Bug#835520: Seconding nine patches & seeking seconds for two more
Hello Sean Whitton,
On Thu, Aug 03, 2017 at 10:55:30AM -0400, Sean Whitton wrote:
> Hello,
>
> I second all of Andreas' patches except the 5th and 8th. I've attached
> the diff to which my second applies.
>
> The 5th and 8th patches introduce a normative requirement to use
> debhelper. This is a first for policy, which up to now only comments
> that using debhelper is "easiest".
This was definitely not my intention, but english is not my native
language.
My point was simply that it seemed to me like people often get
mislead by policy to think if it says update-rc.d has to be called
they interpret is as "Oh, I have to write manual maintainer script
code and directly invoke update-rc.d" when in reality debhelper
adds that automatically (and manually writter maintainer script
code should be avoided whenever possible!).....
Your new wording is likely much better, so thanks for improving it!
(Also, the reason for the patch series style submission was so you
could pick and choose the parts you liked and skip the rest.)
On another note I'd also like to take the opportunity to object
to the changelog description in
https://anonscm.debian.org/git/dbnpolicy/policy.git/commit/?id=b960e5448b3d3eca91a2bab1b0c92f91a161d022
https://anonscm.debian.org/git/dbnpolicy/policy.git/commit/?id=4a86cf0f48fb8279cacacb125e82d08ad38953b4
What I tried to do was to update the description to "current" *sysvinit*
standards. (Where "current" means several releases ago. Don't remember
when we made insserv/startpar non-optional in Debian but it was
definitely pre-jessie.)
While at it I also tried to make it more init system agnostic.
In other words I tried to focus on what the (original) bug title was:
getting rid of the harmful parts.
There was no focus on systemd really, but unfortunately it ended up
unavoidable to mention anyway.
It's absolutely not to be considered to be up to speed with systemd!
Debian policy IMHO needs alot of work still to catch up with the current
systemd best practises. I think another (or several) bug reports should
be opened if you really intend to undertake documenting (debians
interactions with) systemd.
Thanks alot for your work on policy! Happy to see things moving
forward and Russ getting some well deserved help!
Regards,
Andreas Henriksson
PS. Feel free to CC me if there's anything you find it valuable
for me to be part of the discussion.
Reply to: