Anyone had an IBM Deskstar drive fail? - GoFuckYourself.com

rowan · 05-02-2004, 08:36 PM

My 2 year old 60Gb went out a couple of months ago and I've been trying since then occasionally to get the data off it. It just clicks a lot and cycles down then up, like it's trying to calibrate to read the sector properly. About 15% of the sectors went bad all at once.

Sounds terminal doesn't it? Well, this morning I low level formatted it to get rid of all the remaining data before I sent it off for replacement.

Drive works fine now. Absolutely no bad sectors at all. And it still has 100% of its capacity available, it's not just that the LL format quietly remapped the bad sectors out of sight.

One theory I saw was that one of the heads was writing while it was moving across the platter. This guy made a graph showing where his bad sectors were, and they made the pattern of a spiral.

So it's probable that my drive is 100% mechanically sound, but there's some nasty firmware bug waiting to spray white-out over part of it again. I can't return it now since it's a "perfectly" working drive.

erehwon · 05-02-2004, 09:03 PM

Never had any problems with IBM drives, I make it a point to try and buy them exclusively.

rowan · 05-02-2004, 09:19 PM

Quote:

Originally posted by erehwon
Never had any problems with IBM drives, I make it a point to try and buy them exclusively.

How long since you last purchased one? IBM haven't made their own drives for at least a couple of years, I think at one stage they were just rebadging Hitachi, Seagate and Samsung models.

I didn't even realise that Hitachi =<b></b>= IBM drives now, until I purchased a new drive and it looked EXACTLY the same as my dead one.

You've never heard about the Deskstar GXP "screech of death" problems? Mine didn't fail mechanically (yet), but plenty have... I think the percentage of returns was exceeding double figures for a while.

Big Ray · 05-02-2004, 09:20 PM

you mean the deathstar? they suck. will never own another.

toddler · 05-02-2004, 09:20 PM

i've had dozens of death stars fail. dozens.

Phil21 · 05-02-2004, 09:29 PM

IBM never "rebadged" others drives (at least not in recent history). They may have "shared" technology with other drive manuf. (namely: seagate), but they were actually a major assembly OEM provider for other companies.

IBM's storage division was sold off to Hitachi about 2 years ago, many believe due to the Deskstar 75GXP fiasco in which they had inordinant failure rates they did not handle properly (at the time, the drive was *THE* drive in terms of capacity/speed, prior to this IBM had an absolute stellar reliability record).

We luckily never bought any effected 75GXP series drives. However, we have *many* (and I do mean MANY.. over 200) 60GXP and 120GXP models in server use 24/7. We have had exactly one 60GB model fail in over 3 years of using these drives. We still buy hitachi drives from time to time, depending on requirements - although our standard supplier is seagate in 95% of all cases. Honestly, drive failure isn't a huge problem here now that I think about it, I can count on one hand the number of drives we have replaced due to failure, and not even one has resulted in substantial data loss.

Man.. I shouldn't hahahahaha this. :/

-Phil

rowan · 05-02-2004, 09:30 PM

I used to swear by IBM drives also. I have an IBM 8.4Gb from 1997 that has spent its life in two servers (24/7) and it was still going strong when I retired it last year to replace it with a much larger capacity drive. It will probably go into a spare desktop one of these days...

Phil21 · 05-02-2004, 09:33 PM

rowan - definitely. All I used to spec was IBM drives, or seagate in some cases.

I have an old IBM SCSI drive (10,000RPM, one of the quick ones that contended w/ the cheetahs back in the day) that has been in constant use since 1996. The thing runs so hot it hurts to touch it, and makes a REALLY bad whineing sound (has for over 2 years now!). However, still going strong. I am afraid to power it off though.

-Phil

05-02-2004, 08:36 PM	#1
rowan Too lazy to set a custom title Join Date: Mar 2002 Location: Australia Posts: 17,393	Anyone had an IBM Deskstar drive fail? My 2 year old 60Gb went out a couple of months ago and I've been trying since then occasionally to get the data off it. It just clicks a lot and cycles down then up, like it's trying to calibrate to read the sector properly. About 15% of the sectors went bad all at once. Sounds terminal doesn't it? Well, this morning I low level formatted it to get rid of all the remaining data before I sent it off for replacement. Drive works fine now. Absolutely no bad sectors at all. And it still has 100% of its capacity available, it's not just that the LL format quietly remapped the bad sectors out of sight. One theory I saw was that one of the heads was writing while it was moving across the platter. This guy made a graph showing where his bad sectors were, and they made the pattern of a spiral. So it's probable that my drive is 100% mechanically sound, but there's some nasty firmware bug waiting to spray white-out over part of it again. I can't return it now since it's a "perfectly" working drive.

05-02-2004, 09:03 PM	#2
erehwon Confirmed User Industry Role: Join Date: Nov 2003 Location: A secure undisclosed location... Posts: 3,759	Never had any problems with IBM drives, I make it a point to try and buy them exclusively. __________________ Money NEVER $leep$...

05-02-2004, 09:20 PM	#5
toddler Confirmed User Join Date: Jun 2002 Location: austin, tx Posts: 1,911	i've had dozens of death stars fail. dozens. __________________ http://www.flickr.com/photos/zoddler/

05-02-2004, 09:29 PM	#6
Phil21 Confirmed User Join Date: May 2001 Location: ICQ: 25285313 Posts: 993	IBM never "rebadged" others drives (at least not in recent history). They may have "shared" technology with other drive manuf. (namely: seagate), but they were actually a major assembly OEM provider for other companies. IBM's storage division was sold off to Hitachi about 2 years ago, many believe due to the Deskstar 75GXP fiasco in which they had inordinant failure rates they did not handle properly (at the time, the drive was THE drive in terms of capacity/speed, prior to this IBM had an absolute stellar reliability record). We luckily never bought any effected 75GXP series drives. However, we have many (and I do mean MANY.. over 200) 60GXP and 120GXP models in server use 24/7. We have had exactly one 60GB model fail in over 3 years of using these drives. We still buy hitachi drives from time to time, depending on requirements - although our standard supplier is seagate in 95% of all cases. Honestly, drive failure isn't a huge problem here now that I think about it, I can count on one hand the number of drives we have replaced due to failure, and not even one has resulted in substantial data loss. Man.. I shouldn't hahahahaha this. :/ -Phil __________________ Quality affordable hosting.

05-02-2004, 09:33 PM	#8
Phil21 Confirmed User Join Date: May 2001 Location: ICQ: 25285313 Posts: 993	rowan - definitely. All I used to spec was IBM drives, or seagate in some cases. I have an old IBM SCSI drive (10,000RPM, one of the quick ones that contended w/ the cheetahs back in the day) that has been in constant use since 1996. The thing runs so hot it hurts to touch it, and makes a REALLY bad whineing sound (has for over 2 years now!). However, still going strong. I am afraid to power it off though. -Phil __________________ Quality affordable hosting.

05-02-2004, 09:20 PM	#4
Big Ray Confirmed User Join Date: Dec 2003 Posts: 464	you mean the deathstar? they suck. will never own another.

05-02-2004, 09:30 PM	#7
rowan Too lazy to set a custom title Join Date: Mar 2002 Location: Australia Posts: 17,393	I used to swear by IBM drives also. I have an IBM 8.4Gb from 1997 that has spent its life in two servers (24/7) and it was still going strong when I retired it last year to replace it with a much larger capacity drive. It will probably go into a spare desktop one of these days...