On Sun, Oct 01, 2006 at 11:31:15AM +0530, Rahul Sundaram wrote:
> RPM is not designed for downgrades
>One of the core features of rpm from the very beginning was to
>downgrades and uninstalls so the user is able to revert from a bad
>package. So rpm does indeed support downgrades by design.
Except that there is no proper way to revert back changes that is done
through install scripts or triggers and what not. There is also the
thing that QA is never done on downgrades.
Packages can be broken, of course, and any package installing itself
in an inrevocable way in scriplets that isn't undone by %*un scripts
is usually broken irrespective of downgrade or upgrades. You can also
have broken packages breaking upgrade paths, and we have a large list
of such existing examples.
In short broken packages don't indicate any (missing) abilities in
rpm itself. Otherwise for each feature of rpm we could find a broken
package and we'd conclude that rpm is designed for nothing. ;)
--
Axel.Thimm at
ATrpms.net