[GRLUG] At long last...

Wed Mar 3 10:38:35 EST 2010

Michael,

I'm glad someone is able to get some use out of that machine.  It was going
to just sit around and gather dust for me.  I found all I was really in need
of at the moment was a NAS.  That box was the epitome of overkill.

Once my other priorities release their hold on me a touch, I am going to be
trying my hand at making a wireless NAS out of a Router with USB port and
DD-WRT that should suck down a fraction of the electricity.  Now all I need
to find is a good quality (cheap) USB external drive that will work.  I've
only got an old 40GB stolen from a laptop and stuck into an enclosure at the
moment.
--
Dave

On Tue, Mar 2, 2010 at 5:54 PM, Steve Romanow <slestak989 at gmail.com> wrote:

> Ben,
>
> This sort of analysis I'm sure makes you an asset to your company.  Good
> tip.
>
>
>
> On 3/2/10, Ben DeMott <ben.demott at gmail.com> wrote:
> > Are non fsynced writes hard on drives?
> > The reason I ask is I killed at least 3 drives Torrenting back in the
> day...
> > At the time the program didn't cache or perform synchronous writes.
> > 1 block download = 1 block written... you could literally watch each
> > progression and  hear another 'click' of writing...
> > When I stopped torrenting my drive failures stopped...
> > I think it must be hard on the head.. moving back and forth.
> >
> > Also as a side note, when a drive was nearing the end of its operating
> > lifetime or we could hear bearing or excessive head sound... we would
> load
> > up Eurosoft - it has a great test for head speed... it simply causes the
> HDD
> > head to move back and forthe... and tells you the average time taken per
> > head sweep... Running this for an hour or two on an already failing drive
> > and absolute failure is just about guaranteed...
> >
> > Just for fun I have an 8gb Western Digital still in a 400mhz PII running
> in
> > the corner... I don't really do anything with it, other than test old
> opengl
> > code on the box occasionally.
> > Its  loud, and slow - but still going strong! (12 years old) ... yeah I
> was
> > 13 when I got the drive... it was hot stuff.
> >
> > On Tue, Mar 2, 2010 at 5:24 PM, Bob Kline <bob.kline at gmail.com> wrote:
> >
> >> One problem is that as an individual
> >> it's hard to get a sense of the quality
> >> of a drive unless you read about it.
> >> For one single drive, if it fails you just
> >> conclude you were unlucky.
> >>
> >> But I remember years ago getting in
> >> a box of drives at work.  Maybe 20.
> >> And was astounded at how fast they
> >> died.  Maybe half after one year.
> >>
> >> In general, it's hard to get a look at
> >> the bigger picture.  Of 10,000 drives,
> >> what fraction dies after the first year?
> >> The place selling them probably has
> >> a good idea, but I suspect that data
> >> is held close to the vest.  No reason
> >> to give the consumer a tip off that the
> >> particular model of drive is junk - they'll
> >> want their money back.  On a one-by-one
> >> basis you can play games.
> >>
> >> Fortunately, drives are cheap today,
> >> and the quality does seem to be
> >> increasing on the whole.
> >>
> >>    -- Bob
> >>
> >>
> >> On Tue, Mar 2, 2010 at 4:27 PM, Ben DeMott <ben.demott at gmail.com>
> wrote:
> >>
> >>> When I worked in the bowels of a corporate office and did IT work...
> >>> It was just about guaranteed that the hard drives in the Dell
> Mini-Towers
> >>> would fail every 1.5 - 2 years consistently.
> >>> I would say about 50% of them did.
> >>>
> >>> Not one replaced drive - (retail purchased) off the shelf ever failed
> in
> >>> my time...(except for server 15k rpm drives) -> they failed fairly
> >>> regularly.
> >>>
> >>> When I was a young lad, I worked for Geek Squad for awhile... and I
> >>> noticed that If you inspect OEM drives vs Retail drives you will find
> >>> that
> >>> they actually have different Manufacturer origins for most of the HDD
> >>> Suppliers.
> >>> We had to dispose of bad customer drives or give them back... so I've
> >>> physically taken apart probably 200 drives...  The size or fitness of
> the
> >>> parts doesn't appear to be different on similar models of similar sizes
> >>> but
> >>> they were usually manufactured in a different plant... In fact with
> >>> Western
> >>> Digital this is how you determine warranty fitness.  The serial number
> >>> (used
> >>> to) inform you of the Plant the drive was manufactured in. If the
> serial
> >>> number had an 03, 04, 05 in it.. it meant it was OEM -> In the key to
> the
> >>> serial number these same digits referenced "MFG Origin"
> >>> I'm convinced the drives made for OEM Computer manufacturers are
> somehow
> >>> more cheaply built as a whole than drives that come off the retail
> shelf
> >>> for
> >>> consumers.
> >>>
> >>> Just my conspiratorial thoughts on the topic...
> >>>
> >>>
> >>> On Tue, Mar 2, 2010 at 4:03 PM, John J. Foerch
> >>> <jjfoerch at earthlink.net>wrote:
> >>>
> >>>> On Tue, Mar 02, 2010 at 03:43:52PM -0500, Bob Kline wrote:
> >>>> > What's your favorite drive manufacturer?
> >>>> >
> >>>> > These things float around.  Right now I
> >>>> > prefer WD.
> >>>> >
> >>>> >     -- Bob
> >>>>
> >>>>
> >>>> WD all the way.
> >>>>
> >>>> --
> >>>> John Foerch
> >>>> _______________________________________________
> >>>> grlug mailing list
> >>>> grlug at grlug.org
> >>>> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
> >>>>
> >>>
> >>>
> >>> _______________________________________________
> >>> grlug mailing list
> >>> grlug at grlug.org
> >>> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
> >>>
> >>
> >>
> >> _______________________________________________
> >> grlug mailing list
> >> grlug at grlug.org
> >> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
> >>
> >
>
> --
> Sent from my mobile device
>  _______________________________________________
> grlug mailing list
> grlug at grlug.org
> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/defanged-9
Size: 8207 bytes
Desc: not available
Url : http://shinobu.grlug.org/pipermail/grlug/attachments/20100303/f4e3a82a/attachment-0001.bin 