Posts Tagged by Perl

Here’s how to debug a Perl program that’s running remotely on a server that doesn’t have X. It’s pretty simple, these are just notes for me…

You’ve ssh’d into a remote server (headless), to debug a perl script that forks:

$SIG{CHLD} = sub {
  while (1) {
    my $nino = waitpid(-1, WNOHANG);
    last if $nino <= 0;
    if ($?) {
      &log("ERROR: Child process $nino exited with $?");
    }
    $children--;
  }

# let's fork!
my $pid = fork();
if (not defined $pid) {
  &log("ERROR: fork(): resources not available");
}
elsif ($pid > 0) {
  # parent
  $children++;
  next;
}
else {
  # child
  $SIG{CHLD} = $SIG{TERM} = $SIG{INT} = 'DEFAULT';
  do_stuff($arg1, $arg2);
  exit;
}

But when you start debugging, you get output like this (which tells you what to do):

######### Forked, but do not know how to create a new TTY. #########
  Since two debuggers fight for the same TTY, input is severely entangled.

  I know how to switch the output to a different window in xterms, OS/2
  consoles, and Mac OS X Terminal.app only.  For a manual switch, put the name
  of the created TTY in $DB::fork_TTY, or define a function
  DB::get_fork_TTY() returning this.

  On UNIX-like systems one can get the name of a TTY for the given window
  by typing tty, and disconnect the shell from TTY by sleep 1000000.

The easiest solution is to install a minimal X on the target server, then reconnect via ssh, check your DISPLAY variable, then start up the debugger in xterm.

server% sudo aptitude install xinit xterm

client% ssh -X server                # reconnect, to setup DISPLAY

server% env | grep -i disp
DISPLAY=localhost:11.0               # DISPLAY is good

server% xterm &                      # an xterm should display locally

server% perl -d foo.pl               # start debugging

After forking, one or more additional xterms will display for each child, allowing you to debug.