[mythtvnz] HDD Load Cycle Count ticking over too fast

[Stephen Worthington]

On Mon, 25 Jun 2012 23:08:17 +1200, you wrote:

>On 25/06/12 22:57, criggie wrote:
>> On 25/06/12 22:02, David Moore wrote:
>>> This is the problem drive:
>>> Device Model: WDC WD20EARS-00MVWB0
>>> 9 Power_On_Hours 0x0032 095 095 000 Old_age Always - 4206
>>
>>> Older drive:
>>> Device Model: WDC WD10EADS-00L5B1
>>> 12 Power_Cycle_Count 0x0032 099 099 000 Old_age Always - 1219
>>
>>
>> Its obvious - your drives are too new.
>>
>> 9 Power_On_Hours 0x0032 013 013 000 Old_age Always - 76505
>> That's 8.7 years of active power-on time, and counting....
>>
>>
>> Your drive is a WD Ears and they're known for interesting power save
>> attempts. Google about and you'll see. Not as bad as seagate drives that
>> might simply cease to acknowledge your data, or SSDs without functional
>> trim. Best thing to do is either back up your important data, or just
>> don't stress.
>>
>> Remember all hardware breaks eventually.
>>
>>
>
>Agreed. Finding lots with google. As it happens my dud drive is a 
>Seagate. I am working out my backup strategy but I also don't think it 
>makes sense to let a dumb hardware/OS combination wear out a drive too 
>soon. Since simply playing something in myth stops the load cycling dead 
>I may do something to keep the drive active, say, 50% or 75% of the 
>time. I guess this won't wear the motor/bearings more than continuous 
>spin up/down? Definitely less head wear.

What sort of Seagate?  What firmware version?  There is a set of
7200.11 model drives with original firmware that can just brick
themselves on boot.  I had one that did it.  The fix is to flash it
with an updated firmware version, and it will come to life again
without any loss of data at all.

Unfortunately, 7200.11 drives are crap anyway and tend to die soon
with lots of bad sectors, even after you unbrick them.  I think I have
had about 5 of mine do that, and had to be replaced under warranty. I
had one drive that had to be replaced twice!  After that it had got to
the point where I just told the supplier to not bother to send a
7200.11 as a replacement for a 7200.11 because it would just fail in
another few months.  They seemed to accept that and replaced with
other drive types such as 7200.12 or "business class" drives which
have been better.  I believe I do not have any 7200.11s left any more
- 100% failure rate.  The older 7200.10 drives just keep on going, and
the newer 7200.12 drives are better than 7200.11s by far but still
have a high failure rate.  I have not bought any Seagates since
7200.12, due to all those failures.