3 perlipc - Perl interprocess communication (signals, fifos, pipes, safe subprocesses, sockets, and semaphores)
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.
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.
22 For example, to trap an interrupt signal, set up a handler like this:
27 die "Somebody sent me a SIG$signame";
29 $SIG{INT} = 'catch_zap'; # could fail in modules
30 $SIG{INT} = \&catch_zap; # best strategy
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
38 dump - see L<Deferred Signals> below.
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:
46 defined $Config{sig_name} || die "No sigs?";
47 foreach $name (split(' ', $Config{sig_name})) {
53 So to check whether signal 17 and SIGALRM were the same, do just this:
55 print "signal #17 = $signame[17]\n";
57 print "SIGALRM is $signo{ALRM}\n";
60 You may also choose to assign the strings C<'IGNORE'> or C<'DEFAULT'> as
61 the handler, in which case Perl will try to discard the signal or do the
64 On most Unix platforms, the C<CHLD> (sometimes also known as C<CLD>) signal
65 has special behavior with respect to a value of C<'IGNORE'>.
66 Setting C<$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.
72 Some signals can be neither trapped nor ignored, such as
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.)
79 local $SIG{INT} = 'IGNORE';
83 # interrupts still ignored, for now...
86 Sending a signal to a negative process ID means that you send the signal
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):
92 local $SIG{HUP} = 'IGNORE';
94 # snazzy writing of: kill('HUP', -$$)
97 Another interesting signal to send is signal number zero. This doesn't
98 actually affect another process, but instead checks whether it's alive
99 or has changed its UID.
101 unless (kill 0 => $kid_pid) {
102 warn "something wicked happened to $kid_pid";
105 You might also want to employ anonymous functions for simple signal
108 $SIG{INT} = sub { die "\nOutta here!\n" };
110 But that will be problematic for the more complicated handlers that need
111 to reinstall themselves. Because Perl's signal mechanism is currently
112 based on the signal(3) function from the C library, you may sometimes be so
113 misfortunate as to run on systems where that function is "broken", that
114 is, it behaves in the old unreliable SysV way rather than the newer, more
115 reasonable BSD and POSIX fashion. So you'll see defensive people writing
116 signal handlers like this:
120 # loathe sysV: it makes us not only reinstate
121 # the handler, but place it after the wait
122 $SIG{CHLD} = \&REAPER;
124 $SIG{CHLD} = \&REAPER;
125 # now do something that forks...
129 use POSIX ":sys_wait_h";
132 # If a second child dies while in the signal handler caused by the
133 # first death, we won't get another signal. So must loop here else
134 # we will leave the unreaped child as a zombie. And the next time
135 # two children die we get another zombie. And so on.
136 while (($child = waitpid(-1,WNOHANG)) > 0) {
137 $Kid_Status{$child} = $?;
139 $SIG{CHLD} = \&REAPER; # still loathe sysV
141 $SIG{CHLD} = \&REAPER;
142 # do something that forks...
144 Signal handling is also used for timeouts in Unix, While safely
145 protected within an C<eval{}> block, you set a signal handler to trap
146 alarm signals and then schedule to have one delivered to you in some
147 number of seconds. Then try your blocking operation, clearing the alarm
148 when it's done but not before you've exited your C<eval{}> block. If it
149 goes off, you'll use die() to jump out of the block, much as you might
150 using longjmp() or throw() in other languages.
155 local $SIG{ALRM} = sub { die "alarm clock restart" };
157 flock(FH, 2); # blocking write lock
160 if ($@ and $@ !~ /alarm clock restart/) { die }
162 If the operation being timed out is system() or qx(), this technique
163 is liable to generate zombies. If this matters to you, you'll
164 need to do your own fork() and exec(), and kill the errant child process.
166 For more complex signal handling, you might see the standard POSIX
167 module. Lamentably, this is almost entirely undocumented, but
168 the F<t/lib/posix.t> file from the Perl source distribution has some
171 =head2 Handling the SIGHUP Signal in Daemons
173 A process that usually starts when the system boots and shuts down
174 when the system is shut down is called a daemon (Disk And Execution
175 MONitor). If a daemon process has a configuration file which is
176 modified after the process has been started, there should be a way to
177 tell that process to re-read its configuration file, without stopping
178 the process. Many daemons provide this mechanism using the C<SIGHUP>
179 signal handler. When you want to tell the daemon to re-read the file
180 you simply send it the C<SIGHUP> signal.
182 Not all platforms automatically reinstall their (native) signal
183 handlers after a signal delivery. This means that the handler works
184 only the first time the signal is sent. The solution to this problem
185 is to use C<POSIX> signal handlers if available, their behaviour
188 The following example implements a simple daemon, which restarts
189 itself every time the C<SIGHUP> signal is received. The actual code is
190 located in the subroutine C<code()>, which simply prints some debug
191 info to show that it works and should be replaced with the real code.
197 use File::Basename ();
198 use File::Spec::Functions;
202 # make the daemon cross-platform, so exec always calls the script
203 # itself with the right path, no matter how the script was invoked.
204 my $script = File::Basename::basename($0);
205 my $SELF = catfile $FindBin::Bin, $script;
207 # POSIX unmasks the sigprocmask properly
208 my $sigset = POSIX::SigSet->new();
209 my $action = POSIX::SigAction->new('sigHUP_handler',
212 POSIX::sigaction(&POSIX::SIGHUP, $action);
215 print "got SIGHUP\n";
216 exec($SELF, @ARGV) or die "Couldn't restart: $!\n";
223 print "ARGV: @ARGV\n";
235 A named pipe (often referred to as a FIFO) is an old Unix IPC
236 mechanism for processes communicating on the same machine. It works
237 just like a regular, connected anonymous pipes, except that the
238 processes rendezvous using a filename and don't have to be related.
240 To create a named pipe, use the Unix command mknod(1) or on some
241 systems, mkfifo(1). These may not be in your normal path.
243 # system return val is backwards, so && not ||
245 $ENV{PATH} .= ":/etc:/usr/etc";
246 if ( system('mknod', $path, 'p')
247 && system('mkfifo', $path) )
249 die "mk{nod,fifo} $path failed";
253 A fifo is convenient when you want to connect a process to an unrelated
254 one. When you open a fifo, the program will block until there's something
257 For example, let's say you'd like to have your F<.signature> file be a
258 named pipe that has a Perl program on the other end. Now every time any
259 program (like a mailer, news reader, finger program, etc.) tries to read
260 from that file, the reading program will block and your program will
261 supply the new signature. We'll use the pipe-checking file test B<-p>
262 to find out whether anyone (or anything) has accidentally removed our fifo.
265 $FIFO = '.signature';
266 $ENV{PATH} .= ":/etc:/usr/games";
271 system('mknod', $FIFO, 'p')
272 && die "can't mknod $FIFO: $!";
275 # next line blocks until there's a reader
276 open (FIFO, "> $FIFO") || die "can't write $FIFO: $!";
277 print FIFO "John Smith (smith\@host.org)\n", `fortune -s`;
279 sleep 2; # to avoid dup signals
282 =head2 Deferred Signals
284 In Perls before Perl 5.7.3 by installing Perl code to deal with
285 signals, you were exposing yourself to danger from two things. First,
286 few system library functions are re-entrant. If the signal interrupts
287 while Perl is executing one function (like malloc(3) or printf(3)),
288 and your signal handler then calls the same function again, you could
289 get unpredictable behavior--often, a core dump. Second, Perl isn't
290 itself re-entrant at the lowest levels. If the signal interrupts Perl
291 while Perl is changing its own internal data structures, similarly
292 unpredictable behaviour may result.
294 There were two things you could do, knowing this: be paranoid or be
295 pragmatic. The paranoid approach was to do as little as possible in your
296 signal handler. Set an existing integer variable that already has a
297 value, and return. This doesn't help you if you're in a slow system call,
298 which will just restart. That means you have to C<die> to longjump(3) out
299 of the handler. Even this is a little cavalier for the true paranoiac,
300 who avoids C<die> in a handler because the system I<is> out to get you.
301 The pragmatic approach was to say ``I know the risks, but prefer the
302 convenience'', and to do anything you wanted in your signal handler,
303 and be prepared to clean up core dumps now and again.
305 In Perl 5.7.3 and later to avoid these problems signals are
306 "deferred"-- that is when the signal is delivered to the process by
307 the system (to the C code that implements Perl) a flag is set, and the
308 handler returns immediately. Then at strategic "safe" points in the
309 Perl interpreter (e.g. when it is about to execute a new opcode) the
310 flags are checked and the Perl level handler from %SIG is
311 executed. The "deferred" scheme allows much more flexibility in the
312 coding of signal handler as we know Perl interpreter is in a safe
313 state, and that we are not in a system library function when the
314 handler is called. However the implementation does differ from
315 previous Perls in the following ways:
319 =item Long running opcodes
321 As Perl interpreter only looks at the signal flags when it about to
322 execute a new opcode if a signal arrives during a long running opcode
323 (e.g. a regular expression operation on a very large string) then
324 signal will not be seen until operation completes.
326 =item Interrupting IO
328 When a signal is delivered (e.g. INT control-C) the operating system
329 breaks into IO operations like C<read> (used to implement Perls
330 E<lt>E<gt> operator). On older Perls the handler was called
331 immediately (and as C<read> is not "unsafe" this worked well). With
332 the "deferred" scheme the handler is not called immediately, and if
333 Perl is using system's C<stdio> library that library may re-start the
334 C<read> without returning to Perl and giving it a chance to call the
335 %SIG handler. If this happens on your system the solution is to use
336 C<:perlio> layer to do IO - at least on those handles which you want
337 to be able to break into with signals. (The C<:perlio> layer checks
338 the signal flags and calls %SIG handlers before resuming IO operation.)
340 Note that the default in Perl 5.7.3 and later is to automatically use
341 the C<:perlio> layer.
343 =item Signals as "faults"
345 Certain signals e.g. SEGV, ILL, BUS are generated as a result of
346 virtual memory or other "faults". These are normally fatal and there
347 is little a Perl-level handler can do with them. (In particular the
348 old signal scheme was particularly unsafe in such cases.) However if
349 a %SIG handler is set the new scheme simply sets a flag and returns as
350 described above. This may cause the operating system to try the
351 offending machine instruction again and - as nothing has changed - it
352 will generate the signal again. The result of this is a rather odd
353 "loop". In future Perl's signal mechanism may be changed to avoid this
354 - perhaps by simply disallowing %SIG handlers on signals of that
355 type. Until then the work-round is not to set a %SIG handler on those
356 signals. (Which signals they are is operating system dependant.)
358 =item Signals triggered by operating system state
360 On some operating systems certain signal handlers are supposed to "do
361 something" before returning. One example can be CHLD or CLD which
362 indicates a child process has completed. On some operating systems the
363 signal handler is expected to C<wait> for the completed child
364 process. On such systems the deferred signal scheme will not work for
365 those signals (it does not do the C<wait>). Again the failure will
366 look like a loop as the operating system will re-issue the signal as
367 there are un-waited-for completed child processes.
371 =head1 Using open() for IPC
373 Perl's basic open() statement can also be used for unidirectional
374 interprocess communication by either appending or prepending a pipe
375 symbol to the second argument to open(). Here's how to start
376 something up in a child process you intend to write to:
378 open(SPOOLER, "| cat -v | lpr -h 2>/dev/null")
379 || die "can't fork: $!";
380 local $SIG{PIPE} = sub { die "spooler pipe broke" };
381 print SPOOLER "stuff\n";
382 close SPOOLER || die "bad spool: $! $?";
384 And here's how to start up a child process you intend to read from:
386 open(STATUS, "netstat -an 2>&1 |")
387 || die "can't fork: $!";
389 next if /^(tcp|udp)/;
392 close STATUS || die "bad netstat: $! $?";
394 If one can be sure that a particular program is a Perl script that is
395 expecting filenames in @ARGV, the clever programmer can write something
398 % program f1 "cmd1|" - f2 "cmd2|" f3 < tmpfile
400 and irrespective of which shell it's called from, the Perl program will
401 read from the file F<f1>, the process F<cmd1>, standard input (F<tmpfile>
402 in this case), the F<f2> file, the F<cmd2> command, and finally the F<f3>
403 file. Pretty nifty, eh?
405 You might notice that you could use backticks for much the
406 same effect as opening a pipe for reading:
408 print grep { !/^(tcp|udp)/ } `netstat -an 2>&1`;
409 die "bad netstat" if $?;
411 While this is true on the surface, it's much more efficient to process the
412 file one line or record at a time because then you don't have to read the
413 whole thing into memory at once. It also gives you finer control of the
414 whole process, letting you to kill off the child process early if you'd
417 Be careful to check both the open() and the close() return values. If
418 you're I<writing> to a pipe, you should also trap SIGPIPE. Otherwise,
419 think of what happens when you start up a pipe to a command that doesn't
420 exist: the open() will in all likelihood succeed (it only reflects the
421 fork()'s success), but then your output will fail--spectacularly. Perl
422 can't know whether the command worked because your command is actually
423 running in a separate process whose exec() might have failed. Therefore,
424 while readers of bogus commands return just a quick end of file, writers
425 to bogus command will trigger a signal they'd better be prepared to
428 open(FH, "|bogus") or die "can't fork: $!";
429 print FH "bang\n" or die "can't write: $!";
430 close FH or die "can't close: $!";
432 That won't blow up until the close, and it will blow up with a SIGPIPE.
433 To catch it, you could use this:
435 $SIG{PIPE} = 'IGNORE';
436 open(FH, "|bogus") or die "can't fork: $!";
437 print FH "bang\n" or die "can't write: $!";
438 close FH or die "can't close: status=$?";
442 Both the main process and any child processes it forks share the same
443 STDIN, STDOUT, and STDERR filehandles. If both processes try to access
444 them at once, strange things can happen. You may also want to close
445 or reopen the filehandles for the child. You can get around this by
446 opening your pipe with open(), but on some systems this means that the
447 child process cannot outlive the parent.
449 =head2 Background Processes
451 You can run a command in the background with:
455 The command's STDOUT and STDERR (and possibly STDIN, depending on your
456 shell) will be the same as the parent's. You won't need to catch
457 SIGCHLD because of the double-fork taking place (see below for more
460 =head2 Complete Dissociation of Child from Parent
462 In some cases (starting server processes, for instance) you'll want to
463 completely dissociate the child process from the parent. This is
464 often called daemonization. A well behaved daemon will also chdir()
465 to the root directory (so it doesn't prevent unmounting the filesystem
466 containing the directory from which it was launched) and redirect its
467 standard file descriptors from and to F</dev/null> (so that random
468 output doesn't wind up on the user's terminal).
473 chdir '/' or die "Can't chdir to /: $!";
474 open STDIN, '/dev/null' or die "Can't read /dev/null: $!";
475 open STDOUT, '>/dev/null'
476 or die "Can't write to /dev/null: $!";
477 defined(my $pid = fork) or die "Can't fork: $!";
479 setsid or die "Can't start a new session: $!";
480 open STDERR, '>&STDOUT' or die "Can't dup stdout: $!";
483 The fork() has to come before the setsid() to ensure that you aren't a
484 process group leader (the setsid() will fail if you are). If your
485 system doesn't have the setsid() function, open F</dev/tty> and use the
486 C<TIOCNOTTY> ioctl() on it instead. See L<tty(4)> for details.
488 Non-Unix users should check their Your_OS::Process module for other
491 =head2 Safe Pipe Opens
493 Another interesting approach to IPC is making your single program go
494 multiprocess and communicate between (or even amongst) yourselves. The
495 open() function will accept a file argument of either C<"-|"> or C<"|-">
496 to do a very interesting thing: it forks a child connected to the
497 filehandle you've opened. The child is running the same program as the
498 parent. This is useful for safely opening a file when running under an
499 assumed UID or GID, for example. If you open a pipe I<to> minus, you can
500 write to the filehandle you opened and your kid will find it in his
501 STDIN. If you open a pipe I<from> minus, you can read from the filehandle
502 you opened whatever your kid writes to his STDOUT.
504 use English '-no_match_vars';
508 $pid = open(KID_TO_WRITE, "|-");
509 unless (defined $pid) {
510 warn "cannot fork: $!";
511 die "bailing out" if $sleep_count++ > 6;
514 } until defined $pid;
517 print KID_TO_WRITE @some_data;
518 close(KID_TO_WRITE) || warn "kid exited $?";
520 ($EUID, $EGID) = ($UID, $GID); # suid progs only
521 open (FILE, "> /safe/file")
522 || die "can't open /safe/file: $!";
524 print FILE; # child's STDIN is parent's KID
526 exit; # don't forget this
529 Another common use for this construct is when you need to execute
530 something without the shell's interference. With system(), it's
531 straightforward, but you can't use a pipe open or backticks safely.
532 That's because there's no way to stop the shell from getting its hands on
533 your arguments. Instead, use lower-level control to call exec() directly.
535 Here's a safe backtick or pipe open for read:
537 # add error processing as above
538 $pid = open(KID_TO_READ, "-|");
541 while (<KID_TO_READ>) {
542 # do something interesting
544 close(KID_TO_READ) || warn "kid exited $?";
547 ($EUID, $EGID) = ($UID, $GID); # suid only
548 exec($program, @options, @args)
549 || die "can't exec program: $!";
554 And here's a safe pipe open for writing:
556 # add error processing as above
557 $pid = open(KID_TO_WRITE, "|-");
558 $SIG{ALRM} = sub { die "whoops, $program pipe broke" };
564 close(KID_TO_WRITE) || warn "kid exited $?";
567 ($EUID, $EGID) = ($UID, $GID);
568 exec($program, @options, @args)
569 || die "can't exec program: $!";
573 Note that these operations are full Unix forks, which means they may not be
574 correctly implemented on alien systems. Additionally, these are not true
575 multithreading. If you'd like to learn more about threading, see the
576 F<modules> file mentioned below in the SEE ALSO section.
578 =head2 Bidirectional Communication with Another Process
580 While this works reasonably well for unidirectional communication, what
581 about bidirectional communication? The obvious thing you'd like to do
582 doesn't actually work:
584 open(PROG_FOR_READING_AND_WRITING, "| some program |")
586 and if you forget to use the C<use warnings> pragma or the B<-w> flag,
587 then you'll miss out entirely on the diagnostic message:
589 Can't do bidirectional pipe at -e line 1.
591 If you really want to, you can use the standard open2() library function
592 to catch both ends. There's also an open3() for tridirectional I/O so you
593 can also catch your child's STDERR, but doing so would then require an
594 awkward select() loop and wouldn't allow you to use normal Perl input
597 If you look at its source, you'll see that open2() uses low-level
598 primitives like Unix pipe() and exec() calls to create all the connections.
599 While it might have been slightly more efficient by using socketpair(), it
600 would have then been even less portable than it already is. The open2()
601 and open3() functions are unlikely to work anywhere except on a Unix
602 system or some other one purporting to be POSIX compliant.
604 Here's an example of using open2():
608 $pid = open2(*Reader, *Writer, "cat -u -n" );
609 print Writer "stuff\n";
612 The problem with this is that Unix buffering is really going to
613 ruin your day. Even though your C<Writer> filehandle is auto-flushed,
614 and the process on the other end will get your data in a timely manner,
615 you can't usually do anything to force it to give it back to you
616 in a similarly quick fashion. In this case, we could, because we
617 gave I<cat> a B<-u> flag to make it unbuffered. But very few Unix
618 commands are designed to operate over pipes, so this seldom works
619 unless you yourself wrote the program on the other end of the
622 A solution to this is the nonstandard F<Comm.pl> library. It uses
623 pseudo-ttys to make your program behave more reasonably:
626 $ph = open_proc('cat -n');
628 print $ph "a line\n";
629 print "got back ", scalar <$ph>;
632 This way you don't have to have control over the source code of the
633 program you're using. The F<Comm> library also has expect()
634 and interact() functions. Find the library (and we hope its
635 successor F<IPC::Chat>) at your nearest CPAN archive as detailed
636 in the SEE ALSO section below.
638 The newer Expect.pm module from CPAN also addresses this kind of thing.
639 This module requires two other modules from CPAN: IO::Pty and IO::Stty.
640 It sets up a pseudo-terminal to interact with programs that insist on
641 using talking to the terminal device driver. If your system is
642 amongst those supported, this may be your best bet.
644 =head2 Bidirectional Communication with Yourself
646 If you want, you may make low-level pipe() and fork()
647 to stitch this together by hand. This example only
648 talks to itself, but you could reopen the appropriate
649 handles to STDIN and STDOUT and call other processes.
652 # pipe1 - bidirectional communication using two pipe pairs
653 # designed for the socketpair-challenged
654 use IO::Handle; # thousands of lines just for autoflush :-(
655 pipe(PARENT_RDR, CHILD_WTR); # XXX: failure?
656 pipe(CHILD_RDR, PARENT_WTR); # XXX: failure?
657 CHILD_WTR->autoflush(1);
658 PARENT_WTR->autoflush(1);
661 close PARENT_RDR; close PARENT_WTR;
662 print CHILD_WTR "Parent Pid $$ is sending this\n";
663 chomp($line = <CHILD_RDR>);
664 print "Parent Pid $$ just read this: `$line'\n";
665 close CHILD_RDR; close CHILD_WTR;
668 die "cannot fork: $!" unless defined $pid;
669 close CHILD_RDR; close CHILD_WTR;
670 chomp($line = <PARENT_RDR>);
671 print "Child Pid $$ just read this: `$line'\n";
672 print PARENT_WTR "Child Pid $$ is sending this\n";
673 close PARENT_RDR; close PARENT_WTR;
677 But you don't actually have to make two pipe calls. If you
678 have the socketpair() system call, it will do this all for you.
681 # pipe2 - bidirectional communication using socketpair
682 # "the best ones always go both ways"
685 use IO::Handle; # thousands of lines just for autoflush :-(
686 # We say AF_UNIX because although *_LOCAL is the
687 # POSIX 1003.1g form of the constant, many machines
688 # still don't have it.
689 socketpair(CHILD, PARENT, AF_UNIX, SOCK_STREAM, PF_UNSPEC)
690 or die "socketpair: $!";
693 PARENT->autoflush(1);
697 print CHILD "Parent Pid $$ is sending this\n";
698 chomp($line = <CHILD>);
699 print "Parent Pid $$ just read this: `$line'\n";
703 die "cannot fork: $!" unless defined $pid;
705 chomp($line = <PARENT>);
706 print "Child Pid $$ just read this: `$line'\n";
707 print PARENT "Child Pid $$ is sending this\n";
712 =head1 Sockets: Client/Server Communication
714 While not limited to Unix-derived operating systems (e.g., WinSock on PCs
715 provides socket support, as do some VMS libraries), you may not have
716 sockets on your system, in which case this section probably isn't going to do
717 you much good. With sockets, you can do both virtual circuits (i.e., TCP
718 streams) and datagrams (i.e., UDP packets). You may be able to do even more
719 depending on your system.
721 The Perl function calls for dealing with sockets have the same names as
722 the corresponding system calls in C, but their arguments tend to differ
723 for two reasons: first, Perl filehandles work differently than C file
724 descriptors. Second, Perl already knows the length of its strings, so you
725 don't need to pass that information.
727 One of the major problems with old socket code in Perl was that it used
728 hard-coded values for some of the constants, which severely hurt
729 portability. If you ever see code that does anything like explicitly
730 setting C<$AF_INET = 2>, you know you're in for big trouble: An
731 immeasurably superior approach is to use the C<Socket> module, which more
732 reliably grants access to various constants and functions you'll need.
734 If you're not writing a server/client for an existing protocol like
735 NNTP or SMTP, you should give some thought to how your server will
736 know when the client has finished talking, and vice-versa. Most
737 protocols are based on one-line messages and responses (so one party
738 knows the other has finished when a "\n" is received) or multi-line
739 messages and responses that end with a period on an empty line
740 ("\n.\n" terminates a message/response).
742 =head2 Internet Line Terminators
744 The Internet line terminator is "\015\012". Under ASCII variants of
745 Unix, that could usually be written as "\r\n", but under other systems,
746 "\r\n" might at times be "\015\015\012", "\012\012\015", or something
747 completely different. The standards specify writing "\015\012" to be
748 conformant (be strict in what you provide), but they also recommend
749 accepting a lone "\012" on input (but be lenient in what you require).
750 We haven't always been very good about that in the code in this manpage,
751 but unless you're on a Mac, you'll probably be ok.
753 =head2 Internet TCP Clients and Servers
755 Use Internet-domain sockets when you want to do client-server
756 communication that might extend to machines outside of your own system.
758 Here's a sample TCP client using Internet-domain sockets:
763 my ($remote,$port, $iaddr, $paddr, $proto, $line);
765 $remote = shift || 'localhost';
766 $port = shift || 2345; # random port
767 if ($port =~ /\D/) { $port = getservbyname($port, 'tcp') }
768 die "No port" unless $port;
769 $iaddr = inet_aton($remote) || die "no host: $remote";
770 $paddr = sockaddr_in($port, $iaddr);
772 $proto = getprotobyname('tcp');
773 socket(SOCK, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
774 connect(SOCK, $paddr) || die "connect: $!";
775 while (defined($line = <SOCK>)) {
779 close (SOCK) || die "close: $!";
782 And here's a corresponding server to go along with it. We'll
783 leave the address as INADDR_ANY so that the kernel can choose
784 the appropriate interface on multihomed hosts. If you want sit
785 on a particular interface (like the external side of a gateway
786 or firewall machine), you should fill this in with your real address
791 BEGIN { $ENV{PATH} = '/usr/ucb:/bin' }
794 my $EOL = "\015\012";
796 sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" }
798 my $port = shift || 2345;
799 my $proto = getprotobyname('tcp');
801 ($port) = $port =~ /^(\d+)$/ or die "invalid port";
803 socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
804 setsockopt(Server, SOL_SOCKET, SO_REUSEADDR,
805 pack("l", 1)) || die "setsockopt: $!";
806 bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!";
807 listen(Server,SOMAXCONN) || die "listen: $!";
809 logmsg "server started on port $port";
813 $SIG{CHLD} = \&REAPER;
815 for ( ; $paddr = accept(Client,Server); close Client) {
816 my($port,$iaddr) = sockaddr_in($paddr);
817 my $name = gethostbyaddr($iaddr,AF_INET);
819 logmsg "connection from $name [",
820 inet_ntoa($iaddr), "]
823 print Client "Hello there, $name, it's now ",
824 scalar localtime, $EOL;
827 And here's a multithreaded version. It's multithreaded in that
828 like most typical servers, it spawns (forks) a slave server to
829 handle the client request so that the master server can quickly
830 go back to service a new client.
834 BEGIN { $ENV{PATH} = '/usr/ucb:/bin' }
837 my $EOL = "\015\012";
839 sub spawn; # forward declaration
840 sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" }
842 my $port = shift || 2345;
843 my $proto = getprotobyname('tcp');
845 ($port) = $port =~ /^(\d+)$/ or die "invalid port";
847 socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
848 setsockopt(Server, SOL_SOCKET, SO_REUSEADDR,
849 pack("l", 1)) || die "setsockopt: $!";
850 bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!";
851 listen(Server,SOMAXCONN) || die "listen: $!";
853 logmsg "server started on port $port";
858 use POSIX ":sys_wait_h";
861 while (($waitedpid = waitpid(-1,WNOHANG)) > 0) {
862 logmsg "reaped $waitedpid" . ($? ? " with exit $?" : '');
864 $SIG{CHLD} = \&REAPER; # loathe sysV
867 $SIG{CHLD} = \&REAPER;
869 for ( $waitedpid = 0;
870 ($paddr = accept(Client,Server)) || $waitedpid;
871 $waitedpid = 0, close Client)
873 next if $waitedpid and not $paddr;
874 my($port,$iaddr) = sockaddr_in($paddr);
875 my $name = gethostbyaddr($iaddr,AF_INET);
877 logmsg "connection from $name [",
878 inet_ntoa($iaddr), "]
883 print "Hello there, $name, it's now ", scalar localtime, $EOL;
884 exec '/usr/games/fortune' # XXX: `wrong' line terminators
885 or confess "can't exec fortune: $!";
893 unless (@_ == 0 && $coderef && ref($coderef) eq 'CODE') {
894 confess "usage: spawn CODEREF";
898 if (!defined($pid = fork)) {
899 logmsg "cannot fork: $!";
903 return; # I'm the parent
905 # else I'm the child -- go spawn
907 open(STDIN, "<&Client") || die "can't dup client to stdin";
908 open(STDOUT, ">&Client") || die "can't dup client to stdout";
909 ## open(STDERR, ">&STDOUT") || die "can't dup stdout to stderr";
913 This server takes the trouble to clone off a child version via fork() for
914 each incoming request. That way it can handle many requests at once,
915 which you might not always want. Even if you don't fork(), the listen()
916 will allow that many pending connections. Forking servers have to be
917 particularly careful about cleaning up their dead children (called
918 "zombies" in Unix parlance), because otherwise you'll quickly fill up your
921 We suggest that you use the B<-T> flag to use taint checking (see L<perlsec>)
922 even if we aren't running setuid or setgid. This is always a good idea
923 for servers and other programs run on behalf of someone else (like CGI
924 scripts), because it lessens the chances that people from the outside will
925 be able to compromise your system.
927 Let's look at another TCP client. This one connects to the TCP "time"
928 service on a number of different machines and shows how far their clocks
929 differ from the system on which it's being run:
935 my $SECS_of_70_YEARS = 2208988800;
936 sub ctime { scalar localtime(shift) }
938 my $iaddr = gethostbyname('localhost');
939 my $proto = getprotobyname('tcp');
940 my $port = getservbyname('time', 'tcp');
941 my $paddr = sockaddr_in(0, $iaddr);
945 printf "%-24s %8s %s\n", "localhost", 0, ctime(time());
947 foreach $host (@ARGV) {
948 printf "%-24s ", $host;
949 my $hisiaddr = inet_aton($host) || die "unknown host";
950 my $hispaddr = sockaddr_in($port, $hisiaddr);
951 socket(SOCKET, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
952 connect(SOCKET, $hispaddr) || die "bind: $!";
954 read(SOCKET, $rtime, 4);
956 my $histime = unpack("N", $rtime) - $SECS_of_70_YEARS ;
957 printf "%8d %s\n", $histime - time, ctime($histime);
960 =head2 Unix-Domain TCP Clients and Servers
962 That's fine for Internet-domain clients and servers, but what about local
963 communications? While you can use the same setup, sometimes you don't
964 want to. Unix-domain sockets are local to the current host, and are often
965 used internally to implement pipes. Unlike Internet domain sockets, Unix
966 domain sockets can show up in the file system with an ls(1) listing.
969 srw-rw-rw- 1 root 0 Oct 31 07:23 /dev/log
971 You can test for these with Perl's B<-S> file test:
973 unless ( -S '/dev/log' ) {
974 die "something's wicked with the log system";
977 Here's a sample Unix-domain client:
982 my ($rendezvous, $line);
984 $rendezvous = shift || '/tmp/catsock';
985 socket(SOCK, PF_UNIX, SOCK_STREAM, 0) || die "socket: $!";
986 connect(SOCK, sockaddr_un($rendezvous)) || die "connect: $!";
987 while (defined($line = <SOCK>)) {
992 And here's a corresponding server. You don't have to worry about silly
993 network terminators here because Unix domain sockets are guaranteed
994 to be on the localhost, and thus everything works right.
1001 BEGIN { $ENV{PATH} = '/usr/ucb:/bin' }
1002 sub spawn; # forward declaration
1003 sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" }
1005 my $NAME = '/tmp/catsock';
1006 my $uaddr = sockaddr_un($NAME);
1007 my $proto = getprotobyname('tcp');
1009 socket(Server,PF_UNIX,SOCK_STREAM,0) || die "socket: $!";
1011 bind (Server, $uaddr) || die "bind: $!";
1012 listen(Server,SOMAXCONN) || die "listen: $!";
1014 logmsg "server started on $NAME";
1018 use POSIX ":sys_wait_h";
1021 while (($waitedpid = waitpid(-1,WNOHANG)) > 0) {
1022 logmsg "reaped $waitedpid" . ($? ? " with exit $?" : '');
1024 $SIG{CHLD} = \&REAPER; # loathe sysV
1027 $SIG{CHLD} = \&REAPER;
1030 for ( $waitedpid = 0;
1031 accept(Client,Server) || $waitedpid;
1032 $waitedpid = 0, close Client)
1035 logmsg "connection on $NAME";
1037 print "Hello there, it's now ", scalar localtime, "\n";
1038 exec '/usr/games/fortune' or die "can't exec fortune: $!";
1043 my $coderef = shift;
1045 unless (@_ == 0 && $coderef && ref($coderef) eq 'CODE') {
1046 confess "usage: spawn CODEREF";
1050 if (!defined($pid = fork)) {
1051 logmsg "cannot fork: $!";
1054 logmsg "begat $pid";
1055 return; # I'm the parent
1057 # else I'm the child -- go spawn
1059 open(STDIN, "<&Client") || die "can't dup client to stdin";
1060 open(STDOUT, ">&Client") || die "can't dup client to stdout";
1061 ## open(STDERR, ">&STDOUT") || die "can't dup stdout to stderr";
1065 As you see, it's remarkably similar to the Internet domain TCP server, so
1066 much so, in fact, that we've omitted several duplicate functions--spawn(),
1067 logmsg(), ctime(), and REAPER()--which are exactly the same as in the
1070 So why would you ever want to use a Unix domain socket instead of a
1071 simpler named pipe? Because a named pipe doesn't give you sessions. You
1072 can't tell one process's data from another's. With socket programming,
1073 you get a separate session for each client: that's why accept() takes two
1076 For example, let's say that you have a long running database server daemon
1077 that you want folks from the World Wide Web to be able to access, but only
1078 if they go through a CGI interface. You'd have a small, simple CGI
1079 program that does whatever checks and logging you feel like, and then acts
1080 as a Unix-domain client and connects to your private server.
1082 =head1 TCP Clients with IO::Socket
1084 For those preferring a higher-level interface to socket programming, the
1085 IO::Socket module provides an object-oriented approach. IO::Socket is
1086 included as part of the standard Perl distribution as of the 5.004
1087 release. If you're running an earlier version of Perl, just fetch
1088 IO::Socket from CPAN, where you'll also find modules providing easy
1089 interfaces to the following systems: DNS, FTP, Ident (RFC 931), NIS and
1090 NISPlus, NNTP, Ping, POP3, SMTP, SNMP, SSLeay, Telnet, and Time--just
1093 =head2 A Simple Client
1095 Here's a client that creates a TCP connection to the "daytime"
1096 service at port 13 of the host name "localhost" and prints out everything
1097 that the server there cares to provide.
1101 $remote = IO::Socket::INET->new(
1103 PeerAddr => "localhost",
1104 PeerPort => "daytime(13)",
1106 or die "cannot connect to daytime port at localhost";
1107 while ( <$remote> ) { print }
1109 When you run this program, you should get something back that
1112 Wed May 14 08:40:46 MDT 1997
1114 Here are what those parameters to the C<new> constructor mean:
1120 This is which protocol to use. In this case, the socket handle returned
1121 will be connected to a TCP socket, because we want a stream-oriented
1122 connection, that is, one that acts pretty much like a plain old file.
1123 Not all sockets are this of this type. For example, the UDP protocol
1124 can be used to make a datagram socket, used for message-passing.
1128 This is the name or Internet address of the remote host the server is
1129 running on. We could have specified a longer name like C<"www.perl.com">,
1130 or an address like C<"204.148.40.9">. For demonstration purposes, we've
1131 used the special hostname C<"localhost">, which should always mean the
1132 current machine you're running on. The corresponding Internet address
1133 for localhost is C<"127.1">, if you'd rather use that.
1137 This is the service name or port number we'd like to connect to.
1138 We could have gotten away with using just C<"daytime"> on systems with a
1139 well-configured system services file,[FOOTNOTE: The system services file
1140 is in I</etc/services> under Unix] but just in case, we've specified the
1141 port number (13) in parentheses. Using just the number would also have
1142 worked, but constant numbers make careful programmers nervous.
1146 Notice how the return value from the C<new> constructor is used as
1147 a filehandle in the C<while> loop? That's what's called an indirect
1148 filehandle, a scalar variable containing a filehandle. You can use
1149 it the same way you would a normal filehandle. For example, you
1150 can read one line from it this way:
1154 all remaining lines from is this way:
1158 and send a line of data to it this way:
1160 print $handle "some data\n";
1162 =head2 A Webget Client
1164 Here's a simple client that takes a remote host to fetch a document
1165 from, and then a list of documents to get from that host. This is a
1166 more interesting client than the previous one because it first sends
1167 something to the server before fetching the server's response.
1171 unless (@ARGV > 1) { die "usage: $0 host document ..." }
1172 $host = shift(@ARGV);
1175 foreach $document ( @ARGV ) {
1176 $remote = IO::Socket::INET->new( Proto => "tcp",
1178 PeerPort => "http(80)",
1180 unless ($remote) { die "cannot connect to http daemon on $host" }
1181 $remote->autoflush(1);
1182 print $remote "GET $document HTTP/1.0" . $BLANK;
1183 while ( <$remote> ) { print }
1187 The web server handing the "http" service, which is assumed to be at
1188 its standard port, number 80. If the web server you're trying to
1189 connect to is at a different port (like 1080 or 8080), you should specify
1190 as the named-parameter pair, C<< PeerPort => 8080 >>. The C<autoflush>
1191 method is used on the socket because otherwise the system would buffer
1192 up the output we sent it. (If you're on a Mac, you'll also need to
1193 change every C<"\n"> in your code that sends data over the network to
1194 be a C<"\015\012"> instead.)
1196 Connecting to the server is only the first part of the process: once you
1197 have the connection, you have to use the server's language. Each server
1198 on the network has its own little command language that it expects as
1199 input. The string that we send to the server starting with "GET" is in
1200 HTTP syntax. In this case, we simply request each specified document.
1201 Yes, we really are making a new connection for each document, even though
1202 it's the same host. That's the way you always used to have to speak HTTP.
1203 Recent versions of web browsers may request that the remote server leave
1204 the connection open a little while, but the server doesn't have to honor
1207 Here's an example of running that program, which we'll call I<webget>:
1209 % webget www.perl.com /guanaco.html
1210 HTTP/1.1 404 File Not Found
1211 Date: Thu, 08 May 1997 18:02:32 GMT
1212 Server: Apache/1.2b6
1214 Content-type: text/html
1216 <HEAD><TITLE>404 File Not Found</TITLE></HEAD>
1217 <BODY><H1>File Not Found</H1>
1218 The requested URL /guanaco.html was not found on this server.<P>
1221 Ok, so that's not very interesting, because it didn't find that
1222 particular document. But a long response wouldn't have fit on this page.
1224 For a more fully-featured version of this program, you should look to
1225 the I<lwp-request> program included with the LWP modules from CPAN.
1227 =head2 Interactive Client with IO::Socket
1229 Well, that's all fine if you want to send one command and get one answer,
1230 but what about setting up something fully interactive, somewhat like
1231 the way I<telnet> works? That way you can type a line, get the answer,
1232 type a line, get the answer, etc.
1234 This client is more complicated than the two we've done so far, but if
1235 you're on a system that supports the powerful C<fork> call, the solution
1236 isn't that rough. Once you've made the connection to whatever service
1237 you'd like to chat with, call C<fork> to clone your process. Each of
1238 these two identical process has a very simple job to do: the parent
1239 copies everything from the socket to standard output, while the child
1240 simultaneously copies everything from standard input to the socket.
1241 To accomplish the same thing using just one process would be I<much>
1242 harder, because it's easier to code two processes to do one thing than it
1243 is to code one process to do two things. (This keep-it-simple principle
1244 a cornerstones of the Unix philosophy, and good software engineering as
1245 well, which is probably why it's spread to other systems.)
1252 my ($host, $port, $kidpid, $handle, $line);
1254 unless (@ARGV == 2) { die "usage: $0 host port" }
1255 ($host, $port) = @ARGV;
1257 # create a tcp connection to the specified host and port
1258 $handle = IO::Socket::INET->new(Proto => "tcp",
1261 or die "can't connect to port $port on $host: $!";
1263 $handle->autoflush(1); # so output gets there right away
1264 print STDERR "[Connected to $host:$port]\n";
1266 # split the program into two processes, identical twins
1267 die "can't fork: $!" unless defined($kidpid = fork());
1269 # the if{} block runs only in the parent process
1271 # copy the socket to standard output
1272 while (defined ($line = <$handle>)) {
1275 kill("TERM", $kidpid); # send SIGTERM to child
1277 # the else{} block runs only in the child process
1279 # copy standard input to the socket
1280 while (defined ($line = <STDIN>)) {
1281 print $handle $line;
1285 The C<kill> function in the parent's C<if> block is there to send a
1286 signal to our child process (current running in the C<else> block)
1287 as soon as the remote server has closed its end of the connection.
1289 If the remote server sends data a byte at time, and you need that
1290 data immediately without waiting for a newline (which might not happen),
1291 you may wish to replace the C<while> loop in the parent with the
1295 while (sysread($handle, $byte, 1) == 1) {
1299 Making a system call for each byte you want to read is not very efficient
1300 (to put it mildly) but is the simplest to explain and works reasonably
1303 =head1 TCP Servers with IO::Socket
1305 As always, setting up a server is little bit more involved than running a client.
1306 The model is that the server creates a special kind of socket that
1307 does nothing but listen on a particular port for incoming connections.
1308 It does this by calling the C<< IO::Socket::INET->new() >> method with
1309 slightly different arguments than the client did.
1315 This is which protocol to use. Like our clients, we'll
1316 still specify C<"tcp"> here.
1321 port in the C<LocalPort> argument, which we didn't do for the client.
1322 This is service name or port number for which you want to be the
1323 server. (Under Unix, ports under 1024 are restricted to the
1324 superuser.) In our sample, we'll use port 9000, but you can use
1325 any port that's not currently in use on your system. If you try
1326 to use one already in used, you'll get an "Address already in use"
1327 message. Under Unix, the C<netstat -a> command will show
1328 which services current have servers.
1332 The C<Listen> parameter is set to the maximum number of
1333 pending connections we can accept until we turn away incoming clients.
1334 Think of it as a call-waiting queue for your telephone.
1335 The low-level Socket module has a special symbol for the system maximum, which
1340 The C<Reuse> parameter is needed so that we restart our server
1341 manually without waiting a few minutes to allow system buffers to
1346 Once the generic server socket has been created using the parameters
1347 listed above, the server then waits for a new client to connect
1348 to it. The server blocks in the C<accept> method, which eventually accepts a
1349 bidirectional connection from the remote client. (Make sure to autoflush
1350 this handle to circumvent buffering.)
1352 To add to user-friendliness, our server prompts the user for commands.
1353 Most servers don't do this. Because of the prompt without a newline,
1354 you'll have to use the C<sysread> variant of the interactive client above.
1356 This server accepts one of five different commands, sending output
1357 back to the client. Note that unlike most network servers, this one
1358 only handles one incoming client at a time. Multithreaded servers are
1359 covered in Chapter 6 of the Camel.
1361 Here's the code. We'll
1365 use Net::hostent; # for OO version of gethostbyaddr
1367 $PORT = 9000; # pick something not in use
1369 $server = IO::Socket::INET->new( Proto => 'tcp',
1371 Listen => SOMAXCONN,
1374 die "can't setup server" unless $server;
1375 print "[Server $0 accepting clients]\n";
1377 while ($client = $server->accept()) {
1378 $client->autoflush(1);
1379 print $client "Welcome to $0; type help for command list.\n";
1380 $hostinfo = gethostbyaddr($client->peeraddr);
1381 printf "[Connect from %s]\n", $hostinfo->name || $client->peerhost;
1382 print $client "Command? ";
1383 while ( <$client>) {
1384 next unless /\S/; # blank line
1385 if (/quit|exit/i) { last; }
1386 elsif (/date|time/i) { printf $client "%s\n", scalar localtime; }
1387 elsif (/who/i ) { print $client `who 2>&1`; }
1388 elsif (/cookie/i ) { print $client `/usr/games/fortune 2>&1`; }
1389 elsif (/motd/i ) { print $client `cat /etc/motd 2>&1`; }
1391 print $client "Commands: quit date who cookie motd\n";
1394 print $client "Command? ";
1399 =head1 UDP: Message Passing
1401 Another kind of client-server setup is one that uses not connections, but
1402 messages. UDP communications involve much lower overhead but also provide
1403 less reliability, as there are no promises that messages will arrive at
1404 all, let alone in order and unmangled. Still, UDP offers some advantages
1405 over TCP, including being able to "broadcast" or "multicast" to a whole
1406 bunch of destination hosts at once (usually on your local subnet). If you
1407 find yourself overly concerned about reliability and start building checks
1408 into your message system, then you probably should use just TCP to start
1411 Note that UDP datagrams are I<not> a bytestream and should not be treated
1412 as such. This makes using I/O mechanisms with internal buffering
1413 like stdio (i.e. print() and friends) especially cumbersome. Use syswrite(),
1414 or better send(), like in the example below.
1416 Here's a UDP program similar to the sample Internet TCP client given
1417 earlier. However, instead of checking one host at a time, the UDP version
1418 will check many of them asynchronously by simulating a multicast and then
1419 using select() to do a timed-out wait for I/O. To do something similar
1420 with TCP, you'd have to use a different socket handle for each host.
1427 my ( $count, $hisiaddr, $hispaddr, $histime,
1428 $host, $iaddr, $paddr, $port, $proto,
1429 $rin, $rout, $rtime, $SECS_of_70_YEARS);
1431 $SECS_of_70_YEARS = 2208988800;
1433 $iaddr = gethostbyname(hostname());
1434 $proto = getprotobyname('udp');
1435 $port = getservbyname('time', 'udp');
1436 $paddr = sockaddr_in(0, $iaddr); # 0 means let kernel pick
1438 socket(SOCKET, PF_INET, SOCK_DGRAM, $proto) || die "socket: $!";
1439 bind(SOCKET, $paddr) || die "bind: $!";
1442 printf "%-12s %8s %s\n", "localhost", 0, scalar localtime time;
1446 $hisiaddr = inet_aton($host) || die "unknown host";
1447 $hispaddr = sockaddr_in($port, $hisiaddr);
1448 defined(send(SOCKET, 0, 0, $hispaddr)) || die "send $host: $!";
1452 vec($rin, fileno(SOCKET), 1) = 1;
1454 # timeout after 10.0 seconds
1455 while ($count && select($rout = $rin, undef, undef, 10.0)) {
1457 ($hispaddr = recv(SOCKET, $rtime, 4, 0)) || die "recv: $!";
1458 ($port, $hisiaddr) = sockaddr_in($hispaddr);
1459 $host = gethostbyaddr($hisiaddr, AF_INET);
1460 $histime = unpack("N", $rtime) - $SECS_of_70_YEARS ;
1461 printf "%-12s ", $host;
1462 printf "%8d %s\n", $histime - time, scalar localtime($histime);
1466 Note that this example does not include any retries and may consequently
1467 fail to contact a reachable host. The most prominent reason for this
1468 is congestion of the queues on the sending host if the number of
1469 list of hosts to contact is sufficiently large.
1473 While System V IPC isn't so widely used as sockets, it still has some
1474 interesting uses. You can't, however, effectively use SysV IPC or
1475 Berkeley mmap() to have shared memory so as to share a variable amongst
1476 several processes. That's because Perl would reallocate your string when
1477 you weren't wanting it to.
1479 Here's a small example showing shared memory usage.
1481 use IPC::SysV qw(IPC_PRIVATE IPC_RMID S_IRWXU);
1484 $id = shmget(IPC_PRIVATE, $size, S_IRWXU) || die "$!";
1485 print "shm key $id\n";
1487 $message = "Message #1";
1488 shmwrite($id, $message, 0, 60) || die "$!";
1489 print "wrote: '$message'\n";
1490 shmread($id, $buff, 0, 60) || die "$!";
1491 print "read : '$buff'\n";
1493 # the buffer of shmread is zero-character end-padded.
1494 substr($buff, index($buff, "\0")) = '';
1495 print "un" unless $buff eq $message;
1498 print "deleting shm $id\n";
1499 shmctl($id, IPC_RMID, 0) || die "$!";
1501 Here's an example of a semaphore:
1503 use IPC::SysV qw(IPC_CREAT);
1506 $id = semget($IPC_KEY, 10, 0666 | IPC_CREAT ) || die "$!";
1507 print "shm key $id\n";
1509 Put this code in a separate file to be run in more than one process.
1510 Call the file F<take>:
1512 # create a semaphore
1515 $id = semget($IPC_KEY, 0 , 0 );
1516 die if !defined($id);
1522 # wait for semaphore to be zero
1524 $opstring1 = pack("s!s!s!", $semnum, $semop, $semflag);
1526 # Increment the semaphore count
1528 $opstring2 = pack("s!s!s!", $semnum, $semop, $semflag);
1529 $opstring = $opstring1 . $opstring2;
1531 semop($id,$opstring) || die "$!";
1533 Put this code in a separate file to be run in more than one process.
1534 Call this file F<give>:
1536 # 'give' the semaphore
1537 # run this in the original process and you will see
1538 # that the second process continues
1541 $id = semget($IPC_KEY, 0, 0);
1542 die if !defined($id);
1547 # Decrement the semaphore count
1549 $opstring = pack("s!s!s!", $semnum, $semop, $semflag);
1551 semop($id,$opstring) || die "$!";
1553 The SysV IPC code above was written long ago, and it's definitely
1554 clunky looking. For a more modern look, see the IPC::SysV module
1555 which is included with Perl starting from Perl 5.005.
1557 A small example demonstrating SysV message queues:
1559 use IPC::SysV qw(IPC_PRIVATE IPC_RMID IPC_CREAT S_IRWXU);
1561 my $id = msgget(IPC_PRIVATE, IPC_CREAT | S_IRWXU);
1563 my $sent = "message";
1569 if (msgsnd($id, pack("l! a*", $type_sent, $sent), 0)) {
1570 if (msgrcv($id, $rcvd, 60, 0, 0)) {
1571 ($type_rcvd, $rcvd) = unpack("l! a*", $rcvd);
1572 if ($rcvd eq $sent) {
1578 die "# msgrcv failed\n";
1581 die "# msgsnd failed\n";
1583 msgctl($id, IPC_RMID, 0) || die "# msgctl failed: $!\n";
1585 die "# msgget failed\n";
1590 Most of these routines quietly but politely return C<undef> when they
1591 fail instead of causing your program to die right then and there due to
1592 an uncaught exception. (Actually, some of the new I<Socket> conversion
1593 functions croak() on bad arguments.) It is therefore essential to
1594 check return values from these functions. Always begin your socket
1595 programs this way for optimal success, and don't forget to add B<-T>
1596 taint checking flag to the #! line for servers:
1605 All these routines create system-specific portability problems. As noted
1606 elsewhere, Perl is at the mercy of your C libraries for much of its system
1607 behaviour. It's probably safest to assume broken SysV semantics for
1608 signals and to stick with simple TCP and UDP socket operations; e.g., don't
1609 try to pass open file descriptors over a local UDP datagram socket if you
1610 want your code to stand a chance of being portable.
1612 As mentioned in the signals section, because few vendors provide C
1613 libraries that are safely re-entrant, the prudent programmer will do
1614 little else within a handler beyond setting a numeric variable that
1615 already exists; or, if locked into a slow (restarting) system call,
1616 using die() to raise an exception and longjmp(3) out. In fact, even
1617 these may in some cases cause a core dump. It's probably best to avoid
1618 signals except where they are absolutely inevitable. This
1619 will be addressed in a future release of Perl.
1623 Tom Christiansen, with occasional vestiges of Larry Wall's original
1624 version and suggestions from the Perl Porters.
1628 There's a lot more to networking than this, but this should get you
1631 For intrepid programmers, the indispensable textbook is I<Unix Network
1632 Programming> by W. Richard Stevens (published by Addison-Wesley). Note
1633 that most books on networking address networking from the perspective of
1634 a C programmer; translation to Perl is left as an exercise for the reader.
1636 The IO::Socket(3) manpage describes the object library, and the Socket(3)
1637 manpage describes the low-level interface to sockets. Besides the obvious
1638 functions in L<perlfunc>, you should also check out the F<modules> file
1639 at your nearest CPAN site. (See L<perlmodlib> or best yet, the F<Perl
1640 FAQ> for a description of what CPAN is and where to get it.)
1642 Section 5 of the F<modules> file is devoted to "Networking, Device Control
1643 (modems), and Interprocess Communication", and contains numerous unbundled
1644 modules numerous networking modules, Chat and Expect operations, CGI
1645 programming, DCE, FTP, IPC, NNTP, Proxy, Ptty, RPC, SNMP, SMTP, Telnet,
1646 Threads, and ToolTalk--just to name a few.