faxrunqd still busted on SUNOS

Gert Doering (gert@greenie.muc.de)
Thu, 20 Feb 1997 11:12:30 +0100


Hi,

Ulrich Pfeifer wrote:
> For your information, the Perl5 pumpkin holders word on this:
> 
> According to Chip Salzenberg:
> 
> > This tells me only that the system() operator ought to do the moral
> > equivalent of C<local $SIG{CHLD} = 'DFL'>.  (You can't expect to grab
> > the status of the child in a SIGCHLD handler and also have system()
> > know what it is.)

Yes, that's about the conclusion I came to. The new faxrunqd does not use
system, but direct fork()/exec() combination.

Also, it doesn't use signals, as perl's "wait()" function is more than
appropriate for the task.

These design issues *enormously* simplify the rewrite.

gert

-- 
USENET is *not* the non-clickable part of WWW!
                                                           //www.muc.de/~gert/
Gert Doering - Munich, Germany                             gert@greenie.muc.de
fax: +49-89-3545980                         gert.doering@physik.tu-muenchen.de