[Mageia-dev] [soft-commits] [4312] services: Update NEWS

[Colin Guthrie]

'Twas brillig, and Thierry Vignaud at 29/04/12 14:04 did gyre and gimble:
> On 29 April 2012 14:26, Colin Guthrie <mageia at colin.guthr.ie> wrote:
>>> According to your detailed changelog in the previous commit,
>>> I actually doubt this is for drakxservices.
>>> If it's about the "first boot", then this should have been logged in
>>> the installer NEWS
>>
>> Hmmm, if you say so. I just figured that anything that runs when stage1
>> and stage2 are not involved is not anything to do with the "installer".
>>
>> Conceptually, I can see how first boot could be considered part of the
>> "install" process, so feel free to correct as you see fit seeing as this
>> somewhat subjective distinction is not something I have any real desire
>> to master or argue over :) I'll just try and guess better next time :p
> 
> Well it was obvious that it cannot be about drakxservices since
> the context was "first boot" and drakxservices isn't in the game at this
> stage.
> I though you was speaking about services set up by the installer
> for the first boot.
> 
> But now, with the actual bug report, I see it was affecting the harddrake
> service.

At least I try to be wrong in interesting ways :D

> So it's in the right NEWS file but the description is wrong, it should
> speak about service_harddrake then since drakxservice hasn't
> the same timeout as harddrake and it's the later that has the issue
> not the former.

Yup, that makes more sense. I guess I was just using drakxservices as a
catchall term... my bad.

> In that case, the $::isStandalone won't be enought as it will be set
> for both harddrake service & drakxservices.
> We could set another global variable in harddrake ($::booting ?)
> and test for it in services.pm.
> WDYT?

Possibly, but I'm not sure it actually matters right now. In the GUI
version I think we actually do the start/stop in a weird way anyway via
/sbin/service and backtick execution[1] rather than this helper
function. That should really be fixed, but when I looked at this and
tried to change it, it had a bigger impact on how the GUI works (as it
grabs the output and displays it and systemct restart doesn't have much
in the way of interesting output to show here). I was not in the mood to
rework GUIs when doing the systemd stuff, so hopefully this will annoy
someone enough in the future and get fixed then. For now I don't see
this as critical.

Col


1.
http://svnweb.mageia.org/soft/drakx/trunk/perl-install/services.pm?view=markup#l275

-- 

Colin Guthrie
colin(at)mageia.org
http://colin.guthr.ie/

Day Job:
  Tribalogic Limited http://www.tribalogic.net/
Open Source:
  Mageia Contributor http://www.mageia.org/
  PulseAudio Hacker http://www.pulseaudio.org/
  Trac Hacker http://trac.edgewall.org/