Re: the ongoing xfree86 buildd saga
Thomas Bushnell BSG <firstname.lastname@example.org> writes:
> I'm not complaining about the slow archs, and the m68k buildd failure
> will surely be noticed. It has not, however, been retried. Why?
It will most likely be retried manually on a large system. The
wanna-build state is then usualy left as "building". There is hardly
any point in requeing the package just to have a different buildd
pick it up and fail with the same problem.
> I'm asking for *information*. How do I find out what the plans are?
A good glass ball helps. Some patience and experience help too.
There is no way of finding this out short of asking the admin of the
buildd that failed. Standard plans of this sort are not publicised,
they are just implemented.
Screwing up in the postrm of a package in Build-Depends is about the
worst you can do to the current buildd. It is pretty sure to cripple
them all. The inability of buildds to rebuild their chroots from
scratch doesn't help. Shit happens, stop whining and improve the
buildd. You are welcome to join the alioth project.