Dear Michael, thanks for clarifying the situation.
> serverfault solution isn't exactly wrong, it is incomplete, it > does not cover a situation when you have slow-to-appear devices. > This solution, however, fixes the problem in this bugreport, > a problem which I introduced when I tried to address the problem > with slow-to-appear devices. Ofcourse it is better to be able > to boot from a degraded raid than to be able to boot from slow > devices, esp. since the latter had a workaround. So in this > sense, serverfault solution will fix _this_ bug. so, couldn't you patch the package to fix the problem for all people that do not have "slow-to-appear devices"? > But I also dislike doing work which is being thrown away by others, > since this is a pure waste of time/energy, and time is a very scarse > resource. I don't want to do any work if I know this work will be > thrown away, and here, debian-installer people did throw my work > a) without a good reason and b) without actual rights for that. Of course what you describe is extremely disappointing for you. Since I don't know what exactly happened: There should be a decision committee which listens to both sides and decides _for Debian_. > So I stopped maintaining all software which is related to debian-installer, > because now I know it is just a waste of time. As simple as that, and > there's nothing like dislike of someone in there, there's nothing > personal. It's not obvious to me why a bug in mdadm can't be fixed because there is a conflict with the d-i maintainers. Why can d-i maintainers decide about mdadm? > In short, I don't maintain mdadm anymore, so there's no reason to > ask me about it. That would be very sad because it's a really important package. Thanks & Greetings, Paul
