Barracuda XT, Barracuda, Barracuda Green, DiamondMax and other desktop SATA and IDE drives

[Robinio]

Hi all,

I've got an ST31000524AS drive which started giving odd pauses, then crashed and wouldn't boot into windows.

I installed an OS on another drive, checked the drive with seatools and it got errors on the short DST and the long generic test with the codes A9A9E1EA & A9A9E1DA respectively.

I ran HDDScan to check the smart info, it showed warnings on:

Num

Attribute Name

Value

Worst

Raw-Hex

Threshold

5

Reallocation Sector Count

099

099

00000000-0033

036

197

Current Pending Sector Count

100

100

00000000-0003

000

198

Uncorrectable Sector Count

100

100

00000000-0003

000

Chkdsk in windows crashed a couple of times when trying to run it to repair it.

I tried swapping sata cables and moving it to different ports, but chkdsk would run through and the drive would still be the same.

I submitted an RMA for it then grabbed what data I needed off it, then ran Active@'s killdisk to Wipe it, performing a full zero write to the whole disk.

After wiping it the drive stopped appearing in device manager, so I rebooted into the bios and it was missing there too.

For curiosity's sake, I pulled the drive out and reseated it again, it detected again and I thought I'd give it a final check in windows.

Seatools now reports the drive is fine in the short DST, I'm half way through the long generic test as I write this which the drive had failed previously.

HDDscan shows the smart errors gone too, so it must have been able to reallocate the faulty sectors.

Num

Attribute Name

Value

Worst

Raw-Hex

Threshold

5

Reallocation Sector Count

097

097

00000000-009F

036

197

Current Pending Sector Count

100

100

00000000-0000

000

198

Uncorrectable Sector Count

100

100

00000000-0000

000

So now I'm not sure what to do.

Do I keep the drive, put a partition back on and use it as normal, or do I continue with th RMA and send it in?

Any advice would be much appreciated.

Robin.

Edit:

Just adding the full SMART info, as it seems to have a few other items which are probably related. I could be wrong on this, but this drive (out of the 4 conventional HDDs in this PC) is the only one to have any hex values for the highlighted items:

Num	Attribute Name	Value	Worst	Raw(hex)	Threshold
001	Raw Read Error Rate	109	088	00000173-6BF3	006
003	Spin Up Time	100	100	00000000-0000	000
004	Start/Stop Count	100	100	00000000-01E3	020
005	Reallocation Sector Count	097	097	00000000-009F	036
007	Seek Error Rate	075	060	00000220-87F1	030
009	PowerOn Hours Count	099	099	00000000-0574	000
010	Spin Retry Count	100	100	00000000-0000	097
012	Device Power Cycle Count	100	100	00000000-01E2	020
183	Unknown Attribute	100	100	00000000-0000	000
184	End to End Error	100	100	00000000-0000	099
187	Reported Uncorrectable Error	001	001	00000000-01B6	000
188	Command Timeout	100	097	0018001A-001B	000
189	High Fly Writes	100	100	00000000-0000	000
190	Airflow Temperature	064	052	36 C	045
194	HDA Temperature	036	048	36 C	000
195	Hardware ECC Recovered	054	035	00000173-6BF3	000
197	Current Pending Sector Count	100	100	00000000-0000	000
198	Uncorrectable Sector Count	100	100	00000000-0000	000
199	UltraDMA CRC Error Count	200	200	00000000-0000	000
240	Head flying hours	100	253	F7350000-0A25	000
241	Unknown Attribute	100	253	0000C02D-83C1	000
242	Unknown Attribute	100	253	0000D115-33D1	000

[Robinio]

It's now finished the long test and it says it's passed.

I've installed HDD sentinel to check what it says, it gives the health of the drive 24% and says the drive has 159 bad sectors which have been reallocated. It also says there were 438 data transfer errors.

I don't trust the drive but not sure seagate will replace it as it is :/

[fzabkar]

Your SeaTools error code uniquely identifies your drive's serial number and the number of the failing test. AIUI, if you provide this code when requesting an RMA, then Seagate will honour your warranty, even if they cannot reproduce the error. In any case, a drive with 159 (= 0x9F) reallocated sectors, plus 438 (= 0x1B6) Reported Uncorrectable Errors, is not to be trusted.

BTW, before the zero-fill operation you had 51 (= 0x33) reallocated sectors. By the time you read my post you will probably have a lot more. The reason that the Current Pending Sector Count has dropped to zero is that a sector remains pending until the OS writes to it. When this happens the drive knows that the existing data are no longer of any value. It then retests the sector and returns it to service if good, or replaces it with a spare if bad.

[Robinio]

Thanks for the reply and the information, it's much appreciated.

If they think they'll replace the drive just from the code then I'll send it back.