[rescue] Moving Sun servers?

[Mark Benson]

On 29 Apr 2011, at 09:28, Mike Loewen wrote:

>   Back in the '80s, I remember people telling me that after a hard drive had
been running in one orientation for a long time, it was unwise to turn the
system so the drive was in a different orientation.  Was there any truth to
this, or was it just an urban myth?  Something about the bearings, as I
recall.

There might have been some credence in this back in the early days. Pre-FDB
(fluid dynamic bearings) the bearings in a drive would wear differently
depending which way the drives were used. It's possible that could cause
'bedding in' in one direction that'd either lead to minuscule misalignment of
the platters or, more commonly, stiction (basically the platters seize up
solid if you move the drive into a different orientation). Pretty much all
wear issues associated with that went away with FDBs, so I'd guess by the late
90s/early 2000s it stopped being an issue. I'd still say it was rare unless
the drive had a *lot* of miles on the clock.

It's not something that's ever bothered me in the last 10 year that's for
sure. I've used modern drives any number of ways up, then moved them, it
doesn't seem to bother them one iota. Drives dying on me these day almost
always are from dead sectors on the platters. Not had one lock on me in a long
time.

--