Re: [Fwd: issue with -j 2]

[Antoine Petitet]

Hi Paul,

Thanks for your swift reply. The kernel is on this machine
2.4.21-37.EL #1 SMP
I agree with you that this is rather unlikely that the problem
is in GNU Make. I have searched the web quite a bit for a
problem that could be related to no avail. I have searched on
the machine itself who was killing the process, but could not
find it (no messages in the system logs). What is disturbing too
is that the object file is indeed produced, but the main process
still dies.

I tried multiple times on another machine the same command
but could not reproduce the problem there. The machine is
running a kernel 2.6.9-5.ELsmp. This is a Red Hat Enterprise
Linux WS release 4 (Nahant) distribution.

Best wishes,

Antoine


Paul Smith wrote:

> On Thu, 2009-02-19 at 13:23 +0100, Antoine Petitet wrote:
>  
>
> > Hi Folks,
> >
> > I sent this message to help mailing alias. I should have sent it
> > maybe to this alias, eventhough, it is not obvious to me at this
> > point that the issue is in GNU make.
> >    
>
> This is not good at all.  This means that somehow make lost track of a
> child process: that child process died and was somehow reaped, but make
> either didn't catch the signal (which should not be possible) or else
> make didn't handle it properly when it was caught... which seems
> unlikely since I'd think we'd have a lot more instances of this
> happening.
>
> I wonder if you have the chance to try this on anything newer than Red
> Hat Enterprise 3.  That's pretty old.  Is it still using a 2.4 kernel (I
> know RHEL2 used a 2.4 kernel and RHEL4 used a 2.6 kernel, but I have no
> idea what RHEL3 used)?
>
>