[quantumstate]

I've been a fan of Seagate drives for a long time, but they have finally dropped a bomb on us. The 7200.11 class drives in all sizes have had an extraordinarily high failure rate since they came out, manifesting as disappearing from POST autosize or coming up with a size of zero anywhere from an hour to four months after install.


Most still think this is a firmware problem, but I began to suspect differently when after months of these reports, Seagate still hasn't officially acknowledged the problem. They stopped responding to support emails in early December; when called they promise a response within 2 hours which never comes; and RMAs are replaced with the same model and firmware. This behavior adds up to a much more serious problem than firmware.


Seagate closed a forum thread about this problem, and are not responding to questions there. In response customers started a new thread here , and today the following post was made:

Quote:
Hi,


Today I got back my HDD from Data recovery service. Since they ask 1000 USD and I dont want to spend so much for 2 weeks Seagate HDD usage, I got my HDD back.

The guy at recovery service was rather kind. He says he gets Seagate HDD every day for recovery. Since they have clean labs, they do some head replacements. They get some parts directly from Seagate. He does not advice to do at home, as dust may complicate things. So guys work in clean environment! He says in his experience, HDD from 500 Gb are not reliable, tend to fail. Most problem is head misalignment.


Here are his comments after evaluation:


"Our data recovery specialist had done an evaluation on your hard disk. There are multiple file structure damage and read write errors on the hard disk due to a head misalignment. The firmware and translator on the hard disk is also corrupted.


We will need to use an identical head to replace the head in the hard disk our clean lab, reload the firmware and fix the translator to enable data retrieval.


Data should be recoverable if translator corruption is not too severe."

Some are skeptical and say that the recovery company is just trying to make a buck. But this is the final nail in the coffin, AFAIC:
Seagate Ships Desktop Hard Drive With World's Highest Areal Density -- 500GB Per Disk


What?! 7200.12 is out now, when 7200.11 is only a couple months old? Tells me that 7200.11 is fatally flawed, so FWIW to you friends.

 

[Daravon]

Good to know, but I plan on buying nothing but Green Powers until something similarly quiet, cool and fast comes out. Which might already have happened for all I know.

 

[quantumstate]

The WD's seem to have none of the level of complaints out there that the Seagate .11's have. I asked WD support about Seagate's problems and he said that lately Seagate has just been slapping together drives and rushing them out there; that more often than not a new firmware rev comes out after shipping, as opposed to WD. FWIW. I've always used Seagates before, and WD used to have a terrible rep, but that seems to have reversed now.


In studying the Green drives, they do use far less power than Seagate's .11's, although the max capacity is only 1 TB. (Did I just say 'only'?) However there is very little difference overall between the Green and Black drives, and I am torn. The only substantive difference is idle power.


I guess what I'll do is wait until a WD 1.5 or 2TB comes out and buy then, hoping this ragged Seagate holds on till then, and keeping backups. When the time comes, I'll probably be looking for the AV flavor of the drive.

 

[DAP]

Head "misalignment"??!? No such thing. There is no alignment done to the heads, therefore there is no way for them to become "misaligned". It is easy for a head to become damaged and require replacement if one wishes to recover data, but no way for it to become "misaligned".

Sounds to me like you were talking to a sales type person who has no concept of the technical side of the business.

 

[quantumstate]

First of all, that quote was obtained from the referenced thread, and was by someone else. Second, oh yes head misalignment.


Remember how these disks are constructed: Two heads on each surface, mounted on a T-frame. Each head represents a 'cylinder' of data through all the platters, and each bit of a byte is written by one of the heads (simplified), so the whole byte is written at once. Maybe a T-frame wasn't spot-welded securely and goes caddy-wompus, or its base mounting wasn't tightened so the heads get misaligned in relation to one another.


I suspect that head misalignment or glue delamination is exactly what is going on here. If it were firmware, don't you think that Seagate would immediately fix it and stem the bleeding? What they are actually doing is stonewalling and hiding. Why? This is the most damaging course of action, and so for them to resort to it must mean it's their only choice. Oh yeah, they're replacing the drives, but I say this only to buy time until the .12's start shipping in numbers. They hope to leave as many .11's out there as possible because this could ruin the company.

 

[tux99]



Quote:
Originally Posted by quantumstate /forum/post/15475556


The WD's seem to have none of the level of complaints out there that the Seagate .11's have. I asked WD support about Seagate's problems and he said that lately Seagate has just been slapping together drives and rushing them out there; that more often than not a new firmware rev comes out after shipping, as opposed to WD. FWIW. I've always used Seagates before, and WD used to have a terrible rep, but that seems to have reversed now.


In studying the Green drives, they do use far less power than Seagate's .11's, although the max capacity is only 1 TB. (Did I just say 'only'?) However there is very little difference overall between the Green and Black drives, and I am torn. The only substantive difference is idle power.


I guess what I'll do is wait until a WD 1.5 or 2TB comes out and buy then, hoping this ragged Seagate holds on till then, and keeping backups. When the time comes, I'll probably be looking for the AV flavor of the drive.

I have four of those WD GP AV drives you link to, the 1TB version, set up as a raid5 with Linux software raid and jfs on top of it (I had a couple of threads about advice for the best set up on this forum going, you might remember them).

I bought them last september, so far they have been flawless, they mostly contain media files and I keep a md5sum for all files on them, which I occasionally check, so far not a single file has had any corruption (that's what you would expect, but I'm a bit paranoid about these high density drives and silent data corruption).


Also they are very quiet and stay at 32-35 degrees celsius all the time, even under load, with no dedicated cooling.

 

[quantumstate]

Sign the petition:
http://www.p3titiononline.com/ST310003/petition.html

(change the 3 in petition to e)

 

[drkdiggler]

tux99:


Check out the Data Scrubbing portion of the Gentoo Linux Wiki - RAID/Software article . I have implemented a weekly scrub on my RAID5 array (5x750GB) to help calm my similar worries.

 

[quantumstate]

Beginning to look like my theory is confirmed. Apparently Seagate is in the process of recalling all the 7200.11 series from wholesalers and will be replacing them with 7200.12 series. This would not be a firmware problem. The CEO has just been forced out over this mess.


So if you can hang on for a bit, get yours replaced with a .12 .


BTW, it also appears that an early warning may be when you smartctl -a /dev/sda a very large number of errors (like over 100k) for:

188 Unknown_Attribute

 

[k_ross]



Quote:
Originally Posted by quantumstate /forum/post/15551972


Apparently Seagate is in the process of recalling all the 7200.11 series from wholesalers and will be replacing them with 7200.12 series.

Interesting, as I have a 1TB 7200.11. Link?

Quote:
The CEO has just been forced out over this mess.

Also interesting. Link?

 

[quantumstate]

Hey Seattle, Lake City here!


Former is a rumor. A poster over on msfn said his friend works for a wholesaler, and they've been ordered to return all .11's. It could be that Seagate intends to upgrade the firmware, but like I say if this were a firmware fix, Seagate would not be behaving thisaway. I say it is much more serious.


For the latter, just look around. He was just fired. And I can guess why, because of the timing.


Ya, I have the 1TB too, but I'm holding onto my red arse until WD comes out with their 1.5TB, which is supposed to be any day now.

 

[quantumstate]

Seagate board ousts CEO.

COO & president resigns:
http://blogs.barrons.com/techtraderd...comment-185653

 

[k_ross]

I think what I'll do is get a second 1 TB drive, and mirror them with RAID 1. It would suck to have to rebuild that much data. Nothing on it is irreplaceable, but it would take forever to put all the movie files back on it.


Of course I may need to add more drives anyway, if I decide to go ahead and get the R5000 modification to my Dish Network receiver, and start recording lots of HD content. That's assuming the MythTV mods for the R5000 actually work as well as they are supposed to.


In that case, I may get two more drives instead of one, and start building a RAID 5 array. Ick, three drives in a HTPC box, too noisy. NAS here I come!


-- Kevin

 

[MichaelZ]

I've got a few of the 7200.11 drives from 1tb to 1.5 tb and they've all worked pretty well for me. Although I don't beat these drives that bad they've been solid for last six months or so. I'll gladly take an upgrade, if free, for a better drive, well maybe!

I rarely trust any drives so I am keeping my eye on these and doing appropriate backups.

 

[mattbrown]

Aside from the reference to 7200.11, anything to be concerned about here?

[email protected]:~$ sudo smartctl -a /dev/sda1

smartctl version 5.38 [x86_64-unknown-linux-gnu] Copyright (C) 2002-8 Bruce Allen

Home page is http://smartmontools.sourceforge.net/


=== START OF INFORMATION SECTION ===

Model Family: Seagate Barracuda 7200.11

Device Model: ST3500320AS

Serial Number: 5QM2YQAT

Firmware Version: SD15

User Capacity: 500,107,862,016 bytes

Device is: In smartctl database [for details use: -P show]

ATA Version is: 8

ATA Standard is: ATA-8-ACS revision 4

Local Time is: Wed Jan 14 20:52:03 2009 EST

SMART support is: Available - device has SMART capability.

SMART support is: Disabled


SMART Disabled. Use option -s with argument 'on' to enable it.

 

[quantumstate]



Quote:
Originally Posted by k_ross /forum/post/15557647


Of course I may need to add more drives anyway, if I decide to go ahead and get the R5000 modification to my Dish Network receiver, and start recording lots of HD content. That's assuming the MythTV mods for the R5000 actually work as well as they are supposed to.

Not too noisy if they're the Caviar Green. That's what I'm going to as soon as the 1.5TB comes out. May go to RAID5 or 10.


The R5000 records well in Myth now, although I've had some problems with channel changing. Still experimenting, and not sure whether it's an actual problem. You can't Watch TV with it because it always sends a turnon to the box first, and this adds so much dwell by the time the channel is changed, that Myth times out back to the menu. I've suggested to Alan to check how HD-PVR and HDHomeRun do it. Wonderful HD though, and nice small H.264 files without transcoding.


Michael, I've lost all confidence in Seagate now, and if I can get my drive upgraded I'm selling. Although the new gen .12 drives are now failing the SAME WAY! Party-time!!


Matt, you have Smart turned off. Check in /etc/rc2.d and see whether you have S20smartmontools. If not, put a symlink there and start it. Then run the smartctl command.

 

[Gradius2]



Quote:
Originally Posted by mattbrown /forum/post/15557915


Aside from the reference to 7200.11, anything to be concerned about here?


=== START OF INFORMATION SECTION ===

Model Family: Seagate Barracuda 7200.11

Device Model: ST3500320AS

Serial Number: 5QM2YQAT

Firmware Version: SD15

LOL, that one is faulty too. I have four 750GB SD15, three them already died. Yours will die too, they last (with lucky) just 5 months max.


I lost like $20k and months of hard works on them. I almost cried.

 

[tux99]

There is an good article about Seagate drive issues here too:

http://www.theregister.co.uk/2009/01...ailure_plague/


Some of the user comments are quite interesting too.


I sweared on Seagate until a few years ago, but the current generation of drives from WD like the VelociRaptor as a boot disk and the WD-GP drives for large capacity data storage has won me over, I think WD is CURRENTLY making the best drives (from a technology and reliability point of view).


This might be all different again in a year or so, but I doubt Seagate will recover quickly from this desaster, remember IBM's "deathstar" drives, IBM had to give up making HDDs after that and sold the HDD business off to Hitachi.

I still wouldn't buy a Hitachi HDD, out of mistrust, that some of those IBM flaws are still lurking somewhere inside them.

 

[Troubleshooter]

I believe I read that the Deathstar line is being shuffled off to Toshiba now

Tough making margins in disk these days I think and even if the IBM drives didn't face that horrible time period, I'm sure IBM is glad to be rid of the disk line! They've all produced bad drives over the years. Though I did have a full height 5 1/4 5MB drive from Winchester that was rock solid as long as you parked the heads before moving it. It was circa 1982 and it fired right up in 2003 after years in attics and garages. It was in a Wang PC that probably could have taken a .45 slug and kept on chuggin

-Trouble

 

[DAP]

 http://hardware.slashdot.org/article.../01/17/0115207


Segate is now claiming that it is a firmware bug & no data is actually lost, just inaccessible.

They claim to have a firmware fix, but you have to e-mail them to get it (follow the links).