On 24 August 2015 at 12:36, Axel Beckert <a...@debian.org> wrote:
> Hi Felipe,
>
> Felipe Sateler wrote:
>> On 23 August 2015 at 12:18, Axel Beckert <a...@debian.org> wrote:
>> > from my point of view, this bug report is a false positive.
>>
>> Indeed.
>
> Thanks for the confirmation!
>
>> Hmm, interesting. It had not occurred to me that uninstalling the
>> package will not remove the init script until purge.
>
> Yep, that's occassionally a burden. Then again, there are many cases
> where I'm happy that init scripts are conffiles and can be edited by
> the local admin.
>
>> Please silence the lintian warning, so that you will not get bugged
>> about this until we figure out a way to detect this case.
>
> Will do.
>
> Shortest workaround is probably to mention this false positive in the
> long description of the tag and to suggest a lintian override in that
> case. (Then again, this may lead to too eager overriding.)
Long term the problem will disappear because the runlevel S patch will
go away. In the meantime we are investigating adding support to
dh_systemd for preserving the mask after remove but before purge. This
way you would be able to drop the post{inst,rm} snippet you currently
have and let dh_systemd handle it; at the same time the lintian
warning would not trigger (because the mask would be in the package
list). But we are not there yet.
--
Saludos,
Felipe Sateler
