develooper Front page | perl.perl5.porters | Postings from December 2017

Re: [perl #127794] Strange behavior when forking in BEGIN

Thread Previous | Thread Next
From:
Chad Granum
Date:
December 18, 2017 15:24
Subject:
Re: [perl #127794] Strange behavior when forking in BEGIN
Message ID:
CAJFr3ktX23YvNOOBgnFdgwwmZ9RXMbZDgnu1y=suMYuFFpYCmQ@mail.gmail.com
I have code that forks in BEGIN and works around this issue, I would oppose
making fork in begin fatal. That said I would be fine with a warning which
I could disable int he cases I really need to.

That said, I would greatly prefer some kind of fix, though I understand
that is probably not gonna happen.

-Chad

On Sun, Dec 17, 2017 at 5:18 AM, slaven@rezic.de via RT <
perlbug-followup@perl.org> wrote:

> Dana Sun, 17 Dec 2017 02:10:02 -0800, hv reče:
> > On Sat, 16 Dec 2017 00:40:06 -0800, zefram@fysh.org wrote:
> > > This kind of strangeness is only what should be expected from forking
> > > in a BEGIN block.  This ticket should be closed.
> >
> > What who should expect - someone intimately familiar with perl
> > internals, or someone who uses perl and (on a good day) may have read
> > `perldoc -f fork`?
> >
> > The preceding discussion suggests to me that an attempt to fork() at
> > BEGIN time should yield at least a warning (and maybe a fatal error)
> > to reflect the fact perl will not be able to honour a reasonable non-
> > expert's expectations of what that means.
>
> Probably there are legitimate uses of fork() in a BEGIN block --- I would
> expect that a fork+exec here is harmless.
>
> Regards,
>     Slaven
>
>
>
> ---
> via perlbug:  queue: perl5 status: open
> https://rt.perl.org/Ticket/Display.html?id=127794
>

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About