On Thu, 14 Jul 2016, Goswin von Brederlow wrote: > > I can't be responsible for every bootstrapping tool out there. > > But in the bugreport you noticed the problem, figured out its an > ordering problem that every bootstrap program has and then just told > debootstrap about it and nobody else. So yes, you are basically > repsonsible.
Sorry, but no. Using chown in a postinst does not make me responsible for faulty debootstrap-like programs who configure base-files and base-passwd in the same dpkg run. The responsaible for a package having a bug is always the maintainer. In the bug I quoted, this was re-discovered by Holger, who had an automatic setup to discover this kind of things. Are you going to blame Holger as well for *not* having an equivalent setup to discover bugs in cdebootstrap? Maybe you should suggest him politely to add a setup for cdebootstrap as well instead. > And that is the problem of not fixing this at the source, > you get many bootstrap programs that need fixing. You seem to speak as if there were a ton of different bootstrap programs. How many of them are there actually? Because if there are only two of them you are making a lot of fuss about nothing. WHy don't you just do the right thing and reassign this bug to cdebootstrap, if you actually detected the problem there?
