Reply
Yottabyte
fzabkar
Posts: 4,660
Registered: ‎01-27-2009
0

Re: RAID problems with 7200.11 drives: unified thread

[ Edited ]
Here is a thread at HDD Guru involving a Seagate drive with a strange firmware issue:

http://forum.hddguru.com/seagate-st31000324ns-wrong-and-t14574.html

The drive's label indicates that it is an ES.2 model with ES.2 firmware (ST31000324NS, SN04), yet the drive is reporting that it is a 7200.11 model with 7200.11 firmware (ST31000340AS, SD35). The PCB appears to be original -- the date codes on the label matches those on the chips.

The ES.2 model number looks like a typo (maybe it should be ST31000342NS). Regardless, it still begs the question, what is actually different about the two model series? Are the enterprise drives just regular drives with different firmware? Do they have better spec'ed parts? What makes them RAID-worthy?

BTW, I believe that SD35 firmware was not publicly released as a field upgrade, which then begs another question, who updated the drive?

Message Edited by fzabkar on 26-01-2010 06:53 PM
Kilobyte
axelkloth
Posts: 43
Registered: ‎10-02-2009
0

Re: RAID problems with 7200.11 drives: unified thread

Very interesting - that would explain why I saw similar issues with the 7200.11 and the ES.2 - and it would also explain why they are available in similar capacities, at prices well below the Cheetah. Same hardware, similar if not identical firmware, just different target markets. Can someone physically verify they are the same hardware?
Petabyte
HughR
Posts: 421
Registered: ‎01-01-2009

Re: RAID problems with 7200.11 drives: unified thread

I have always assumed that the 7200.11 and ES.2 drives had the same hardware.  (Of course not all 7200.11 drives are the same, so my belief has been that each ES.2 drive was the same as some 7200.11 drive.)  Note the word "assume".  I have no evidence but common sense.

 

Why two lines for the same device?  Well, they are not identical: slightly different firmware (ES.2 has time-bounded  error recovery, required for RAID), different marketing, different warranties.  The key goal is to achieve "price differentiation".  This is the goal of any producer in a commodity business.

 

When products are commodities, their price reflects the cost of production; if some vendor's cost of production is high, he cannot even get that without losing market to lower-cost producers.  Price differentiation allows a vendor to start charging based on the value to the customer rather than the cost of production.

 

There is a very good chance that Seagate would do something to prevent ES.2 firmware working in a 7200.11 drive.  Preventing the opposite isn't very important but might be a side-effect.

 

This is also why Seagate probably makes it impossible to put the 7200.11 drives into a mode where error recovery is time-bounded.

 

What surprises me is that even ES.2 drives don't (always?  sometimes?  as often as 7200.11 drives?) work well in RAID setups.  This means that the problem is not just the traditional unbounded error recovery time.  If there is a single underlying problem, and the ES.2 problem is the same as the 7200.11 problem, then things are really grave for Seagate.  They have not solved it in a year!

Petabyte
HughR
Posts: 421
Registered: ‎01-01-2009
0

Re: RAID problems with 7200.11 drives: unified thread

I just noticed this "known issues" page for a Linux Kernel ATA wiki: http://ata.wiki.kernel.org/index.php/Known_issues#Seagate_harddrives_which_time_out_FLUSH_CACHE_when...

 

I wonder if it is true.  I wonder what the source of the information is.  I wonder if it is relevant to this thread.

Yottabyte
fzabkar
Posts: 4,660
Registered: ‎01-27-2009
0

Re: RAID problems with 7200.11 drives: unified thread

It appears that the NCQ bug was corrected in the SD1A update. The Linux community appear to have benchmarked the affected drives before and after the update, so I would say that the problem was real.

Kilobyte
axelkloth
Posts: 43
Registered: ‎10-02-2009
0

NCQ

Hugh: I doubt that the NCQ bug had anything to do with the issues that have been reported here. One of my first tests was to disable NCQ on the server and the RAID controller, and it had no effect whatsoever. I believe that the disks' firmware versions all are very buggy, with the NCQ bug being one of the few that has been fixed and now works under most circumstances.
Yottabyte
fzabkar
Posts: 4,660
Registered: ‎01-27-2009
0

Re: RAID problems with 7200.11 drives: unified thread

Here is another peculiar thread:

Seagate ST31000340NS ES.2 (SN0x firmware) seen by BIOS as ST31000340AS (SD35 firmware)?

http://forum.hddguru.com/seagate-st31000340ns-seen-bios-st31000340as-t14694.html

Kilobyte
Undermoose
Posts: 22
Registered: ‎10-18-2009
0

Re: RAID problems with 7200.11 drives: unified thread

I've posted on this thread before, so this is a followup to let everyone know that the array is running perfectly. Not even a dropout or a verify, and I've had several system BSOD from a bad sound card driver install on my Windows 7 rig.

6 disk raid 10 - using 6 x 7200.11 1.5tb drives on ICH9R.

I'm using 9.5.6.1001 Intel Rapid Storage Tech driver from Station-Drivers.com.
Regular Visitor
maxima
Posts: 8
Registered: ‎12-18-2009
0

Re: RAID problems with 7200.11 drives: unified thread

Visitor
jbenslen
Posts: 1
Registered: ‎02-02-2010
0

Re: RAID problems with 7200.11 drives: unified thread

So, since the release of the new firmware would it be safe to upgrade my ST3500320AS to the newest firmware and run it in a raid? Found this harddrive laying around not being used but since all I seem to find is problems dealing with the 7200.11 drives I didnt know if the problems been fixed yet.