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 (Safe 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 a child 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 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.
108 You may be able to determine the cause of failure using C<%!>.
110 unless (kill 0 => $pid or $!{EPERM}) {
111 warn "$pid looks dead";
114 You might also want to employ anonymous functions for simple signal
117 $SIG{INT} = sub { die "\nOutta here!\n" };
119 But that will be problematic for the more complicated handlers that need
120 to reinstall themselves. Because Perl's signal mechanism is currently
121 based on the signal(3) function from the C library, you may sometimes be so
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:
129 # loathe sysV: it makes us not only reinstate
130 # the handler, but place it after the wait
131 $SIG{CHLD} = \&REAPER;
133 $SIG{CHLD} = \&REAPER;
134 # now do something that forks...
138 use POSIX ":sys_wait_h";
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.
145 while (($child = waitpid(-1,WNOHANG)) > 0) {
146 $Kid_Status{$child} = $?;
148 $SIG{CHLD} = \&REAPER; # still loathe sysV
150 $SIG{CHLD} = \&REAPER;
151 # do something that forks...
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.
164 local $SIG{ALRM} = sub { die "alarm clock restart" };
166 flock(FH, 2); # blocking write lock
169 if ($@ and $@ !~ /alarm clock restart/) { die }
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.
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
180 =head2 Handling the SIGHUP Signal in Daemons
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.
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
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.
206 use File::Basename ();
207 use File::Spec::Functions;
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;
216 # POSIX unmasks the sigprocmask properly
217 my $sigset = POSIX::SigSet->new();
218 my $action = POSIX::SigAction->new('sigHUP_handler',
221 POSIX::sigaction(&POSIX::SIGHUP, $action);
224 print "got SIGHUP\n";
225 exec($SELF, @ARGV) or die "Couldn't restart: $!\n";
232 print "ARGV: @ARGV\n";
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
246 just like a regular, connected anonymous pipes, except that the
247 processes rendezvous using a filename and don't have to be related.
249 To create a named pipe, use the C<POSIX::mkfifo()> function.
251 use POSIX qw(mkfifo);
252 mkfifo($path, 0700) or die "mkfifo $path failed: $!";
254 You can also use the Unix command mknod(1) or on some
255 systems, mkfifo(1). These may not be in your normal path.
257 # system return val is backwards, so && not ||
259 $ENV{PATH} .= ":/etc:/usr/etc";
260 if ( system('mknod', $path, 'p')
261 && system('mkfifo', $path) )
263 die "mk{nod,fifo} $path failed";
267 A fifo is convenient when you want to connect a process to an unrelated
268 one. When you open a fifo, the program will block until there's something
271 For example, let's say you'd like to have your F<.signature> file be a
272 named pipe that has a Perl program on the other end. Now every time any
273 program (like a mailer, news reader, finger program, etc.) tries to read
274 from that file, the reading program will block and your program will
275 supply the new signature. We'll use the pipe-checking file test B<-p>
276 to find out whether anyone (or anything) has accidentally removed our fifo.
279 $FIFO = '.signature';
285 POSIX::mkfifo($FIFO, 0700)
286 or die "can't mkfifo $FIFO: $!";
289 # next line blocks until there's a reader
290 open (FIFO, "> $FIFO") || die "can't write $FIFO: $!";
291 print FIFO "John Smith (smith\@host.org)\n", `fortune -s`;
293 sleep 2; # to avoid dup signals
296 =head2 Deferred Signals (Safe Signals)
298 In Perls before Perl 5.7.3 by installing Perl code to deal with
299 signals, you were exposing yourself to danger from two things. First,
300 few system library functions are re-entrant. If the signal interrupts
301 while Perl is executing one function (like malloc(3) or printf(3)),
302 and your signal handler then calls the same function again, you could
303 get unpredictable behavior--often, a core dump. Second, Perl isn't
304 itself re-entrant at the lowest levels. If the signal interrupts Perl
305 while Perl is changing its own internal data structures, similarly
306 unpredictable behaviour may result.
308 There were two things you could do, knowing this: be paranoid or be
309 pragmatic. The paranoid approach was to do as little as possible in your
310 signal handler. Set an existing integer variable that already has a
311 value, and return. This doesn't help you if you're in a slow system call,
312 which will just restart. That means you have to C<die> to longjmp(3) out
313 of the handler. Even this is a little cavalier for the true paranoiac,
314 who avoids C<die> in a handler because the system I<is> out to get you.
315 The pragmatic approach was to say "I know the risks, but prefer the
316 convenience", and to do anything you wanted in your signal handler,
317 and be prepared to clean up core dumps now and again.
319 In Perl 5.7.3 and later to avoid these problems signals are
320 "deferred"-- that is when the signal is delivered to the process by
321 the system (to the C code that implements Perl) a flag is set, and the
322 handler returns immediately. Then at strategic "safe" points in the
323 Perl interpreter (e.g. when it is about to execute a new opcode) the
324 flags are checked and the Perl level handler from %SIG is
325 executed. The "deferred" scheme allows much more flexibility in the
326 coding of signal handler as we know Perl interpreter is in a safe
327 state, and that we are not in a system library function when the
328 handler is called. However the implementation does differ from
329 previous Perls in the following ways:
333 =item Long running opcodes
335 As Perl interpreter only looks at the signal flags when it about to
336 execute a new opcode if a signal arrives during a long running opcode
337 (e.g. a regular expression operation on a very large string) then
338 signal will not be seen until operation completes.
340 =item Interrupting IO
342 When a signal is delivered (e.g. INT control-C) the operating system
343 breaks into IO operations like C<read> (used to implement Perls
344 E<lt>E<gt> operator). On older Perls the handler was called
345 immediately (and as C<read> is not "unsafe" this worked well). With
346 the "deferred" scheme the handler is not called immediately, and if
347 Perl is using system's C<stdio> library that library may re-start the
348 C<read> without returning to Perl and giving it a chance to call the
349 %SIG handler. If this happens on your system the solution is to use
350 C<:perlio> layer to do IO - at least on those handles which you want
351 to be able to break into with signals. (The C<:perlio> layer checks
352 the signal flags and calls %SIG handlers before resuming IO operation.)
354 Note that the default in Perl 5.7.3 and later is to automatically use
355 the C<:perlio> layer.
357 Note that some networking library functions like gethostbyname() are
358 known to have their own implementations of timeouts which may conflict
359 with your timeouts. If you are having problems with such functions,
360 you can try using the POSIX sigaction() function, which bypasses the
361 Perl safe signals (note that this means subjecting yourself to
362 possible memory corruption, as described above). Instead of setting
365 local $SIG{ALRM} = sub { die "alarm" };
367 try something like the following:
369 use POSIX qw(SIGALRM);
370 POSIX::sigaction(SIGALRM,
371 POSIX::SigAction->new(sub { die "alarm" }))
372 or die "Error setting SIGALRM handler: $!\n";
374 =item Restartable system calls
376 On systems that supported it, older versions of Perl used the
377 SA_RESTART flag when installing %SIG handlers. This meant that
378 restartable system calls would continue rather than returning when
379 a signal arrived. In order to deliver deferred signals promptly,
380 Perl 5.7.3 and later do I<not> use SA_RESTART. Consequently,
381 restartable system calls can fail (with $! set to C<EINTR>) in places
382 where they previously would have succeeded.
384 Note that the default C<:perlio> layer will retry C<read>, C<write>
385 and C<close> as described above and that interrupted C<wait> and
386 C<waitpid> calls will always be retried.
388 =item Signals as "faults"
390 Certain signals e.g. SEGV, ILL, BUS are generated as a result of
391 virtual memory or other "faults". These are normally fatal and there
392 is little a Perl-level handler can do with them. (In particular the
393 old signal scheme was particularly unsafe in such cases.) However if
394 a %SIG handler is set the new scheme simply sets a flag and returns as
395 described above. This may cause the operating system to try the
396 offending machine instruction again and - as nothing has changed - it
397 will generate the signal again. The result of this is a rather odd
398 "loop". In future Perl's signal mechanism may be changed to avoid this
399 - perhaps by simply disallowing %SIG handlers on signals of that
400 type. Until then the work-round is not to set a %SIG handler on those
401 signals. (Which signals they are is operating system dependent.)
403 =item Signals triggered by operating system state
405 On some operating systems certain signal handlers are supposed to "do
406 something" before returning. One example can be CHLD or CLD which
407 indicates a child process has completed. On some operating systems the
408 signal handler is expected to C<wait> for the completed child
409 process. On such systems the deferred signal scheme will not work for
410 those signals (it does not do the C<wait>). Again the failure will
411 look like a loop as the operating system will re-issue the signal as
412 there are un-waited-for completed child processes.
416 If you want the old signal behaviour back regardless of possible
417 memory corruption, set the environment variable C<PERL_SIGNALS> to
418 C<"unsafe"> (a new feature since Perl 5.8.1).
420 =head1 Using open() for IPC
422 Perl's basic open() statement can also be used for unidirectional
423 interprocess communication by either appending or prepending a pipe
424 symbol to the second argument to open(). Here's how to start
425 something up in a child process you intend to write to:
427 open(SPOOLER, "| cat -v | lpr -h 2>/dev/null")
428 || die "can't fork: $!";
429 local $SIG{PIPE} = sub { die "spooler pipe broke" };
430 print SPOOLER "stuff\n";
431 close SPOOLER || die "bad spool: $! $?";
433 And here's how to start up a child process you intend to read from:
435 open(STATUS, "netstat -an 2>&1 |")
436 || die "can't fork: $!";
438 next if /^(tcp|udp)/;
441 close STATUS || die "bad netstat: $! $?";
443 If one can be sure that a particular program is a Perl script that is
444 expecting filenames in @ARGV, the clever programmer can write something
447 % program f1 "cmd1|" - f2 "cmd2|" f3 < tmpfile
449 and irrespective of which shell it's called from, the Perl program will
450 read from the file F<f1>, the process F<cmd1>, standard input (F<tmpfile>
451 in this case), the F<f2> file, the F<cmd2> command, and finally the F<f3>
452 file. Pretty nifty, eh?
454 You might notice that you could use backticks for much the
455 same effect as opening a pipe for reading:
457 print grep { !/^(tcp|udp)/ } `netstat -an 2>&1`;
458 die "bad netstat" if $?;
460 While this is true on the surface, it's much more efficient to process the
461 file one line or record at a time because then you don't have to read the
462 whole thing into memory at once. It also gives you finer control of the
463 whole process, letting you to kill off the child process early if you'd
466 Be careful to check both the open() and the close() return values. If
467 you're I<writing> to a pipe, you should also trap SIGPIPE. Otherwise,
468 think of what happens when you start up a pipe to a command that doesn't
469 exist: the open() will in all likelihood succeed (it only reflects the
470 fork()'s success), but then your output will fail--spectacularly. Perl
471 can't know whether the command worked because your command is actually
472 running in a separate process whose exec() might have failed. Therefore,
473 while readers of bogus commands return just a quick end of file, writers
474 to bogus command will trigger a signal they'd better be prepared to
477 open(FH, "|bogus") or die "can't fork: $!";
478 print FH "bang\n" or die "can't write: $!";
479 close FH or die "can't close: $!";
481 That won't blow up until the close, and it will blow up with a SIGPIPE.
482 To catch it, you could use this:
484 $SIG{PIPE} = 'IGNORE';
485 open(FH, "|bogus") or die "can't fork: $!";
486 print FH "bang\n" or die "can't write: $!";
487 close FH or die "can't close: status=$?";
491 Both the main process and any child processes it forks share the same
492 STDIN, STDOUT, and STDERR filehandles. If both processes try to access
493 them at once, strange things can happen. You may also want to close
494 or reopen the filehandles for the child. You can get around this by
495 opening your pipe with open(), but on some systems this means that the
496 child process cannot outlive the parent.
498 =head2 Background Processes
500 You can run a command in the background with:
504 The command's STDOUT and STDERR (and possibly STDIN, depending on your
505 shell) will be the same as the parent's. You won't need to catch
506 SIGCHLD because of the double-fork taking place (see below for more
509 =head2 Complete Dissociation of Child from Parent
511 In some cases (starting server processes, for instance) you'll want to
512 completely dissociate the child process from the parent. This is
513 often called daemonization. A well behaved daemon will also chdir()
514 to the root directory (so it doesn't prevent unmounting the filesystem
515 containing the directory from which it was launched) and redirect its
516 standard file descriptors from and to F</dev/null> (so that random
517 output doesn't wind up on the user's terminal).
522 chdir '/' or die "Can't chdir to /: $!";
523 open STDIN, '/dev/null' or die "Can't read /dev/null: $!";
524 open STDOUT, '>/dev/null'
525 or die "Can't write to /dev/null: $!";
526 defined(my $pid = fork) or die "Can't fork: $!";
528 setsid or die "Can't start a new session: $!";
529 open STDERR, '>&STDOUT' or die "Can't dup stdout: $!";
532 The fork() has to come before the setsid() to ensure that you aren't a
533 process group leader (the setsid() will fail if you are). If your
534 system doesn't have the setsid() function, open F</dev/tty> and use the
535 C<TIOCNOTTY> ioctl() on it instead. See L<tty(4)> for details.
537 Non-Unix users should check their Your_OS::Process module for other
540 =head2 Safe Pipe Opens
542 Another interesting approach to IPC is making your single program go
543 multiprocess and communicate between (or even amongst) yourselves. The
544 open() function will accept a file argument of either C<"-|"> or C<"|-">
545 to do a very interesting thing: it forks a child connected to the
546 filehandle you've opened. The child is running the same program as the
547 parent. This is useful for safely opening a file when running under an
548 assumed UID or GID, for example. If you open a pipe I<to> minus, you can
549 write to the filehandle you opened and your kid will find it in his
550 STDIN. If you open a pipe I<from> minus, you can read from the filehandle
551 you opened whatever your kid writes to his STDOUT.
553 use English '-no_match_vars';
557 $pid = open(KID_TO_WRITE, "|-");
558 unless (defined $pid) {
559 warn "cannot fork: $!";
560 die "bailing out" if $sleep_count++ > 6;
563 } until defined $pid;
566 print KID_TO_WRITE @some_data;
567 close(KID_TO_WRITE) || warn "kid exited $?";
569 ($EUID, $EGID) = ($UID, $GID); # suid progs only
570 open (FILE, "> /safe/file")
571 || die "can't open /safe/file: $!";
573 print FILE; # child's STDIN is parent's KID
575 exit; # don't forget this
578 Another common use for this construct is when you need to execute
579 something without the shell's interference. With system(), it's
580 straightforward, but you can't use a pipe open or backticks safely.
581 That's because there's no way to stop the shell from getting its hands on
582 your arguments. Instead, use lower-level control to call exec() directly.
584 Here's a safe backtick or pipe open for read:
586 # add error processing as above
587 $pid = open(KID_TO_READ, "-|");
590 while (<KID_TO_READ>) {
591 # do something interesting
593 close(KID_TO_READ) || warn "kid exited $?";
596 ($EUID, $EGID) = ($UID, $GID); # suid only
597 exec($program, @options, @args)
598 || die "can't exec program: $!";
603 And here's a safe pipe open for writing:
605 # add error processing as above
606 $pid = open(KID_TO_WRITE, "|-");
607 $SIG{PIPE} = sub { die "whoops, $program pipe broke" };
613 close(KID_TO_WRITE) || warn "kid exited $?";
616 ($EUID, $EGID) = ($UID, $GID);
617 exec($program, @options, @args)
618 || die "can't exec program: $!";
622 Since Perl 5.8.0, you can also use the list form of C<open> for pipes :
625 open KID_PS, "-|", "ps", "aux" or die $!;
627 forks the ps(1) command (without spawning a shell, as there are more than
628 three arguments to open()), and reads its standard output via the
629 C<KID_PS> filehandle. The corresponding syntax to write to command
630 pipes (with C<"|-"> in place of C<"-|">) is also implemented.
632 Note that these operations are full Unix forks, which means they may not be
633 correctly implemented on alien systems. Additionally, these are not true
634 multithreading. If you'd like to learn more about threading, see the
635 F<modules> file mentioned below in the SEE ALSO section.
637 =head2 Bidirectional Communication with Another Process
639 While this works reasonably well for unidirectional communication, what
640 about bidirectional communication? The obvious thing you'd like to do
641 doesn't actually work:
643 open(PROG_FOR_READING_AND_WRITING, "| some program |")
645 and if you forget to use the C<use warnings> pragma or the B<-w> flag,
646 then you'll miss out entirely on the diagnostic message:
648 Can't do bidirectional pipe at -e line 1.
650 If you really want to, you can use the standard open2() library function
651 to catch both ends. There's also an open3() for tridirectional I/O so you
652 can also catch your child's STDERR, but doing so would then require an
653 awkward select() loop and wouldn't allow you to use normal Perl input
656 If you look at its source, you'll see that open2() uses low-level
657 primitives like Unix pipe() and exec() calls to create all the connections.
658 While it might have been slightly more efficient by using socketpair(), it
659 would have then been even less portable than it already is. The open2()
660 and open3() functions are unlikely to work anywhere except on a Unix
661 system or some other one purporting to be POSIX compliant.
663 Here's an example of using open2():
667 $pid = open2(*Reader, *Writer, "cat -u -n" );
668 print Writer "stuff\n";
671 The problem with this is that Unix buffering is really going to
672 ruin your day. Even though your C<Writer> filehandle is auto-flushed,
673 and the process on the other end will get your data in a timely manner,
674 you can't usually do anything to force it to give it back to you
675 in a similarly quick fashion. In this case, we could, because we
676 gave I<cat> a B<-u> flag to make it unbuffered. But very few Unix
677 commands are designed to operate over pipes, so this seldom works
678 unless you yourself wrote the program on the other end of the
681 A solution to this is the nonstandard F<Comm.pl> library. It uses
682 pseudo-ttys to make your program behave more reasonably:
685 $ph = open_proc('cat -n');
687 print $ph "a line\n";
688 print "got back ", scalar <$ph>;
691 This way you don't have to have control over the source code of the
692 program you're using. The F<Comm> library also has expect()
693 and interact() functions. Find the library (and we hope its
694 successor F<IPC::Chat>) at your nearest CPAN archive as detailed
695 in the SEE ALSO section below.
697 The newer Expect.pm module from CPAN also addresses this kind of thing.
698 This module requires two other modules from CPAN: IO::Pty and IO::Stty.
699 It sets up a pseudo-terminal to interact with programs that insist on
700 using talking to the terminal device driver. If your system is
701 amongst those supported, this may be your best bet.
703 =head2 Bidirectional Communication with Yourself
705 If you want, you may make low-level pipe() and fork()
706 to stitch this together by hand. This example only
707 talks to itself, but you could reopen the appropriate
708 handles to STDIN and STDOUT and call other processes.
711 # pipe1 - bidirectional communication using two pipe pairs
712 # designed for the socketpair-challenged
713 use IO::Handle; # thousands of lines just for autoflush :-(
714 pipe(PARENT_RDR, CHILD_WTR); # XXX: failure?
715 pipe(CHILD_RDR, PARENT_WTR); # XXX: failure?
716 CHILD_WTR->autoflush(1);
717 PARENT_WTR->autoflush(1);
720 close PARENT_RDR; close PARENT_WTR;
721 print CHILD_WTR "Parent Pid $$ is sending this\n";
722 chomp($line = <CHILD_RDR>);
723 print "Parent Pid $$ just read this: `$line'\n";
724 close CHILD_RDR; close CHILD_WTR;
727 die "cannot fork: $!" unless defined $pid;
728 close CHILD_RDR; close CHILD_WTR;
729 chomp($line = <PARENT_RDR>);
730 print "Child Pid $$ just read this: `$line'\n";
731 print PARENT_WTR "Child Pid $$ is sending this\n";
732 close PARENT_RDR; close PARENT_WTR;
736 But you don't actually have to make two pipe calls. If you
737 have the socketpair() system call, it will do this all for you.
740 # pipe2 - bidirectional communication using socketpair
741 # "the best ones always go both ways"
744 use IO::Handle; # thousands of lines just for autoflush :-(
745 # We say AF_UNIX because although *_LOCAL is the
746 # POSIX 1003.1g form of the constant, many machines
747 # still don't have it.
748 socketpair(CHILD, PARENT, AF_UNIX, SOCK_STREAM, PF_UNSPEC)
749 or die "socketpair: $!";
752 PARENT->autoflush(1);
756 print CHILD "Parent Pid $$ is sending this\n";
757 chomp($line = <CHILD>);
758 print "Parent Pid $$ just read this: `$line'\n";
762 die "cannot fork: $!" unless defined $pid;
764 chomp($line = <PARENT>);
765 print "Child Pid $$ just read this: `$line'\n";
766 print PARENT "Child Pid $$ is sending this\n";
771 =head1 Sockets: Client/Server Communication
773 While not limited to Unix-derived operating systems (e.g., WinSock on PCs
774 provides socket support, as do some VMS libraries), you may not have
775 sockets on your system, in which case this section probably isn't going to do
776 you much good. With sockets, you can do both virtual circuits (i.e., TCP
777 streams) and datagrams (i.e., UDP packets). You may be able to do even more
778 depending on your system.
780 The Perl function calls for dealing with sockets have the same names as
781 the corresponding system calls in C, but their arguments tend to differ
782 for two reasons: first, Perl filehandles work differently than C file
783 descriptors. Second, Perl already knows the length of its strings, so you
784 don't need to pass that information.
786 One of the major problems with old socket code in Perl was that it used
787 hard-coded values for some of the constants, which severely hurt
788 portability. If you ever see code that does anything like explicitly
789 setting C<$AF_INET = 2>, you know you're in for big trouble: An
790 immeasurably superior approach is to use the C<Socket> module, which more
791 reliably grants access to various constants and functions you'll need.
793 If you're not writing a server/client for an existing protocol like
794 NNTP or SMTP, you should give some thought to how your server will
795 know when the client has finished talking, and vice-versa. Most
796 protocols are based on one-line messages and responses (so one party
797 knows the other has finished when a "\n" is received) or multi-line
798 messages and responses that end with a period on an empty line
799 ("\n.\n" terminates a message/response).
801 =head2 Internet Line Terminators
803 The Internet line terminator is "\015\012". Under ASCII variants of
804 Unix, that could usually be written as "\r\n", but under other systems,
805 "\r\n" might at times be "\015\015\012", "\012\012\015", or something
806 completely different. The standards specify writing "\015\012" to be
807 conformant (be strict in what you provide), but they also recommend
808 accepting a lone "\012" on input (but be lenient in what you require).
809 We haven't always been very good about that in the code in this manpage,
810 but unless you're on a Mac, you'll probably be ok.
812 =head2 Internet TCP Clients and Servers
814 Use Internet-domain sockets when you want to do client-server
815 communication that might extend to machines outside of your own system.
817 Here's a sample TCP client using Internet-domain sockets:
822 my ($remote,$port, $iaddr, $paddr, $proto, $line);
824 $remote = shift || 'localhost';
825 $port = shift || 2345; # random port
826 if ($port =~ /\D/) { $port = getservbyname($port, 'tcp') }
827 die "No port" unless $port;
828 $iaddr = inet_aton($remote) || die "no host: $remote";
829 $paddr = sockaddr_in($port, $iaddr);
831 $proto = getprotobyname('tcp');
832 socket(SOCK, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
833 connect(SOCK, $paddr) || die "connect: $!";
834 while (defined($line = <SOCK>)) {
838 close (SOCK) || die "close: $!";
841 And here's a corresponding server to go along with it. We'll
842 leave the address as INADDR_ANY so that the kernel can choose
843 the appropriate interface on multihomed hosts. If you want sit
844 on a particular interface (like the external side of a gateway
845 or firewall machine), you should fill this in with your real address
850 BEGIN { $ENV{PATH} = '/usr/ucb:/bin' }
853 my $EOL = "\015\012";
855 sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" }
857 my $port = shift || 2345;
858 my $proto = getprotobyname('tcp');
860 ($port) = $port =~ /^(\d+)$/ or die "invalid port";
862 socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
863 setsockopt(Server, SOL_SOCKET, SO_REUSEADDR,
864 pack("l", 1)) || die "setsockopt: $!";
865 bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!";
866 listen(Server,SOMAXCONN) || die "listen: $!";
868 logmsg "server started on port $port";
872 $SIG{CHLD} = \&REAPER;
874 for ( ; $paddr = accept(Client,Server); close Client) {
875 my($port,$iaddr) = sockaddr_in($paddr);
876 my $name = gethostbyaddr($iaddr,AF_INET);
878 logmsg "connection from $name [",
879 inet_ntoa($iaddr), "]
882 print Client "Hello there, $name, it's now ",
883 scalar localtime, $EOL;
886 And here's a multithreaded version. It's multithreaded in that
887 like most typical servers, it spawns (forks) a slave server to
888 handle the client request so that the master server can quickly
889 go back to service a new client.
893 BEGIN { $ENV{PATH} = '/usr/ucb:/bin' }
896 my $EOL = "\015\012";
898 sub spawn; # forward declaration
899 sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" }
901 my $port = shift || 2345;
902 my $proto = getprotobyname('tcp');
904 ($port) = $port =~ /^(\d+)$/ or die "invalid port";
906 socket(Server, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
907 setsockopt(Server, SOL_SOCKET, SO_REUSEADDR,
908 pack("l", 1)) || die "setsockopt: $!";
909 bind(Server, sockaddr_in($port, INADDR_ANY)) || die "bind: $!";
910 listen(Server,SOMAXCONN) || die "listen: $!";
912 logmsg "server started on port $port";
917 use POSIX ":sys_wait_h";
921 local $!; # don't let waitpid() overwrite current error
922 while ((my $pid = waitpid(-1,WNOHANG)) > 0 && WIFEXITED($?)) {
923 logmsg "reaped $waitedpid" . ($? ? " with exit $?" : '');
925 $SIG{CHLD} = \&REAPER; # loathe sysV
928 $SIG{CHLD} = \&REAPER;
931 $paddr = accept(Client, Server) || do {
932 # try again if accept() returned because a signal was received
936 my ($port, $iaddr) = sockaddr_in($paddr);
937 my $name = gethostbyaddr($iaddr, AF_INET);
939 logmsg "connection from $name [",
945 print "Hello there, $name, it's now ", scalar localtime, $EOL;
946 exec '/usr/games/fortune' # XXX: `wrong' line terminators
947 or confess "can't exec fortune: $!";
955 unless (@_ == 0 && $coderef && ref($coderef) eq 'CODE') {
956 confess "usage: spawn CODEREF";
960 if (! defined($pid = fork)) {
961 logmsg "cannot fork: $!";
966 return; # I'm the parent
968 # else I'm the child -- go spawn
970 open(STDIN, "<&Client") || die "can't dup client to stdin";
971 open(STDOUT, ">&Client") || die "can't dup client to stdout";
972 ## open(STDERR, ">&STDOUT") || die "can't dup stdout to stderr";
976 This server takes the trouble to clone off a child version via fork()
977 for each incoming request. That way it can handle many requests at
978 once, which you might not always want. Even if you don't fork(), the
979 listen() will allow that many pending connections. Forking servers
980 have to be particularly careful about cleaning up their dead children
981 (called "zombies" in Unix parlance), because otherwise you'll quickly
982 fill up your process table. The REAPER subroutine is used here to
983 call waitpid() for any child processes that have finished, thereby
984 ensuring that they terminate cleanly and don't join the ranks of the
987 Within the while loop we call accept() and check to see if it returns
988 a false value. This would normally indicate a system error that needs
989 to be reported. However the introduction of safe signals (see
990 L</Deferred Signals (Safe Signals)> above) in Perl 5.7.3 means that
991 accept() may also be interrupted when the process receives a signal.
992 This typically happens when one of the forked sub-processes exits and
993 notifies the parent process with a CHLD signal.
995 If accept() is interrupted by a signal then $! will be set to EINTR.
996 If this happens then we can safely continue to the next iteration of
997 the loop and another call to accept(). It is important that your
998 signal handling code doesn't modify the value of $! or this test will
999 most likely fail. In the REAPER subroutine we create a local version
1000 of $! before calling waitpid(). When waitpid() sets $! to ECHILD (as
1001 it inevitably does when it has no more children waiting), it will
1002 update the local copy leaving the original unchanged.
1004 We suggest that you use the B<-T> flag to use taint checking (see L<perlsec>)
1005 even if we aren't running setuid or setgid. This is always a good idea
1006 for servers and other programs run on behalf of someone else (like CGI
1007 scripts), because it lessens the chances that people from the outside will
1008 be able to compromise your system.
1010 Let's look at another TCP client. This one connects to the TCP "time"
1011 service on a number of different machines and shows how far their clocks
1012 differ from the system on which it's being run:
1018 my $SECS_of_70_YEARS = 2208988800;
1019 sub ctime { scalar localtime(shift) }
1021 my $iaddr = gethostbyname('localhost');
1022 my $proto = getprotobyname('tcp');
1023 my $port = getservbyname('time', 'tcp');
1024 my $paddr = sockaddr_in(0, $iaddr);
1028 printf "%-24s %8s %s\n", "localhost", 0, ctime(time());
1030 foreach $host (@ARGV) {
1031 printf "%-24s ", $host;
1032 my $hisiaddr = inet_aton($host) || die "unknown host";
1033 my $hispaddr = sockaddr_in($port, $hisiaddr);
1034 socket(SOCKET, PF_INET, SOCK_STREAM, $proto) || die "socket: $!";
1035 connect(SOCKET, $hispaddr) || die "bind: $!";
1037 read(SOCKET, $rtime, 4);
1039 my $histime = unpack("N", $rtime) - $SECS_of_70_YEARS;
1040 printf "%8d %s\n", $histime - time, ctime($histime);
1043 =head2 Unix-Domain TCP Clients and Servers
1045 That's fine for Internet-domain clients and servers, but what about local
1046 communications? While you can use the same setup, sometimes you don't
1047 want to. Unix-domain sockets are local to the current host, and are often
1048 used internally to implement pipes. Unlike Internet domain sockets, Unix
1049 domain sockets can show up in the file system with an ls(1) listing.
1052 srw-rw-rw- 1 root 0 Oct 31 07:23 /dev/log
1054 You can test for these with Perl's B<-S> file test:
1056 unless ( -S '/dev/log' ) {
1057 die "something's wicked with the log system";
1060 Here's a sample Unix-domain client:
1065 my ($rendezvous, $line);
1067 $rendezvous = shift || 'catsock';
1068 socket(SOCK, PF_UNIX, SOCK_STREAM, 0) || die "socket: $!";
1069 connect(SOCK, sockaddr_un($rendezvous)) || die "connect: $!";
1070 while (defined($line = <SOCK>)) {
1075 And here's a corresponding server. You don't have to worry about silly
1076 network terminators here because Unix domain sockets are guaranteed
1077 to be on the localhost, and thus everything works right.
1084 BEGIN { $ENV{PATH} = '/usr/ucb:/bin' }
1085 sub spawn; # forward declaration
1086 sub logmsg { print "$0 $$: @_ at ", scalar localtime, "\n" }
1088 my $NAME = 'catsock';
1089 my $uaddr = sockaddr_un($NAME);
1090 my $proto = getprotobyname('tcp');
1092 socket(Server,PF_UNIX,SOCK_STREAM,0) || die "socket: $!";
1094 bind (Server, $uaddr) || die "bind: $!";
1095 listen(Server,SOMAXCONN) || die "listen: $!";
1097 logmsg "server started on $NAME";
1101 use POSIX ":sys_wait_h";
1104 while (($waitedpid = waitpid(-1,WNOHANG)) > 0) {
1105 logmsg "reaped $waitedpid" . ($? ? " with exit $?" : '');
1107 $SIG{CHLD} = \&REAPER; # loathe sysV
1110 $SIG{CHLD} = \&REAPER;
1113 for ( $waitedpid = 0;
1114 accept(Client,Server) || $waitedpid;
1115 $waitedpid = 0, close Client)
1118 logmsg "connection on $NAME";
1120 print "Hello there, it's now ", scalar localtime, "\n";
1121 exec '/usr/games/fortune' or die "can't exec fortune: $!";
1126 my $coderef = shift;
1128 unless (@_ == 0 && $coderef && ref($coderef) eq 'CODE') {
1129 confess "usage: spawn CODEREF";
1133 if (!defined($pid = fork)) {
1134 logmsg "cannot fork: $!";
1137 logmsg "begat $pid";
1138 return; # I'm the parent
1140 # else I'm the child -- go spawn
1142 open(STDIN, "<&Client") || die "can't dup client to stdin";
1143 open(STDOUT, ">&Client") || die "can't dup client to stdout";
1144 ## open(STDERR, ">&STDOUT") || die "can't dup stdout to stderr";
1148 As you see, it's remarkably similar to the Internet domain TCP server, so
1149 much so, in fact, that we've omitted several duplicate functions--spawn(),
1150 logmsg(), ctime(), and REAPER()--which are exactly the same as in the
1153 So why would you ever want to use a Unix domain socket instead of a
1154 simpler named pipe? Because a named pipe doesn't give you sessions. You
1155 can't tell one process's data from another's. With socket programming,
1156 you get a separate session for each client: that's why accept() takes two
1159 For example, let's say that you have a long running database server daemon
1160 that you want folks from the World Wide Web to be able to access, but only
1161 if they go through a CGI interface. You'd have a small, simple CGI
1162 program that does whatever checks and logging you feel like, and then acts
1163 as a Unix-domain client and connects to your private server.
1165 =head1 TCP Clients with IO::Socket
1167 For those preferring a higher-level interface to socket programming, the
1168 IO::Socket module provides an object-oriented approach. IO::Socket is
1169 included as part of the standard Perl distribution as of the 5.004
1170 release. If you're running an earlier version of Perl, just fetch
1171 IO::Socket from CPAN, where you'll also find modules providing easy
1172 interfaces to the following systems: DNS, FTP, Ident (RFC 931), NIS and
1173 NISPlus, NNTP, Ping, POP3, SMTP, SNMP, SSLeay, Telnet, and Time--just
1176 =head2 A Simple Client
1178 Here's a client that creates a TCP connection to the "daytime"
1179 service at port 13 of the host name "localhost" and prints out everything
1180 that the server there cares to provide.
1184 $remote = IO::Socket::INET->new(
1186 PeerAddr => "localhost",
1187 PeerPort => "daytime(13)",
1189 or die "cannot connect to daytime port at localhost";
1190 while ( <$remote> ) { print }
1192 When you run this program, you should get something back that
1195 Wed May 14 08:40:46 MDT 1997
1197 Here are what those parameters to the C<new> constructor mean:
1203 This is which protocol to use. In this case, the socket handle returned
1204 will be connected to a TCP socket, because we want a stream-oriented
1205 connection, that is, one that acts pretty much like a plain old file.
1206 Not all sockets are this of this type. For example, the UDP protocol
1207 can be used to make a datagram socket, used for message-passing.
1211 This is the name or Internet address of the remote host the server is
1212 running on. We could have specified a longer name like C<"www.perl.com">,
1213 or an address like C<"204.148.40.9">. For demonstration purposes, we've
1214 used the special hostname C<"localhost">, which should always mean the
1215 current machine you're running on. The corresponding Internet address
1216 for localhost is C<"127.1">, if you'd rather use that.
1220 This is the service name or port number we'd like to connect to.
1221 We could have gotten away with using just C<"daytime"> on systems with a
1222 well-configured system services file,[FOOTNOTE: The system services file
1223 is in I</etc/services> under Unix] but just in case, we've specified the
1224 port number (13) in parentheses. Using just the number would also have
1225 worked, but constant numbers make careful programmers nervous.
1229 Notice how the return value from the C<new> constructor is used as
1230 a filehandle in the C<while> loop? That's what's called an indirect
1231 filehandle, a scalar variable containing a filehandle. You can use
1232 it the same way you would a normal filehandle. For example, you
1233 can read one line from it this way:
1237 all remaining lines from is this way:
1241 and send a line of data to it this way:
1243 print $handle "some data\n";
1245 =head2 A Webget Client
1247 Here's a simple client that takes a remote host to fetch a document
1248 from, and then a list of documents to get from that host. This is a
1249 more interesting client than the previous one because it first sends
1250 something to the server before fetching the server's response.
1254 unless (@ARGV > 1) { die "usage: $0 host document ..." }
1255 $host = shift(@ARGV);
1258 foreach $document ( @ARGV ) {
1259 $remote = IO::Socket::INET->new( Proto => "tcp",
1261 PeerPort => "http(80)",
1263 unless ($remote) { die "cannot connect to http daemon on $host" }
1264 $remote->autoflush(1);
1265 print $remote "GET $document HTTP/1.0" . $BLANK;
1266 while ( <$remote> ) { print }
1270 The web server handing the "http" service, which is assumed to be at
1271 its standard port, number 80. If the web server you're trying to
1272 connect to is at a different port (like 1080 or 8080), you should specify
1273 as the named-parameter pair, C<< PeerPort => 8080 >>. The C<autoflush>
1274 method is used on the socket because otherwise the system would buffer
1275 up the output we sent it. (If you're on a Mac, you'll also need to
1276 change every C<"\n"> in your code that sends data over the network to
1277 be a C<"\015\012"> instead.)
1279 Connecting to the server is only the first part of the process: once you
1280 have the connection, you have to use the server's language. Each server
1281 on the network has its own little command language that it expects as
1282 input. The string that we send to the server starting with "GET" is in
1283 HTTP syntax. In this case, we simply request each specified document.
1284 Yes, we really are making a new connection for each document, even though
1285 it's the same host. That's the way you always used to have to speak HTTP.
1286 Recent versions of web browsers may request that the remote server leave
1287 the connection open a little while, but the server doesn't have to honor
1290 Here's an example of running that program, which we'll call I<webget>:
1292 % webget www.perl.com /guanaco.html
1293 HTTP/1.1 404 File Not Found
1294 Date: Thu, 08 May 1997 18:02:32 GMT
1295 Server: Apache/1.2b6
1297 Content-type: text/html
1299 <HEAD><TITLE>404 File Not Found</TITLE></HEAD>
1300 <BODY><H1>File Not Found</H1>
1301 The requested URL /guanaco.html was not found on this server.<P>
1304 Ok, so that's not very interesting, because it didn't find that
1305 particular document. But a long response wouldn't have fit on this page.
1307 For a more fully-featured version of this program, you should look to
1308 the I<lwp-request> program included with the LWP modules from CPAN.
1310 =head2 Interactive Client with IO::Socket
1312 Well, that's all fine if you want to send one command and get one answer,
1313 but what about setting up something fully interactive, somewhat like
1314 the way I<telnet> works? That way you can type a line, get the answer,
1315 type a line, get the answer, etc.
1317 This client is more complicated than the two we've done so far, but if
1318 you're on a system that supports the powerful C<fork> call, the solution
1319 isn't that rough. Once you've made the connection to whatever service
1320 you'd like to chat with, call C<fork> to clone your process. Each of
1321 these two identical process has a very simple job to do: the parent
1322 copies everything from the socket to standard output, while the child
1323 simultaneously copies everything from standard input to the socket.
1324 To accomplish the same thing using just one process would be I<much>
1325 harder, because it's easier to code two processes to do one thing than it
1326 is to code one process to do two things. (This keep-it-simple principle
1327 a cornerstones of the Unix philosophy, and good software engineering as
1328 well, which is probably why it's spread to other systems.)
1335 my ($host, $port, $kidpid, $handle, $line);
1337 unless (@ARGV == 2) { die "usage: $0 host port" }
1338 ($host, $port) = @ARGV;
1340 # create a tcp connection to the specified host and port
1341 $handle = IO::Socket::INET->new(Proto => "tcp",
1344 or die "can't connect to port $port on $host: $!";
1346 $handle->autoflush(1); # so output gets there right away
1347 print STDERR "[Connected to $host:$port]\n";
1349 # split the program into two processes, identical twins
1350 die "can't fork: $!" unless defined($kidpid = fork());
1352 # the if{} block runs only in the parent process
1354 # copy the socket to standard output
1355 while (defined ($line = <$handle>)) {
1358 kill("TERM", $kidpid); # send SIGTERM to child
1360 # the else{} block runs only in the child process
1362 # copy standard input to the socket
1363 while (defined ($line = <STDIN>)) {
1364 print $handle $line;
1368 The C<kill> function in the parent's C<if> block is there to send a
1369 signal to our child process (current running in the C<else> block)
1370 as soon as the remote server has closed its end of the connection.
1372 If the remote server sends data a byte at time, and you need that
1373 data immediately without waiting for a newline (which might not happen),
1374 you may wish to replace the C<while> loop in the parent with the
1378 while (sysread($handle, $byte, 1) == 1) {
1382 Making a system call for each byte you want to read is not very efficient
1383 (to put it mildly) but is the simplest to explain and works reasonably
1386 =head1 TCP Servers with IO::Socket
1388 As always, setting up a server is little bit more involved than running a client.
1389 The model is that the server creates a special kind of socket that
1390 does nothing but listen on a particular port for incoming connections.
1391 It does this by calling the C<< IO::Socket::INET->new() >> method with
1392 slightly different arguments than the client did.
1398 This is which protocol to use. Like our clients, we'll
1399 still specify C<"tcp"> here.
1404 port in the C<LocalPort> argument, which we didn't do for the client.
1405 This is service name or port number for which you want to be the
1406 server. (Under Unix, ports under 1024 are restricted to the
1407 superuser.) In our sample, we'll use port 9000, but you can use
1408 any port that's not currently in use on your system. If you try
1409 to use one already in used, you'll get an "Address already in use"
1410 message. Under Unix, the C<netstat -a> command will show
1411 which services current have servers.
1415 The C<Listen> parameter is set to the maximum number of
1416 pending connections we can accept until we turn away incoming clients.
1417 Think of it as a call-waiting queue for your telephone.
1418 The low-level Socket module has a special symbol for the system maximum, which
1423 The C<Reuse> parameter is needed so that we restart our server
1424 manually without waiting a few minutes to allow system buffers to
1429 Once the generic server socket has been created using the parameters
1430 listed above, the server then waits for a new client to connect
1431 to it. The server blocks in the C<accept> method, which eventually accepts a
1432 bidirectional connection from the remote client. (Make sure to autoflush
1433 this handle to circumvent buffering.)
1435 To add to user-friendliness, our server prompts the user for commands.
1436 Most servers don't do this. Because of the prompt without a newline,
1437 you'll have to use the C<sysread> variant of the interactive client above.
1439 This server accepts one of five different commands, sending output
1440 back to the client. Note that unlike most network servers, this one
1441 only handles one incoming client at a time. Multithreaded servers are
1442 covered in Chapter 6 of the Camel.
1444 Here's the code. We'll
1448 use Net::hostent; # for OO version of gethostbyaddr
1450 $PORT = 9000; # pick something not in use
1452 $server = IO::Socket::INET->new( Proto => 'tcp',
1454 Listen => SOMAXCONN,
1457 die "can't setup server" unless $server;
1458 print "[Server $0 accepting clients]\n";
1460 while ($client = $server->accept()) {
1461 $client->autoflush(1);
1462 print $client "Welcome to $0; type help for command list.\n";
1463 $hostinfo = gethostbyaddr($client->peeraddr);
1464 printf "[Connect from %s]\n", $hostinfo ? $hostinfo->name : $client->peerhost;
1465 print $client "Command? ";
1466 while ( <$client>) {
1467 next unless /\S/; # blank line
1468 if (/quit|exit/i) { last; }
1469 elsif (/date|time/i) { printf $client "%s\n", scalar localtime; }
1470 elsif (/who/i ) { print $client `who 2>&1`; }
1471 elsif (/cookie/i ) { print $client `/usr/games/fortune 2>&1`; }
1472 elsif (/motd/i ) { print $client `cat /etc/motd 2>&1`; }
1474 print $client "Commands: quit date who cookie motd\n";
1477 print $client "Command? ";
1482 =head1 UDP: Message Passing
1484 Another kind of client-server setup is one that uses not connections, but
1485 messages. UDP communications involve much lower overhead but also provide
1486 less reliability, as there are no promises that messages will arrive at
1487 all, let alone in order and unmangled. Still, UDP offers some advantages
1488 over TCP, including being able to "broadcast" or "multicast" to a whole
1489 bunch of destination hosts at once (usually on your local subnet). If you
1490 find yourself overly concerned about reliability and start building checks
1491 into your message system, then you probably should use just TCP to start
1494 Note that UDP datagrams are I<not> a bytestream and should not be treated
1495 as such. This makes using I/O mechanisms with internal buffering
1496 like stdio (i.e. print() and friends) especially cumbersome. Use syswrite(),
1497 or better send(), like in the example below.
1499 Here's a UDP program similar to the sample Internet TCP client given
1500 earlier. However, instead of checking one host at a time, the UDP version
1501 will check many of them asynchronously by simulating a multicast and then
1502 using select() to do a timed-out wait for I/O. To do something similar
1503 with TCP, you'd have to use a different socket handle for each host.
1510 my ( $count, $hisiaddr, $hispaddr, $histime,
1511 $host, $iaddr, $paddr, $port, $proto,
1512 $rin, $rout, $rtime, $SECS_of_70_YEARS);
1514 $SECS_of_70_YEARS = 2208988800;
1516 $iaddr = gethostbyname(hostname());
1517 $proto = getprotobyname('udp');
1518 $port = getservbyname('time', 'udp');
1519 $paddr = sockaddr_in(0, $iaddr); # 0 means let kernel pick
1521 socket(SOCKET, PF_INET, SOCK_DGRAM, $proto) || die "socket: $!";
1522 bind(SOCKET, $paddr) || die "bind: $!";
1525 printf "%-12s %8s %s\n", "localhost", 0, scalar localtime time;
1529 $hisiaddr = inet_aton($host) || die "unknown host";
1530 $hispaddr = sockaddr_in($port, $hisiaddr);
1531 defined(send(SOCKET, 0, 0, $hispaddr)) || die "send $host: $!";
1535 vec($rin, fileno(SOCKET), 1) = 1;
1537 # timeout after 10.0 seconds
1538 while ($count && select($rout = $rin, undef, undef, 10.0)) {
1540 ($hispaddr = recv(SOCKET, $rtime, 4, 0)) || die "recv: $!";
1541 ($port, $hisiaddr) = sockaddr_in($hispaddr);
1542 $host = gethostbyaddr($hisiaddr, AF_INET);
1543 $histime = unpack("N", $rtime) - $SECS_of_70_YEARS;
1544 printf "%-12s ", $host;
1545 printf "%8d %s\n", $histime - time, scalar localtime($histime);
1549 Note that this example does not include any retries and may consequently
1550 fail to contact a reachable host. The most prominent reason for this
1551 is congestion of the queues on the sending host if the number of
1552 list of hosts to contact is sufficiently large.
1556 While System V IPC isn't so widely used as sockets, it still has some
1557 interesting uses. You can't, however, effectively use SysV IPC or
1558 Berkeley mmap() to have shared memory so as to share a variable amongst
1559 several processes. That's because Perl would reallocate your string when
1560 you weren't wanting it to.
1562 Here's a small example showing shared memory usage.
1564 use IPC::SysV qw(IPC_PRIVATE IPC_RMID S_IRUSR S_IWUSR);
1567 $id = shmget(IPC_PRIVATE, $size, S_IRUSR|S_IWUSR) || die "$!";
1568 print "shm key $id\n";
1570 $message = "Message #1";
1571 shmwrite($id, $message, 0, 60) || die "$!";
1572 print "wrote: '$message'\n";
1573 shmread($id, $buff, 0, 60) || die "$!";
1574 print "read : '$buff'\n";
1576 # the buffer of shmread is zero-character end-padded.
1577 substr($buff, index($buff, "\0")) = '';
1578 print "un" unless $buff eq $message;
1581 print "deleting shm $id\n";
1582 shmctl($id, IPC_RMID, 0) || die "$!";
1584 Here's an example of a semaphore:
1586 use IPC::SysV qw(IPC_CREAT);
1589 $id = semget($IPC_KEY, 10, 0666 | IPC_CREAT ) || die "$!";
1590 print "shm key $id\n";
1592 Put this code in a separate file to be run in more than one process.
1593 Call the file F<take>:
1595 # create a semaphore
1598 $id = semget($IPC_KEY, 0 , 0 );
1599 die if !defined($id);
1605 # wait for semaphore to be zero
1607 $opstring1 = pack("s!s!s!", $semnum, $semop, $semflag);
1609 # Increment the semaphore count
1611 $opstring2 = pack("s!s!s!", $semnum, $semop, $semflag);
1612 $opstring = $opstring1 . $opstring2;
1614 semop($id,$opstring) || die "$!";
1616 Put this code in a separate file to be run in more than one process.
1617 Call this file F<give>:
1619 # 'give' the semaphore
1620 # run this in the original process and you will see
1621 # that the second process continues
1624 $id = semget($IPC_KEY, 0, 0);
1625 die if !defined($id);
1630 # Decrement the semaphore count
1632 $opstring = pack("s!s!s!", $semnum, $semop, $semflag);
1634 semop($id,$opstring) || die "$!";
1636 The SysV IPC code above was written long ago, and it's definitely
1637 clunky looking. For a more modern look, see the IPC::SysV module
1638 which is included with Perl starting from Perl 5.005.
1640 A small example demonstrating SysV message queues:
1642 use IPC::SysV qw(IPC_PRIVATE IPC_RMID IPC_CREAT S_IRUSR S_IWUSR);
1644 my $id = msgget(IPC_PRIVATE, IPC_CREAT | S_IRUSR | S_IWUSR);
1646 my $sent = "message";
1647 my $type_sent = 1234;
1652 if (msgsnd($id, pack("l! a*", $type_sent, $sent), 0)) {
1653 if (msgrcv($id, $rcvd, 60, 0, 0)) {
1654 ($type_rcvd, $rcvd) = unpack("l! a*", $rcvd);
1655 if ($rcvd eq $sent) {
1661 die "# msgrcv failed\n";
1664 die "# msgsnd failed\n";
1666 msgctl($id, IPC_RMID, 0) || die "# msgctl failed: $!\n";
1668 die "# msgget failed\n";
1673 Most of these routines quietly but politely return C<undef> when they
1674 fail instead of causing your program to die right then and there due to
1675 an uncaught exception. (Actually, some of the new I<Socket> conversion
1676 functions croak() on bad arguments.) It is therefore essential to
1677 check return values from these functions. Always begin your socket
1678 programs this way for optimal success, and don't forget to add B<-T>
1679 taint checking flag to the #! line for servers:
1688 All these routines create system-specific portability problems. As noted
1689 elsewhere, Perl is at the mercy of your C libraries for much of its system
1690 behaviour. It's probably safest to assume broken SysV semantics for
1691 signals and to stick with simple TCP and UDP socket operations; e.g., don't
1692 try to pass open file descriptors over a local UDP datagram socket if you
1693 want your code to stand a chance of being portable.
1697 Tom Christiansen, with occasional vestiges of Larry Wall's original
1698 version and suggestions from the Perl Porters.
1702 There's a lot more to networking than this, but this should get you
1705 For intrepid programmers, the indispensable textbook is I<Unix
1706 Network Programming, 2nd Edition, Volume 1> by W. Richard Stevens
1707 (published by Prentice-Hall). Note that most books on networking
1708 address the subject from the perspective of a C programmer; translation
1709 to Perl is left as an exercise for the reader.
1711 The IO::Socket(3) manpage describes the object library, and the Socket(3)
1712 manpage describes the low-level interface to sockets. Besides the obvious
1713 functions in L<perlfunc>, you should also check out the F<modules> file
1714 at your nearest CPAN site. (See L<perlmodlib> or best yet, the F<Perl
1715 FAQ> for a description of what CPAN is and where to get it.)
1717 Section 5 of the F<modules> file is devoted to "Networking, Device Control
1718 (modems), and Interprocess Communication", and contains numerous unbundled
1719 modules numerous networking modules, Chat and Expect operations, CGI
1720 programming, DCE, FTP, IPC, NNTP, Proxy, Ptty, RPC, SNMP, SMTP, Telnet,
1721 Threads, and ToolTalk--just to name a few.