Re: segfault with just running "rpm -qa"

[Jeff Johnson <n3npq nc rr com>]

Robert P. J. Day wrote:

> On Wed, 24 Sep 2003, Paul Heinlein wrote:
>
>  
>
> > On Wed, 24 Sep 2003, Stephen Smoogen wrote:
> >
> >    
> >
> > > My first course of action would be to see if the Red Hat default
> > > 2.4.x kernel showed this problem. If it doesnt then I would say it
> > > is something in the latest test NPTL items that eneds to be looked
> > > at. I would also get the latest rpm from either rawhide or rpm.org.
> > > Try to get it update itself.
> > >      
> > The easy way to test if NPTL is the problem is to use the
> > LD_ASSUME_KERNEL hack:
> >
> >  LD_ASSUME_KERNEL=2.2.5 rpm -qa
> >    
>
> tried it, didn't make a difference.  this *appeared* to start just
> after i moved to the 2.6.0-test5-bk10 kernel, although i can't be
> certain of the exact correspondence.
>
> my previous version was the bk3 kernel, i'll reboot to that and
> see what happens.  argh.

I am running arjanv's
   
http://people.redhat.com/arjanv/2.5/SRPMS.kernel/kernel-smp-2.6.0-0.test5.1.38.src.rpm
rebuilt with some usb fiddles (that have nothing to do with nptl or 
rpm-4.2.1) sucessfully.

Note carefully that "successfully" (for me) does not mean I recommend 
2.6.0 to anyone, there
are still problems with usb, probably more, that I am blissfully unaware of.

However, I do not see NPTL or O_DIRECT problems while using latest rpm-4.2.1
packages from the usual places while running -test4.1.38 (or 36 or 34).

Nice job Arjan, I forgive you O_DIRECT! ;-)

YMMV, everyone's does whil testing bleeding edge software.

73 de Jeff