Barracuda ATA IV / ST360021A

[shaun]

Hi there,

I have this Barracuda IV here which won't image. In the first 6 GB or so, it will fail every 128-256MBytes, while it runs through from 6 to 10GB without errors.
(Haven't tested any further)
This makes me wonder if I've a weak head here or something even worse. Its a Apple-labelled drive, just if it matters - maybe a non-stock FW or something.

regards

[drc]

Define "fail"

[shaun]

Sorry, I obviously forgot to mention the actual problem

Although it is seeking quite loud while copying, it repeatedly seeks for several times and even louder when it comes to one of the locations where it does what I called "failing", no data is transferred and after some attempts the drive stays busy, not responding to further commands until power-cycled. The locations are reproducable the same, every approx 128MB in the first 2 Gigabytes or something, every 256MB between 2 and 6 GB and as I wrote no faults between 6 and 10GB.

I haven't yet tried whether it has a serial terminal so I cannot tell if or what it reports.

[shaun]

Here's the serial dump... I was able to image from 6,000,000k up to 17,900,000k without errors, now I'm giving the area > 18G a try.

<here the drive is powered on>

Interface task reset
1024k x 16 buffer detected
SNOWMASS - 2_Disk 5.05 07-23-01 10:24 gzymm
ATReset Power Cycling

Ref 00F3 - ATReset
Head Mask 0007 - Switch to full int.
Spin Ready

Sweeping 189 Cyls
PHMaster with Non-conforming Slave
ATReset
HMaster with ATReset
HMaster with Non-conforming ATReset
HMaster with Non-conforming Slave

<here the errors start>

AT Er 00 Nwt Er 48 RdWr 381d.00.0224
ATA St d0 Er 00 Op c8 e, 222f780,f0 7f 03
Niwot: 0222b380 4a b380.0.222 8000 0634 02dc 0000

CE Log EC=48 Rtype=4 OV=0 STStatus0
AT Er 00 Nwt Er 47 RdWr 381d.00.0225
ATA St d0 Er 00 Op c8 e, 222f780,f0 7e 03
Niwot: 0222f7f1 1a f7f1.0.222 8000 09bd 02dc 0000

CE Log EC=47 Rtype=4 OV=0 STStatus0
AT Er 00 Nwt Er 43 RdWr 381d.00.0226
ATA St d0 Er 00 Op c8 e, 222f780,f0 7d 03
Niwot: 0222f7f2 1a f7f2.0.222 8000 09be 02dc 0000
AutoRd Err 43 at 381d.00.0226

CE Log EC=43 Rtype=5 OV=0 STStatus0
AT Er 00 Nwt Er 43 RdWr 321c.01.010c
ATA St d0 Er 40 Op c8 e, 222f780,f0 00 03
Niwot: 000c1100 44 1100.0.00c 010b 0000 000c 110b
ATReset
HMaster with ATReset
HMaster with Non-conforming ATReset
HMaster with Non-conforming Slave
AT Er 00 Nwt Er 48 RdWr 381d.00.0224
ATA St d0 Er 00 Op c8 e, 222f780,f0 7f 03
Niwot: 0222f780 4a f780.0.222 8000 0400 033e 0000
AutoRd Err 48 at 381d.00.0224

CE Log EC=48 Rtype=5 OV=0 STStatus0
AT Er 00 Nwt Er 48 RdWr 1454.01.010c
ATA St d0 Er 10 Op c8 e, 222f780,f0 00 03
Niwot: 000c2200 44 2200.0.00c 010b 0000 000c 110b
ATReset
HMaster with ATReset
HMaster with Non-conforming ATReset
HMaster with Non-conforming Slave
AT Er 00 Nwt Er 37 RdWr 381d.00.0224
ATA St d0 Er 00 Op c8 e, 222f780,f0 89 03
Niwot: 0222f780 4a f780.0.222 8000 0400 033e 0000
AutoRd Err 48 at 381d.00.0224

[rameez]

PHMaster with Non-conforming Slave
This indicates pcb problem as well .

[BlackST]

seems more a degradation to me...

[shaun]

Hi there,

I also though of degradation, but I've not yet had a case like this with some oddly distributed areas of several 100-1000s of sectors which are completely unreadable. Either evenly spaced complete failures (like a dead head) or changing offsets where the failures start, sometimes readable by repeated attempts.

The nonconforming slave message is probably because no jumper was installed while my serial terminal cable was plugged in, so the drive was thinking it is a master with a slave connected