LinHES Forums
http://forums.linhes.org/

HOWTO: Seagate 7200.11 HDDs - read to prevent DATA LOSS!
http://forums.linhes.org/viewtopic.php?f=11&t=19453		 Page 1 of 1
Author:  graysky [ Tue Jan 20, 2009 2:34 pm ]
Post subject:  HOWTO: Seagate 7200.11 HDDs - read to prevent DATA LOSS!
Background Information
This howto is specifically aimed at the owners of Seagate 11th generation HDDs because of an extremely serious firmware problem. I don't want to copy/paste the whole thing, so first off, read this Anandtech article describing the issue.

I have copy/pasted the following from Seagate's knowledge base which is the list of affected HDDs by model number... there are 21 of them under the Seagate and Maxtor brands!

So why did I post this in the HOWTO section? In this post, I will teach how to use smartmontools to accomplish two goals:

1) Show you the model of your HDD (Seagate or not)
2) Show you the version of the firmware you're running

Why am I doing this? Seagate's own KB is too windows-centric and doesn't offer a LINUX user any utility to diagnose this serious issue!

Anyway, here is the list of affected drives, read on to see how to install and use smartmontools:

"Barracuda 7200.11"
ST31000340AS
ST3750330AS
ST3640330AS
ST3500320AS
ST31500341AS
ST31000333AS
ST3640323AS
ST3640623AS
ST3320613AS
ST3320813AS
ST3160813AS

"Barracuda ES.2 SATA"
ST31000340NS
ST3750330NS
ST3500320NS
ST3250310NS

"DiamondMax 22"
STM31000340AS
STM3750330AS
STM3500320AS
STM31000334AS
STM3320614AS
STM3160813AS

I would NOT update the firmware at this time (20-Jan-09)! Keep an eye on the Anantech article (hopefully they keep it updated).

A while ago, I posted instructions to install smartmontools onto R5.5 for the purposes of reading the the SMART info on your HDD. You can also use smartmontools to read your HDD's model and firmware (among other things). Here is my original post. I will copy/paste the installation instructions here for you so you don't need to read it.

Installing Smartmontools on KM R5.5

Code:
# aptitude update
# aptitude install -s smartmontools


The -s switch is to simulate the installation. Read the output from aptitude and make sure it doesn't want to do anything crazy like removing 40 files or the like. When you're happy with the output, go ahead and remove the -s switch and do the install for real:

Code:
# aptitude install smartmontools


Use Smartmontools to Figure-out Your HDD Model Number and Firemware Version

Just run the following (assuming you're using a SATA HDD). If you're using an IDE drive, substitute 'hda' for 'sda' in the following command:

Code:
# smartctrl -a /dev/sda | grep Model


Again, if you have an IDE drive, you'd use 'hda' in that command. If you have two HDDs in your system, just change the last letter accordingly.. For example, sda = 1st hdd and sdb = 2nd hdd, etc. Also, this command is CASE SENSITIVE (note that the word 'Model' is capitalized).

Here is an example output from my system:
Code:
smartctl -a /dev/sda | grep Model
Model Family:     Seagate Barracuda 7200.11
Device Model:     ST3750330AS


Now to read the firmware, use the following command:

Code:
# smartctl -a /dev/sda | grep Firmware


Here is an example output from my system:
Code:
# smartctl -a /dev/sda | grep Firmware
Firmware Version: SD15


Anyway, just wanted to share the info and a way to diagnose things under KM.
Author:  graysky [ Sat Jan 24, 2009 8:40 am ]
Post subject: 
I just updated my ST3750330AS and ST3640323AS drives to their latest respective firmwares and was able to reboot and use them just fine. I think the issue is solved. Antech is also reported green lights on the latest firmware releases. Go to the Seagate kb linked above and d/l your firmware update!

Code:
smartctl -i /dev/sda
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
Serial Number:    XXXXXXXX
Device Model:     ST3750330AS
Firmware Version: SD1A
User Capacity:    750,156,374,016 bytes


Code:
# smartctl -i /dev/sdb
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 ===
Device Model:     ST3640323AS
Serial Number:    XXXXXXXX
Firmware Version: SD1B
User Capacity:    640,135,028,736 bytes


Note that some models are using a modified version of the original SD1A which was the one known to brick the HDDs.... why in the world would Seagate release a 'refreshed' version is beyond me! In other words, they should have given the refreshed version a new version number since so many people have been warned not to install the original SD1A.
Author:  graysky [ Tue Jan 27, 2009 3:32 pm ]
Post subject: 
Seagate engineer explains Barracuda firmware issues - article from guru3d you might find interesting regarding this debacle.
Author:  Liv2Cod [ Tue Jan 27, 2009 7:50 pm ]
Post subject: 
I saw Fry's selling the 1.5T drives for $129 this weekend. Not a lot of takers, tho... :lol:
Author:  cecil [ Wed Jan 28, 2009 1:06 am ]
Post subject: 
:lol: :lol: :lol:
Author:  graysky [ Wed Jan 28, 2009 3:55 pm ]
Post subject: 
Liv2Cod wrote:
I saw Fry's selling the 1.5T drives for $129 this weekend. Not a lot of takers, tho... :lol:


From what I've read, the SD1B firmware fix is great. 11-1/2 gigs per dollar is pretty sweet.
Author:  cliffsjunk [ Sun Feb 01, 2009 2:12 am ]
Post subject: 
From what I read, this sounds like it might be more than just a firmware problem, but they are trying to pass it off as firmware to avoid the cost of replacing the drives.

Take this with a grain of salt as I am suspicious by nature (;>) ...

Cliff
Author:  graysky [ Sun Feb 01, 2009 5:51 am ]
Post subject: 
@cliffsj - hmmm... do you have any links that support that assertion? Have a look at this interview with a seagate engineer:

Quote:
The problem with the 7200.11 series bricking, which has been in the news for the last month, was what really got the ball rolling. The Seagate employee says that is an old problem that was difficult to diagnose. A log or journal is written to in the firmware when certain events occur. If this reaches 320 entries and the drive is powered down, it will produce errors during initialization and not report information to the BIOS. Engineers quickly began work on a new firmware update to prevent this from happening.

Normally, a customer would go through the usual process of contacting customer support for the new preventative update and “this firmware had to go through five different checks to make sure it applies to the specific conditions to qualify sending to a customer, before now. 5 chances for us to go 'your drive needs the other (or none) firmware update'.” However, management, in order to quell the possibility of liability for drive failures, pushed a general public release of the firmware. “Suddenly, it's down to one check, and even that was more designed for a contingency just in case the wrong firmware was sent out.”
Author:  cliffsjunk [ Wed Feb 18, 2009 8:33 am ]
Post subject: 
graysky wrote:
@cliffsj - hmmm... do you have any links that support that assertion?


The data from some bricked drives are recoverable and from some are not. Now why would that be if it is just a log full error?

I remember reading that part of the issue was a particular chip (buffer ram?) was not being detected correctly sometimes and that smells like bad hardware unless they can fix it by adding wait states in software.

Sorry I really don't remember where I read these things.
Page 1 of 1 All times are UTC - 6 hours
Powered by phpBB® Forum Software © phpBB Group
http://www.phpbb.com/