Re: Fixing 131478 in debootstrap
On Sun, 2002-03-03 at 21:54, Chris Tillman wrote:
> Since everyone's mystified about the Malformed release bug, and we're
> NMUing debootstrap, how about applying the patch in 131478? I don't
> really understand the issues the bug writer is talking about, but it
> sure seems like races and such can be a source for mysterious bugs.
> AFAICT the patch doesn't change the functionality. Why not try it?
It looks like the piece of code affected by the patch isn't executed
until after the point at which the "Malformed.." error is generated. So
I don't think it will help here, though the patch seems benign enough in
I must admit, I wasn't really aware that this problem was still with us.
My first inclination would be to attack it by having debootstrap
redirect wget's stderr to somewhere other than /dev/null, so that in the
event of download problems the user can at least figure out what's going
on. Debootstrap itself probably won't be able to spot failed downloads
of the Release file because its error checking relies on $PIPESTATUS
which ash doesn't seem to support.