> > From: Matthew Jacob <mjacob@feral.com>
> >
> > # mv sh sh.old
> > # mv sh.new sh
> > # sh
> >
> > ^C^C^C
> >
> > I just probably toasted myself. This was a shell built today, but
> > statically linked with a march 8 libc.a.
> 
> Well, but .. reports were that the problem could be reliably repeated
> and avoided depending on which kernel one booted.
> 
> Oh .. what kernel were you running that sh(1) on?  It's possible that
> the kernel problem could affect sh(1).

last Sunday sources hang. I'm now getting out last saturday now that CVS
is opened again.

> 
> Furthermore, any hypothetical sh(1) malfunction wouldn't stop you
> from, say, doing a telnet(1) or ssh(1) in. Were other sessions affected
> or was just the one sh(1) process wedged?

I was still at single user.


Once I've correctly bracketed then I'll look at what changed and figure
out. Some time tomorrow.