[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: debsign process



On Wed, May 30, 2001 at 04:04:11PM -0400, Ben Collins wrote:
> > > If the debsign process fails the first sign, because you enter the
> > > invalid passphrase for your gpgkey, no *.changes are created.
> > > 
> > > This seems unlogic! Shouldn't it create *.changes and *.dsc files,
> > > and then start signing, so if it fails, one can run debsign on
> > > the *.changes file manually?
> > 
> > Agreed!  Please file this as a wishlist bug against dpkg-dev (the
> > program involved is dpkg-buildpackage).
> 
> Maybe the debsign should be renamed dpkg-sign, and added to dpkg-dev.
> That way it can be called seperately (and we combine code).
> 
> Also, FYI, someone is offering to recreate debsign as a C program, so
> that it only has to ask for the passphrase once. This will help a lot,
> especially if/when package signing is an everyday occurence.

Fun, fun, fun.  There are three different signing programs around at
present, AFAIK: dpkg-buildpackage, which signs the .dsc and .changes
files; debsign, which emulates the signing part of dpkg-buildpackage,
and debsigs, which signs the control.tar.gz and data.tar.gz within the
.deb itself.  None of them are safe to cache the passphrase (which
should require a setuid-root binary to allocate safe memory; I note
that mutt does not do this, though).  Which one would be rewritten?

Nevertheless, I like the idea of debsign (or an enhanced version
thereof) becoming part of dpkg-dev.

   Julian

-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

         Julian Gilbey, Dept of Maths, Queen Mary, Univ. of London
       Debian GNU/Linux Developer,  see http://people.debian.org/~jdg
  Donate free food to the world's hungry: see http://www.thehungersite.com/



Reply to: