This message is from the T13 list server.

On Tuesday 27 April 2004 00:15, Pat LaVarre wrote:
> This message is from the T13 list server.
>
> > I asked about this originally,
>
> Hi, thanks for sharing your personal history of pain.
>
> > > The natural implicit result is then a maximally accelerated test til
> > > failure: read, park, repeat, indefinitely.
> >
> > ... that is exactly what these drives do, with "advanced power
> > management" enabled (which it is by default) ...
> >
> > the implications ... are making me (and many other people) blue.
>
> I haven't yet heard these devices are operating out of spec.
>
> Yes, if operating within spec actually promises short life in a system,
> then naturally I see first the customers of that system, then the vendor
> of that system, then the supplier of the devices, will feel pain.

That is precisely the problem.  Some drives auto-park by default every 10 
seconds.  Most UN*X type systems (I think including MacOS X) sync things to 
the disk every 30 or 60 seconds.  This means that the drive parks and then 
very soon unparks, repeatedly.  We've seen many cases where the "load cycle 
count" number multiplied by 30 is almost identical to the total operation 
hours of the drive.

The party line from drive vendors seems to be that "the drive is only rated 
for 300000 load/unloads".  Well, gee, thanks, I do know how to read.  Do the 
arithmetic.  That means the drive is, in the worst case, rated for 10*300000 
seconds, or 34.7 days.  99.9% of people would not consider a disk lifetime of 
34.7 days even vaguely reasonable.  To be blunt, it's asinine.  And, yet,   
THIS IS WHAT THE DRIVES DO OUT OF THE BOX.

To make matters worse, while one brand of drives can be made to not do this by 
turning off "advanced power management," with another brand turning off this 
misfeature means that the drive can leave the head over the same spot on the 
disk indefinitely -- and I don't think I need to remind people here what 
happened to some IBM drives a few years ago when they exhibited *that* 
behavior.

The only sure-fire solution I've found is to make the software read a random 
sector periodically, to keep the drive form auto-parking, and to keep the 
head moving a bit.  It is my considered opinion that I should not have to do 
this to make the hardware reliable.

I guess what I was looking for here is:

1) Do people think this behavior in a disk drive is sane?

2) Are there known reliability issues with existing or past drives that have 
had this behavior?

3) Is there some other workaround that I've missed?

4) Maybe some drive vendors could comment on what they do, and why.

Reply via email to