[Macpartners] OS X Finder hanging proble

[Kerem B Limon]

Quoting "John C. Welch" <jwelch at MIT.EDU>:

> On 10/7/04 5:40 PM, "Kerem B Limon" <k_limon at MIT.EDU> wrote:
> 
> > PS: BTW, this is one of the greatest shortcomings Apple refuses to fix
> (next
> > to
> > the single button mouse), thanks to the 'it's a black box, don't let the
> user
> > know what's going on inside' philosophy. A $0.02 hard drive access LED
> would
> > help indicate whether there is hard disk activity at the time this happens
> and
> > whether it's continuing to get stuck on a corrupt sector or the like.
> 
> And what is that going to tell you?
> 
> That the arm is moving?
> 
> That it's reading?
> 
> That it's writing?

Hard drive access LEDs, protocol-wise, are not as ambiguous binary information
providers as they may seem. That is, while there are only two states--'on' and
'off', the rate at which they turn on and off, the amount they stay on, etc.
convey far more useful information. And these are not only intuitive, they are
very easy to describe to a layperson to remotely troubleshoot.

Incidentally, the LED only turns on when there is activity in the form of a read
or write actually happening, not when there is purely mechanical work being
done such as an arm moving. 'Seeking' turns it on, because it involves not only
the arm moving into position, but a read happening to actually position the
head accurately.

As anyone who's done this kind of troubleshooting knows, there is a very
particular and dis+tinguishable way the LED stays on solid when the machine is
'stuck' trying to access a hard drive; for example, if the drive is
electronically dead and the LED stays on stuck while the machine takes a long
time to POST, you can tell it's the drive and not something else. If the
flickering LED while copying a file suddenly gets stuck solid while your shell
also hangs, it could indicate a corrupt sector right there. More importantly, I
can tell if it is the particular app I am using that is failing to write vs.
the storage media that is failing. Same thing is true with CD/DVD-ROM and
writer lights that are often hidden behind the cosmetics of an Apple chassis.

Sure, I'd like a whole cockpit-full of unequivocal indicators being the space
cadet that I am, but that sure ain't gonna happen unless I put them in (as I do
on some of my machines). In the meanwhile, one relevant, intuitive indicator is
better than none.

> 
> It's a *nix box, you're going to have nigh - constant drive "activity". Swap
> files, etc. And, at login, it is normal for there to be a lot of disk
> activity *anyway*

True, but with sufficient RAM (as in most modern machines), there should be less
rigorous and repetitive activity when something is not being actively done (in
the fore- or background). Janet's case in particular does not involve login,
but accessing a particular folder mid-session, so an LED would be useful in
troubleshooting what is going on, given her user interface is useless at that
stage.

> 
> Now, as to the *problem*....
> 
> Janet...did you run fsck -fy, or just fsck -f? The difference is important.
> 
> Second...did you actually run diskwarrior on this machine by booting from a
> CD? An aborted install can cause a lot of disk structure problems.
> 
> The fact that it happens with new users indicates it's completely out of the
> /Users realm.
> 
> I would run a current version of Diskwarrior (at least 3.0, most current is
> 3.0.2) on the drive, from the DW CD and see if that does the trick.
> 
> john
> 
> -- 
> Socks should match your pants, and your belt should match your shoes. After
> that, if anyone complains, tell Œem they should be happy you¹re wearing any
> clothes at all.
> 
> 
> 
> _______________________________________________
> Macpartners mailing list
> Macpartners at mit.edu
> http://mailman.mit.edu/mailman/listinfo/macpartners
> 

Kerem

Kerem B. Limon
kerem.limon at mit.edu /e-mail