[ale] Which large capacity drives are you having the best luck with?

[Ron Frazier]

That is a definite disadvantage of this type of product.  It's not as
bad with a single PC / drive since you can just reboot overnight and
continue using the unit the next day.  You might be able to get some of
the same advantages in a running system by running sort of a continuous
"defrag" operation.  This would periodically, say every 4-6 months, read
each sector of data and relocate is to another sector, just like defrag
does.  This wouldn't necessarily defrag the drive, and if done badly
could fragment it worse.  But it would make sure every piece of data is
read and rewritten 2 - 3 times / year.  That should prevent data decay
or "bit rot", assuming the sector is readable at all.  You still
wouldn't get the extensive read retries for bad sectors, or Dynastat
data recovery, or extensive surface analysis.  However, if the algorithm
were designed correctly, you could make sure each sector gets written
over time, and that the fragmentation didn't get out of hand.  All these
writes would give the firmware a chance to reallocate the sectors if
needed, and raise a warning flag, without shutting the machine down.  It
would be neat to find or create a utility that would "maintain" the
drive in the background and improve the reliability.

Searching for data scrubbing on Google yields some interesting results
(although I'm not sure that's the right term), including this "Data
Scrub with Linux RAID or Die" article:

http://www.ashtech.net/~syntax/blog/archives/53-Data-Scrub-with-Linux-RAID-or-Die.html

Typing this into Google also yields some interesting results but it's
hard to find thorough and authoritative data.

(hard drive OR hdd) (reliability OR maintenance)

Here are some links that looked good after a cursory glance.  I haven't
had time to read them all, they just looked interesting:

http://www.pcguide.com/ref/hdd/geom/error_Read.htm - Read Error
Severities and Error Management Logic - This site has huge quantities of
info, and looks like it's pretty good.  I only spent 5 minutes looking
at it though.  Every page has many many links to other pages.

http://www.pcguide.com/ref/hdd/index.htm - Hard Disk Drives - From the
same site.

http://www.pcguide.com/ref/hdd/perf/qual/index-c.html - Hard Disk
Quality and Reliability.  From the same site.

http://www.linuxjournal.com/article/193
Disk Maintenance Under Linux - Disk Recovery

Ron

On Thu, 2011-01-06 at 12:11 -0500, Derek Atkins wrote:

> My only complaint about Spinrite (having not used it, yet) is that it
> requires rebooting my system.  It's clearly not a tool that can be run
> on an actively running system to help prevent disk failures.  I would
> have to either shut down my system to run Spinrite, or I'd have to pull
> out my disks (leaving my RAID1 diminished) to pop the drive into another
> machine, run Spinrite, then replace it and resync the RAID..  Then pull
> the other drive.  But that would seem to defeat the purpose of Spinrite;
> if I pull a good drive and leave in the bad drive, the raid sync would
> fail (or sync bad data).
> 
> > RinL
> 
> -derek

-- 

(PS - If you email me and don't get a quick response, you might want to
call on the phone.  I get about 300 emails per day from alternate energy
mailing lists and such.  I don't always see new messages very quickly.)

Ron Frazier

770-205-9422 (O)   Leave a message.
linuxdude AT c3energy.com