[TriLUG] do hard disks have a shelf life?

Thomas Gardner tmg at pobox.com
Thu Sep 6 17:50:46 EDT 2012


Of course, what Joe is not telling us is that when he says he is
``labeling'' his disks, what he means by that is that he takes the top
off of the case and uses his electric engraver to write all this good
information on the uppermost platter's surface.  That way, it'll never
be lost, of course.  You've been remembering to squirt a generous
splash of WD-40 all around in there before closing it up like I told
you to, haven't you Joe?  :-)

Seriously, mostly all I ever use is old equipment, especially disk
drives since I don't need much, and buying last year's technology
saves money.  I've never had any trouble, but then again, the disks
they were using for ballast on the Mayflower that I generally tend to
use might be completely different beasts than whatever you put on your
shelf a couple years ago.

I once had an FS get so horribly corrupted so very quickly, I
theorized (at the time) that what really happened was is that, well,
over time any magnet will tend to deteriorate and the field slowly
weaken, so I thought that maybe the stuff that hadn't been written for
a long time (it was my $HOMEs drive so some of the oldest stuff hadn't
been written to for many OS flavor changes and many, many years) just
got so weak that it couldn't be read reliably any more.  These are,
after all, very weak, tiny little magnets that make up our ones and
zeros on the surface of a platter.

Of course, my choice was to either believe that or believe that the
Linux kernel (or maybe fsck) had some serious FS bug (*GASP*) that
just cascaded out of control and corrupted everything.  I suspect the
thought of the latter was too much for me to bear, so I came up with
the former to cover.  Very convenient that I came up with a theory
that I couldn't prove, eh?  The disk this happened to was a disk that
was probably at least a decade old at the time.  I just recreated the
FS (on the same disk), restored from backup and went on my merry way.
Never again had a lick of trouble from it, except for it being too
small, which is all that finally dun it in.

I also bought a pair of disks (in fact, I think I'm using them now)
when I didn't really have to.  They were already old when I bought
them (new, but old).  They were to replace a drive that I thought was
going to die (I think it was making noises, ur sumsuch).  Well, the
long and short of it is that the disk I thought was going to die
soldiered on for years before finally succumbing to whatever ailed it.
 The disks that sat on the shelf for (I dunno, maybe 2 or 3 or maybe
even more) years, plus who knows how long before I even bought them,
worked fine when I finally got around to using them.  Again, though,
this is probably a whole different generation of device than what
you're talking about, so I don't know if the comparison is valid or
not.

$0.02,
tg.

On 9/6/12, Aaron Joyner <aaron at joyner.ws> wrote:
> To your precise question about the surface of the disk degrading over
> time, I don't think so... but I know just the guy to ask.  I'll get
> back to you on that.
>
> In general, in an abstract perfect world, solid state components
> should have a very long shelf life, in the low to medium single digit
> number of decades.  Unfortunately, when that perfect world meets
> reality, some solid state electronics do have a much shorter shelf
> life (and commensurately, operational life), usually due to lower
> quality materials or chemical combinations used in manufacturing.  The
> most classic case of this is capacitor failure on motherboards
> (http://en.wikipedia.org/wiki/Capacitor_plague).  Other components are
> susceptible to similar breakdown of components over time.  This can be
> exacerbated if they're exposed to temperature changes.  One nice
> aspect of a warm running hard drive is that it stays at a fairly
> constant temperature.  I doubt this is exactly your case, but if
> hypothetically you'd left those drives on a shelf by a window, and
> they heated and cooled each day, they would eventually develop small
> fractures in the traces of the PCBs, leading to premature failure.
>
> In summary, they probably shouldn't, but they probably do.
>
> Aaron S. Joyner
>
>
> On Thu, Sep 6, 2012 at 4:20 PM, Joseph Mack NA3T <jmack at wm7d.net> wrote:
>> For home use I replace disks after about 2yrs and after checking them
>> with
>> badblocks and smartctl, put them on the shelf for projects. The disks are
>> labelled with the date of first use, their read and write rate and the
>> date
>> of the last badblocks and smartctl run
>>
>> In the last 4 weeks or so, I put into action, 5 disks 3-8yrs old, that
>> haven't been used for maybe 2-5 yrs. Everyone of them failed within a
>> week.
>> One of them is a 1.5TByte disk that the highest 300G of sectors (is this
>> inside or outside?) is unwritable. Others failed with clicks of death,
>> bad
>> lba's according to smartctl or just not booting.
>>
>> It's hard to imagine that electronics would go bad in 2-5 yrs sitting on
>> a
>> shelf. Does the surface of the disk degrade with time?
>>
>> Thanks Joe
>>
>> --
>> Joseph Mack NA3T EME(B,D), FM05lw North Carolina
>> jmack (at) wm7d (dot) net - azimuthal equidistant map
>> generator at http://www.wm7d.net/azproj.shtml
>> Homepage http://www.austintek.com/ It's GNU/Linux!
>> --
>> This message was sent to: Aaron S. Joyner <aaron at joyner.ws>
>> To unsubscribe, send a blank message to trilug-leave at trilug.org from that
>> address.
>> TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug
>> Unsubscribe or edit options on the web  :
>> http://www.trilug.org/mailman/options/trilug/aaron%40joyner.ws
>> TriLUG FAQ          :
>> http://www.trilug.org/wiki/Frequently_Asked_Questions
> --
> This message was sent to: Thomas Gardner <tmg at pobox.com>
> To unsubscribe, send a blank message to trilug-leave at trilug.org from that
> address.
> TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug
> Unsubscribe or edit options on the web	:
> http://www.trilug.org/mailman/options/trilug/tmg%40pobox.com
> TriLUG FAQ          : http://www.trilug.org/wiki/Frequently_Asked_Questions
>



More information about the TriLUG mailing list