Reply
Byte
Cami
Posts: 6
Registered: ‎06-03-2012
0
Accepted Solution

Drive passes Seatools but I'm not convinced.

 


Hi

I've been starting to worry about my 2TB samsung drive. 

 

It's been giving very slow tranfer rates. I also have a 1TB Samsung, which is tranferring as fast as 50MB/s where as my 2TB has slowed to 2MB/s on some copies. Max it's reached recently is 17MB/s. 

 

I moved the location of it thinking maybe it was just the connections and then tried coying off a USB drive. The 1TB copied at at least 11MB/s where as my 2TB wouldn't get over 1MB/s. 

 

It passed all the tests in Seatools though. So I ran HDD Scan. The SMART check pulled up a few errors on HDD scan. 

 

I'll attach the page here. 

 

Under warranty until 2014

Yottabyte
fzabkar
Posts: 4,689
Registered: ‎01-27-2009

Re: Drive passes Seatools but I'm not convinced.

Your drive has 80 (= 0x50) bad sectors which have been replaced with spares, and two which are waiting to be replaced (pending reallocation).

I would suggest that you run a read benchmark using HD Tune. The shape of the curve will tell us a lot about the state of the drive. As for your results via USB, a 1MB/s transfer rate is what you would expect via USB1.1, in which case HD Tune's benchmark would be a flat line.

Byte
Cami
Posts: 6
Registered: ‎06-03-2012
0

Re: Drive passes Seatools but I'm not convinced.

[ Edited ]

Thanks for the reply. 

 

I think this is what you asked for. 

http://dl.dropbox.com/u/52986617/Capte.JPG

 

The USB speeds were rough glances at Teracopy's speed. I didn't move the USB port but it was much, much fast going to my other drive.

 

EDIT: here's the health reading in HD TUNE

http://dl.dropbox.com/u/52986617/Capte2.JPG

 

Also, i've already ordered a "Seagate ST2000DL003 3.5 inch 2TB" to transfer all my data ASAP.

Yottabyte
fzabkar
Posts: 4,689
Registered: ‎01-27-2009
0

Re: Drive passes Seatools but I'm not convinced.

Your HD Tune benchmark is very bad. A good graph (via USB) would have a smooth, flat read "curve". The access time graph would consist of a band of data points with a width equal to the time for one complete rotation of the platters. For a 5400 RPM drive, the width would be 11 msec. In your case there are three bands. The first is the normal one, while the second and third bands correspond to read retries. A good drive would have only one band with very little scatter.

I would backup your data and replace the drive.

Byte
Cami
Posts: 6
Registered: ‎06-03-2012
0

Re: Drive passes Seatools but I'm not convinced.

[ Edited ]

Thanks, I suspected something was wrong with it. 

 

It's connected directly via SATA. I also did a test with my good samsung drive. Didn't save it but the minimum speed was about 50MB/s compared to 3MB/s here. Although my good one is an F1 7200rpm where as this is a green 5900/5400rpm. 

 

So will Seagate still sort it under warranty? Even though it passes all the Seatools tests? They say it's covered till 2014 but they also say to provide the Seatools results.

EDIT: 

Re-reading the thread, I think there may be slight misunderstanding. The drive has always been connected via SATA. In a HP Micro Server. When I mentioned USB. What I'd done was plug a memory stick in and tried to copy from the USB stick to the 2 HDD's... Just to clarify.

 

EDIT:

Warranty Status
(*see details below) Options
(*see details below)

 HD204UI In Warranty
Expiration 11-Oct-2014 In Warranty


*Warranty Status & *Options

In Warranty
This drive is covered by a Seagate warranty.

If the drive is defective, it may be returned for exchange. You will not receive your original drive in return. Please do not return a healthy drive. Seagate reserves the right to return a good drive back to its owner and to charge for all associated shipping costs.

Test with SeaTools to verify the drive is failing and save the SeaTools Test Code result number to validate the drive failed SeaTools testing.

Please back up all data before returning the drive.

Yottabyte
fzabkar
Posts: 4,689
Registered: ‎01-27-2009
0

Re: Drive passes Seatools but I'm not convinced.

Sometimes the low transfer rate can be due to the OS switching from DMA mode to PIO mode as a consequence of encountering an excessive number of CRC errrors.

Otherwise, Seagate does provide a way to RMA a drive in the absence of a SeaTools test code:
http://www.seagate.com/staticfiles/support/seatools/seatools-test-codes.html

In your case you would choose "poor performance", code C5CB9DAA. It would be best to contact Tech Support, though.

Byte
Cami
Posts: 6
Registered: ‎06-03-2012
0

Re: Drive passes Seatools but I'm not convinced.

[ Edited ]

OK Well I hadn't done much apart from emailing tech support until now. I was waiting for my replacement drive so I could copy all the data.

 

Once I copied the data, I formatted the drive. 

 

I think the HD Tune benchmark is possibly healthier now? Old is the left. On the right is what it's like now after I formatted it.

 

https://dl.dropbox.com/u/52986617/Capturecc.JPG

 

EDIT: Although it's still throwing up some warnings under the health tab 

https://dl.dropbox.com/u/52986617/Captursdade.JPG

Yottabyte
fzabkar
Posts: 4,689
Registered: ‎01-27-2009
0

Re: Drive passes Seatools but I'm not convinced.

The reallocated sector count has increased by 24. Two of those would be the previously "pending" sectors, but ISTM that this number can only grow. Therefore I wouldn't trust your drive. That said, I find it strange that the normalised value hasn't budged from 100. I've seen this behaviour in other Samsung models and I'm wondering whether it is some kind of SMART bug.

Byte
Cami
Posts: 6
Registered: ‎06-03-2012
0

Re: Drive passes Seatools but I'm not convinced.

Just to let you know, I got an email from Seagate saying the replacement is on it's way. 

 

So I guess there was something wrong with it after all. 

 

I wonder if I'll get a Samsung or a Seagate. Hoping for a Seagate actually.

Byte
Cami
Posts: 6
Registered: ‎06-03-2012
0

Re: Drive passes Seatools but I'm not convinced.

Got a nice seagate barracuda as a replacement. Seems to be working fine. 

 

Thanks again for the help.