Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Discuss] Striping is bad



On Sat, Feb 18, 2012 at 9:45 PM, Mark Komarinski <mkomarinski at wayga.org> wrote:
> On 2/18/2012 10:45 PM, markw at mohawksoft.com wrote:
>>
>> If you are saying that a catastrophic failure of a storage device does not
>> reflect poorly upon the manufacturer, I suggest you rethink your position.
>>
>> If I were driving home in a Kia and it died with no symptoms, i.e. was
>> running perfectly with no "check engine" light as well as properly
>> maintained, would you NOT blame the manufacturer?
>
> If a Kia went for $100, I'd buy two in case one died unexpectedly.

Sounds like 'mirroring' (RAID-1) to me :)

yep, RAID initially stood for 'Redundant Array of Inexpensive Disks'.
So using anything
above RAID-0 gives some level of redundancy. ... I did use striping
with great results
years ago.  But we would RAID-5 2 boxes of 7 disks, and stripe over
the RAID-5 arrays.
Back in that day, 8G 5.25" Seagate disks were 'high tech', and we had
a hardware RAID-5 controllers,
and to get enough room for some seismic data, we did stripe these
controllers together.

We monitored the hardware RAID controllers, and swapped out drives at
the first sign of
issues. ... Still we backed up data regularly.

I understand the frustration with apparent 'bad drives', but even back
then getting 'priority'
Enterprise level drives, we ran into some production runs that had
short life issues.  When
it happened, we negotiated with the vendors for 'pre-positioned on
site replacements' being
made available, because we counted downtime in many $$/hour.

Life happens. ...  Don't beat yourself up.  Just learn and go on.

Yea, the drives could be  better, they always can.  But you never know
when they can go out
on you, so you have to be ready no matter what drive.  Assume they
will go out 'way to early'.
Just be prepared.

My latest thing was wanting to go to 'green' drives (lower power, even
at some performance),
but my luck with them has been pretty bad.  They have all had failures
of various kinds.
I just keep an eye on where I can get 'reasonable' replacements
quickly, try to keep my
backups updated, and go on.  I live in a 'dusty' environment, and it
just eats drives, no matter
what I do.



BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org