Dustin Kirkland <kirkl...@canonical.com> writes: > On Thu, 2010-02-11 at 15:54 -0800, Raphael Geissert wrote:
>> I don't think anything will be merged unless the concern about the >> ubuntu- specific exclusion of some of the tests is also addressed. It >> would be even better if ubuntu helped with the implementation of the >> 'lsb-header' parameter to further reduce the divergence. > The way this patch is implemented, I don't think that it's necessarily > Ubuntu specific. > It checks to see if the init.d/$script under examination is a symlink > to /lib/init/upstart-job. If it is, then it skips the traditional LSB > sysvinit style checks. As explained by Raphael and by Steve in the Launchpad bug, the patch is not acceptable for Debian as-is because it bypasses the check for an update-rc.d invocation in the maintainer scripts for upstart jobs. This invocation continues to be required in Debian. I'm checking in a modified version of this patch that avoids checking the syntax of the init script if it's a symlink to upstart-job, since doing so doesn't really make any sense, but which does not suppress the check for update-rc.d. I'm afraid you will need to handle the latter in an Ubuntu-specific manner for the time being until there's some better solution that works for both distributions. Separately, I'm not sure that Debian can really reasonably accept packages with this sort of symlink yet given dependency-based boot, but that's a separate problem that I'm not sure we need to also address in this bug. If that does need some sort of check, it shouldn't be a bevy of confusing syntax errors for upstart-job anyway. -- Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
