Re: Service message: Daylight saving time ends

[rick.baird@xxxxxxxxxxxxxxx]

Right around the time IBM came out with the new risc boxes, they had disk 
drives were unusually prone to sticktion.

I remember seeing an IBM knowledge base document (for SEs) on how to fix 
the problem in the field (Justified Machine Slapping - I like that term!!), and 
that once they got them spinning, they could 
fix the problem via PTF.

What did the PTF do?   At IPL, the cpu would send several strong jolts of 
juice to the drives to get them spinning. 

Ain't modern technology wunnerful?

Rick


------------original message---------------
Better solution to the Sticktion problem is the one IBM uses. 

    a) Grasp the drive in your hand as if the platters were being held 
by the tips of your fingers.
    b) Rotate the drive rapidly away from you and stop abruptly.  This 
should cause the platters to rotate with the cover and want to continue 
to rotate when you stop the cover. 
    c) Repate b) several times.

    This avoids pounding on the drive and actually gets them 'un-stuck' 
quite often.

    I wouldn reccomend the rapid power cycle option for several 
reasons.  Good power supplies will have a 'crowbar' circuit that won't 
power up if all voltages haven't fallen to zero.  This delays things, 
nullifying the effect.  Not-so-good power supplies don't have this 
circuit and as such the magic smoke could easily be let out of them or 
some other connected componant.  Additionally most SCSI drives power up 
in sequence by their ID so as to limit surges to the power supplies so 
quick power cycling won't reach the stuck drive at all.

    Good Luck!

      - Larry

PaulMmn wrote: