Re: Default suspend on live images?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]



On Tue, 2018-03-13 at 13:58 +0100, Kamil Paral wrote:
> On Fri, Mar 9, 2018 at 6:10 AM, Adam Williamson <adamwill@xxxxxxxxxxxxxxxxx>
> wrote:
> 
> > Hi folks!
> > 
> > So, Fedora 28 has inherited an upstream GNOME change to default to
> > suspending after 20 minutes:
> > 
> > https://bugzilla.gnome.org/show_bug.cgi?id=681869
> > 
> > I believe this will also apply to live images, unless we suppress it.
> > Should we suppress it? Suspending live images does not work well and
> > is, I think, expressly not really 'supported'.
> > 
> 
> Today, we have tested what happens when anaconda installation takes longer
> than the auto-suspend timeout. The good news is that anaconda inhibits idle
> timer, meaning the installation is not suspended during the installation
> phase. The bad news is that once the installation is complete, and I mean
> *immediately* after that, the system suspends. You can't even say whether
> the installation finished or not. At one instant you see an "almost
> finished" anaconda screen, and a split second later your system "shuts
> down". (It really looks like the power cable was plugged off). There's no
> screen dimming, no grace period to let the user react (as there is when the
> screen is about to lock down). So overall this is a poor experience. In
> this particular case, it's related to anaconda auto-exiting after
> installation is complete:
> https://bugzilla.redhat.com/show_bug.cgi?id=1553935
> But the same thing would happen even if it didn't auto-exit, just released
> the idle lock.
> 
> It seems the exact same thing (immediate suspend) will happen to you once
> your movie player stops playing a movie, or when your YouTube video ends,
> etc. Frantisek filed a bug about it here:
> https://gitlab.gnome.org/GNOME/gnome-settings-daemon/issues/14

Aha! That explains something for sure: I hit this when adjusting openQA
for the 'auto-quit' bug, but wasn't sure exactly what was going on, so
didn't want to bring it up and further complicate the thread.

So of course we didn't notice this before because *once install
completed* you had to click the quit button, and using the
mouse/keyboard to do that would reset the idle timer.

So it sounds like, basically, there's a bug that could be described as
"the idle timer should be reset to 0 when it's uninhibited"? It seems a
bit silly for the idle timer to count up the whole time inhibition is
in place, then immediately "kick in" the moment the inhibition is
dropped (if this happens without any other action which *itself* resets
the idle counter), at least to me.
-- 
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net
http://www.happyassassin.net
_______________________________________________
desktop mailing list -- desktop@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to desktop-leave@xxxxxxxxxxxxxxxxxxxxxxx




[Index of Archives]     [Fedora Users]     [Fedora KDE]     [Fedora Announce]     [Fedora Docs]     [Fedora Config]     [PAM]     [Red Hat Development]     [Red Hat 9]     [Gimp]     [Yosemite News]

  Powered by Linux