in reply to Re: Re: Re: forking memory problem
in thread forking memory problem
"If message isn't in perldiag, it probably isn't generated by Perl. Consult your operating system documentation instead."
Of course it is not a Perl generated message, Perl already crashed upon memory access violation. This is a message generated by operating system, but it is caused by Perl.
"The code snippet provided leaves open the possibility of a runaway loop."
I actually never said that possibility is exculded. What I said is that, run away loop is irrelevent here (The error message does not indicate this). If his code has a run away loop, with the run away loop alone, sacco's code will quickly use up resource and can no longer fork (so his fork will fail, and get the message I mentioned. I saw this message before on both unix and win32 when I forgot to waitpid and kept forking.)
On the other hand, even if it is caused by run away loop, it is still a Perl bug. Perl shall not crash on run away loop.
|
|---|