Re: sudden power loss (WAS: sudden power loss on tibook)
> > > have you found any interactions with any daemons such as pbbuttonsd?
> >
> > A long time ago people were talking about turning off NoTapTyping or
> > something similarly named -- I tried that and it didn't make a
> > difference.
> >
>
> On a lark I am testing by stopping pbbuttonsd completely after boot. I don't
> use the notap while typing feature, but the ordinary trackpad notap isn't
> getting set by pbbuttonsd as it used to be, so I know there's already some
> unusual behavior.
Probably unrelated. I didn't get any conclusive cause for the poweroff
phenomenon from the reports I've seen - IIRC stopping pbbuttonsd was one
of the things people tried before, as was stopping cpu frequency
switchers. Nothing helped.
> > > I'm very sick of booting to fsck with "you have not checked your
> > > > filesystem for X thousand days" since the clock gets reset to 1904!
> > >
> > > That's one difference on my Wallstreet II -- the clock isn't getting
> > reset
> > > when the power goes off.
> >
> > I wonder if that means we have different problems? I'm not really
> > sure.
> >
>
>
> I'll search the archives for the sleep thing -- but since I leave the
> PowerBook on, connected to AC, the sleep issue hasn't been a big issue for
> me. And the fact that my clock isn't getting reset may just reflect a
> difference in hardware.
Maybe the PMU on the Wallstreet II doesn't reset the date when killing
power to the machine. Maybe the date is kept elsewhere in a separate
chip there. The power loss most probably results from communication
between CPU and PMU breaking down, which should never be preempted by user
programs (speaking of which - does your kernel have preemption enabled?).
> The sudden power loss while actively using the machine sounds EXACTLY the
> same, but unfortunately I won't know for a few days whether stopping
> pbbuttonsd has had any effect.
Good luck with that,
Michael
Reply to: