Subject: | |
From: | |
Reply To: | |
Date: | Tue, 7 Aug 2001 15:23:33 -0700 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Ted Ashton wrote:
>
> Thus it was written in the epistle of Mark Bixby,
> > Ted Ashton wrote:
> > >
> > > First, the warning. I have discovered that at least in some circumstances,
> > > MPE::IMAGE does not play well with system(). I would appreciate it if
> > > someone could confirm that the following (be sure to change the database
> > > name and password) hangs on the system() call:
> > >
> > > use MPE::IMAGE qw(:all);
> > > my $db = DbOpen('TEST.DATABASE','MYPASS',5);
> > > print "Made it thus far . . .\n";
> > > system('callci','showjob');
> >
> > The Perl system() function is going to have to fork a shell, and fork cannot
> > handle open TurboIMAGE databases.
>
> Bing! I'd forgotten that. Thanks! Evidently it can't handle forking when
> an IMAGE database *has been open*, as adding a DbClose() before the system()
> call doesn't solve the problem. Does that fit with what others understand
> about the problem?
But according to the stack trace, we haven't even called POSIX system() or
fork() yet. It is the Perl interpreter that has decided to call sleep().
I think that to understand why Perl is calling sleep(), you will need to look
at the Perl interpreter source code.
--
[log in to unmask]
Remainder of .sig suppressed to conserve scarce California electrons...
* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
|
|
|