SPUG: Dying Children

Dan Ebert mathin at mathin.com
Mon Sep 20 14:47:18 CDT 2004

Ha.  I think I have solved my own question so I thought I'd post what I
found out.

The child processes have a decent sized chunk of data (a hash with up to
500,000 keys).  When an 'exit' is called Perl does some memory
de-allocation which takes time, hence the delay.

You can get around this by calling POSIX::_exit(0).  (I know you are not
supposed to use &_ sub routines from a module :), but this was suggested
by the Camel book as a way of avoiding the execution of END blocks, so
what the hell.)  It bypasses the destructor processing (and END blocks).

I think that since the child is a separate process, the system just
reclaims the memory when it dies anyway.  I kept a close eye on the
memory usage with top while the script was running and I didn't see any

If anyone sees any problems I'm overlooking here I'd appreciate hearing
about them.


Immigration is the sincerest form of flattery.
	- Unknown

On Mon, 20 Sep 2004, Dan Ebert wrote:

> I have a script which forks of a child process to write a file while the
> parent script goes on to collect the data for the next file.  I'm keeping
> track of the number of file writing children so I don't have too many
> writers at one time.  I have a debug print statement right before the
> exit in the child process:
> print "done writing file\n";
> exit;
> I am noticing that there are sometimes several seconds between the print
> and the actual death of the child.  This is delaying the forking of the
> next child which is slowing the script down.  Has anyone seen this before?
> Thanks,
> Dan.
> ----------------------------------------------------------
> Immigration is the sincerest form of flattery.
> 	- Unknown
> ----------------------------------------------------------
> _____________________________________________________________
> Seattle Perl Users Group Mailing List
> POST TO: spug-list at mail.pm.org  http://spugwiki.perlocity.org
> ACCOUNT CONFIG: http://mail.pm.org/mailman/listinfo/spug-list
> MEETINGS: 3rd Tuesdays, Location Unknown
> WEB PAGE: http://www.seattleperl.org

More information about the spug-list mailing list