11 |
root |
1.2 |
|
12 |
root |
1.38 |
$SELF # receiving/own port id in rcv callbacks |
13 |
|
|
|
14 |
root |
1.48 |
# initialise the node so it can send/receive messages |
15 |
root |
1.72 |
configure; |
16 |
root |
1.48 |
|
17 |
root |
1.75 |
# ports are message destinations |
18 |
root |
1.38 |
|
19 |
|
|
# sending messages |
20 |
root |
1.2 |
snd $port, type => data...; |
21 |
root |
1.38 |
snd $port, @msg; |
22 |
|
|
snd @msg_with_first_element_being_a_port; |
23 |
root |
1.2 |
|
24 |
root |
1.50 |
# creating/using ports, the simple way |
25 |
root |
1.73 |
my $simple_port = port { my @msg = @_ }; |
26 |
root |
1.22 |
|
27 |
root |
1.52 |
# creating/using ports, tagged message matching |
28 |
root |
1.38 |
my $port = port; |
29 |
root |
1.73 |
rcv $port, ping => sub { snd $_[0], "pong" }; |
30 |
|
|
rcv $port, pong => sub { warn "pong received\n" }; |
31 |
root |
1.2 |
|
32 |
root |
1.48 |
# create a port on another node |
33 |
|
|
my $port = spawn $node, $initfunc, @initdata; |
34 |
|
|
|
35 |
root |
1.116 |
# destroy a port again |
36 |
root |
1.101 |
kil $port; # "normal" kill |
37 |
|
|
kil $port, my_error => "everything is broken"; # error kill |
38 |
|
|
|
39 |
root |
1.35 |
# monitoring |
40 |
root |
1.129 |
mon $port, $cb->(@msg) # callback is invoked on death |
41 |
|
|
mon $port, $localport # kill localport on abnormal death |
42 |
|
|
mon $port, $localport, @msg # send message on death |
43 |
root |
1.35 |
|
44 |
root |
1.101 |
# 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 |
|
|
}; |
51 |
|
|
|
52 |
root |
1.143 |
# distributed database - modification |
53 |
|
|
db_set $family => $subkey [=> $value] # add a subkey |
54 |
|
|
db_del $family => $subkey... # delete one or more subkeys |
55 |
|
|
db_reg $family => $port [=> $value] # register a port |
56 |
|
|
|
57 |
|
|
# distributed database - queries |
58 |
|
|
db_family $family => $cb->(\%familyhash) |
59 |
|
|
db_keys $family => $cb->(\@keys) |
60 |
|
|
db_values $family => $cb->(\@values) |
61 |
|
|
|
62 |
|
|
# distributed database - monitoring a family |
63 |
|
|
db_mon $family => $cb->(\%familyhash, \@added, \@changed, \@deleted) |
64 |
|
|
|
65 |
root |
1.1 |
=head1 DESCRIPTION |
66 |
|
|
|
67 |
root |
1.2 |
This module (-family) implements a simple message passing framework. |
68 |
|
|
|
69 |
|
|
Despite its simplicity, you can securely message other processes running |
70 |
root |
1.67 |
on the same or other hosts, and you can supervise entities remotely. |
71 |
root |
1.2 |
|
72 |
root |
1.23 |
For an introduction to this module family, see the L<AnyEvent::MP::Intro> |
73 |
root |
1.67 |
manual page and the examples under F<eg/>. |
74 |
root |
1.23 |
|
75 |
root |
1.2 |
=head1 CONCEPTS |
76 |
|
|
|
77 |
|
|
=over 4 |
78 |
|
|
|
79 |
|
|
=item port |
80 |
|
|
|
81 |
root |
1.79 |
Not to be confused with a TCP port, a "port" is something you can send |
82 |
|
|
messages to (with the C<snd> function). |
83 |
root |
1.29 |
|
84 |
root |
1.53 |
Ports allow you to register C<rcv> handlers that can match all or just |
85 |
root |
1.64 |
some messages. Messages send to ports will not be queued, regardless of |
86 |
|
|
anything was listening for them or not. |
87 |
root |
1.2 |
|
88 |
root |
1.119 |
Ports are represented by (printable) strings called "port IDs". |
89 |
|
|
|
90 |
root |
1.67 |
=item port ID - C<nodeid#portname> |
91 |
root |
1.2 |
|
92 |
root |
1.123 |
A port ID is the concatenation of a node ID, a hash-mark (C<#>) |
93 |
|
|
as separator, and a port name (a printable string of unspecified |
94 |
|
|
format created by AnyEvent::MP). |
95 |
root |
1.2 |
|
96 |
|
|
=item node |
97 |
|
|
|
98 |
root |
1.53 |
A node is a single process containing at least one port - the node port, |
99 |
root |
1.67 |
which enables nodes to manage each other remotely, and to create new |
100 |
root |
1.53 |
ports. |
101 |
root |
1.2 |
|
102 |
root |
1.67 |
Nodes are either public (have one or more listening ports) or private |
103 |
|
|
(no listening ports). Private nodes cannot talk to other private nodes |
104 |
root |
1.119 |
currently, but all nodes can talk to public nodes. |
105 |
|
|
|
106 |
|
|
Nodes is represented by (printable) strings called "node IDs". |
107 |
root |
1.2 |
|
108 |
root |
1.117 |
=item node ID - C<[A-Za-z0-9_\-.:]*> |
109 |
root |
1.2 |
|
110 |
root |
1.64 |
A node ID is a string that uniquely identifies the node within a |
111 |
|
|
network. Depending on the configuration used, node IDs can look like a |
112 |
|
|
hostname, a hostname and a port, or a random string. AnyEvent::MP itself |
113 |
root |
1.119 |
doesn't interpret node IDs in any way except to uniquely identify a node. |
114 |
root |
1.64 |
|
115 |
|
|
=item binds - C<ip:port> |
116 |
|
|
|
117 |
|
|
Nodes can only talk to each other by creating some kind of connection to |
118 |
|
|
each other. To do this, nodes should listen on one or more local transport |
119 |
root |
1.119 |
endpoints - binds. |
120 |
|
|
|
121 |
|
|
Currently, only standard C<ip:port> specifications can be used, which |
122 |
|
|
specify TCP ports to listen on. So a bind is basically just a tcp socket |
123 |
root |
1.147 |
in listening mode that accepts connections from other nodes. |
124 |
root |
1.64 |
|
125 |
root |
1.83 |
=item seed nodes |
126 |
root |
1.64 |
|
127 |
root |
1.119 |
When a node starts, it knows nothing about the network it is in - it |
128 |
|
|
needs to connect to at least one other node that is already in the |
129 |
|
|
network. These other nodes are called "seed nodes". |
130 |
|
|
|
131 |
|
|
Seed nodes themselves are not special - they are seed nodes only because |
132 |
|
|
some other node I<uses> them as such, but any node can be used as seed |
133 |
root |
1.146 |
node for other nodes, and eahc node can use a different set of seed nodes. |
134 |
root |
1.83 |
|
135 |
|
|
In addition to discovering the network, seed nodes are also used to |
136 |
root |
1.147 |
maintain the network - all nodes using the same seed node are part of the |
137 |
|
|
same network. If a network is split into multiple subnets because e.g. the |
138 |
|
|
network link between the parts goes down, then using the same seed nodes |
139 |
|
|
for all nodes ensures that eventually the subnets get merged again. |
140 |
root |
1.83 |
|
141 |
|
|
Seed nodes are expected to be long-running, and at least one seed node |
142 |
root |
1.85 |
should always be available. They should also be relatively responsive - a |
143 |
|
|
seed node that blocks for long periods will slow down everybody else. |
144 |
root |
1.83 |
|
145 |
root |
1.119 |
For small networks, it's best if every node uses the same set of seed |
146 |
|
|
nodes. For large networks, it can be useful to specify "regional" seed |
147 |
|
|
nodes for most nodes in an area, and use all seed nodes as seed nodes for |
148 |
|
|
each other. What's important is that all seed nodes connections form a |
149 |
|
|
complete graph, so that the network cannot split into separate subnets |
150 |
|
|
forever. |
151 |
|
|
|
152 |
|
|
Seed nodes are represented by seed IDs. |
153 |
|
|
|
154 |
|
|
=item seed IDs - C<host:port> |
155 |
root |
1.83 |
|
156 |
root |
1.119 |
Seed IDs are transport endpoint(s) (usually a hostname/IP address and a |
157 |
elmex |
1.96 |
TCP port) of nodes that should be used as seed nodes. |
158 |
root |
1.29 |
|
159 |
root |
1.119 |
=item global nodes |
160 |
|
|
|
161 |
|
|
An AEMP network needs a discovery service - nodes need to know how to |
162 |
|
|
connect to other nodes they only know by name. In addition, AEMP offers a |
163 |
|
|
distributed "group database", which maps group names to a list of strings |
164 |
|
|
- for example, to register worker ports. |
165 |
|
|
|
166 |
|
|
A network needs at least one global node to work, and allows every node to |
167 |
|
|
be a global node. |
168 |
|
|
|
169 |
|
|
Any node that loads the L<AnyEvent::MP::Global> module becomes a global |
170 |
|
|
node and tries to keep connections to all other nodes. So while it can |
171 |
|
|
make sense to make every node "global" in small networks, it usually makes |
172 |
|
|
sense to only make seed nodes into global nodes in large networks (nodes |
173 |
root |
1.146 |
keep connections to seed nodes and global nodes, so making them the same |
174 |
root |
1.119 |
reduces overhead). |
175 |
root |
1.67 |
|
176 |
root |
1.2 |
=back |
177 |
|
|
|
178 |
root |
1.3 |
=head1 VARIABLES/FUNCTIONS |
179 |
root |
1.2 |
|
180 |
|
|
=over 4 |
181 |
|
|
|
182 |
root |
1.1 |
=cut |
183 |
|
|
|
184 |
|
|
package AnyEvent::MP; |
185 |
|
|
|
186 |
root |
1.121 |
use AnyEvent::MP::Config (); |
187 |
root |
1.44 |
use AnyEvent::MP::Kernel; |
188 |
root |
1.144 |
use AnyEvent::MP::Kernel qw( |
189 |
|
|
%NODE %PORT %PORT_DATA $UNIQ $RUNIQ $ID |
190 |
|
|
add_node load_func |
191 |
|
|
|
192 |
|
|
NODE $NODE |
193 |
|
|
configure |
194 |
|
|
node_of port_is_local |
195 |
|
|
snd kil |
196 |
|
|
db_set db_del |
197 |
|
|
db_mon db_family db_keys db_values |
198 |
|
|
); |
199 |
root |
1.2 |
|
200 |
root |
1.1 |
use common::sense; |
201 |
|
|
|
202 |
root |
1.2 |
use Carp (); |
203 |
|
|
|
204 |
root |
1.141 |
use AnyEvent (); |
205 |
root |
1.124 |
use Guard (); |
206 |
root |
1.1 |
|
207 |
root |
1.2 |
use base "Exporter"; |
208 |
|
|
|
209 |
root |
1.152 |
our $VERSION = '2.02'; # also in MP/Config.pm |
210 |
root |
1.43 |
|
211 |
root |
1.8 |
our @EXPORT = qw( |
212 |
root |
1.72 |
configure |
213 |
root |
1.144 |
|
214 |
root |
1.153 |
NODE $NODE |
215 |
root |
1.144 |
*SELF |
216 |
|
|
|
217 |
root |
1.153 |
node_of port_is_local |
218 |
|
|
|
219 |
|
|
snd kil |
220 |
root |
1.144 |
port rcv mon mon_guard psub peval spawn cal |
221 |
root |
1.124 |
db_set db_del db_reg |
222 |
root |
1.128 |
db_mon db_family db_keys db_values |
223 |
root |
1.144 |
|
224 |
|
|
after |
225 |
root |
1.8 |
); |
226 |
root |
1.2 |
|
227 |
root |
1.22 |
our $SELF; |
228 |
|
|
|
229 |
|
|
sub _self_die() { |
230 |
|
|
my $msg = $@; |
231 |
|
|
$msg =~ s/\n+$// unless ref $msg; |
232 |
|
|
kil $SELF, die => $msg; |
233 |
|
|
} |
234 |
|
|
|
235 |
|
|
=item $thisnode = NODE / $NODE |
236 |
|
|
|
237 |
root |
1.67 |
The C<NODE> function returns, and the C<$NODE> variable contains, the node |
238 |
root |
1.64 |
ID of the node running in the current process. This value is initialised by |
239 |
root |
1.72 |
a call to C<configure>. |
240 |
root |
1.22 |
|
241 |
root |
1.63 |
=item $nodeid = node_of $port |
242 |
root |
1.22 |
|
243 |
root |
1.67 |
Extracts and returns the node ID from a port ID or a node ID. |
244 |
root |
1.34 |
|
245 |
root |
1.144 |
=item $is_local = port_is_local $port |
246 |
|
|
|
247 |
|
|
Returns true iff the port is a local port. |
248 |
|
|
|
249 |
root |
1.78 |
=item configure $profile, key => value... |
250 |
|
|
|
251 |
root |
1.72 |
=item configure key => value... |
252 |
root |
1.34 |
|
253 |
root |
1.64 |
Before a node can talk to other nodes on the network (i.e. enter |
254 |
root |
1.72 |
"distributed mode") it has to configure itself - the minimum a node needs |
255 |
root |
1.64 |
to know is its own name, and optionally it should know the addresses of |
256 |
|
|
some other nodes in the network to discover other nodes. |
257 |
root |
1.34 |
|
258 |
root |
1.121 |
This function configures a node - it must be called exactly once (or |
259 |
|
|
never) before calling other AnyEvent::MP functions. |
260 |
|
|
|
261 |
root |
1.108 |
The key/value pairs are basically the same ones as documented for the |
262 |
root |
1.127 |
F<aemp> command line utility (sans the set/del prefix), with these additions: |
263 |
root |
1.121 |
|
264 |
|
|
=over 4 |
265 |
|
|
|
266 |
|
|
=item norc => $boolean (default false) |
267 |
|
|
|
268 |
|
|
If true, then the rc file (e.g. F<~/.perl-anyevent-mp>) will I<not> |
269 |
root |
1.145 |
be consulted - all configuration options must be specified in the |
270 |
root |
1.121 |
C<configure> call. |
271 |
root |
1.108 |
|
272 |
root |
1.121 |
=item force => $boolean (default false) |
273 |
|
|
|
274 |
|
|
IF true, then the values specified in the C<configure> will take |
275 |
|
|
precedence over any values configured via the rc file. The default is for |
276 |
|
|
the rc file to override any options specified in the program. |
277 |
|
|
|
278 |
|
|
=back |
279 |
root |
1.34 |
|
280 |
root |
1.72 |
=over 4 |
281 |
|
|
|
282 |
|
|
=item step 1, gathering configuration from profiles |
283 |
|
|
|
284 |
|
|
The function first looks up a profile in the aemp configuration (see the |
285 |
|
|
L<aemp> commandline utility). The profile name can be specified via the |
286 |
root |
1.78 |
named C<profile> parameter or can simply be the first parameter). If it is |
287 |
|
|
missing, then the nodename (F<uname -n>) will be used as profile name. |
288 |
root |
1.34 |
|
289 |
root |
1.72 |
The profile data is then gathered as follows: |
290 |
root |
1.69 |
|
291 |
elmex |
1.77 |
First, all remaining key => value pairs (all of which are conveniently |
292 |
root |
1.72 |
undocumented at the moment) will be interpreted as configuration |
293 |
|
|
data. Then they will be overwritten by any values specified in the global |
294 |
|
|
default configuration (see the F<aemp> utility), then the chain of |
295 |
|
|
profiles chosen by the profile name (and any C<parent> attributes). |
296 |
|
|
|
297 |
|
|
That means that the values specified in the profile have highest priority |
298 |
|
|
and the values specified directly via C<configure> have lowest priority, |
299 |
|
|
and can only be used to specify defaults. |
300 |
root |
1.49 |
|
301 |
root |
1.64 |
If the profile specifies a node ID, then this will become the node ID of |
302 |
root |
1.122 |
this process. If not, then the profile name will be used as node ID, with |
303 |
root |
1.126 |
a unique randoms tring (C</%u>) appended. |
304 |
root |
1.122 |
|
305 |
root |
1.126 |
The node ID can contain some C<%> sequences that are expanded: C<%n> |
306 |
|
|
is expanded to the local nodename, C<%u> is replaced by a random |
307 |
|
|
strign to make the node unique. For example, the F<aemp> commandline |
308 |
|
|
utility uses C<aemp/%n/%u> as nodename, which might expand to |
309 |
|
|
C<aemp/cerebro/ZQDGSIkRhEZQDGSIkRhE>. |
310 |
root |
1.64 |
|
311 |
root |
1.72 |
=item step 2, bind listener sockets |
312 |
|
|
|
313 |
root |
1.64 |
The next step is to look up the binds in the profile, followed by binding |
314 |
|
|
aemp protocol listeners on all binds specified (it is possible and valid |
315 |
root |
1.148 |
to have no binds, meaning that the node cannot be contacted from the |
316 |
root |
1.64 |
outside. This means the node cannot talk to other nodes that also have no |
317 |
|
|
binds, but it can still talk to all "normal" nodes). |
318 |
|
|
|
319 |
root |
1.70 |
If the profile does not specify a binds list, then a default of C<*> is |
320 |
root |
1.72 |
used, meaning the node will bind on a dynamically-assigned port on every |
321 |
|
|
local IP address it finds. |
322 |
|
|
|
323 |
|
|
=item step 3, connect to seed nodes |
324 |
root |
1.64 |
|
325 |
root |
1.119 |
As the last step, the seed ID list from the profile is passed to the |
326 |
root |
1.64 |
L<AnyEvent::MP::Global> module, which will then use it to keep |
327 |
root |
1.72 |
connectivity with at least one node at any point in time. |
328 |
root |
1.64 |
|
329 |
root |
1.72 |
=back |
330 |
|
|
|
331 |
root |
1.87 |
Example: become a distributed node using the local node name as profile. |
332 |
root |
1.72 |
This should be the most common form of invocation for "daemon"-type nodes. |
333 |
root |
1.34 |
|
334 |
root |
1.72 |
configure |
335 |
root |
1.34 |
|
336 |
root |
1.126 |
Example: become a semi-anonymous node. This form is often used for |
337 |
|
|
commandline clients. |
338 |
root |
1.34 |
|
339 |
root |
1.126 |
configure nodeid => "myscript/%n/%u"; |
340 |
root |
1.72 |
|
341 |
root |
1.120 |
Example: configure a node using a profile called seed, which is suitable |
342 |
root |
1.72 |
for a seed node as it binds on all local addresses on a fixed port (4040, |
343 |
|
|
customary for aemp). |
344 |
|
|
|
345 |
|
|
# use the aemp commandline utility |
346 |
root |
1.122 |
# aemp profile seed binds '*:4040' |
347 |
root |
1.72 |
|
348 |
|
|
# then use it |
349 |
|
|
configure profile => "seed"; |
350 |
root |
1.34 |
|
351 |
root |
1.72 |
# or simply use aemp from the shell again: |
352 |
|
|
# aemp run profile seed |
353 |
root |
1.34 |
|
354 |
root |
1.72 |
# or provide a nicer-to-remember nodeid |
355 |
|
|
# aemp run profile seed nodeid "$(hostname)" |
356 |
root |
1.34 |
|
357 |
root |
1.22 |
=item $SELF |
358 |
|
|
|
359 |
|
|
Contains the current port id while executing C<rcv> callbacks or C<psub> |
360 |
|
|
blocks. |
361 |
root |
1.3 |
|
362 |
root |
1.67 |
=item *SELF, SELF, %SELF, @SELF... |
363 |
root |
1.22 |
|
364 |
|
|
Due to some quirks in how perl exports variables, it is impossible to |
365 |
root |
1.67 |
just export C<$SELF>, all the symbols named C<SELF> are exported by this |
366 |
root |
1.22 |
module, but only C<$SELF> is currently used. |
367 |
root |
1.3 |
|
368 |
root |
1.33 |
=item snd $port, type => @data |
369 |
root |
1.3 |
|
370 |
root |
1.33 |
=item snd $port, @msg |
371 |
root |
1.3 |
|
372 |
root |
1.67 |
Send the given message to the given port, which can identify either a |
373 |
|
|
local or a remote port, and must be a port ID. |
374 |
root |
1.8 |
|
375 |
root |
1.67 |
While the message can be almost anything, it is highly recommended to |
376 |
|
|
use a string as first element (a port ID, or some word that indicates a |
377 |
|
|
request type etc.) and to consist if only simple perl values (scalars, |
378 |
|
|
arrays, hashes) - if you think you need to pass an object, think again. |
379 |
|
|
|
380 |
|
|
The message data logically becomes read-only after a call to this |
381 |
|
|
function: modifying any argument (or values referenced by them) is |
382 |
|
|
forbidden, as there can be considerable time between the call to C<snd> |
383 |
|
|
and the time the message is actually being serialised - in fact, it might |
384 |
|
|
never be copied as within the same process it is simply handed to the |
385 |
|
|
receiving port. |
386 |
root |
1.3 |
|
387 |
|
|
The type of data you can transfer depends on the transport protocol: when |
388 |
|
|
JSON is used, then only strings, numbers and arrays and hashes consisting |
389 |
|
|
of those are allowed (no objects). When Storable is used, then anything |
390 |
|
|
that Storable can serialise and deserialise is allowed, and for the local |
391 |
root |
1.67 |
node, anything can be passed. Best rely only on the common denominator of |
392 |
|
|
these. |
393 |
root |
1.3 |
|
394 |
root |
1.22 |
=item $local_port = port |
395 |
root |
1.2 |
|
396 |
root |
1.50 |
Create a new local port object and returns its port ID. Initially it has |
397 |
|
|
no callbacks set and will throw an error when it receives messages. |
398 |
root |
1.10 |
|
399 |
root |
1.50 |
=item $local_port = port { my @msg = @_ } |
400 |
root |
1.15 |
|
401 |
root |
1.50 |
Creates a new local port, and returns its ID. Semantically the same as |
402 |
|
|
creating a port and calling C<rcv $port, $callback> on it. |
403 |
root |
1.15 |
|
404 |
root |
1.50 |
The block will be called for every message received on the port, with the |
405 |
|
|
global variable C<$SELF> set to the port ID. Runtime errors will cause the |
406 |
|
|
port to be C<kil>ed. The message will be passed as-is, no extra argument |
407 |
|
|
(i.e. no port ID) will be passed to the callback. |
408 |
root |
1.15 |
|
409 |
root |
1.50 |
If you want to stop/destroy the port, simply C<kil> it: |
410 |
root |
1.15 |
|
411 |
root |
1.50 |
my $port = port { |
412 |
|
|
my @msg = @_; |
413 |
|
|
... |
414 |
|
|
kil $SELF; |
415 |
root |
1.15 |
}; |
416 |
root |
1.10 |
|
417 |
|
|
=cut |
418 |
|
|
|
419 |
root |
1.33 |
sub rcv($@); |
420 |
|
|
|
421 |
root |
1.132 |
my $KILME = sub { |
422 |
root |
1.150 |
(my $tag = substr $_[0], 0, 30) =~ s/([^\x20-\x7e])/./g; |
423 |
root |
1.135 |
kil $SELF, unhandled_message => "no callback found for message '$tag'"; |
424 |
root |
1.132 |
}; |
425 |
root |
1.50 |
|
426 |
root |
1.22 |
sub port(;&) { |
427 |
root |
1.123 |
my $id = $UNIQ . ++$ID; |
428 |
root |
1.22 |
my $port = "$NODE#$id"; |
429 |
|
|
|
430 |
root |
1.132 |
rcv $port, shift || $KILME; |
431 |
root |
1.10 |
|
432 |
root |
1.22 |
$port |
433 |
root |
1.10 |
} |
434 |
|
|
|
435 |
root |
1.50 |
=item rcv $local_port, $callback->(@msg) |
436 |
root |
1.31 |
|
437 |
root |
1.50 |
Replaces the default callback on the specified port. There is no way to |
438 |
|
|
remove the default callback: use C<sub { }> to disable it, or better |
439 |
|
|
C<kil> the port when it is no longer needed. |
440 |
root |
1.3 |
|
441 |
root |
1.33 |
The global C<$SELF> (exported by this module) contains C<$port> while |
442 |
root |
1.50 |
executing the callback. Runtime errors during callback execution will |
443 |
|
|
result in the port being C<kil>ed. |
444 |
root |
1.22 |
|
445 |
root |
1.133 |
The default callback receives all messages not matched by a more specific |
446 |
root |
1.50 |
C<tag> match. |
447 |
root |
1.22 |
|
448 |
root |
1.50 |
=item rcv $local_port, tag => $callback->(@msg_without_tag), ... |
449 |
root |
1.3 |
|
450 |
root |
1.54 |
Register (or replace) callbacks to be called on messages starting with the |
451 |
|
|
given tag on the given port (and return the port), or unregister it (when |
452 |
|
|
C<$callback> is C<$undef> or missing). There can only be one callback |
453 |
|
|
registered for each tag. |
454 |
root |
1.3 |
|
455 |
root |
1.50 |
The original message will be passed to the callback, after the first |
456 |
|
|
element (the tag) has been removed. The callback will use the same |
457 |
|
|
environment as the default callback (see above). |
458 |
root |
1.3 |
|
459 |
root |
1.36 |
Example: create a port and bind receivers on it in one go. |
460 |
|
|
|
461 |
|
|
my $port = rcv port, |
462 |
root |
1.50 |
msg1 => sub { ... }, |
463 |
|
|
msg2 => sub { ... }, |
464 |
root |
1.36 |
; |
465 |
|
|
|
466 |
|
|
Example: create a port, bind receivers and send it in a message elsewhere |
467 |
|
|
in one go: |
468 |
|
|
|
469 |
|
|
snd $otherport, reply => |
470 |
|
|
rcv port, |
471 |
root |
1.50 |
msg1 => sub { ... }, |
472 |
root |
1.36 |
... |
473 |
|
|
; |
474 |
|
|
|
475 |
root |
1.54 |
Example: temporarily register a rcv callback for a tag matching some port |
476 |
root |
1.102 |
(e.g. for an rpc reply) and unregister it after a message was received. |
477 |
root |
1.54 |
|
478 |
|
|
rcv $port, $otherport => sub { |
479 |
|
|
my @reply = @_; |
480 |
|
|
|
481 |
|
|
rcv $SELF, $otherport; |
482 |
|
|
}; |
483 |
|
|
|
484 |
root |
1.3 |
=cut |
485 |
|
|
|
486 |
|
|
sub rcv($@) { |
487 |
root |
1.33 |
my $port = shift; |
488 |
root |
1.75 |
my ($nodeid, $portid) = split /#/, $port, 2; |
489 |
root |
1.3 |
|
490 |
root |
1.144 |
$nodeid eq $NODE |
491 |
root |
1.33 |
or Carp::croak "$port: rcv can only be called on local ports, caught"; |
492 |
root |
1.22 |
|
493 |
root |
1.50 |
while (@_) { |
494 |
|
|
if (ref $_[0]) { |
495 |
|
|
if (my $self = $PORT_DATA{$portid}) { |
496 |
|
|
"AnyEvent::MP::Port" eq ref $self |
497 |
|
|
or Carp::croak "$port: rcv can only be called on message matching ports, caught"; |
498 |
root |
1.33 |
|
499 |
root |
1.103 |
$self->[0] = shift; |
500 |
root |
1.50 |
} else { |
501 |
|
|
my $cb = shift; |
502 |
|
|
$PORT{$portid} = sub { |
503 |
|
|
local $SELF = $port; |
504 |
|
|
eval { &$cb }; _self_die if $@; |
505 |
|
|
}; |
506 |
|
|
} |
507 |
|
|
} elsif (defined $_[0]) { |
508 |
|
|
my $self = $PORT_DATA{$portid} ||= do { |
509 |
root |
1.103 |
my $self = bless [$PORT{$portid} || sub { }, { }, $port], "AnyEvent::MP::Port"; |
510 |
root |
1.50 |
|
511 |
|
|
$PORT{$portid} = sub { |
512 |
|
|
local $SELF = $port; |
513 |
|
|
|
514 |
|
|
if (my $cb = $self->[1]{$_[0]}) { |
515 |
|
|
shift; |
516 |
|
|
eval { &$cb }; _self_die if $@; |
517 |
|
|
} else { |
518 |
|
|
&{ $self->[0] }; |
519 |
root |
1.33 |
} |
520 |
|
|
}; |
521 |
root |
1.50 |
|
522 |
|
|
$self |
523 |
root |
1.33 |
}; |
524 |
|
|
|
525 |
root |
1.50 |
"AnyEvent::MP::Port" eq ref $self |
526 |
|
|
or Carp::croak "$port: rcv can only be called on message matching ports, caught"; |
527 |
root |
1.22 |
|
528 |
root |
1.50 |
my ($tag, $cb) = splice @_, 0, 2; |
529 |
root |
1.33 |
|
530 |
root |
1.50 |
if (defined $cb) { |
531 |
|
|
$self->[1]{$tag} = $cb; |
532 |
root |
1.33 |
} else { |
533 |
root |
1.50 |
delete $self->[1]{$tag}; |
534 |
root |
1.33 |
} |
535 |
root |
1.22 |
} |
536 |
root |
1.3 |
} |
537 |
root |
1.31 |
|
538 |
root |
1.33 |
$port |
539 |
root |
1.2 |
} |
540 |
|
|
|
541 |
root |
1.101 |
=item peval $port, $coderef[, @args] |
542 |
|
|
|
543 |
root |
1.151 |
Evaluates the given C<$codref> within the context of C<$port>, that is, |
544 |
root |
1.145 |
when the code throws an exception the C<$port> will be killed. |
545 |
root |
1.101 |
|
546 |
|
|
Any remaining args will be passed to the callback. Any return values will |
547 |
|
|
be returned to the caller. |
548 |
|
|
|
549 |
|
|
This is useful when you temporarily want to execute code in the context of |
550 |
|
|
a port. |
551 |
|
|
|
552 |
|
|
Example: create a port and run some initialisation code in it's context. |
553 |
|
|
|
554 |
|
|
my $port = port { ... }; |
555 |
|
|
|
556 |
|
|
peval $port, sub { |
557 |
|
|
init |
558 |
|
|
or die "unable to init"; |
559 |
|
|
}; |
560 |
|
|
|
561 |
|
|
=cut |
562 |
|
|
|
563 |
|
|
sub peval($$) { |
564 |
|
|
local $SELF = shift; |
565 |
|
|
my $cb = shift; |
566 |
|
|
|
567 |
|
|
if (wantarray) { |
568 |
|
|
my @res = eval { &$cb }; |
569 |
|
|
_self_die if $@; |
570 |
|
|
@res |
571 |
|
|
} else { |
572 |
|
|
my $res = eval { &$cb }; |
573 |
|
|
_self_die if $@; |
574 |
|
|
$res |
575 |
|
|
} |
576 |
|
|
} |
577 |
|
|
|
578 |
root |
1.22 |
=item $closure = psub { BLOCK } |
579 |
root |
1.2 |
|
580 |
root |
1.22 |
Remembers C<$SELF> and creates a closure out of the BLOCK. When the |
581 |
|
|
closure is executed, sets up the environment in the same way as in C<rcv> |
582 |
|
|
callbacks, i.e. runtime errors will cause the port to get C<kil>ed. |
583 |
|
|
|
584 |
root |
1.101 |
The effect is basically as if it returned C<< sub { peval $SELF, sub { |
585 |
root |
1.114 |
BLOCK }, @_ } >>. |
586 |
root |
1.101 |
|
587 |
root |
1.22 |
This is useful when you register callbacks from C<rcv> callbacks: |
588 |
|
|
|
589 |
|
|
rcv delayed_reply => sub { |
590 |
|
|
my ($delay, @reply) = @_; |
591 |
|
|
my $timer = AE::timer $delay, 0, psub { |
592 |
|
|
snd @reply, $SELF; |
593 |
|
|
}; |
594 |
|
|
}; |
595 |
root |
1.3 |
|
596 |
root |
1.8 |
=cut |
597 |
root |
1.3 |
|
598 |
root |
1.22 |
sub psub(&) { |
599 |
|
|
my $cb = shift; |
600 |
root |
1.3 |
|
601 |
root |
1.22 |
my $port = $SELF |
602 |
|
|
or Carp::croak "psub can only be called from within rcv or psub callbacks, not"; |
603 |
root |
1.1 |
|
604 |
root |
1.22 |
sub { |
605 |
|
|
local $SELF = $port; |
606 |
root |
1.2 |
|
607 |
root |
1.22 |
if (wantarray) { |
608 |
|
|
my @res = eval { &$cb }; |
609 |
|
|
_self_die if $@; |
610 |
|
|
@res |
611 |
|
|
} else { |
612 |
|
|
my $res = eval { &$cb }; |
613 |
|
|
_self_die if $@; |
614 |
|
|
$res |
615 |
|
|
} |
616 |
|
|
} |
617 |
root |
1.2 |
} |
618 |
|
|
|
619 |
root |
1.67 |
=item $guard = mon $port, $rcvport # kill $rcvport when $port dies |
620 |
root |
1.36 |
|
621 |
root |
1.67 |
=item $guard = mon $port # kill $SELF when $port dies |
622 |
root |
1.32 |
|
623 |
root |
1.139 |
=item $guard = mon $port, $cb->(@reason) # call $cb when $port dies |
624 |
|
|
|
625 |
root |
1.67 |
=item $guard = mon $port, $rcvport, @msg # send a message when $port dies |
626 |
root |
1.32 |
|
627 |
root |
1.42 |
Monitor the given port and do something when the port is killed or |
628 |
|
|
messages to it were lost, and optionally return a guard that can be used |
629 |
|
|
to stop monitoring again. |
630 |
|
|
|
631 |
root |
1.139 |
The first two forms distinguish between "normal" and "abnormal" kil's: |
632 |
root |
1.32 |
|
633 |
root |
1.139 |
In the first form (another port given), if the C<$port> is C<kil>'ed with |
634 |
|
|
a non-empty reason, the other port (C<$rcvport>) will be kil'ed with the |
635 |
|
|
same reason. That is, on "normal" kil's nothing happens, while under all |
636 |
|
|
other conditions, the other port is killed with the same reason. |
637 |
root |
1.32 |
|
638 |
root |
1.139 |
The second form (kill self) is the same as the first form, except that |
639 |
root |
1.36 |
C<$rvport> defaults to C<$SELF>. |
640 |
|
|
|
641 |
root |
1.139 |
The remaining forms don't distinguish between "normal" and "abnormal" kil's |
642 |
|
|
- it's up to the callback or receiver to check whether the C<@reason> is |
643 |
|
|
empty and act accordingly. |
644 |
|
|
|
645 |
|
|
In the third form (callback), the callback is simply called with any |
646 |
|
|
number of C<@reason> elements (empty @reason means that the port was deleted |
647 |
|
|
"normally"). Note also that I<< the callback B<must> never die >>, so use |
648 |
|
|
C<eval> if unsure. |
649 |
|
|
|
650 |
|
|
In the last form (message), a message of the form C<$rcvport, @msg, |
651 |
|
|
@reason> will be C<snd>. |
652 |
root |
1.32 |
|
653 |
root |
1.79 |
Monitoring-actions are one-shot: once messages are lost (and a monitoring |
654 |
root |
1.139 |
alert was raised), they are removed and will not trigger again, even if it |
655 |
|
|
turns out that the port is still alive. |
656 |
root |
1.79 |
|
657 |
root |
1.139 |
As a rule of thumb, monitoring requests should always monitor a remote |
658 |
|
|
port locally (using a local C<$rcvport> or a callback). The reason is that |
659 |
|
|
kill messages might get lost, just like any other message. Another less |
660 |
|
|
obvious reason is that even monitoring requests can get lost (for example, |
661 |
|
|
when the connection to the other node goes down permanently). When |
662 |
|
|
monitoring a port locally these problems do not exist. |
663 |
root |
1.37 |
|
664 |
root |
1.79 |
C<mon> effectively guarantees that, in the absence of hardware failures, |
665 |
|
|
after starting the monitor, either all messages sent to the port will |
666 |
|
|
arrive, or the monitoring action will be invoked after possible message |
667 |
|
|
loss has been detected. No messages will be lost "in between" (after |
668 |
|
|
the first lost message no further messages will be received by the |
669 |
|
|
port). After the monitoring action was invoked, further messages might get |
670 |
|
|
delivered again. |
671 |
|
|
|
672 |
|
|
Inter-host-connection timeouts and monitoring depend on the transport |
673 |
|
|
used. The only transport currently implemented is TCP, and AnyEvent::MP |
674 |
|
|
relies on TCP to detect node-downs (this can take 10-15 minutes on a |
675 |
elmex |
1.96 |
non-idle connection, and usually around two hours for idle connections). |
676 |
root |
1.79 |
|
677 |
|
|
This means that monitoring is good for program errors and cleaning up |
678 |
|
|
stuff eventually, but they are no replacement for a timeout when you need |
679 |
|
|
to ensure some maximum latency. |
680 |
|
|
|
681 |
root |
1.32 |
Example: call a given callback when C<$port> is killed. |
682 |
|
|
|
683 |
|
|
mon $port, sub { warn "port died because of <@_>\n" }; |
684 |
|
|
|
685 |
|
|
Example: kill ourselves when C<$port> is killed abnormally. |
686 |
|
|
|
687 |
root |
1.36 |
mon $port; |
688 |
root |
1.32 |
|
689 |
root |
1.36 |
Example: send us a restart message when another C<$port> is killed. |
690 |
root |
1.32 |
|
691 |
|
|
mon $port, $self => "restart"; |
692 |
|
|
|
693 |
|
|
=cut |
694 |
|
|
|
695 |
|
|
sub mon { |
696 |
root |
1.75 |
my ($nodeid, $port) = split /#/, shift, 2; |
697 |
root |
1.32 |
|
698 |
root |
1.75 |
my $node = $NODE{$nodeid} || add_node $nodeid; |
699 |
root |
1.32 |
|
700 |
root |
1.41 |
my $cb = @_ ? shift : $SELF || Carp::croak 'mon: called with one argument only, but $SELF not set,'; |
701 |
root |
1.32 |
|
702 |
|
|
unless (ref $cb) { |
703 |
|
|
if (@_) { |
704 |
|
|
# send a kill info message |
705 |
root |
1.41 |
my (@msg) = ($cb, @_); |
706 |
root |
1.32 |
$cb = sub { snd @msg, @_ }; |
707 |
|
|
} else { |
708 |
|
|
# simply kill other port |
709 |
|
|
my $port = $cb; |
710 |
|
|
$cb = sub { kil $port, @_ if @_ }; |
711 |
|
|
} |
712 |
|
|
} |
713 |
|
|
|
714 |
|
|
$node->monitor ($port, $cb); |
715 |
|
|
|
716 |
|
|
defined wantarray |
717 |
root |
1.124 |
and ($cb += 0, Guard::guard { $node->unmonitor ($port, $cb) }) |
718 |
root |
1.32 |
} |
719 |
|
|
|
720 |
|
|
=item $guard = mon_guard $port, $ref, $ref... |
721 |
|
|
|
722 |
|
|
Monitors the given C<$port> and keeps the passed references. When the port |
723 |
|
|
is killed, the references will be freed. |
724 |
|
|
|
725 |
|
|
Optionally returns a guard that will stop the monitoring. |
726 |
|
|
|
727 |
|
|
This function is useful when you create e.g. timers or other watchers and |
728 |
root |
1.67 |
want to free them when the port gets killed (note the use of C<psub>): |
729 |
root |
1.32 |
|
730 |
|
|
$port->rcv (start => sub { |
731 |
root |
1.67 |
my $timer; $timer = mon_guard $port, AE::timer 1, 1, psub { |
732 |
root |
1.32 |
undef $timer if 0.9 < rand; |
733 |
|
|
}); |
734 |
|
|
}); |
735 |
|
|
|
736 |
|
|
=cut |
737 |
|
|
|
738 |
|
|
sub mon_guard { |
739 |
|
|
my ($port, @refs) = @_; |
740 |
|
|
|
741 |
root |
1.36 |
#TODO: mon-less form? |
742 |
|
|
|
743 |
root |
1.32 |
mon $port, sub { 0 && @refs } |
744 |
|
|
} |
745 |
|
|
|
746 |
root |
1.33 |
=item kil $port[, @reason] |
747 |
root |
1.32 |
|
748 |
|
|
Kill the specified port with the given C<@reason>. |
749 |
|
|
|
750 |
root |
1.107 |
If no C<@reason> is specified, then the port is killed "normally" - |
751 |
|
|
monitor callback will be invoked, but the kil will not cause linked ports |
752 |
|
|
(C<mon $mport, $lport> form) to get killed. |
753 |
root |
1.32 |
|
754 |
root |
1.107 |
If a C<@reason> is specified, then linked ports (C<mon $mport, $lport> |
755 |
|
|
form) get killed with the same reason. |
756 |
root |
1.32 |
|
757 |
|
|
Runtime errors while evaluating C<rcv> callbacks or inside C<psub> blocks |
758 |
|
|
will be reported as reason C<< die => $@ >>. |
759 |
|
|
|
760 |
|
|
Transport/communication errors are reported as C<< transport_error => |
761 |
|
|
$message >>. |
762 |
|
|
|
763 |
root |
1.133 |
Common idioms: |
764 |
|
|
|
765 |
|
|
# silently remove yourself, do not kill linked ports |
766 |
|
|
kil $SELF; |
767 |
|
|
|
768 |
|
|
# report a failure in some detail |
769 |
|
|
kil $SELF, failure_mode_1 => "it failed with too high temperature"; |
770 |
|
|
|
771 |
|
|
# do not waste much time with killing, just die when something goes wrong |
772 |
|
|
open my $fh, "<file" |
773 |
|
|
or die "file: $!"; |
774 |
root |
1.38 |
|
775 |
|
|
=item $port = spawn $node, $initfunc[, @initdata] |
776 |
|
|
|
777 |
|
|
Creates a port on the node C<$node> (which can also be a port ID, in which |
778 |
|
|
case it's the node where that port resides). |
779 |
|
|
|
780 |
root |
1.67 |
The port ID of the newly created port is returned immediately, and it is |
781 |
|
|
possible to immediately start sending messages or to monitor the port. |
782 |
root |
1.38 |
|
783 |
root |
1.67 |
After the port has been created, the init function is called on the remote |
784 |
|
|
node, in the same context as a C<rcv> callback. This function must be a |
785 |
|
|
fully-qualified function name (e.g. C<MyApp::Chat::Server::init>). To |
786 |
|
|
specify a function in the main program, use C<::name>. |
787 |
root |
1.38 |
|
788 |
|
|
If the function doesn't exist, then the node tries to C<require> |
789 |
|
|
the package, then the package above the package and so on (e.g. |
790 |
|
|
C<MyApp::Chat::Server>, C<MyApp::Chat>, C<MyApp>) until the function |
791 |
|
|
exists or it runs out of package names. |
792 |
|
|
|
793 |
|
|
The init function is then called with the newly-created port as context |
794 |
root |
1.82 |
object (C<$SELF>) and the C<@initdata> values as arguments. It I<must> |
795 |
|
|
call one of the C<rcv> functions to set callbacks on C<$SELF>, otherwise |
796 |
|
|
the port might not get created. |
797 |
root |
1.38 |
|
798 |
root |
1.67 |
A common idiom is to pass a local port, immediately monitor the spawned |
799 |
|
|
port, and in the remote init function, immediately monitor the passed |
800 |
|
|
local port. This two-way monitoring ensures that both ports get cleaned up |
801 |
|
|
when there is a problem. |
802 |
root |
1.38 |
|
803 |
root |
1.80 |
C<spawn> guarantees that the C<$initfunc> has no visible effects on the |
804 |
|
|
caller before C<spawn> returns (by delaying invocation when spawn is |
805 |
|
|
called for the local node). |
806 |
|
|
|
807 |
root |
1.38 |
Example: spawn a chat server port on C<$othernode>. |
808 |
|
|
|
809 |
|
|
# this node, executed from within a port context: |
810 |
|
|
my $server = spawn $othernode, "MyApp::Chat::Server::connect", $SELF; |
811 |
|
|
mon $server; |
812 |
|
|
|
813 |
|
|
# init function on C<$othernode> |
814 |
|
|
sub connect { |
815 |
|
|
my ($srcport) = @_; |
816 |
|
|
|
817 |
|
|
mon $srcport; |
818 |
|
|
|
819 |
|
|
rcv $SELF, sub { |
820 |
|
|
... |
821 |
|
|
}; |
822 |
|
|
} |
823 |
|
|
|
824 |
|
|
=cut |
825 |
|
|
|
826 |
|
|
sub _spawn { |
827 |
|
|
my $port = shift; |
828 |
|
|
my $init = shift; |
829 |
|
|
|
830 |
root |
1.82 |
# rcv will create the actual port |
831 |
root |
1.38 |
local $SELF = "$NODE#$port"; |
832 |
|
|
eval { |
833 |
|
|
&{ load_func $init } |
834 |
|
|
}; |
835 |
|
|
_self_die if $@; |
836 |
|
|
} |
837 |
|
|
|
838 |
|
|
sub spawn(@) { |
839 |
root |
1.75 |
my ($nodeid, undef) = split /#/, shift, 2; |
840 |
root |
1.38 |
|
841 |
root |
1.123 |
my $id = $RUNIQ . ++$ID; |
842 |
root |
1.38 |
|
843 |
root |
1.39 |
$_[0] =~ /::/ |
844 |
|
|
or Carp::croak "spawn init function must be a fully-qualified name, caught"; |
845 |
|
|
|
846 |
root |
1.75 |
snd_to_func $nodeid, "AnyEvent::MP::_spawn" => $id, @_; |
847 |
root |
1.38 |
|
848 |
root |
1.75 |
"$nodeid#$id" |
849 |
root |
1.38 |
} |
850 |
|
|
|
851 |
root |
1.121 |
|
852 |
root |
1.59 |
=item after $timeout, @msg |
853 |
|
|
|
854 |
|
|
=item after $timeout, $callback |
855 |
|
|
|
856 |
|
|
Either sends the given message, or call the given callback, after the |
857 |
|
|
specified number of seconds. |
858 |
|
|
|
859 |
root |
1.67 |
This is simply a utility function that comes in handy at times - the |
860 |
|
|
AnyEvent::MP author is not convinced of the wisdom of having it, though, |
861 |
|
|
so it may go away in the future. |
862 |
root |
1.59 |
|
863 |
|
|
=cut |
864 |
|
|
|
865 |
|
|
sub after($@) { |
866 |
|
|
my ($timeout, @action) = @_; |
867 |
|
|
|
868 |
|
|
my $t; $t = AE::timer $timeout, 0, sub { |
869 |
|
|
undef $t; |
870 |
|
|
ref $action[0] |
871 |
|
|
? $action[0]() |
872 |
|
|
: snd @action; |
873 |
|
|
}; |
874 |
|
|
} |
875 |
|
|
|
876 |
root |
1.129 |
#=item $cb2 = timeout $seconds, $cb[, @args] |
877 |
|
|
|
878 |
root |
1.87 |
=item cal $port, @msg, $callback[, $timeout] |
879 |
|
|
|
880 |
|
|
A simple form of RPC - sends a message to the given C<$port> with the |
881 |
|
|
given contents (C<@msg>), but adds a reply port to the message. |
882 |
|
|
|
883 |
|
|
The reply port is created temporarily just for the purpose of receiving |
884 |
|
|
the reply, and will be C<kil>ed when no longer needed. |
885 |
|
|
|
886 |
|
|
A reply message sent to the port is passed to the C<$callback> as-is. |
887 |
|
|
|
888 |
|
|
If an optional time-out (in seconds) is given and it is not C<undef>, |
889 |
|
|
then the callback will be called without any arguments after the time-out |
890 |
|
|
elapsed and the port is C<kil>ed. |
891 |
|
|
|
892 |
root |
1.98 |
If no time-out is given (or it is C<undef>), then the local port will |
893 |
|
|
monitor the remote port instead, so it eventually gets cleaned-up. |
894 |
root |
1.87 |
|
895 |
|
|
Currently this function returns the temporary port, but this "feature" |
896 |
|
|
might go in future versions unless you can make a convincing case that |
897 |
|
|
this is indeed useful for something. |
898 |
|
|
|
899 |
|
|
=cut |
900 |
|
|
|
901 |
|
|
sub cal(@) { |
902 |
|
|
my $timeout = ref $_[-1] ? undef : pop; |
903 |
|
|
my $cb = pop; |
904 |
|
|
|
905 |
|
|
my $port = port { |
906 |
|
|
undef $timeout; |
907 |
|
|
kil $SELF; |
908 |
|
|
&$cb; |
909 |
|
|
}; |
910 |
|
|
|
911 |
|
|
if (defined $timeout) { |
912 |
|
|
$timeout = AE::timer $timeout, 0, sub { |
913 |
|
|
undef $timeout; |
914 |
|
|
kil $port; |
915 |
|
|
$cb->(); |
916 |
|
|
}; |
917 |
|
|
} else { |
918 |
|
|
mon $_[0], sub { |
919 |
|
|
kil $port; |
920 |
|
|
$cb->(); |
921 |
|
|
}; |
922 |
|
|
} |
923 |
|
|
|
924 |
|
|
push @_, $port; |
925 |
|
|
&snd; |
926 |
|
|
|
927 |
|
|
$port |
928 |
|
|
} |
929 |
|
|
|
930 |
root |
1.8 |
=back |
931 |
|
|
|
932 |
root |
1.124 |
=head1 DISTRIBUTED DATABASE |
933 |
|
|
|
934 |
|
|
AnyEvent::MP comes with a simple distributed database. The database will |
935 |
root |
1.131 |
be mirrored asynchronously on all global nodes. Other nodes bind to one |
936 |
|
|
of the global nodes for their needs. Every node has a "local database" |
937 |
|
|
which contains all the values that are set locally. All local databases |
938 |
|
|
are merged together to form the global database, which can be queried. |
939 |
|
|
|
940 |
|
|
The database structure is that of a two-level hash - the database hash |
941 |
|
|
contains hashes which contain values, similarly to a perl hash of hashes, |
942 |
|
|
i.e.: |
943 |
root |
1.124 |
|
944 |
root |
1.131 |
$DATABASE{$family}{$subkey} = $value |
945 |
root |
1.124 |
|
946 |
|
|
The top level hash key is called "family", and the second-level hash key |
947 |
root |
1.126 |
is called "subkey" or simply "key". |
948 |
root |
1.124 |
|
949 |
root |
1.125 |
The family must be alphanumeric, i.e. start with a letter and consist |
950 |
|
|
of letters, digits, underscores and colons (C<[A-Za-z][A-Za-z0-9_:]*>, |
951 |
|
|
pretty much like Perl module names. |
952 |
root |
1.124 |
|
953 |
root |
1.125 |
As the family namespace is global, it is recommended to prefix family names |
954 |
root |
1.124 |
with the name of the application or module using it. |
955 |
|
|
|
956 |
root |
1.126 |
The subkeys must be non-empty strings, with no further restrictions. |
957 |
root |
1.125 |
|
958 |
root |
1.124 |
The values should preferably be strings, but other perl scalars should |
959 |
root |
1.131 |
work as well (such as C<undef>, arrays and hashes). |
960 |
root |
1.124 |
|
961 |
root |
1.126 |
Every database entry is owned by one node - adding the same family/subkey |
962 |
root |
1.124 |
combination on multiple nodes will not cause discomfort for AnyEvent::MP, |
963 |
|
|
but the result might be nondeterministic, i.e. the key might have |
964 |
|
|
different values on different nodes. |
965 |
|
|
|
966 |
root |
1.126 |
Different subkeys in the same family can be owned by different nodes |
967 |
|
|
without problems, and in fact, this is the common method to create worker |
968 |
|
|
pools. For example, a worker port for image scaling might do this: |
969 |
root |
1.124 |
|
970 |
root |
1.126 |
db_set my_image_scalers => $port; |
971 |
root |
1.124 |
|
972 |
root |
1.126 |
And clients looking for an image scaler will want to get the |
973 |
root |
1.129 |
C<my_image_scalers> keys from time to time: |
974 |
|
|
|
975 |
|
|
db_keys my_image_scalers => sub { |
976 |
|
|
@ports = @{ $_[0] }; |
977 |
|
|
}; |
978 |
|
|
|
979 |
|
|
Or better yet, they want to monitor the database family, so they always |
980 |
|
|
have a reasonable up-to-date copy: |
981 |
|
|
|
982 |
|
|
db_mon my_image_scalers => sub { |
983 |
|
|
@ports = keys %{ $_[0] }; |
984 |
|
|
}; |
985 |
|
|
|
986 |
|
|
In general, you can set or delete single subkeys, but query and monitor |
987 |
|
|
whole families only. |
988 |
root |
1.126 |
|
989 |
root |
1.129 |
If you feel the need to monitor or query a single subkey, try giving it |
990 |
|
|
it's own family. |
991 |
root |
1.126 |
|
992 |
|
|
=over |
993 |
|
|
|
994 |
root |
1.137 |
=item $guard = db_set $family => $subkey [=> $value] |
995 |
root |
1.126 |
|
996 |
|
|
Sets (or replaces) a key to the database - if C<$value> is omitted, |
997 |
|
|
C<undef> is used instead. |
998 |
|
|
|
999 |
root |
1.137 |
When called in non-void context, C<db_set> returns a guard that |
1000 |
|
|
automatically calls C<db_del> when it is destroyed. |
1001 |
|
|
|
1002 |
root |
1.130 |
=item db_del $family => $subkey... |
1003 |
root |
1.124 |
|
1004 |
root |
1.130 |
Deletes one or more subkeys from the database family. |
1005 |
root |
1.124 |
|
1006 |
root |
1.137 |
=item $guard = db_reg $family => $port => $value |
1007 |
|
|
|
1008 |
|
|
=item $guard = db_reg $family => $port |
1009 |
|
|
|
1010 |
|
|
=item $guard = db_reg $family |
1011 |
|
|
|
1012 |
|
|
Registers a port in the given family and optionally returns a guard to |
1013 |
|
|
remove it. |
1014 |
|
|
|
1015 |
|
|
This function basically does the same as: |
1016 |
root |
1.124 |
|
1017 |
root |
1.137 |
db_set $family => $port => $value |
1018 |
|
|
|
1019 |
|
|
Except that the port is monitored and automatically removed from the |
1020 |
|
|
database family when it is kil'ed. |
1021 |
|
|
|
1022 |
|
|
If C<$value> is missing, C<undef> is used. If C<$port> is missing, then |
1023 |
|
|
C<$SELF> is used. |
1024 |
|
|
|
1025 |
|
|
This function is most useful to register a port in some port group (which |
1026 |
|
|
is just another name for a database family), and have it removed when the |
1027 |
|
|
port is gone. This works best when the port is a local port. |
1028 |
|
|
|
1029 |
|
|
=cut |
1030 |
|
|
|
1031 |
|
|
sub db_reg($$;$) { |
1032 |
|
|
my $family = shift; |
1033 |
|
|
my $port = @_ ? shift : $SELF; |
1034 |
|
|
|
1035 |
|
|
my $clr = sub { db_del $family => $port }; |
1036 |
|
|
mon $port, $clr; |
1037 |
|
|
|
1038 |
|
|
db_set $family => $port => $_[0]; |
1039 |
|
|
|
1040 |
|
|
defined wantarray |
1041 |
|
|
and &Guard::guard ($clr) |
1042 |
|
|
} |
1043 |
root |
1.124 |
|
1044 |
root |
1.129 |
=item db_family $family => $cb->(\%familyhash) |
1045 |
|
|
|
1046 |
|
|
Queries the named database C<$family> and call the callback with the |
1047 |
|
|
family represented as a hash. You can keep and freely modify the hash. |
1048 |
|
|
|
1049 |
|
|
=item db_keys $family => $cb->(\@keys) |
1050 |
|
|
|
1051 |
|
|
Same as C<db_family>, except it only queries the family I<subkeys> and passes |
1052 |
|
|
them as array reference to the callback. |
1053 |
|
|
|
1054 |
|
|
=item db_values $family => $cb->(\@values) |
1055 |
|
|
|
1056 |
|
|
Same as C<db_family>, except it only queries the family I<values> and passes them |
1057 |
|
|
as array reference to the callback. |
1058 |
|
|
|
1059 |
root |
1.143 |
=item $guard = db_mon $family => $cb->(\%familyhash, \@added, \@changed, \@deleted) |
1060 |
root |
1.128 |
|
1061 |
root |
1.146 |
Creates a monitor on the given database family. Each time a key is |
1062 |
|
|
set or is deleted the callback is called with a hash containing the |
1063 |
root |
1.130 |
database family and three lists of added, changed and deleted subkeys, |
1064 |
|
|
respectively. If no keys have changed then the array reference might be |
1065 |
|
|
C<undef> or even missing. |
1066 |
|
|
|
1067 |
root |
1.132 |
If not called in void context, a guard object is returned that, when |
1068 |
|
|
destroyed, stops the monitor. |
1069 |
|
|
|
1070 |
root |
1.130 |
The family hash reference and the key arrays belong to AnyEvent::MP and |
1071 |
|
|
B<must not be modified or stored> by the callback. When in doubt, make a |
1072 |
|
|
copy. |
1073 |
|
|
|
1074 |
|
|
As soon as possible after the monitoring starts, the callback will be |
1075 |
|
|
called with the intiial contents of the family, even if it is empty, |
1076 |
|
|
i.e. there will always be a timely call to the callback with the current |
1077 |
|
|
contents. |
1078 |
root |
1.128 |
|
1079 |
|
|
It is possible that the callback is called with a change event even though |
1080 |
|
|
the subkey is already present and the value has not changed. |
1081 |
|
|
|
1082 |
|
|
The monitoring stops when the guard object is destroyed. |
1083 |
|
|
|
1084 |
|
|
Example: on every change to the family "mygroup", print out all keys. |
1085 |
|
|
|
1086 |
|
|
my $guard = db_mon mygroup => sub { |
1087 |
root |
1.130 |
my ($family, $a, $c, $d) = @_; |
1088 |
root |
1.128 |
print "mygroup members: ", (join " ", keys %$family), "\n"; |
1089 |
|
|
}; |
1090 |
|
|
|
1091 |
|
|
Exmaple: wait until the family "My::Module::workers" is non-empty. |
1092 |
|
|
|
1093 |
|
|
my $guard; $guard = db_mon My::Module::workers => sub { |
1094 |
root |
1.130 |
my ($family, $a, $c, $d) = @_; |
1095 |
root |
1.128 |
return unless %$family; |
1096 |
|
|
undef $guard; |
1097 |
|
|
print "My::Module::workers now nonempty\n"; |
1098 |
|
|
}; |
1099 |
|
|
|
1100 |
root |
1.146 |
Example: print all changes to the family "AnyEvent::Fantasy::Module". |
1101 |
root |
1.128 |
|
1102 |
root |
1.146 |
my $guard = db_mon AnyEvent::Fantasy::Module => sub { |
1103 |
root |
1.130 |
my ($family, $a, $c, $d) = @_; |
1104 |
root |
1.128 |
|
1105 |
root |
1.130 |
print "+$_=$family->{$_}\n" for @$a; |
1106 |
|
|
print "*$_=$family->{$_}\n" for @$c; |
1107 |
|
|
print "-$_=$family->{$_}\n" for @$d; |
1108 |
root |
1.128 |
}; |
1109 |
|
|
|
1110 |
root |
1.124 |
=cut |
1111 |
|
|
|
1112 |
|
|
=back |
1113 |
|
|
|
1114 |
root |
1.26 |
=head1 AnyEvent::MP vs. Distributed Erlang |
1115 |
|
|
|
1116 |
root |
1.35 |
AnyEvent::MP got lots of its ideas from distributed Erlang (Erlang node |
1117 |
|
|
== aemp node, Erlang process == aemp port), so many of the documents and |
1118 |
|
|
programming techniques employed by Erlang apply to AnyEvent::MP. Here is a |
1119 |
root |
1.27 |
sample: |
1120 |
|
|
|
1121 |
root |
1.95 |
http://www.erlang.se/doc/programming_rules.shtml |
1122 |
|
|
http://erlang.org/doc/getting_started/part_frame.html # chapters 3 and 4 |
1123 |
|
|
http://erlang.org/download/erlang-book-part1.pdf # chapters 5 and 6 |
1124 |
|
|
http://erlang.org/download/armstrong_thesis_2003.pdf # chapters 4 and 5 |
1125 |
root |
1.27 |
|
1126 |
|
|
Despite the similarities, there are also some important differences: |
1127 |
root |
1.26 |
|
1128 |
|
|
=over 4 |
1129 |
|
|
|
1130 |
root |
1.65 |
=item * Node IDs are arbitrary strings in AEMP. |
1131 |
root |
1.26 |
|
1132 |
root |
1.65 |
Erlang relies on special naming and DNS to work everywhere in the same |
1133 |
|
|
way. AEMP relies on each node somehow knowing its own address(es) (e.g. by |
1134 |
root |
1.99 |
configuration or DNS), and possibly the addresses of some seed nodes, but |
1135 |
|
|
will otherwise discover other nodes (and their IDs) itself. |
1136 |
root |
1.27 |
|
1137 |
root |
1.54 |
=item * Erlang has a "remote ports are like local ports" philosophy, AEMP |
1138 |
root |
1.51 |
uses "local ports are like remote ports". |
1139 |
|
|
|
1140 |
|
|
The failure modes for local ports are quite different (runtime errors |
1141 |
|
|
only) then for remote ports - when a local port dies, you I<know> it dies, |
1142 |
|
|
when a connection to another node dies, you know nothing about the other |
1143 |
|
|
port. |
1144 |
|
|
|
1145 |
|
|
Erlang pretends remote ports are as reliable as local ports, even when |
1146 |
|
|
they are not. |
1147 |
|
|
|
1148 |
|
|
AEMP encourages a "treat remote ports differently" philosophy, with local |
1149 |
|
|
ports being the special case/exception, where transport errors cannot |
1150 |
|
|
occur. |
1151 |
|
|
|
1152 |
root |
1.26 |
=item * Erlang uses processes and a mailbox, AEMP does not queue. |
1153 |
|
|
|
1154 |
root |
1.119 |
Erlang uses processes that selectively receive messages out of order, and |
1155 |
|
|
therefore needs a queue. AEMP is event based, queuing messages would serve |
1156 |
|
|
no useful purpose. For the same reason the pattern-matching abilities |
1157 |
|
|
of AnyEvent::MP are more limited, as there is little need to be able to |
1158 |
elmex |
1.77 |
filter messages without dequeuing them. |
1159 |
root |
1.26 |
|
1160 |
root |
1.119 |
This is not a philosophical difference, but simply stems from AnyEvent::MP |
1161 |
|
|
being event-based, while Erlang is process-based. |
1162 |
|
|
|
1163 |
root |
1.146 |
You can have a look at L<Coro::MP> for a more Erlang-like process model on |
1164 |
root |
1.119 |
top of AEMP and Coro threads. |
1165 |
root |
1.26 |
|
1166 |
|
|
=item * Erlang sends are synchronous, AEMP sends are asynchronous. |
1167 |
|
|
|
1168 |
root |
1.119 |
Sending messages in Erlang is synchronous and blocks the process until |
1169 |
root |
1.147 |
a connection has been established and the message sent (and so does not |
1170 |
root |
1.119 |
need a queue that can overflow). AEMP sends return immediately, connection |
1171 |
|
|
establishment is handled in the background. |
1172 |
root |
1.26 |
|
1173 |
root |
1.51 |
=item * Erlang suffers from silent message loss, AEMP does not. |
1174 |
root |
1.26 |
|
1175 |
root |
1.99 |
Erlang implements few guarantees on messages delivery - messages can get |
1176 |
|
|
lost without any of the processes realising it (i.e. you send messages a, |
1177 |
|
|
b, and c, and the other side only receives messages a and c). |
1178 |
root |
1.26 |
|
1179 |
root |
1.117 |
AEMP guarantees (modulo hardware errors) correct ordering, and the |
1180 |
|
|
guarantee that after one message is lost, all following ones sent to the |
1181 |
|
|
same port are lost as well, until monitoring raises an error, so there are |
1182 |
|
|
no silent "holes" in the message sequence. |
1183 |
root |
1.26 |
|
1184 |
root |
1.119 |
If you want your software to be very reliable, you have to cope with |
1185 |
|
|
corrupted and even out-of-order messages in both Erlang and AEMP. AEMP |
1186 |
|
|
simply tries to work better in common error cases, such as when a network |
1187 |
|
|
link goes down. |
1188 |
|
|
|
1189 |
root |
1.26 |
=item * Erlang can send messages to the wrong port, AEMP does not. |
1190 |
|
|
|
1191 |
root |
1.119 |
In Erlang it is quite likely that a node that restarts reuses an Erlang |
1192 |
|
|
process ID known to other nodes for a completely different process, |
1193 |
|
|
causing messages destined for that process to end up in an unrelated |
1194 |
|
|
process. |
1195 |
root |
1.26 |
|
1196 |
root |
1.119 |
AEMP does not reuse port IDs, so old messages or old port IDs floating |
1197 |
root |
1.26 |
around in the network will not be sent to an unrelated port. |
1198 |
|
|
|
1199 |
|
|
=item * Erlang uses unprotected connections, AEMP uses secure |
1200 |
|
|
authentication and can use TLS. |
1201 |
|
|
|
1202 |
root |
1.66 |
AEMP can use a proven protocol - TLS - to protect connections and |
1203 |
root |
1.26 |
securely authenticate nodes. |
1204 |
|
|
|
1205 |
root |
1.28 |
=item * The AEMP protocol is optimised for both text-based and binary |
1206 |
|
|
communications. |
1207 |
|
|
|
1208 |
root |
1.66 |
The AEMP protocol, unlike the Erlang protocol, supports both programming |
1209 |
root |
1.119 |
language independent text-only protocols (good for debugging), and binary, |
1210 |
root |
1.67 |
language-specific serialisers (e.g. Storable). By default, unless TLS is |
1211 |
|
|
used, the protocol is actually completely text-based. |
1212 |
root |
1.28 |
|
1213 |
|
|
It has also been carefully designed to be implementable in other languages |
1214 |
root |
1.66 |
with a minimum of work while gracefully degrading functionality to make the |
1215 |
root |
1.28 |
protocol simple. |
1216 |
|
|
|
1217 |
root |
1.35 |
=item * AEMP has more flexible monitoring options than Erlang. |
1218 |
|
|
|
1219 |
root |
1.119 |
In Erlang, you can chose to receive I<all> exit signals as messages or |
1220 |
|
|
I<none>, there is no in-between, so monitoring single Erlang processes is |
1221 |
|
|
difficult to implement. |
1222 |
|
|
|
1223 |
|
|
Monitoring in AEMP is more flexible than in Erlang, as one can choose |
1224 |
|
|
between automatic kill, exit message or callback on a per-port basis. |
1225 |
root |
1.35 |
|
1226 |
root |
1.37 |
=item * Erlang tries to hide remote/local connections, AEMP does not. |
1227 |
root |
1.35 |
|
1228 |
root |
1.67 |
Monitoring in Erlang is not an indicator of process death/crashes, in the |
1229 |
|
|
same way as linking is (except linking is unreliable in Erlang). |
1230 |
root |
1.37 |
|
1231 |
|
|
In AEMP, you don't "look up" registered port names or send to named ports |
1232 |
|
|
that might or might not be persistent. Instead, you normally spawn a port |
1233 |
root |
1.67 |
on the remote node. The init function monitors you, and you monitor the |
1234 |
|
|
remote port. Since both monitors are local to the node, they are much more |
1235 |
|
|
reliable (no need for C<spawn_link>). |
1236 |
root |
1.37 |
|
1237 |
|
|
This also saves round-trips and avoids sending messages to the wrong port |
1238 |
|
|
(hard to do in Erlang). |
1239 |
root |
1.35 |
|
1240 |
root |
1.26 |
=back |
1241 |
|
|
|
1242 |
root |
1.46 |
=head1 RATIONALE |
1243 |
|
|
|
1244 |
|
|
=over 4 |
1245 |
|
|
|
1246 |
root |
1.67 |
=item Why strings for port and node IDs, why not objects? |
1247 |
root |
1.46 |
|
1248 |
|
|
We considered "objects", but found that the actual number of methods |
1249 |
root |
1.67 |
that can be called are quite low. Since port and node IDs travel over |
1250 |
root |
1.46 |
the network frequently, the serialising/deserialising would add lots of |
1251 |
root |
1.67 |
overhead, as well as having to keep a proxy object everywhere. |
1252 |
root |
1.46 |
|
1253 |
|
|
Strings can easily be printed, easily serialised etc. and need no special |
1254 |
|
|
procedures to be "valid". |
1255 |
|
|
|
1256 |
root |
1.110 |
And as a result, a port with just a default receiver consists of a single |
1257 |
root |
1.117 |
code reference stored in a global hash - it can't become much cheaper. |
1258 |
root |
1.47 |
|
1259 |
root |
1.67 |
=item Why favour JSON, why not a real serialising format such as Storable? |
1260 |
root |
1.46 |
|
1261 |
|
|
In fact, any AnyEvent::MP node will happily accept Storable as framing |
1262 |
|
|
format, but currently there is no way to make a node use Storable by |
1263 |
root |
1.67 |
default (although all nodes will accept it). |
1264 |
root |
1.46 |
|
1265 |
|
|
The default framing protocol is JSON because a) JSON::XS is many times |
1266 |
|
|
faster for small messages and b) most importantly, after years of |
1267 |
|
|
experience we found that object serialisation is causing more problems |
1268 |
root |
1.67 |
than it solves: Just like function calls, objects simply do not travel |
1269 |
root |
1.46 |
easily over the network, mostly because they will always be a copy, so you |
1270 |
|
|
always have to re-think your design. |
1271 |
|
|
|
1272 |
|
|
Keeping your messages simple, concentrating on data structures rather than |
1273 |
|
|
objects, will keep your messages clean, tidy and efficient. |
1274 |
|
|
|
1275 |
|
|
=back |
1276 |
|
|
|
1277 |
root |
1.137 |
=head1 PORTING FROM AnyEvent::MP VERSION 1.X |
1278 |
|
|
|
1279 |
root |
1.139 |
AEMP version 2 has a few major incompatible changes compared to version 1: |
1280 |
root |
1.137 |
|
1281 |
|
|
=over 4 |
1282 |
|
|
|
1283 |
|
|
=item AnyEvent::MP::Global no longer has group management functions. |
1284 |
|
|
|
1285 |
root |
1.140 |
At least not officially - the grp_* functions are still exported and might |
1286 |
|
|
work, but they will be removed in some later release. |
1287 |
|
|
|
1288 |
root |
1.137 |
AnyEvent::MP now comes with a distributed database that is more |
1289 |
root |
1.139 |
powerful. Its database families map closely to port groups, but the API |
1290 |
|
|
has changed (the functions are also now exported by AnyEvent::MP). Here is |
1291 |
|
|
a rough porting guide: |
1292 |
root |
1.137 |
|
1293 |
|
|
grp_reg $group, $port # old |
1294 |
|
|
db_reg $group, $port # new |
1295 |
|
|
|
1296 |
|
|
$list = grp_get $group # old |
1297 |
|
|
db_keys $group, sub { my $list = shift } # new |
1298 |
|
|
|
1299 |
|
|
grp_mon $group, $cb->(\@ports, $add, $del) # old |
1300 |
|
|
db_mon $group, $cb->(\%ports, $add, $change, $del) # new |
1301 |
|
|
|
1302 |
root |
1.139 |
C<grp_reg> is a no-brainer (just replace by C<db_reg>), but C<grp_get> is |
1303 |
|
|
no longer instant, because the local node might not have a copy of the |
1304 |
|
|
group. You can either modify your code to allow for a callback, or use |
1305 |
|
|
C<db_mon> to keep an updated copy of the group: |
1306 |
root |
1.137 |
|
1307 |
|
|
my $local_group_copy; |
1308 |
root |
1.139 |
db_mon $group => sub { $local_group_copy = $_[0] }; |
1309 |
root |
1.137 |
|
1310 |
root |
1.139 |
# now "keys %$local_group_copy" always returns the most up-to-date |
1311 |
root |
1.137 |
# list of ports in the group. |
1312 |
|
|
|
1313 |
root |
1.139 |
C<grp_mon> can be replaced by C<db_mon> with minor changes - C<db_mon> |
1314 |
|
|
passes a hash as first argument, and an extra C<$chg> argument that can be |
1315 |
|
|
ignored: |
1316 |
root |
1.137 |
|
1317 |
|
|
db_mon $group => sub { |
1318 |
root |
1.149 |
my ($ports, $add, $chg, $del) = @_; |
1319 |
root |
1.137 |
$ports = [keys %$ports]; |
1320 |
|
|
|
1321 |
|
|
# now $ports, $add and $del are the same as |
1322 |
|
|
# were originally passed by grp_mon. |
1323 |
|
|
... |
1324 |
|
|
}; |
1325 |
|
|
|
1326 |
|
|
=item Nodes not longer connect to all other nodes. |
1327 |
|
|
|
1328 |
|
|
In AEMP 1.x, every node automatically loads the L<AnyEvent::MP::Global> |
1329 |
|
|
module, which in turn would create connections to all other nodes in the |
1330 |
|
|
network (helped by the seed nodes). |
1331 |
|
|
|
1332 |
|
|
In version 2.x, global nodes still connect to all other global nodes, but |
1333 |
|
|
other nodes don't - now every node either is a global node itself, or |
1334 |
|
|
attaches itself to another global node. |
1335 |
|
|
|
1336 |
|
|
If a node isn't a global node itself, then it attaches itself to one |
1337 |
|
|
of its seed nodes. If that seed node isn't a global node yet, it will |
1338 |
|
|
automatically be upgraded to a global node. |
1339 |
|
|
|
1340 |
|
|
So in many cases, nothing needs to be changed - one just has to make sure |
1341 |
|
|
that all seed nodes are meshed together with the other seed nodes (as with |
1342 |
root |
1.139 |
AEMP 1.x), and other nodes specify them as seed nodes. This is most easily |
1343 |
|
|
achieved by specifying the same set of seed nodes for all nodes in the |
1344 |
|
|
network. |
1345 |
root |
1.137 |
|
1346 |
|
|
Not opening a connection to every other node is usually an advantage, |
1347 |
|
|
except when you need the lower latency of an already established |
1348 |
|
|
connection. To ensure a node establishes a connection to another node, |
1349 |
|
|
you can monitor the node port (C<mon $node, ...>), which will attempt to |
1350 |
root |
1.138 |
create the connection (and notify you when the connection fails). |
1351 |
root |
1.137 |
|
1352 |
root |
1.138 |
=item Listener-less nodes (nodes without binds) are gone. |
1353 |
root |
1.137 |
|
1354 |
root |
1.138 |
And are not coming back, at least not in their old form. If no C<binds> |
1355 |
root |
1.139 |
are specified for a node, AnyEvent::MP assumes a default of C<*:*>. |
1356 |
root |
1.137 |
|
1357 |
|
|
There are vague plans to implement some form of routing domains, which |
1358 |
|
|
might or might not bring back listener-less nodes, but don't count on it. |
1359 |
|
|
|
1360 |
|
|
The fact that most connections are now optional somewhat mitigates this, |
1361 |
|
|
as a node can be effectively unreachable from the outside without any |
1362 |
|
|
problems, as long as it isn't a global node and only reaches out to other |
1363 |
|
|
nodes (as opposed to being contacted from other nodes). |
1364 |
|
|
|
1365 |
root |
1.138 |
=item $AnyEvent::MP::Kernel::WARN has gone. |
1366 |
|
|
|
1367 |
|
|
AnyEvent has acquired a logging framework (L<AnyEvent::Log>), and AEMP now |
1368 |
|
|
uses this, and so should your programs. |
1369 |
|
|
|
1370 |
|
|
Every module now documents what kinds of messages it generates, with |
1371 |
|
|
AnyEvent::MP acting as a catch all. |
1372 |
|
|
|
1373 |
|
|
On the positive side, this means that instead of setting |
1374 |
root |
1.139 |
C<PERL_ANYEVENT_MP_WARNLEVEL>, you can get away by setting C<AE_VERBOSE> - |
1375 |
root |
1.138 |
much less to type. |
1376 |
|
|
|
1377 |
root |
1.137 |
=back |
1378 |
|
|
|
1379 |
root |
1.139 |
=head1 LOGGING |
1380 |
|
|
|
1381 |
root |
1.146 |
AnyEvent::MP does not normally log anything by itself, but since it is the |
1382 |
root |
1.151 |
root of the context hierarchy for AnyEvent::MP modules, it will receive |
1383 |
root |
1.139 |
all log messages by submodules. |
1384 |
|
|
|
1385 |
root |
1.1 |
=head1 SEE ALSO |
1386 |
|
|
|
1387 |
root |
1.68 |
L<AnyEvent::MP::Intro> - a gentle introduction. |
1388 |
|
|
|
1389 |
|
|
L<AnyEvent::MP::Kernel> - more, lower-level, stuff. |
1390 |
|
|
|
1391 |
root |
1.113 |
L<AnyEvent::MP::Global> - network maintenance and port groups, to find |
1392 |
root |
1.68 |
your applications. |
1393 |
|
|
|
1394 |
root |
1.105 |
L<AnyEvent::MP::DataConn> - establish data connections between nodes. |
1395 |
|
|
|
1396 |
root |
1.81 |
L<AnyEvent::MP::LogCatcher> - simple service to display log messages from |
1397 |
|
|
all nodes. |
1398 |
|
|
|
1399 |
root |
1.1 |
L<AnyEvent>. |
1400 |
|
|
|
1401 |
|
|
=head1 AUTHOR |
1402 |
|
|
|
1403 |
|
|
Marc Lehmann <schmorp@schmorp.de> |
1404 |
|
|
http://home.schmorp.de/ |
1405 |
|
|
|
1406 |
|
|
=cut |
1407 |
|
|
|
1408 |
|
|
1 |
1409 |
|
|
|