Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.mac.apps > #21600

Re: pointer can't penetrate link area

Show all headers | View raw

Jolly Roger <jollyroger@pobox.com> wrote:

> On 2014-03-11, DushanM <dushanm@nnips.net> wrote:
> > Jolly Roger <jollyroger@pobox.com> wrote:
> >
> >> On 2014-03-08, DushanM <dushanm@nnips.net> wrote:
> >> > Jolly Roger <jollyroger@pobox.com> wrote:
> >> >
> >> >> On 2014-03-06, DushanM <dushanm@nnips.net> wrote:
> >> >> > DushanM <dushanm@nnips.net> wrote:
> >> >> >
> >> >> >> DushanM <dushanm@nnips.net> wrote:
> >> >> >> 
> >> >> >>  <snip>
> >> >> >> > 
> >> >> >> > Oops - I spoke too soon: this weekend the problem was back.  Now
> >> >> >> > I'll have to see whether it was just the restarting that made it
> >> >> >> > go away temporarily, or whether clearing out the cache was
> >> >> >> > involved.  And if the cache, maybe deleting the System cache will
> >> >> >> > fix the problem for good. Also, after the problem disappears I'll
> >> >> >> > be on the lookout for what was just done before it re-appeared.
> >> >> >> > It'll be Thursday before I can check that, and will report back.
> >> >> >> > 
> >> >> >> > - Dushan
> >> >> >> 
> >> >> >> I did a bit more experimentation and observed this:
> >> >> >> 
> >> >> >> # a restart is all that's needed to make the pointer-freeze problem
> >> >> >>   disappear in my wife's account; no cache deleting is necessary
> >> >> >> 
> >> >> >> # the problem seems to stay gone so long as I stay on my wife's
> >> >> >> account, tho statistics are still sparse,
> >> >> >> 
> >> >> >> # but when I switch to my account and then back to my wife's, the
> >> >> >> problem reappears
> >> >> >> 
> >> >> >> # meanwhile, I've done nothing in either account except invoke
> >> >> >> Firefox
> >> >> >> 
> >> >> >> I'll experiment more this w.e.
> >> >> >> 
> >> >> >> - Dushan
> >> >> >
> >> >> > I've little to add to the above, except that I don't have to do
> >> >> > anything at all to re-introduce the pointer problem in my wife's
> >> >> > account except switch to my account, then immediately back: the
> >> >> > problem's there.  So apparently the switching itself was responsible,
> >> >> > probably when account data was being re-assembled.  So what is done
> >> >> > in that process that is not done during restart (the restart is
> >> >> > directly to my wife's account)?
> >> >> 
> >> >> Something must get cached. You previously said that deleting cache
> >> >> files temporarily fixed the issue, right?
> >> >> 
> >> >> I believe you can find out which cache files are being deleted by
> >> >> examining the system log after deleting them. That may be a place to
> >> >> start. You could try manually (selectively) deleting certain ones or
> >> >> groups of them to see which one(s) are at fault.
> >> >
> >> > Oh, maybe I see what you're driving at: when accounts get switched, data
> >> > in the 'from' account gets cached, then re-introduced when that account
> >> > is re-entered.  And this does not happen - or at least not identically -
> >> > during a restart.  Is that the general idea?
> >> 
> >> My theory is because deleting caches fixes the problem, that's what is
> >> happening.
> >> 
> >> > I can certainly try inventorying the cache files before and after a
> >> > restart and an account switch, and comparing.  But then it would seem as
> >> > if the switching operation itself somehow carries the infection, or at
> >> > least activates it, since it wasn't active before the switch after a
> >> > restart.  The mechanism for this is more and more puzzling to me.
> >> >
> >> > I try the cache inspection approach and see what I learn.
> >> >
> >> > Thanks for bearing with me on this problem.
> >> 
> >> I think if you can figure out which specific cache file(s) are involved,
> >> that will lead you to the root caused (and the solution).
> >
> > Yesterday I spent several hours trying our different things, and learned
> > a little, but not what caused the problem or how to fix it.
> >
> > 0 in each case below the initial condition was the same: restart into my
> >   wife's acct, with no sign of the problem;
> >
> > 1 I manually deleled every cache I could find on my wife's acct, both
> >   system and user, switched to my acct and back, and the problem came
> >   back;
> >
> > 2 ... then did the same thing using Cocktail, in case I'd missed some-
> >   thing, and got identical results;
> >
> > 3 did nothing (no cache clearing), switched to the Guest acct and back,
> >   and the problem was still gone;
> >
> > 4 did nothing (no cache clearing), logged out of wife's acct then logged
> >   back into it, and the problem was still gone.
> >
> > These observations indicate that in fact clearing cache does not fix the
> > problem; only restarting does.
> >
> > To summarize:
> > -- switching from wife's acct to my acct, and _only_ to my acct, brought
> >    on the problem
> > -- restarting is the only way I found to get rid of the problem - tempo-
> >    rarily until I switch to my acct
> > -- I did not see deleting cache have any visible effect on the above be-
> >    havior
> 
> That sounds right to me.
> 
> > So the question seems to be, how does a momentary switch to my acct con-
> > taminate my wife's on return?
> 
> Great question! Sheesh... I'm running out of ideas as to what the cause
> could be. 
> 
> Have you tried making note of the exact times you log in/out/restart and
> looking at the console and system logs for entries with those time
> stamps to see if you might get a clue as to some error condition that
> might be at play?

No, I haven't, but that's a thought.  I could compare switching to/from
my acct with to/from another acct and see if there's a difference.  If
so, that should tell us something.
 
> > Can you suggest more experiments?  I thought of one: create a new acct
> > and see what effect it has on switching to it and back.  
> 
> That would be a good test. If it passes, you conceivably migrate your
> data from your old account to a new account to (hopefully) rid yourself
> of whatever is causing the problem.

I'll be travelling for a few days, so won't get to try these things
until this w.e.  I'll let you know the results next Monday.

- Dushan

Back to comp.sys.mac.apps | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

pointer can't penetrate link area Dushan Mitrovich <dushanm@nnips.net> - 2014-02-10 17:15 -0700
  Re: pointer can't penetrate link area philo  <philo@privacy.net> - 2014-02-10 18:21 -0600
    Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-11 12:18 -0700
      Re: pointer can't penetrate link area philo  <philo@privacy.net> - 2014-02-11 16:20 -0600
        Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-13 19:29 -0700
          Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-17 17:24 -0700
            Re: pointer can't penetrate link area Jolly Roger <jollyroger@pobox.com> - 2014-02-18 00:59 +0000
              Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-18 00:10 -0700
                Re: pointer can't penetrate link area Jolly Roger <jollyroger@pobox.com> - 2014-02-18 16:44 +0000
                Re: pointer can't penetrate link area Barry Margolin <barmar@alum.mit.edu> - 2014-02-18 12:31 -0500
                Re: pointer can't penetrate link area Jolly Roger <jollyroger@pobox.com> - 2014-02-18 19:25 +0000
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-22 00:48 -0700
                Re: pointer can't penetrate link area Jolly Roger <jollyroger@pobox.com> - 2014-02-22 16:00 +0000
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-24 19:45 -0700
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-28 12:21 -0700
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-03-06 16:01 -0700
                Re: pointer can't penetrate link area Jolly Roger <jollyroger@pobox.com> - 2014-03-07 01:01 +0000
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-03-08 14:50 -0700
                Re: pointer can't penetrate link area Jolly Roger <jollyroger@pobox.com> - 2014-03-08 22:32 +0000
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-03-10 18:45 -0600
                Re: pointer can't penetrate link area Jolly Roger <jollyroger@pobox.com> - 2014-03-11 01:19 +0000
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-03-18 14:12 -0600
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-03-18 14:12 -0600
                Re: pointer can't penetrate link area - SOLVED dushanm@nnips.net (DushanM) - 2014-03-21 21:58 -0600
                Re: pointer can't penetrate link area - SOLVED Jolly Roger <jollyroger@pobox.com> - 2014-03-22 18:28 +0000
                Re: pointer can't penetrate link area - UNSOLVED dushanm@nnips.net (DushanM) - 2014-03-24 20:35 -0600
                Re: pointer can't penetrate link area - RE-SOLVED dushanm@nnips.net (DushanM) - 2014-04-03 13:37 -0600
                Re: pointer can't penetrate link area - RE-SOLVED Jolly Roger <jollyroger@pobox.com> - 2014-04-03 21:10 +0000
                Re: pointer can't penetrate link area - RE-SOLVED dushanm@nnips.net (DushanM) - 2014-04-09 16:09 -0600
                Re: pointer can't penetrate link area dushanm@nnips.net (DushanM) - 2014-02-22 00:47 -0700
  Re: pointer can't penetrate link area dorayme <do_ray_me@bigpond.com> - 2014-02-11 11:29 +1100

csiph-web