Commit | Line | Data |
a0d0e21e |
1 | =head1 NAME |
2 | |
184e9718 |
3 | perlipc - Perl interprocess communication (signals, fifos, pipes, safe subprocesses, sockets, and semaphores) |
a0d0e21e |
4 | |
5 | =head1 DESCRIPTION |
6 | |
4633a7c4 |
7 | The basic IPC facilities of Perl are built out of the good old Unix |
8 | signals, named pipes, pipe opens, the Berkeley socket routines, and SysV |
9 | IPC calls. Each is used in slightly different situations. |
10 | |
11 | =head1 Signals |
12 | |
490f90af |
13 | Perl uses a simple signal handling model: the %SIG hash contains names |
14 | or references of user-installed signal handlers. These handlers will |
15 | be called with an argument which is the name of the signal that |
16 | triggered it. A signal may be generated intentionally from a |
17 | particular keyboard sequence like control-C or control-Z, sent to you |
18 | from another process, or triggered automatically by the kernel when |
19 | special events transpire, like a child process exiting, your process |
20 | running out of stack space, or hitting file size limit. |
4633a7c4 |
21 | |
a11adca0 |
22 | For example, to trap an interrupt signal, set up a handler like this: |
4633a7c4 |
23 | |
24 | sub catch_zap { |
25 | my $signame = shift; |
26 | $shucks++; |
27 | die "Somebody sent me a SIG$signame"; |
54310121 |
28 | } |
4633a7c4 |
29 | $SIG{INT} = 'catch_zap'; # could fail in modules |
30 | $SIG{INT} = \&catch_zap; # best strategy |
31 | |
490f90af |
32 | Prior to Perl 5.7.3 it was necessary to do as little as you possibly |
33 | could in your handler; notice how all we do is set a global variable |
34 | and then raise an exception. That's because on most systems, |
35 | libraries are not re-entrant; particularly, memory allocation and I/O |
36 | routines are not. That meant that doing nearly I<anything> in your |
37 | handler could in theory trigger a memory fault and subsequent core |
ec488bcf |
38 | dump - see L</Deferred Signals (Safe Signals)> below. |
a11adca0 |
39 | |
4633a7c4 |
40 | The names of the signals are the ones listed out by C<kill -l> on your |
41 | system, or you can retrieve them from the Config module. Set up an |
42 | @signame list indexed by number to get the name and a %signo table |
43 | indexed by name to get the number: |
44 | |
45 | use Config; |
46 | defined $Config{sig_name} || die "No sigs?"; |
47 | foreach $name (split(' ', $Config{sig_name})) { |
48 | $signo{$name} = $i; |
49 | $signame[$i] = $name; |
50 | $i++; |
54310121 |
51 | } |
4633a7c4 |
52 | |
6a3992aa |
53 | So to check whether signal 17 and SIGALRM were the same, do just this: |
4633a7c4 |
54 | |
55 | print "signal #17 = $signame[17]\n"; |
54310121 |
56 | if ($signo{ALRM}) { |
4633a7c4 |
57 | print "SIGALRM is $signo{ALRM}\n"; |
54310121 |
58 | } |
4633a7c4 |
59 | |
60 | You may also choose to assign the strings C<'IGNORE'> or C<'DEFAULT'> as |
61 | the handler, in which case Perl will try to discard the signal or do the |
f648820c |
62 | default thing. |
63 | |
19799a22 |
64 | On most Unix platforms, the C<CHLD> (sometimes also known as C<CLD>) signal |
f648820c |
65 | has special behavior with respect to a value of C<'IGNORE'>. |
66 | Setting C<$SIG{CHLD}> to C<'IGNORE'> on such a platform has the effect of |
67 | not creating zombie processes when the parent process fails to C<wait()> |
68 | on its child processes (i.e. child processes are automatically reaped). |
69 | Calling C<wait()> with C<$SIG{CHLD}> set to C<'IGNORE'> usually returns |
70 | C<-1> on such platforms. |
71 | |
72 | Some signals can be neither trapped nor ignored, such as |
4633a7c4 |
73 | the KILL and STOP (but not the TSTP) signals. One strategy for |
74 | temporarily ignoring signals is to use a local() statement, which will be |
75 | automatically restored once your block is exited. (Remember that local() |
76 | values are "inherited" by functions called from within that block.) |
77 | |
78 | sub precious { |
79 | local $SIG{INT} = 'IGNORE'; |
80 | &more_functions; |
54310121 |
81 | } |
4633a7c4 |
82 | sub more_functions { |
83 | # interrupts still ignored, for now... |
54310121 |
84 | } |
4633a7c4 |
85 | |
86 | Sending a signal to a negative process ID means that you send the signal |
fb73857a |
87 | to the entire Unix process-group. This code sends a hang-up signal to all |
88 | processes in the current process group (and sets $SIG{HUP} to IGNORE so |
89 | it doesn't kill itself): |
4633a7c4 |
90 | |
91 | { |
92 | local $SIG{HUP} = 'IGNORE'; |
93 | kill HUP => -$$; |
94 | # snazzy writing of: kill('HUP', -$$) |
95 | } |
a0d0e21e |
96 | |
4633a7c4 |
97 | Another interesting signal to send is signal number zero. This doesn't |
1e9c1022 |
98 | actually affect a child process, but instead checks whether it's alive |
54310121 |
99 | or has changed its UID. |
a0d0e21e |
100 | |
4633a7c4 |
101 | unless (kill 0 => $kid_pid) { |
102 | warn "something wicked happened to $kid_pid"; |
54310121 |
103 | } |
a0d0e21e |
104 | |
1e9c1022 |
105 | When directed at a process whose UID is not identical to that |
106 | of the sending process, signal number zero may fail because |
107 | you lack permission to send the signal, even though the process is alive. |
bf003f36 |
108 | You may be able to determine the cause of failure using C<%!>. |
1e9c1022 |
109 | |
bf003f36 |
110 | unless (kill 0 => $pid or $!{EPERM}) { |
1e9c1022 |
111 | warn "$pid looks dead"; |
112 | } |
113 | |
4633a7c4 |
114 | You might also want to employ anonymous functions for simple signal |
115 | handlers: |
a0d0e21e |
116 | |
4633a7c4 |
117 | $SIG{INT} = sub { die "\nOutta here!\n" }; |
a0d0e21e |
118 | |
4633a7c4 |
119 | But that will be problematic for the more complicated handlers that need |
54310121 |
120 | to reinstall themselves. Because Perl's signal mechanism is currently |
184e9718 |
121 | based on the signal(3) function from the C library, you may sometimes be so |
4633a7c4 |
122 | misfortunate as to run on systems where that function is "broken", that |
123 | is, it behaves in the old unreliable SysV way rather than the newer, more |
124 | reasonable BSD and POSIX fashion. So you'll see defensive people writing |
125 | signal handlers like this: |
a0d0e21e |
126 | |
54310121 |
127 | sub REAPER { |
4633a7c4 |
128 | $waitedpid = wait; |
6a3992aa |
129 | # loathe sysV: it makes us not only reinstate |
130 | # the handler, but place it after the wait |
54310121 |
131 | $SIG{CHLD} = \&REAPER; |
4633a7c4 |
132 | } |
133 | $SIG{CHLD} = \&REAPER; |
134 | # now do something that forks... |
135 | |
816229cf |
136 | or better still: |
4633a7c4 |
137 | |
6a3992aa |
138 | use POSIX ":sys_wait_h"; |
54310121 |
139 | sub REAPER { |
4633a7c4 |
140 | my $child; |
816229cf |
141 | # If a second child dies while in the signal handler caused by the |
142 | # first death, we won't get another signal. So must loop here else |
143 | # we will leave the unreaped child as a zombie. And the next time |
144 | # two children die we get another zombie. And so on. |
1450d070 |
145 | while (($child = waitpid(-1,WNOHANG)) > 0) { |
4633a7c4 |
146 | $Kid_Status{$child} = $?; |
54310121 |
147 | } |
6a3992aa |
148 | $SIG{CHLD} = \&REAPER; # still loathe sysV |
4633a7c4 |
149 | } |
150 | $SIG{CHLD} = \&REAPER; |
151 | # do something that forks... |
152 | |
153 | Signal handling is also used for timeouts in Unix, While safely |
154 | protected within an C<eval{}> block, you set a signal handler to trap |
155 | alarm signals and then schedule to have one delivered to you in some |
156 | number of seconds. Then try your blocking operation, clearing the alarm |
157 | when it's done but not before you've exited your C<eval{}> block. If it |
158 | goes off, you'll use die() to jump out of the block, much as you might |
159 | using longjmp() or throw() in other languages. |
160 | |
161 | Here's an example: |
162 | |
54310121 |
163 | eval { |
4633a7c4 |
164 | local $SIG{ALRM} = sub { die "alarm clock restart" }; |
54310121 |
165 | alarm 10; |
4633a7c4 |
166 | flock(FH, 2); # blocking write lock |
54310121 |
167 | alarm 0; |
4633a7c4 |
168 | }; |
169 | if ($@ and $@ !~ /alarm clock restart/) { die } |
170 | |
8a4f6ac2 |
171 | If the operation being timed out is system() or qx(), this technique |
172 | is liable to generate zombies. If this matters to you, you'll |
173 | need to do your own fork() and exec(), and kill the errant child process. |
174 | |
4633a7c4 |
175 | For more complex signal handling, you might see the standard POSIX |
176 | module. Lamentably, this is almost entirely undocumented, but |
177 | the F<t/lib/posix.t> file from the Perl source distribution has some |
178 | examples in it. |
179 | |
28494392 |
180 | =head2 Handling the SIGHUP Signal in Daemons |
181 | |
182 | A process that usually starts when the system boots and shuts down |
183 | when the system is shut down is called a daemon (Disk And Execution |
184 | MONitor). If a daemon process has a configuration file which is |
185 | modified after the process has been started, there should be a way to |
186 | tell that process to re-read its configuration file, without stopping |
187 | the process. Many daemons provide this mechanism using the C<SIGHUP> |
188 | signal handler. When you want to tell the daemon to re-read the file |
189 | you simply send it the C<SIGHUP> signal. |
190 | |
3031ea75 |
191 | Not all platforms automatically reinstall their (native) signal |
192 | handlers after a signal delivery. This means that the handler works |
193 | only the first time the signal is sent. The solution to this problem |
194 | is to use C<POSIX> signal handlers if available, their behaviour |
195 | is well-defined. |
28494392 |
196 | |
197 | The following example implements a simple daemon, which restarts |
198 | itself every time the C<SIGHUP> signal is received. The actual code is |
199 | located in the subroutine C<code()>, which simply prints some debug |
200 | info to show that it works and should be replaced with the real code. |
201 | |
202 | #!/usr/bin/perl -w |
d6fd60d6 |
203 | |
28494392 |
204 | use POSIX (); |
205 | use FindBin (); |
206 | use File::Basename (); |
207 | use File::Spec::Functions; |
d6fd60d6 |
208 | |
28494392 |
209 | $|=1; |
d6fd60d6 |
210 | |
28494392 |
211 | # make the daemon cross-platform, so exec always calls the script |
212 | # itself with the right path, no matter how the script was invoked. |
213 | my $script = File::Basename::basename($0); |
214 | my $SELF = catfile $FindBin::Bin, $script; |
d6fd60d6 |
215 | |
28494392 |
216 | # POSIX unmasks the sigprocmask properly |
217 | my $sigset = POSIX::SigSet->new(); |
218 | my $action = POSIX::SigAction->new('sigHUP_handler', |
219 | $sigset, |
220 | &POSIX::SA_NODEFER); |
221 | POSIX::sigaction(&POSIX::SIGHUP, $action); |
d6fd60d6 |
222 | |
28494392 |
223 | sub sigHUP_handler { |
224 | print "got SIGHUP\n"; |
225 | exec($SELF, @ARGV) or die "Couldn't restart: $!\n"; |
226 | } |
d6fd60d6 |
227 | |
28494392 |
228 | code(); |
d6fd60d6 |
229 | |
28494392 |
230 | sub code { |
231 | print "PID: $$\n"; |
232 | print "ARGV: @ARGV\n"; |
233 | my $c = 0; |
234 | while (++$c) { |
235 | sleep 2; |
236 | print "$c\n"; |
237 | } |
238 | } |
239 | __END__ |
240 | |
241 | |
4633a7c4 |
242 | =head1 Named Pipes |
243 | |
244 | A named pipe (often referred to as a FIFO) is an old Unix IPC |
245 | mechanism for processes communicating on the same machine. It works |
54310121 |
246 | just like a regular, connected anonymous pipes, except that the |
4633a7c4 |
247 | processes rendezvous using a filename and don't have to be related. |
248 | |
249 | To create a named pipe, use the Unix command mknod(1) or on some |
250 | systems, mkfifo(1). These may not be in your normal path. |
251 | |
252 | # system return val is backwards, so && not || |
253 | # |
254 | $ENV{PATH} .= ":/etc:/usr/etc"; |
54310121 |
255 | if ( system('mknod', $path, 'p') |
4633a7c4 |
256 | && system('mkfifo', $path) ) |
257 | { |
5a964f20 |
258 | die "mk{nod,fifo} $path failed"; |
54310121 |
259 | } |
4633a7c4 |
260 | |
261 | |
262 | A fifo is convenient when you want to connect a process to an unrelated |
263 | one. When you open a fifo, the program will block until there's something |
54310121 |
264 | on the other end. |
4633a7c4 |
265 | |
266 | For example, let's say you'd like to have your F<.signature> file be a |
267 | named pipe that has a Perl program on the other end. Now every time any |
6a3992aa |
268 | program (like a mailer, news reader, finger program, etc.) tries to read |
4633a7c4 |
269 | from that file, the reading program will block and your program will |
6a3992aa |
270 | supply the new signature. We'll use the pipe-checking file test B<-p> |
4633a7c4 |
271 | to find out whether anyone (or anything) has accidentally removed our fifo. |
272 | |
273 | chdir; # go home |
274 | $FIFO = '.signature'; |
275 | $ENV{PATH} .= ":/etc:/usr/games"; |
276 | |
277 | while (1) { |
278 | unless (-p $FIFO) { |
279 | unlink $FIFO; |
54310121 |
280 | system('mknod', $FIFO, 'p') |
4633a7c4 |
281 | && die "can't mknod $FIFO: $!"; |
54310121 |
282 | } |
4633a7c4 |
283 | |
284 | # next line blocks until there's a reader |
285 | open (FIFO, "> $FIFO") || die "can't write $FIFO: $!"; |
286 | print FIFO "John Smith (smith\@host.org)\n", `fortune -s`; |
287 | close FIFO; |
6a3992aa |
288 | sleep 2; # to avoid dup signals |
4633a7c4 |
289 | } |
a0d0e21e |
290 | |
ffc145e8 |
291 | =head2 Deferred Signals (Safe Signals) |
5a964f20 |
292 | |
490f90af |
293 | In Perls before Perl 5.7.3 by installing Perl code to deal with |
294 | signals, you were exposing yourself to danger from two things. First, |
295 | few system library functions are re-entrant. If the signal interrupts |
296 | while Perl is executing one function (like malloc(3) or printf(3)), |
297 | and your signal handler then calls the same function again, you could |
298 | get unpredictable behavior--often, a core dump. Second, Perl isn't |
299 | itself re-entrant at the lowest levels. If the signal interrupts Perl |
300 | while Perl is changing its own internal data structures, similarly |
301 | unpredictable behaviour may result. |
5a964f20 |
302 | |
a11adca0 |
303 | There were two things you could do, knowing this: be paranoid or be |
304 | pragmatic. The paranoid approach was to do as little as possible in your |
5a964f20 |
305 | signal handler. Set an existing integer variable that already has a |
306 | value, and return. This doesn't help you if you're in a slow system call, |
307 | which will just restart. That means you have to C<die> to longjump(3) out |
308 | of the handler. Even this is a little cavalier for the true paranoiac, |
309 | who avoids C<die> in a handler because the system I<is> out to get you. |
a11adca0 |
310 | The pragmatic approach was to say ``I know the risks, but prefer the |
311 | convenience'', and to do anything you wanted in your signal handler, |
312 | and be prepared to clean up core dumps now and again. |
313 | |
490f90af |
314 | In Perl 5.7.3 and later to avoid these problems signals are |
315 | "deferred"-- that is when the signal is delivered to the process by |
316 | the system (to the C code that implements Perl) a flag is set, and the |
317 | handler returns immediately. Then at strategic "safe" points in the |
318 | Perl interpreter (e.g. when it is about to execute a new opcode) the |
319 | flags are checked and the Perl level handler from %SIG is |
320 | executed. The "deferred" scheme allows much more flexibility in the |
321 | coding of signal handler as we know Perl interpreter is in a safe |
322 | state, and that we are not in a system library function when the |
323 | handler is called. However the implementation does differ from |
324 | previous Perls in the following ways: |
5a964f20 |
325 | |
a11adca0 |
326 | =over 4 |
5a964f20 |
327 | |
a11adca0 |
328 | =item Long running opcodes |
329 | |
490f90af |
330 | As Perl interpreter only looks at the signal flags when it about to |
331 | execute a new opcode if a signal arrives during a long running opcode |
332 | (e.g. a regular expression operation on a very large string) then |
333 | signal will not be seen until operation completes. |
a11adca0 |
334 | |
335 | =item Interrupting IO |
336 | |
490f90af |
337 | When a signal is delivered (e.g. INT control-C) the operating system |
338 | breaks into IO operations like C<read> (used to implement Perls |
339 | E<lt>E<gt> operator). On older Perls the handler was called |
340 | immediately (and as C<read> is not "unsafe" this worked well). With |
341 | the "deferred" scheme the handler is not called immediately, and if |
342 | Perl is using system's C<stdio> library that library may re-start the |
343 | C<read> without returning to Perl and giving it a chance to call the |
344 | %SIG handler. If this happens on your system the solution is to use |
345 | C<:perlio> layer to do IO - at least on those handles which you want |
346 | to be able to break into with signals. (The C<:perlio> layer checks |
347 | the signal flags and calls %SIG handlers before resuming IO operation.) |
348 | |
349 | Note that the default in Perl 5.7.3 and later is to automatically use |
350 | the C<:perlio> layer. |
a11adca0 |
351 | |
91d81acc |
352 | Note that some networking library functions like gethostbyname() are |
353 | known to have their own implementations of timeouts which may conflict |
354 | with your timeouts. If you are having problems with such functions, |
355 | you can try using the POSIX sigaction() function, which bypasses the |
356 | Perl safe signals (note that this means subjecting yourself to |
357 | possible memory corruption, as described above). Instead of setting |
358 | C<$SIG{ALRM}> try something like the following: |
359 | |
360 | use POSIX; |
361 | sigaction SIGALRM, new POSIX::SigAction sub { die "alarm\n" } |
362 | or die "Error setting SIGALRM handler: $!\n"; |
363 | |
9ce5b4ad |
364 | =item Restartable system calls |
365 | |
366 | On systems that supported it, older versions of Perl used the |
367 | SA_RESTART flag when installing %SIG handlers. This meant that |
368 | restartable system calls would continue rather than returning when |
369 | a signal arrived. In order to deliver deferred signals promptly, |
370 | Perl 5.7.3 and later do I<not> use SA_RESTART. Consequently, |
371 | restartable system calls can fail (with $! set to C<EINTR>) in places |
372 | where they previously would have succeeded. |
373 | |
374 | Note that the default C<:perlio> layer will retry C<read>, C<write> |
375 | and C<close> as described above and that interrupted C<wait> and |
376 | C<waitpid> calls will always be retried. |
377 | |
a11adca0 |
378 | =item Signals as "faults" |
379 | |
490f90af |
380 | Certain signals e.g. SEGV, ILL, BUS are generated as a result of |
381 | virtual memory or other "faults". These are normally fatal and there |
382 | is little a Perl-level handler can do with them. (In particular the |
383 | old signal scheme was particularly unsafe in such cases.) However if |
384 | a %SIG handler is set the new scheme simply sets a flag and returns as |
385 | described above. This may cause the operating system to try the |
386 | offending machine instruction again and - as nothing has changed - it |
387 | will generate the signal again. The result of this is a rather odd |
388 | "loop". In future Perl's signal mechanism may be changed to avoid this |
389 | - perhaps by simply disallowing %SIG handlers on signals of that |
390 | type. Until then the work-round is not to set a %SIG handler on those |
fa11829f |
391 | signals. (Which signals they are is operating system dependent.) |
a11adca0 |
392 | |
393 | =item Signals triggered by operating system state |
394 | |
490f90af |
395 | On some operating systems certain signal handlers are supposed to "do |
396 | something" before returning. One example can be CHLD or CLD which |
397 | indicates a child process has completed. On some operating systems the |
398 | signal handler is expected to C<wait> for the completed child |
399 | process. On such systems the deferred signal scheme will not work for |
400 | those signals (it does not do the C<wait>). Again the failure will |
401 | look like a loop as the operating system will re-issue the signal as |
402 | there are un-waited-for completed child processes. |
a11adca0 |
403 | |
818c4caa |
404 | =back |
a0d0e21e |
405 | |
4ffa73a3 |
406 | If you want the old signal behaviour back regardless of possible |
407 | memory corruption, set the environment variable C<PERL_SIGNALS> to |
45c0772f |
408 | C<"unsafe"> (a new feature since Perl 5.8.1). |
4ffa73a3 |
409 | |
4633a7c4 |
410 | =head1 Using open() for IPC |
411 | |
490f90af |
412 | Perl's basic open() statement can also be used for unidirectional |
413 | interprocess communication by either appending or prepending a pipe |
414 | symbol to the second argument to open(). Here's how to start |
415 | something up in a child process you intend to write to: |
4633a7c4 |
416 | |
54310121 |
417 | open(SPOOLER, "| cat -v | lpr -h 2>/dev/null") |
4633a7c4 |
418 | || die "can't fork: $!"; |
419 | local $SIG{PIPE} = sub { die "spooler pipe broke" }; |
420 | print SPOOLER "stuff\n"; |
421 | close SPOOLER || die "bad spool: $! $?"; |
422 | |
423 | And here's how to start up a child process you intend to read from: |
424 | |
425 | open(STATUS, "netstat -an 2>&1 |") |
426 | || die "can't fork: $!"; |
427 | while (<STATUS>) { |
428 | next if /^(tcp|udp)/; |
429 | print; |
54310121 |
430 | } |
a2eb9003 |
431 | close STATUS || die "bad netstat: $! $?"; |
4633a7c4 |
432 | |
433 | If one can be sure that a particular program is a Perl script that is |
434 | expecting filenames in @ARGV, the clever programmer can write something |
435 | like this: |
436 | |
5a964f20 |
437 | % program f1 "cmd1|" - f2 "cmd2|" f3 < tmpfile |
4633a7c4 |
438 | |
439 | and irrespective of which shell it's called from, the Perl program will |
440 | read from the file F<f1>, the process F<cmd1>, standard input (F<tmpfile> |
441 | in this case), the F<f2> file, the F<cmd2> command, and finally the F<f3> |
442 | file. Pretty nifty, eh? |
443 | |
54310121 |
444 | You might notice that you could use backticks for much the |
4633a7c4 |
445 | same effect as opening a pipe for reading: |
446 | |
447 | print grep { !/^(tcp|udp)/ } `netstat -an 2>&1`; |
448 | die "bad netstat" if $?; |
449 | |
450 | While this is true on the surface, it's much more efficient to process the |
451 | file one line or record at a time because then you don't have to read the |
19799a22 |
452 | whole thing into memory at once. It also gives you finer control of the |
4633a7c4 |
453 | whole process, letting you to kill off the child process early if you'd |
454 | like. |
455 | |
456 | Be careful to check both the open() and the close() return values. If |
457 | you're I<writing> to a pipe, you should also trap SIGPIPE. Otherwise, |
458 | think of what happens when you start up a pipe to a command that doesn't |
459 | exist: the open() will in all likelihood succeed (it only reflects the |
460 | fork()'s success), but then your output will fail--spectacularly. Perl |
461 | can't know whether the command worked because your command is actually |
462 | running in a separate process whose exec() might have failed. Therefore, |
6a3992aa |
463 | while readers of bogus commands return just a quick end of file, writers |
4633a7c4 |
464 | to bogus command will trigger a signal they'd better be prepared to |
465 | handle. Consider: |
466 | |
5a964f20 |
467 | open(FH, "|bogus") or die "can't fork: $!"; |
468 | print FH "bang\n" or die "can't write: $!"; |
469 | close FH or die "can't close: $!"; |
470 | |
471 | That won't blow up until the close, and it will blow up with a SIGPIPE. |
472 | To catch it, you could use this: |
473 | |
474 | $SIG{PIPE} = 'IGNORE'; |
475 | open(FH, "|bogus") or die "can't fork: $!"; |
476 | print FH "bang\n" or die "can't write: $!"; |
477 | close FH or die "can't close: status=$?"; |
4633a7c4 |
478 | |
68dc0745 |
479 | =head2 Filehandles |
480 | |
5a964f20 |
481 | Both the main process and any child processes it forks share the same |
482 | STDIN, STDOUT, and STDERR filehandles. If both processes try to access |
45bc9206 |
483 | them at once, strange things can happen. You may also want to close |
5a964f20 |
484 | or reopen the filehandles for the child. You can get around this by |
485 | opening your pipe with open(), but on some systems this means that the |
486 | child process cannot outlive the parent. |
68dc0745 |
487 | |
488 | =head2 Background Processes |
489 | |
490 | You can run a command in the background with: |
491 | |
7b05b7e3 |
492 | system("cmd &"); |
68dc0745 |
493 | |
494 | The command's STDOUT and STDERR (and possibly STDIN, depending on your |
495 | shell) will be the same as the parent's. You won't need to catch |
496 | SIGCHLD because of the double-fork taking place (see below for more |
497 | details). |
498 | |
499 | =head2 Complete Dissociation of Child from Parent |
500 | |
501 | In some cases (starting server processes, for instance) you'll want to |
893af57a |
502 | completely dissociate the child process from the parent. This is |
503 | often called daemonization. A well behaved daemon will also chdir() |
504 | to the root directory (so it doesn't prevent unmounting the filesystem |
505 | containing the directory from which it was launched) and redirect its |
506 | standard file descriptors from and to F</dev/null> (so that random |
507 | output doesn't wind up on the user's terminal). |
508 | |
509 | use POSIX 'setsid'; |
510 | |
511 | sub daemonize { |
512 | chdir '/' or die "Can't chdir to /: $!"; |
513 | open STDIN, '/dev/null' or die "Can't read /dev/null: $!"; |
514 | open STDOUT, '>/dev/null' |
515 | or die "Can't write to /dev/null: $!"; |
516 | defined(my $pid = fork) or die "Can't fork: $!"; |
517 | exit if $pid; |
518 | setsid or die "Can't start a new session: $!"; |
519 | open STDERR, '>&STDOUT' or die "Can't dup stdout: $!"; |
520 | } |
5a964f20 |
521 | |
893af57a |
522 | The fork() has to come before the setsid() to ensure that you aren't a |
523 | process group leader (the setsid() will fail if you are). If your |
524 | system doesn't have the setsid() function, open F</dev/tty> and use the |
525 | C<TIOCNOTTY> ioctl() on it instead. See L<tty(4)> for details. |
5a964f20 |
526 | |
893af57a |
527 | Non-Unix users should check their Your_OS::Process module for other |
528 | solutions. |
68dc0745 |
529 | |
4633a7c4 |
530 | =head2 Safe Pipe Opens |
531 | |
532 | Another interesting approach to IPC is making your single program go |
533 | multiprocess and communicate between (or even amongst) yourselves. The |
534 | open() function will accept a file argument of either C<"-|"> or C<"|-"> |
535 | to do a very interesting thing: it forks a child connected to the |
536 | filehandle you've opened. The child is running the same program as the |
537 | parent. This is useful for safely opening a file when running under an |
538 | assumed UID or GID, for example. If you open a pipe I<to> minus, you can |
539 | write to the filehandle you opened and your kid will find it in his |
540 | STDIN. If you open a pipe I<from> minus, you can read from the filehandle |
541 | you opened whatever your kid writes to his STDOUT. |
542 | |
a1ce9542 |
543 | use English '-no_match_vars'; |
4633a7c4 |
544 | my $sleep_count = 0; |
545 | |
54310121 |
546 | do { |
c07a80fd |
547 | $pid = open(KID_TO_WRITE, "|-"); |
4633a7c4 |
548 | unless (defined $pid) { |
549 | warn "cannot fork: $!"; |
550 | die "bailing out" if $sleep_count++ > 6; |
551 | sleep 10; |
54310121 |
552 | } |
4633a7c4 |
553 | } until defined $pid; |
554 | |
555 | if ($pid) { # parent |
c07a80fd |
556 | print KID_TO_WRITE @some_data; |
557 | close(KID_TO_WRITE) || warn "kid exited $?"; |
4633a7c4 |
558 | } else { # child |
559 | ($EUID, $EGID) = ($UID, $GID); # suid progs only |
54310121 |
560 | open (FILE, "> /safe/file") |
4633a7c4 |
561 | || die "can't open /safe/file: $!"; |
562 | while (<STDIN>) { |
563 | print FILE; # child's STDIN is parent's KID |
54310121 |
564 | } |
4633a7c4 |
565 | exit; # don't forget this |
54310121 |
566 | } |
4633a7c4 |
567 | |
568 | Another common use for this construct is when you need to execute |
569 | something without the shell's interference. With system(), it's |
54310121 |
570 | straightforward, but you can't use a pipe open or backticks safely. |
4633a7c4 |
571 | That's because there's no way to stop the shell from getting its hands on |
572 | your arguments. Instead, use lower-level control to call exec() directly. |
573 | |
54310121 |
574 | Here's a safe backtick or pipe open for read: |
4633a7c4 |
575 | |
576 | # add error processing as above |
c07a80fd |
577 | $pid = open(KID_TO_READ, "-|"); |
4633a7c4 |
578 | |
579 | if ($pid) { # parent |
c07a80fd |
580 | while (<KID_TO_READ>) { |
4633a7c4 |
581 | # do something interesting |
54310121 |
582 | } |
c07a80fd |
583 | close(KID_TO_READ) || warn "kid exited $?"; |
4633a7c4 |
584 | |
585 | } else { # child |
586 | ($EUID, $EGID) = ($UID, $GID); # suid only |
587 | exec($program, @options, @args) |
588 | || die "can't exec program: $!"; |
589 | # NOTREACHED |
54310121 |
590 | } |
4633a7c4 |
591 | |
592 | |
593 | And here's a safe pipe open for writing: |
594 | |
595 | # add error processing as above |
c07a80fd |
596 | $pid = open(KID_TO_WRITE, "|-"); |
76c0e0db |
597 | $SIG{PIPE} = sub { die "whoops, $program pipe broke" }; |
4633a7c4 |
598 | |
599 | if ($pid) { # parent |
600 | for (@data) { |
c07a80fd |
601 | print KID_TO_WRITE; |
54310121 |
602 | } |
c07a80fd |
603 | close(KID_TO_WRITE) || warn "kid exited $?"; |
4633a7c4 |
604 | |
605 | } else { # child |
606 | ($EUID, $EGID) = ($UID, $GID); |
607 | exec($program, @options, @args) |
608 | || die "can't exec program: $!"; |
609 | # NOTREACHED |
54310121 |
610 | } |
4633a7c4 |
611 | |
307eac13 |
612 | Since Perl 5.8.0, you can also use the list form of C<open> for pipes : |
613 | the syntax |
614 | |
615 | open KID_PS, "-|", "ps", "aux" or die $!; |
616 | |
617 | forks the ps(1) command (without spawning a shell, as there are more than |
618 | three arguments to open()), and reads its standard output via the |
ca585e4d |
619 | C<KID_PS> filehandle. The corresponding syntax to read from command |
620 | pipes (with C<"|-"> in place of C<"-|">) is also implemented. |
307eac13 |
621 | |
4633a7c4 |
622 | Note that these operations are full Unix forks, which means they may not be |
623 | correctly implemented on alien systems. Additionally, these are not true |
54310121 |
624 | multithreading. If you'd like to learn more about threading, see the |
184e9718 |
625 | F<modules> file mentioned below in the SEE ALSO section. |
4633a7c4 |
626 | |
7b05b7e3 |
627 | =head2 Bidirectional Communication with Another Process |
4633a7c4 |
628 | |
629 | While this works reasonably well for unidirectional communication, what |
630 | about bidirectional communication? The obvious thing you'd like to do |
631 | doesn't actually work: |
632 | |
c07a80fd |
633 | open(PROG_FOR_READING_AND_WRITING, "| some program |") |
4633a7c4 |
634 | |
9f1b1f2d |
635 | and if you forget to use the C<use warnings> pragma or the B<-w> flag, |
636 | then you'll miss out entirely on the diagnostic message: |
4633a7c4 |
637 | |
638 | Can't do bidirectional pipe at -e line 1. |
639 | |
640 | If you really want to, you can use the standard open2() library function |
7b05b7e3 |
641 | to catch both ends. There's also an open3() for tridirectional I/O so you |
4633a7c4 |
642 | can also catch your child's STDERR, but doing so would then require an |
643 | awkward select() loop and wouldn't allow you to use normal Perl input |
644 | operations. |
645 | |
646 | If you look at its source, you'll see that open2() uses low-level |
5a964f20 |
647 | primitives like Unix pipe() and exec() calls to create all the connections. |
4633a7c4 |
648 | While it might have been slightly more efficient by using socketpair(), it |
649 | would have then been even less portable than it already is. The open2() |
650 | and open3() functions are unlikely to work anywhere except on a Unix |
651 | system or some other one purporting to be POSIX compliant. |
652 | |
653 | Here's an example of using open2(): |
654 | |
655 | use FileHandle; |
656 | use IPC::Open2; |
5a964f20 |
657 | $pid = open2(*Reader, *Writer, "cat -u -n" ); |
4633a7c4 |
658 | print Writer "stuff\n"; |
659 | $got = <Reader>; |
660 | |
6a3992aa |
661 | The problem with this is that Unix buffering is really going to |
662 | ruin your day. Even though your C<Writer> filehandle is auto-flushed, |
4633a7c4 |
663 | and the process on the other end will get your data in a timely manner, |
6a3992aa |
664 | you can't usually do anything to force it to give it back to you |
54310121 |
665 | in a similarly quick fashion. In this case, we could, because we |
4633a7c4 |
666 | gave I<cat> a B<-u> flag to make it unbuffered. But very few Unix |
667 | commands are designed to operate over pipes, so this seldom works |
54310121 |
668 | unless you yourself wrote the program on the other end of the |
4633a7c4 |
669 | double-ended pipe. |
670 | |
54310121 |
671 | A solution to this is the nonstandard F<Comm.pl> library. It uses |
4633a7c4 |
672 | pseudo-ttys to make your program behave more reasonably: |
673 | |
674 | require 'Comm.pl'; |
675 | $ph = open_proc('cat -n'); |
676 | for (1..10) { |
677 | print $ph "a line\n"; |
678 | print "got back ", scalar <$ph>; |
679 | } |
a0d0e21e |
680 | |
4633a7c4 |
681 | This way you don't have to have control over the source code of the |
54310121 |
682 | program you're using. The F<Comm> library also has expect() |
683 | and interact() functions. Find the library (and we hope its |
4633a7c4 |
684 | successor F<IPC::Chat>) at your nearest CPAN archive as detailed |
184e9718 |
685 | in the SEE ALSO section below. |
a0d0e21e |
686 | |
c8db1d39 |
687 | The newer Expect.pm module from CPAN also addresses this kind of thing. |
688 | This module requires two other modules from CPAN: IO::Pty and IO::Stty. |
689 | It sets up a pseudo-terminal to interact with programs that insist on |
a11adca0 |
690 | using talking to the terminal device driver. If your system is |
c8db1d39 |
691 | amongst those supported, this may be your best bet. |
692 | |
5a964f20 |
693 | =head2 Bidirectional Communication with Yourself |
694 | |
695 | If you want, you may make low-level pipe() and fork() |
696 | to stitch this together by hand. This example only |
697 | talks to itself, but you could reopen the appropriate |
698 | handles to STDIN and STDOUT and call other processes. |
699 | |
700 | #!/usr/bin/perl -w |
701 | # pipe1 - bidirectional communication using two pipe pairs |
702 | # designed for the socketpair-challenged |
703 | use IO::Handle; # thousands of lines just for autoflush :-( |
704 | pipe(PARENT_RDR, CHILD_WTR); # XXX: failure? |
705 | pipe(CHILD_RDR, PARENT_WTR); # XXX: failure? |
706 | CHILD_WTR->autoflush(1); |
707 | PARENT_WTR->autoflush(1); |
708 | |
709 | if ($pid = fork) { |
710 | close PARENT_RDR; close PARENT_WTR; |
711 | print CHILD_WTR "Parent Pid $$ is sending this\n"; |
712 | chomp($line = <CHILD_RDR>); |
713 | print "Parent Pid $$ just read this: `$line'\n"; |
714 | close CHILD_RDR; close CHILD_WTR; |
715 | waitpid($pid,0); |
716 | } else { |
717 | die "cannot fork: $!" unless defined $pid; |
718 | close CHILD_RDR; close CHILD_WTR; |
719 | chomp($line = <PARENT_RDR>); |
720 | print "Child Pid $$ just read this: `$line'\n"; |
721 | print PARENT_WTR "Child Pid $$ is sending this\n"; |
722 | close PARENT_RDR; close PARENT_WTR; |
723 | exit; |
724 | } |
725 | |
a11adca0 |
726 | But you don't actually have to make two pipe calls. If you |
5a964f20 |
727 | have the socketpair() system call, it will do this all for you. |
728 | |
729 | #!/usr/bin/perl -w |
730 | # pipe2 - bidirectional communication using socketpair |
731 | # "the best ones always go both ways" |
732 | |
733 | use Socket; |
734 | use IO::Handle; # thousands of lines just for autoflush :-( |
735 | # We say AF_UNIX because although *_LOCAL is the |
736 | # POSIX 1003.1g form of the constant, many machines |
737 | # still don't have it. |
738 | socketpair(CHILD, PARENT, AF_UNIX, SOCK_STREAM, PF_UNSPEC) |
739 | or die "socketpair: $!"; |
740 | |
741 | CHILD->autoflush(1); |
742 | PARENT->autoflush(1); |
743 | |
744 | if ($pid = fork) { |
745 | close PARENT; |
746 | print CHILD "Parent Pid $$ is sending this\n"; |
747 | chomp($line = <CHILD>); |
748 | print "Parent Pid $$ just read this: `$line'\n"; |
749 | close CHILD; |
750 | waitpid($pid,0); |
751 | } else { |
752 | die "cannot fork: $!" unless defined $pid; |
753 | close CHILD; |
754 | chomp($line = <PARENT>); |
755 | print "Child Pid $$ just read this: `$line'\n"; |
756 | print PARENT "Child Pid $$ is sending this\n"; |
757 | close PARENT; |
758 | exit; |
759 | } |
760 | |
4633a7c4 |
761 | =head1 Sockets: Client/Server Communication |
a0d0e21e |
762 | |
6a3992aa |
763 | While not limited to Unix-derived operating systems (e.g., WinSock on PCs |
4633a7c4 |
764 | provides socket support, as do some VMS libraries), you may not have |
184e9718 |
765 | sockets on your system, in which case this section probably isn't going to do |
6a3992aa |
766 | you much good. With sockets, you can do both virtual circuits (i.e., TCP |
767 | streams) and datagrams (i.e., UDP packets). You may be able to do even more |
4633a7c4 |
768 | depending on your system. |
769 | |
770 | The Perl function calls for dealing with sockets have the same names as |
771 | the corresponding system calls in C, but their arguments tend to differ |
772 | for two reasons: first, Perl filehandles work differently than C file |
773 | descriptors. Second, Perl already knows the length of its strings, so you |
774 | don't need to pass that information. |
a0d0e21e |
775 | |
4633a7c4 |
776 | One of the major problems with old socket code in Perl was that it used |
777 | hard-coded values for some of the constants, which severely hurt |
778 | portability. If you ever see code that does anything like explicitly |
779 | setting C<$AF_INET = 2>, you know you're in for big trouble: An |
780 | immeasurably superior approach is to use the C<Socket> module, which more |
781 | reliably grants access to various constants and functions you'll need. |
a0d0e21e |
782 | |
68dc0745 |
783 | If you're not writing a server/client for an existing protocol like |
784 | NNTP or SMTP, you should give some thought to how your server will |
785 | know when the client has finished talking, and vice-versa. Most |
786 | protocols are based on one-line messages and responses (so one party |
4a6725af |
787 | knows the other has finished when a "\n" is received) or multi-line |
68dc0745 |
788 | messages and responses that end with a period on an empty line |
789 | ("\n.\n" terminates a message/response). |
790 | |
5a964f20 |
791 | =head2 Internet Line Terminators |
792 | |
793 | The Internet line terminator is "\015\012". Under ASCII variants of |
794 | Unix, that could usually be written as "\r\n", but under other systems, |
795 | "\r\n" might at times be "\015\015\012", "\012\012\015", or something |
796 | completely different. The standards specify writing "\015\012" to be |
797 | conformant (be strict in what you provide), but they also recommend |
798 | accepting a lone "\012" on input (but be lenient in what you require). |
799 | We haven't always been very good about that in the code in this manpage, |
800 | but unless you're on a Mac, you'll probably be ok. |
801 | |
4633a7c4 |
802 | =head2 Internet TCP Clients and Servers |
a0d0e21e |
803 | |
4633a7c4 |
804 | Use Internet-domain sockets when you want to do client-server |
805 | communication that might extend to machines outside of your own system. |
806 | |
807 | Here's a sample TCP client using Internet-domain sockets: |
808 | |
809 | #!/usr/bin/perl -w |
4633a7c4 |
810 | use strict; |
811 | use Socket; |
812 | my ($remote,$port, $iaddr, $paddr, $proto, $line); |
813 | |
814 | $remote = shift || 'localhost'; |
815 | $port = shift || 2345; # random port |
816 | if ($port =~ /\D/) { $port = getservbyname($port, 'tcp') } |
817 | die "No port" unless $port; |
818 | $iaddr = inet_aton($remote) || die "no host: $remote"; |
819 | $paddr = sockaddr_in($port, $iaddr); |
820 | |
821 | $proto = getprotobyname('tcp'); |
822 | socket(SOCK, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
823 | connect(SOCK, $paddr) || die "connect: $!"; |
54310121 |
824 | while (defined($line = <SOCK>)) { |
4633a7c4 |
825 | print $line; |
54310121 |
826 | } |
4633a7c4 |
827 | |
828 | close (SOCK) || die "close: $!"; |
829 | exit; |
830 | |
831 | And here's a corresponding server to go along with it. We'll |
832 | leave the address as INADDR_ANY so that the kernel can choose |
54310121 |
833 | the appropriate interface on multihomed hosts. If you want sit |
c07a80fd |
834 | on a particular interface (like the external side of a gateway |
835 | or firewall machine), you should fill this in with your real address |
836 | instead. |
837 | |
838 | #!/usr/bin/perl -Tw |
c07a80fd |
839 | use strict; |
840 | BEGIN { $ENV{PATH} = '/usr/ucb:/bin' } |
841 | use Socket; |
842 | use Carp; |
5865a7df |
843 | my $EOL = "\015\012"; |
c07a80fd |
844 | |
54310121 |
845 | sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" } |
c07a80fd |
846 | |
847 | my $port = shift || 2345; |
848 | my $proto = getprotobyname('tcp'); |
51ee6500 |
849 | |
5865a7df |
850 | ($port) = $port =~ /^(\d+)$/ or die "invalid port"; |
6a3992aa |
851 | |
c07a80fd |
852 | socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
54310121 |
853 | setsockopt(Server, SOL_SOCKET, SO_REUSEADDR, |
c07a80fd |
854 | pack("l", 1)) || die "setsockopt: $!"; |
855 | bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!"; |
856 | listen(Server,SOMAXCONN) || die "listen: $!"; |
857 | |
858 | logmsg "server started on port $port"; |
859 | |
860 | my $paddr; |
861 | |
862 | $SIG{CHLD} = \&REAPER; |
863 | |
864 | for ( ; $paddr = accept(Client,Server); close Client) { |
865 | my($port,$iaddr) = sockaddr_in($paddr); |
866 | my $name = gethostbyaddr($iaddr,AF_INET); |
867 | |
54310121 |
868 | logmsg "connection from $name [", |
869 | inet_ntoa($iaddr), "] |
c07a80fd |
870 | at port $port"; |
871 | |
54310121 |
872 | print Client "Hello there, $name, it's now ", |
5a964f20 |
873 | scalar localtime, $EOL; |
54310121 |
874 | } |
c07a80fd |
875 | |
54310121 |
876 | And here's a multithreaded version. It's multithreaded in that |
877 | like most typical servers, it spawns (forks) a slave server to |
c07a80fd |
878 | handle the client request so that the master server can quickly |
879 | go back to service a new client. |
4633a7c4 |
880 | |
881 | #!/usr/bin/perl -Tw |
4633a7c4 |
882 | use strict; |
883 | BEGIN { $ENV{PATH} = '/usr/ucb:/bin' } |
a0d0e21e |
884 | use Socket; |
4633a7c4 |
885 | use Carp; |
5865a7df |
886 | my $EOL = "\015\012"; |
a0d0e21e |
887 | |
4633a7c4 |
888 | sub spawn; # forward declaration |
54310121 |
889 | sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" } |
a0d0e21e |
890 | |
4633a7c4 |
891 | my $port = shift || 2345; |
892 | my $proto = getprotobyname('tcp'); |
51ee6500 |
893 | |
5865a7df |
894 | ($port) = $port =~ /^(\d+)$/ or die "invalid port"; |
54310121 |
895 | |
c07a80fd |
896 | socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
54310121 |
897 | setsockopt(Server, SOL_SOCKET, SO_REUSEADDR, |
c07a80fd |
898 | pack("l", 1)) || die "setsockopt: $!"; |
899 | bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!"; |
900 | listen(Server,SOMAXCONN) || die "listen: $!"; |
a0d0e21e |
901 | |
4633a7c4 |
902 | logmsg "server started on port $port"; |
a0d0e21e |
903 | |
4633a7c4 |
904 | my $waitedpid = 0; |
905 | my $paddr; |
a0d0e21e |
906 | |
816229cf |
907 | use POSIX ":sys_wait_h"; |
54310121 |
908 | sub REAPER { |
816229cf |
909 | my $child; |
910 | while (($waitedpid = waitpid(-1,WNOHANG)) > 0) { |
911 | logmsg "reaped $waitedpid" . ($? ? " with exit $?" : ''); |
912 | } |
6a3992aa |
913 | $SIG{CHLD} = \&REAPER; # loathe sysV |
4633a7c4 |
914 | } |
915 | |
916 | $SIG{CHLD} = \&REAPER; |
917 | |
54310121 |
918 | for ( $waitedpid = 0; |
919 | ($paddr = accept(Client,Server)) || $waitedpid; |
920 | $waitedpid = 0, close Client) |
4633a7c4 |
921 | { |
6a3992aa |
922 | next if $waitedpid and not $paddr; |
4633a7c4 |
923 | my($port,$iaddr) = sockaddr_in($paddr); |
924 | my $name = gethostbyaddr($iaddr,AF_INET); |
925 | |
54310121 |
926 | logmsg "connection from $name [", |
927 | inet_ntoa($iaddr), "] |
4633a7c4 |
928 | at port $port"; |
a0d0e21e |
929 | |
54310121 |
930 | spawn sub { |
b921b357 |
931 | $|=1; |
5a964f20 |
932 | print "Hello there, $name, it's now ", scalar localtime, $EOL; |
933 | exec '/usr/games/fortune' # XXX: `wrong' line terminators |
4633a7c4 |
934 | or confess "can't exec fortune: $!"; |
935 | }; |
a0d0e21e |
936 | |
54310121 |
937 | } |
a0d0e21e |
938 | |
4633a7c4 |
939 | sub spawn { |
940 | my $coderef = shift; |
a0d0e21e |
941 | |
54310121 |
942 | unless (@_ == 0 && $coderef && ref($coderef) eq 'CODE') { |
4633a7c4 |
943 | confess "usage: spawn CODEREF"; |
a0d0e21e |
944 | } |
4633a7c4 |
945 | |
946 | my $pid; |
947 | if (!defined($pid = fork)) { |
948 | logmsg "cannot fork: $!"; |
949 | return; |
950 | } elsif ($pid) { |
951 | logmsg "begat $pid"; |
6a3992aa |
952 | return; # I'm the parent |
4633a7c4 |
953 | } |
6a3992aa |
954 | # else I'm the child -- go spawn |
4633a7c4 |
955 | |
c07a80fd |
956 | open(STDIN, "<&Client") || die "can't dup client to stdin"; |
957 | open(STDOUT, ">&Client") || die "can't dup client to stdout"; |
4633a7c4 |
958 | ## open(STDERR, ">&STDOUT") || die "can't dup stdout to stderr"; |
959 | exit &$coderef(); |
54310121 |
960 | } |
4633a7c4 |
961 | |
962 | This server takes the trouble to clone off a child version via fork() for |
963 | each incoming request. That way it can handle many requests at once, |
964 | which you might not always want. Even if you don't fork(), the listen() |
965 | will allow that many pending connections. Forking servers have to be |
966 | particularly careful about cleaning up their dead children (called |
967 | "zombies" in Unix parlance), because otherwise you'll quickly fill up your |
968 | process table. |
969 | |
970 | We suggest that you use the B<-T> flag to use taint checking (see L<perlsec>) |
971 | even if we aren't running setuid or setgid. This is always a good idea |
972 | for servers and other programs run on behalf of someone else (like CGI |
973 | scripts), because it lessens the chances that people from the outside will |
974 | be able to compromise your system. |
975 | |
976 | Let's look at another TCP client. This one connects to the TCP "time" |
977 | service on a number of different machines and shows how far their clocks |
978 | differ from the system on which it's being run: |
979 | |
980 | #!/usr/bin/perl -w |
4633a7c4 |
981 | use strict; |
982 | use Socket; |
983 | |
984 | my $SECS_of_70_YEARS = 2208988800; |
54310121 |
985 | sub ctime { scalar localtime(shift) } |
4633a7c4 |
986 | |
54310121 |
987 | my $iaddr = gethostbyname('localhost'); |
988 | my $proto = getprotobyname('tcp'); |
989 | my $port = getservbyname('time', 'tcp'); |
4633a7c4 |
990 | my $paddr = sockaddr_in(0, $iaddr); |
991 | my($host); |
992 | |
993 | $| = 1; |
994 | printf "%-24s %8s %s\n", "localhost", 0, ctime(time()); |
995 | |
996 | foreach $host (@ARGV) { |
997 | printf "%-24s ", $host; |
998 | my $hisiaddr = inet_aton($host) || die "unknown host"; |
999 | my $hispaddr = sockaddr_in($port, $hisiaddr); |
1000 | socket(SOCKET, PF_INET, SOCK_STREAM, $proto) || die "socket: $!"; |
1001 | connect(SOCKET, $hispaddr) || die "bind: $!"; |
1002 | my $rtime = ' '; |
1003 | read(SOCKET, $rtime, 4); |
1004 | close(SOCKET); |
1005 | my $histime = unpack("N", $rtime) - $SECS_of_70_YEARS ; |
1006 | printf "%8d %s\n", $histime - time, ctime($histime); |
a0d0e21e |
1007 | } |
1008 | |
4633a7c4 |
1009 | =head2 Unix-Domain TCP Clients and Servers |
1010 | |
a2eb9003 |
1011 | That's fine for Internet-domain clients and servers, but what about local |
4633a7c4 |
1012 | communications? While you can use the same setup, sometimes you don't |
1013 | want to. Unix-domain sockets are local to the current host, and are often |
54310121 |
1014 | used internally to implement pipes. Unlike Internet domain sockets, Unix |
4633a7c4 |
1015 | domain sockets can show up in the file system with an ls(1) listing. |
1016 | |
5a964f20 |
1017 | % ls -l /dev/log |
4633a7c4 |
1018 | srw-rw-rw- 1 root 0 Oct 31 07:23 /dev/log |
a0d0e21e |
1019 | |
4633a7c4 |
1020 | You can test for these with Perl's B<-S> file test: |
1021 | |
1022 | unless ( -S '/dev/log' ) { |
3ba19564 |
1023 | die "something's wicked with the log system"; |
54310121 |
1024 | } |
4633a7c4 |
1025 | |
1026 | Here's a sample Unix-domain client: |
1027 | |
1028 | #!/usr/bin/perl -w |
4633a7c4 |
1029 | use Socket; |
1030 | use strict; |
1031 | my ($rendezvous, $line); |
1032 | |
2359510d |
1033 | $rendezvous = shift || 'catsock'; |
4633a7c4 |
1034 | socket(SOCK, PF_UNIX, SOCK_STREAM, 0) || die "socket: $!"; |
9607fc9c |
1035 | connect(SOCK, sockaddr_un($rendezvous)) || die "connect: $!"; |
54310121 |
1036 | while (defined($line = <SOCK>)) { |
4633a7c4 |
1037 | print $line; |
54310121 |
1038 | } |
4633a7c4 |
1039 | exit; |
1040 | |
5a964f20 |
1041 | And here's a corresponding server. You don't have to worry about silly |
1042 | network terminators here because Unix domain sockets are guaranteed |
1043 | to be on the localhost, and thus everything works right. |
4633a7c4 |
1044 | |
1045 | #!/usr/bin/perl -Tw |
4633a7c4 |
1046 | use strict; |
1047 | use Socket; |
1048 | use Carp; |
1049 | |
1050 | BEGIN { $ENV{PATH} = '/usr/ucb:/bin' } |
5865a7df |
1051 | sub spawn; # forward declaration |
5a964f20 |
1052 | sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" } |
4633a7c4 |
1053 | |
2359510d |
1054 | my $NAME = 'catsock'; |
4633a7c4 |
1055 | my $uaddr = sockaddr_un($NAME); |
1056 | my $proto = getprotobyname('tcp'); |
1057 | |
c07a80fd |
1058 | socket(Server,PF_UNIX,SOCK_STREAM,0) || die "socket: $!"; |
4633a7c4 |
1059 | unlink($NAME); |
c07a80fd |
1060 | bind (Server, $uaddr) || die "bind: $!"; |
1061 | listen(Server,SOMAXCONN) || die "listen: $!"; |
4633a7c4 |
1062 | |
1063 | logmsg "server started on $NAME"; |
1064 | |
5a964f20 |
1065 | my $waitedpid; |
1066 | |
816229cf |
1067 | use POSIX ":sys_wait_h"; |
5a964f20 |
1068 | sub REAPER { |
816229cf |
1069 | my $child; |
1070 | while (($waitedpid = waitpid(-1,WNOHANG)) > 0) { |
1071 | logmsg "reaped $waitedpid" . ($? ? " with exit $?" : ''); |
1072 | } |
5a964f20 |
1073 | $SIG{CHLD} = \&REAPER; # loathe sysV |
5a964f20 |
1074 | } |
1075 | |
4633a7c4 |
1076 | $SIG{CHLD} = \&REAPER; |
1077 | |
5a964f20 |
1078 | |
54310121 |
1079 | for ( $waitedpid = 0; |
1080 | accept(Client,Server) || $waitedpid; |
1081 | $waitedpid = 0, close Client) |
4633a7c4 |
1082 | { |
1083 | next if $waitedpid; |
1084 | logmsg "connection on $NAME"; |
54310121 |
1085 | spawn sub { |
4633a7c4 |
1086 | print "Hello there, it's now ", scalar localtime, "\n"; |
1087 | exec '/usr/games/fortune' or die "can't exec fortune: $!"; |
1088 | }; |
54310121 |
1089 | } |
4633a7c4 |
1090 | |
5865a7df |
1091 | sub spawn { |
1092 | my $coderef = shift; |
1093 | |
1094 | unless (@_ == 0 && $coderef && ref($coderef) eq 'CODE') { |
1095 | confess "usage: spawn CODEREF"; |
1096 | } |
1097 | |
1098 | my $pid; |
1099 | if (!defined($pid = fork)) { |
1100 | logmsg "cannot fork: $!"; |
1101 | return; |
1102 | } elsif ($pid) { |
1103 | logmsg "begat $pid"; |
1104 | return; # I'm the parent |
1105 | } |
1106 | # else I'm the child -- go spawn |
1107 | |
1108 | open(STDIN, "<&Client") || die "can't dup client to stdin"; |
1109 | open(STDOUT, ">&Client") || die "can't dup client to stdout"; |
1110 | ## open(STDERR, ">&STDOUT") || die "can't dup stdout to stderr"; |
1111 | exit &$coderef(); |
1112 | } |
1113 | |
4633a7c4 |
1114 | As you see, it's remarkably similar to the Internet domain TCP server, so |
1115 | much so, in fact, that we've omitted several duplicate functions--spawn(), |
1116 | logmsg(), ctime(), and REAPER()--which are exactly the same as in the |
1117 | other server. |
1118 | |
1119 | So why would you ever want to use a Unix domain socket instead of a |
1120 | simpler named pipe? Because a named pipe doesn't give you sessions. You |
1121 | can't tell one process's data from another's. With socket programming, |
1122 | you get a separate session for each client: that's why accept() takes two |
1123 | arguments. |
1124 | |
1125 | For example, let's say that you have a long running database server daemon |
1126 | that you want folks from the World Wide Web to be able to access, but only |
1127 | if they go through a CGI interface. You'd have a small, simple CGI |
1128 | program that does whatever checks and logging you feel like, and then acts |
1129 | as a Unix-domain client and connects to your private server. |
1130 | |
7b05b7e3 |
1131 | =head1 TCP Clients with IO::Socket |
1132 | |
1133 | For those preferring a higher-level interface to socket programming, the |
1134 | IO::Socket module provides an object-oriented approach. IO::Socket is |
1135 | included as part of the standard Perl distribution as of the 5.004 |
1136 | release. If you're running an earlier version of Perl, just fetch |
106325ad |
1137 | IO::Socket from CPAN, where you'll also find modules providing easy |
7b05b7e3 |
1138 | interfaces to the following systems: DNS, FTP, Ident (RFC 931), NIS and |
1139 | NISPlus, NNTP, Ping, POP3, SMTP, SNMP, SSLeay, Telnet, and Time--just |
1140 | to name a few. |
1141 | |
1142 | =head2 A Simple Client |
1143 | |
1144 | Here's a client that creates a TCP connection to the "daytime" |
1145 | service at port 13 of the host name "localhost" and prints out everything |
1146 | that the server there cares to provide. |
1147 | |
1148 | #!/usr/bin/perl -w |
1149 | use IO::Socket; |
1150 | $remote = IO::Socket::INET->new( |
1151 | Proto => "tcp", |
1152 | PeerAddr => "localhost", |
1153 | PeerPort => "daytime(13)", |
1154 | ) |
1155 | or die "cannot connect to daytime port at localhost"; |
1156 | while ( <$remote> ) { print } |
1157 | |
1158 | When you run this program, you should get something back that |
1159 | looks like this: |
1160 | |
1161 | Wed May 14 08:40:46 MDT 1997 |
1162 | |
1163 | Here are what those parameters to the C<new> constructor mean: |
1164 | |
13a2d996 |
1165 | =over 4 |
7b05b7e3 |
1166 | |
1167 | =item C<Proto> |
1168 | |
1169 | This is which protocol to use. In this case, the socket handle returned |
1170 | will be connected to a TCP socket, because we want a stream-oriented |
1171 | connection, that is, one that acts pretty much like a plain old file. |
1172 | Not all sockets are this of this type. For example, the UDP protocol |
1173 | can be used to make a datagram socket, used for message-passing. |
1174 | |
1175 | =item C<PeerAddr> |
1176 | |
1177 | This is the name or Internet address of the remote host the server is |
1178 | running on. We could have specified a longer name like C<"www.perl.com">, |
1179 | or an address like C<"204.148.40.9">. For demonstration purposes, we've |
1180 | used the special hostname C<"localhost">, which should always mean the |
1181 | current machine you're running on. The corresponding Internet address |
1182 | for localhost is C<"127.1">, if you'd rather use that. |
1183 | |
1184 | =item C<PeerPort> |
1185 | |
1186 | This is the service name or port number we'd like to connect to. |
1187 | We could have gotten away with using just C<"daytime"> on systems with a |
1188 | well-configured system services file,[FOOTNOTE: The system services file |
1189 | is in I</etc/services> under Unix] but just in case, we've specified the |
1190 | port number (13) in parentheses. Using just the number would also have |
1191 | worked, but constant numbers make careful programmers nervous. |
1192 | |
1193 | =back |
1194 | |
1195 | Notice how the return value from the C<new> constructor is used as |
1196 | a filehandle in the C<while> loop? That's what's called an indirect |
1197 | filehandle, a scalar variable containing a filehandle. You can use |
1198 | it the same way you would a normal filehandle. For example, you |
1199 | can read one line from it this way: |
1200 | |
1201 | $line = <$handle>; |
1202 | |
1203 | all remaining lines from is this way: |
1204 | |
1205 | @lines = <$handle>; |
1206 | |
1207 | and send a line of data to it this way: |
1208 | |
1209 | print $handle "some data\n"; |
1210 | |
1211 | =head2 A Webget Client |
1212 | |
1213 | Here's a simple client that takes a remote host to fetch a document |
1214 | from, and then a list of documents to get from that host. This is a |
1215 | more interesting client than the previous one because it first sends |
1216 | something to the server before fetching the server's response. |
1217 | |
1218 | #!/usr/bin/perl -w |
1219 | use IO::Socket; |
1220 | unless (@ARGV > 1) { die "usage: $0 host document ..." } |
1221 | $host = shift(@ARGV); |
5a964f20 |
1222 | $EOL = "\015\012"; |
1223 | $BLANK = $EOL x 2; |
7b05b7e3 |
1224 | foreach $document ( @ARGV ) { |
1225 | $remote = IO::Socket::INET->new( Proto => "tcp", |
1226 | PeerAddr => $host, |
1227 | PeerPort => "http(80)", |
1228 | ); |
1229 | unless ($remote) { die "cannot connect to http daemon on $host" } |
1230 | $remote->autoflush(1); |
5a964f20 |
1231 | print $remote "GET $document HTTP/1.0" . $BLANK; |
7b05b7e3 |
1232 | while ( <$remote> ) { print } |
1233 | close $remote; |
1234 | } |
1235 | |
1236 | The web server handing the "http" service, which is assumed to be at |
4375e838 |
1237 | its standard port, number 80. If the web server you're trying to |
7b05b7e3 |
1238 | connect to is at a different port (like 1080 or 8080), you should specify |
c47ff5f1 |
1239 | as the named-parameter pair, C<< PeerPort => 8080 >>. The C<autoflush> |
7b05b7e3 |
1240 | method is used on the socket because otherwise the system would buffer |
1241 | up the output we sent it. (If you're on a Mac, you'll also need to |
1242 | change every C<"\n"> in your code that sends data over the network to |
1243 | be a C<"\015\012"> instead.) |
1244 | |
1245 | Connecting to the server is only the first part of the process: once you |
1246 | have the connection, you have to use the server's language. Each server |
1247 | on the network has its own little command language that it expects as |
1248 | input. The string that we send to the server starting with "GET" is in |
1249 | HTTP syntax. In this case, we simply request each specified document. |
1250 | Yes, we really are making a new connection for each document, even though |
1251 | it's the same host. That's the way you always used to have to speak HTTP. |
1252 | Recent versions of web browsers may request that the remote server leave |
1253 | the connection open a little while, but the server doesn't have to honor |
1254 | such a request. |
1255 | |
1256 | Here's an example of running that program, which we'll call I<webget>: |
1257 | |
5a964f20 |
1258 | % webget www.perl.com /guanaco.html |
7b05b7e3 |
1259 | HTTP/1.1 404 File Not Found |
1260 | Date: Thu, 08 May 1997 18:02:32 GMT |
1261 | Server: Apache/1.2b6 |
1262 | Connection: close |
1263 | Content-type: text/html |
1264 | |
1265 | <HEAD><TITLE>404 File Not Found</TITLE></HEAD> |
1266 | <BODY><H1>File Not Found</H1> |
1267 | The requested URL /guanaco.html was not found on this server.<P> |
1268 | </BODY> |
1269 | |
1270 | Ok, so that's not very interesting, because it didn't find that |
1271 | particular document. But a long response wouldn't have fit on this page. |
1272 | |
1273 | For a more fully-featured version of this program, you should look to |
1274 | the I<lwp-request> program included with the LWP modules from CPAN. |
1275 | |
1276 | =head2 Interactive Client with IO::Socket |
1277 | |
1278 | Well, that's all fine if you want to send one command and get one answer, |
1279 | but what about setting up something fully interactive, somewhat like |
1280 | the way I<telnet> works? That way you can type a line, get the answer, |
1281 | type a line, get the answer, etc. |
1282 | |
1283 | This client is more complicated than the two we've done so far, but if |
1284 | you're on a system that supports the powerful C<fork> call, the solution |
1285 | isn't that rough. Once you've made the connection to whatever service |
1286 | you'd like to chat with, call C<fork> to clone your process. Each of |
1287 | these two identical process has a very simple job to do: the parent |
1288 | copies everything from the socket to standard output, while the child |
1289 | simultaneously copies everything from standard input to the socket. |
1290 | To accomplish the same thing using just one process would be I<much> |
1291 | harder, because it's easier to code two processes to do one thing than it |
1292 | is to code one process to do two things. (This keep-it-simple principle |
5a964f20 |
1293 | a cornerstones of the Unix philosophy, and good software engineering as |
1294 | well, which is probably why it's spread to other systems.) |
7b05b7e3 |
1295 | |
1296 | Here's the code: |
1297 | |
1298 | #!/usr/bin/perl -w |
1299 | use strict; |
1300 | use IO::Socket; |
1301 | my ($host, $port, $kidpid, $handle, $line); |
1302 | |
1303 | unless (@ARGV == 2) { die "usage: $0 host port" } |
1304 | ($host, $port) = @ARGV; |
1305 | |
1306 | # create a tcp connection to the specified host and port |
1307 | $handle = IO::Socket::INET->new(Proto => "tcp", |
1308 | PeerAddr => $host, |
1309 | PeerPort => $port) |
1310 | or die "can't connect to port $port on $host: $!"; |
1311 | |
1312 | $handle->autoflush(1); # so output gets there right away |
1313 | print STDERR "[Connected to $host:$port]\n"; |
1314 | |
1315 | # split the program into two processes, identical twins |
1316 | die "can't fork: $!" unless defined($kidpid = fork()); |
1317 | |
1318 | # the if{} block runs only in the parent process |
1319 | if ($kidpid) { |
1320 | # copy the socket to standard output |
1321 | while (defined ($line = <$handle>)) { |
1322 | print STDOUT $line; |
1323 | } |
1324 | kill("TERM", $kidpid); # send SIGTERM to child |
1325 | } |
1326 | # the else{} block runs only in the child process |
1327 | else { |
1328 | # copy standard input to the socket |
1329 | while (defined ($line = <STDIN>)) { |
1330 | print $handle $line; |
1331 | } |
1332 | } |
1333 | |
1334 | The C<kill> function in the parent's C<if> block is there to send a |
1335 | signal to our child process (current running in the C<else> block) |
1336 | as soon as the remote server has closed its end of the connection. |
1337 | |
7b05b7e3 |
1338 | If the remote server sends data a byte at time, and you need that |
1339 | data immediately without waiting for a newline (which might not happen), |
1340 | you may wish to replace the C<while> loop in the parent with the |
1341 | following: |
1342 | |
1343 | my $byte; |
1344 | while (sysread($handle, $byte, 1) == 1) { |
1345 | print STDOUT $byte; |
1346 | } |
1347 | |
1348 | Making a system call for each byte you want to read is not very efficient |
1349 | (to put it mildly) but is the simplest to explain and works reasonably |
1350 | well. |
1351 | |
1352 | =head1 TCP Servers with IO::Socket |
1353 | |
5a964f20 |
1354 | As always, setting up a server is little bit more involved than running a client. |
7b05b7e3 |
1355 | The model is that the server creates a special kind of socket that |
1356 | does nothing but listen on a particular port for incoming connections. |
c47ff5f1 |
1357 | It does this by calling the C<< IO::Socket::INET->new() >> method with |
7b05b7e3 |
1358 | slightly different arguments than the client did. |
1359 | |
13a2d996 |
1360 | =over 4 |
7b05b7e3 |
1361 | |
1362 | =item Proto |
1363 | |
1364 | This is which protocol to use. Like our clients, we'll |
1365 | still specify C<"tcp"> here. |
1366 | |
1367 | =item LocalPort |
1368 | |
1369 | We specify a local |
1370 | port in the C<LocalPort> argument, which we didn't do for the client. |
1371 | This is service name or port number for which you want to be the |
1372 | server. (Under Unix, ports under 1024 are restricted to the |
1373 | superuser.) In our sample, we'll use port 9000, but you can use |
1374 | any port that's not currently in use on your system. If you try |
1375 | to use one already in used, you'll get an "Address already in use" |
19799a22 |
1376 | message. Under Unix, the C<netstat -a> command will show |
7b05b7e3 |
1377 | which services current have servers. |
1378 | |
1379 | =item Listen |
1380 | |
1381 | The C<Listen> parameter is set to the maximum number of |
1382 | pending connections we can accept until we turn away incoming clients. |
1383 | Think of it as a call-waiting queue for your telephone. |
1384 | The low-level Socket module has a special symbol for the system maximum, which |
1385 | is SOMAXCONN. |
1386 | |
1387 | =item Reuse |
1388 | |
1389 | The C<Reuse> parameter is needed so that we restart our server |
1390 | manually without waiting a few minutes to allow system buffers to |
1391 | clear out. |
1392 | |
1393 | =back |
1394 | |
1395 | Once the generic server socket has been created using the parameters |
1396 | listed above, the server then waits for a new client to connect |
d1be9408 |
1397 | to it. The server blocks in the C<accept> method, which eventually accepts a |
1398 | bidirectional connection from the remote client. (Make sure to autoflush |
7b05b7e3 |
1399 | this handle to circumvent buffering.) |
1400 | |
1401 | To add to user-friendliness, our server prompts the user for commands. |
1402 | Most servers don't do this. Because of the prompt without a newline, |
1403 | you'll have to use the C<sysread> variant of the interactive client above. |
1404 | |
1405 | This server accepts one of five different commands, sending output |
1406 | back to the client. Note that unlike most network servers, this one |
1407 | only handles one incoming client at a time. Multithreaded servers are |
f83494b9 |
1408 | covered in Chapter 6 of the Camel. |
7b05b7e3 |
1409 | |
1410 | Here's the code. We'll |
1411 | |
1412 | #!/usr/bin/perl -w |
1413 | use IO::Socket; |
1414 | use Net::hostent; # for OO version of gethostbyaddr |
1415 | |
1416 | $PORT = 9000; # pick something not in use |
1417 | |
1418 | $server = IO::Socket::INET->new( Proto => 'tcp', |
1419 | LocalPort => $PORT, |
1420 | Listen => SOMAXCONN, |
1421 | Reuse => 1); |
1422 | |
1423 | die "can't setup server" unless $server; |
1424 | print "[Server $0 accepting clients]\n"; |
1425 | |
1426 | while ($client = $server->accept()) { |
1427 | $client->autoflush(1); |
1428 | print $client "Welcome to $0; type help for command list.\n"; |
1429 | $hostinfo = gethostbyaddr($client->peeraddr); |
78fc38e1 |
1430 | printf "[Connect from %s]\n", $hostinfo ? $hostinfo->name : $client->peerhost; |
7b05b7e3 |
1431 | print $client "Command? "; |
1432 | while ( <$client>) { |
1433 | next unless /\S/; # blank line |
1434 | if (/quit|exit/i) { last; } |
1435 | elsif (/date|time/i) { printf $client "%s\n", scalar localtime; } |
1436 | elsif (/who/i ) { print $client `who 2>&1`; } |
1437 | elsif (/cookie/i ) { print $client `/usr/games/fortune 2>&1`; } |
1438 | elsif (/motd/i ) { print $client `cat /etc/motd 2>&1`; } |
1439 | else { |
1440 | print $client "Commands: quit date who cookie motd\n"; |
1441 | } |
1442 | } continue { |
1443 | print $client "Command? "; |
1444 | } |
1445 | close $client; |
1446 | } |
1447 | |
1448 | =head1 UDP: Message Passing |
4633a7c4 |
1449 | |
1450 | Another kind of client-server setup is one that uses not connections, but |
1451 | messages. UDP communications involve much lower overhead but also provide |
1452 | less reliability, as there are no promises that messages will arrive at |
1453 | all, let alone in order and unmangled. Still, UDP offers some advantages |
1454 | over TCP, including being able to "broadcast" or "multicast" to a whole |
1455 | bunch of destination hosts at once (usually on your local subnet). If you |
1456 | find yourself overly concerned about reliability and start building checks |
6a3992aa |
1457 | into your message system, then you probably should use just TCP to start |
4633a7c4 |
1458 | with. |
1459 | |
90034919 |
1460 | Note that UDP datagrams are I<not> a bytestream and should not be treated |
1461 | as such. This makes using I/O mechanisms with internal buffering |
1462 | like stdio (i.e. print() and friends) especially cumbersome. Use syswrite(), |
1463 | or better send(), like in the example below. |
1464 | |
4633a7c4 |
1465 | Here's a UDP program similar to the sample Internet TCP client given |
7b05b7e3 |
1466 | earlier. However, instead of checking one host at a time, the UDP version |
4633a7c4 |
1467 | will check many of them asynchronously by simulating a multicast and then |
1468 | using select() to do a timed-out wait for I/O. To do something similar |
1469 | with TCP, you'd have to use a different socket handle for each host. |
1470 | |
1471 | #!/usr/bin/perl -w |
1472 | use strict; |
4633a7c4 |
1473 | use Socket; |
1474 | use Sys::Hostname; |
1475 | |
54310121 |
1476 | my ( $count, $hisiaddr, $hispaddr, $histime, |
1477 | $host, $iaddr, $paddr, $port, $proto, |
4633a7c4 |
1478 | $rin, $rout, $rtime, $SECS_of_70_YEARS); |
1479 | |
1480 | $SECS_of_70_YEARS = 2208988800; |
1481 | |
1482 | $iaddr = gethostbyname(hostname()); |
1483 | $proto = getprotobyname('udp'); |
1484 | $port = getservbyname('time', 'udp'); |
1485 | $paddr = sockaddr_in(0, $iaddr); # 0 means let kernel pick |
1486 | |
1487 | socket(SOCKET, PF_INET, SOCK_DGRAM, $proto) || die "socket: $!"; |
1488 | bind(SOCKET, $paddr) || die "bind: $!"; |
1489 | |
1490 | $| = 1; |
1491 | printf "%-12s %8s %s\n", "localhost", 0, scalar localtime time; |
1492 | $count = 0; |
1493 | for $host (@ARGV) { |
1494 | $count++; |
1495 | $hisiaddr = inet_aton($host) || die "unknown host"; |
1496 | $hispaddr = sockaddr_in($port, $hisiaddr); |
1497 | defined(send(SOCKET, 0, 0, $hispaddr)) || die "send $host: $!"; |
1498 | } |
1499 | |
1500 | $rin = ''; |
1501 | vec($rin, fileno(SOCKET), 1) = 1; |
1502 | |
1503 | # timeout after 10.0 seconds |
1504 | while ($count && select($rout = $rin, undef, undef, 10.0)) { |
1505 | $rtime = ''; |
1506 | ($hispaddr = recv(SOCKET, $rtime, 4, 0)) || die "recv: $!"; |
1507 | ($port, $hisiaddr) = sockaddr_in($hispaddr); |
1508 | $host = gethostbyaddr($hisiaddr, AF_INET); |
1509 | $histime = unpack("N", $rtime) - $SECS_of_70_YEARS ; |
1510 | printf "%-12s ", $host; |
1511 | printf "%8d %s\n", $histime - time, scalar localtime($histime); |
1512 | $count--; |
1513 | } |
1514 | |
90034919 |
1515 | Note that this example does not include any retries and may consequently |
1516 | fail to contact a reachable host. The most prominent reason for this |
1517 | is congestion of the queues on the sending host if the number of |
a31a806a |
1518 | list of hosts to contact is sufficiently large. |
90034919 |
1519 | |
4633a7c4 |
1520 | =head1 SysV IPC |
1521 | |
1522 | While System V IPC isn't so widely used as sockets, it still has some |
1523 | interesting uses. You can't, however, effectively use SysV IPC or |
1524 | Berkeley mmap() to have shared memory so as to share a variable amongst |
1525 | several processes. That's because Perl would reallocate your string when |
1526 | you weren't wanting it to. |
1527 | |
54310121 |
1528 | Here's a small example showing shared memory usage. |
a0d0e21e |
1529 | |
41d6edb2 |
1530 | use IPC::SysV qw(IPC_PRIVATE IPC_RMID S_IRWXU); |
0ade1984 |
1531 | |
a0d0e21e |
1532 | $size = 2000; |
41d6edb2 |
1533 | $id = shmget(IPC_PRIVATE, $size, S_IRWXU) || die "$!"; |
1534 | print "shm key $id\n"; |
a0d0e21e |
1535 | |
1536 | $message = "Message #1"; |
41d6edb2 |
1537 | shmwrite($id, $message, 0, 60) || die "$!"; |
0ade1984 |
1538 | print "wrote: '$message'\n"; |
41d6edb2 |
1539 | shmread($id, $buff, 0, 60) || die "$!"; |
0ade1984 |
1540 | print "read : '$buff'\n"; |
a0d0e21e |
1541 | |
0ade1984 |
1542 | # the buffer of shmread is zero-character end-padded. |
1543 | substr($buff, index($buff, "\0")) = ''; |
1544 | print "un" unless $buff eq $message; |
1545 | print "swell\n"; |
a0d0e21e |
1546 | |
41d6edb2 |
1547 | print "deleting shm $id\n"; |
1548 | shmctl($id, IPC_RMID, 0) || die "$!"; |
a0d0e21e |
1549 | |
1550 | Here's an example of a semaphore: |
1551 | |
0ade1984 |
1552 | use IPC::SysV qw(IPC_CREAT); |
1553 | |
a0d0e21e |
1554 | $IPC_KEY = 1234; |
41d6edb2 |
1555 | $id = semget($IPC_KEY, 10, 0666 | IPC_CREAT ) || die "$!"; |
1556 | print "shm key $id\n"; |
a0d0e21e |
1557 | |
a2eb9003 |
1558 | Put this code in a separate file to be run in more than one process. |
a0d0e21e |
1559 | Call the file F<take>: |
1560 | |
1561 | # create a semaphore |
1562 | |
1563 | $IPC_KEY = 1234; |
41d6edb2 |
1564 | $id = semget($IPC_KEY, 0 , 0 ); |
1565 | die if !defined($id); |
a0d0e21e |
1566 | |
1567 | $semnum = 0; |
1568 | $semflag = 0; |
1569 | |
1570 | # 'take' semaphore |
1571 | # wait for semaphore to be zero |
1572 | $semop = 0; |
41d6edb2 |
1573 | $opstring1 = pack("s!s!s!", $semnum, $semop, $semflag); |
a0d0e21e |
1574 | |
1575 | # Increment the semaphore count |
1576 | $semop = 1; |
41d6edb2 |
1577 | $opstring2 = pack("s!s!s!", $semnum, $semop, $semflag); |
a0d0e21e |
1578 | $opstring = $opstring1 . $opstring2; |
1579 | |
41d6edb2 |
1580 | semop($id,$opstring) || die "$!"; |
a0d0e21e |
1581 | |
a2eb9003 |
1582 | Put this code in a separate file to be run in more than one process. |
a0d0e21e |
1583 | Call this file F<give>: |
1584 | |
4633a7c4 |
1585 | # 'give' the semaphore |
a0d0e21e |
1586 | # run this in the original process and you will see |
1587 | # that the second process continues |
1588 | |
1589 | $IPC_KEY = 1234; |
41d6edb2 |
1590 | $id = semget($IPC_KEY, 0, 0); |
1591 | die if !defined($id); |
a0d0e21e |
1592 | |
1593 | $semnum = 0; |
1594 | $semflag = 0; |
1595 | |
1596 | # Decrement the semaphore count |
1597 | $semop = -1; |
41d6edb2 |
1598 | $opstring = pack("s!s!s!", $semnum, $semop, $semflag); |
a0d0e21e |
1599 | |
41d6edb2 |
1600 | semop($id,$opstring) || die "$!"; |
a0d0e21e |
1601 | |
7b05b7e3 |
1602 | The SysV IPC code above was written long ago, and it's definitely |
0ade1984 |
1603 | clunky looking. For a more modern look, see the IPC::SysV module |
1604 | which is included with Perl starting from Perl 5.005. |
4633a7c4 |
1605 | |
41d6edb2 |
1606 | A small example demonstrating SysV message queues: |
1607 | |
1608 | use IPC::SysV qw(IPC_PRIVATE IPC_RMID IPC_CREAT S_IRWXU); |
1609 | |
1610 | my $id = msgget(IPC_PRIVATE, IPC_CREAT | S_IRWXU); |
1611 | |
1612 | my $sent = "message"; |
e343e2e2 |
1613 | my $type_sent = 1234; |
41d6edb2 |
1614 | my $rcvd; |
1615 | my $type_rcvd; |
1616 | |
1617 | if (defined $id) { |
1618 | if (msgsnd($id, pack("l! a*", $type_sent, $sent), 0)) { |
1619 | if (msgrcv($id, $rcvd, 60, 0, 0)) { |
1620 | ($type_rcvd, $rcvd) = unpack("l! a*", $rcvd); |
1621 | if ($rcvd eq $sent) { |
1622 | print "okay\n"; |
1623 | } else { |
1624 | print "not okay\n"; |
1625 | } |
1626 | } else { |
1627 | die "# msgrcv failed\n"; |
1628 | } |
1629 | } else { |
1630 | die "# msgsnd failed\n"; |
1631 | } |
1632 | msgctl($id, IPC_RMID, 0) || die "# msgctl failed: $!\n"; |
1633 | } else { |
1634 | die "# msgget failed\n"; |
1635 | } |
1636 | |
4633a7c4 |
1637 | =head1 NOTES |
1638 | |
5a964f20 |
1639 | Most of these routines quietly but politely return C<undef> when they |
1640 | fail instead of causing your program to die right then and there due to |
1641 | an uncaught exception. (Actually, some of the new I<Socket> conversion |
1642 | functions croak() on bad arguments.) It is therefore essential to |
1643 | check return values from these functions. Always begin your socket |
1644 | programs this way for optimal success, and don't forget to add B<-T> |
1645 | taint checking flag to the #! line for servers: |
4633a7c4 |
1646 | |
5a964f20 |
1647 | #!/usr/bin/perl -Tw |
4633a7c4 |
1648 | use strict; |
1649 | use sigtrap; |
1650 | use Socket; |
1651 | |
1652 | =head1 BUGS |
1653 | |
1654 | All these routines create system-specific portability problems. As noted |
1655 | elsewhere, Perl is at the mercy of your C libraries for much of its system |
1656 | behaviour. It's probably safest to assume broken SysV semantics for |
6a3992aa |
1657 | signals and to stick with simple TCP and UDP socket operations; e.g., don't |
a2eb9003 |
1658 | try to pass open file descriptors over a local UDP datagram socket if you |
4633a7c4 |
1659 | want your code to stand a chance of being portable. |
1660 | |
4633a7c4 |
1661 | =head1 AUTHOR |
1662 | |
1663 | Tom Christiansen, with occasional vestiges of Larry Wall's original |
7b05b7e3 |
1664 | version and suggestions from the Perl Porters. |
4633a7c4 |
1665 | |
1666 | =head1 SEE ALSO |
1667 | |
7b05b7e3 |
1668 | There's a lot more to networking than this, but this should get you |
1669 | started. |
1670 | |
c04e1326 |
1671 | For intrepid programmers, the indispensable textbook is I<Unix |
1672 | Network Programming, 2nd Edition, Volume 1> by W. Richard Stevens |
1673 | (published by Prentice-Hall). Note that most books on networking |
1674 | address the subject from the perspective of a C programmer; translation |
1675 | to Perl is left as an exercise for the reader. |
7b05b7e3 |
1676 | |
1677 | The IO::Socket(3) manpage describes the object library, and the Socket(3) |
1678 | manpage describes the low-level interface to sockets. Besides the obvious |
1679 | functions in L<perlfunc>, you should also check out the F<modules> file |
1680 | at your nearest CPAN site. (See L<perlmodlib> or best yet, the F<Perl |
1681 | FAQ> for a description of what CPAN is and where to get it.) |
1682 | |
4633a7c4 |
1683 | Section 5 of the F<modules> file is devoted to "Networking, Device Control |
6a3992aa |
1684 | (modems), and Interprocess Communication", and contains numerous unbundled |
4633a7c4 |
1685 | modules numerous networking modules, Chat and Expect operations, CGI |
1686 | programming, DCE, FTP, IPC, NNTP, Proxy, Ptty, RPC, SNMP, SMTP, Telnet, |
1687 | Threads, and ToolTalk--just to name a few. |