Slow EADS - or not?

[silverpig]

I have a 1.5 TB WD Green EADS hard drive, formatted HFS+ journaled in my hackintosh. Everything was great up until a few days ago when I almost filled it up. The programs accessing the drive all hung, the computer was unresponsive, and I was forced to reboot.

Upon multiple reboots, the drive was sometimes not detected at POST, other times it was, but it took a long time, and most times it wasn't mounted in OSX. All other hard drives were fine.

When I did get it mounted, it was SLOW. Like, 1 MB/s on transfers slow. I bought a 1.5 TB EARS and copied the contents of the drive over (it took 3 days), and several of the files were corrupt.

I did a reformat (several) and performance was still slow. I did fscks, disk repairs, and more formats and nothing fixed the performance. I spent three days with it messing around, measuring performance on file copies and xbench, and decided to do an RMA.

Now that my new drive is almost here (an EARS), I decided to do a full wipe of the drive. I reformatted, setting the entire drive contents to be written to 0s. It took a full day to do, and several times the computer was super slow as the hard drive churned and almost locked up all my other programs.

But now performance on the drive is back to where it should be. 200 GB is going to take about 40 minutes to copy over, and xbench tests the drive as being great. Now I'm worried that I'm gonna get charged for my RMA.

Has anyone else noticed this type of behaviour on an EADS? Slowness when the drive gets almost full? I'm going to try filling it up again just to see if I can make the issue happen again.

 

[alaricljs]

Don't know what is available on the Mac, but you need to take a look at what SMART reports. You could have just had a run in with a stack of bad blocks, or something completely different.

 

[silverpig]

SMART says the drive is fine, which kind of doesn't surprise me now, but I have a whack-load of benchmarks and failed boots which say something definitely was wrong with the drive.

I've almost finished filling it back up again. We'll see what happens then.

 

[AstroGuardian]

RMA it. It's definitely a failing drive. Have seen it with a 500GB Western Green. I was used to seeing that kind of lousy disk behavior with the now long gone Maxtors. RMA it and don't waste your time.

 

[sub.mesa]

Post SMART info here; including raw values for ALL SMART variables; i do not believe your SMART data is fine, you probably have Current Pending Sector or UDMA CRC Error Count. You can RMA the drive, but ultimately this likely is lack of HDD maintenance, rather than HDD failure.

 

[silverpig]

The drive is still fairly new. 7153 powered hours as of now and I run it 24/7.

I did a SMART scan yesterday and the disk came up with no errors, but just re-ran it now and got a 707 count in Offline Uncorrectable Sector Count.

location: /dev/disk1
partition count: 1
partitions: Media 5, 
SMART Support: Supported
Last SMART Error: *not set*
Model Family: Western Digital Caviar Green family
Model: WDC WD15EADS-00S2B0
Serial Number: WD-WCAVY1230158
Firmware Version: 01.00A01
Capacity: 1,500,301,910,016 bytes
ATA Version: 8
ATA Standard: Unknown
Health: Failing
Overall Health: PASSED
Offline Data Status: 84
Self Test Status: 0
Offline Data Collection Time: 31380
Data Collection Capabilities: 7B
SMART Capabilities: 3
Error Logging Capabilities: 1
Short Self Test Time: 2
Long Self Test Time: 255
Conveyance Self Test Time: 5
Attribute Log Revision: 16
Attribute Count: 17
Attributes:
ID#   ATTRIBUTE NAME                         FLAG       VALUE   WORST   THRESH  TYPE         UPDATED     FAILED    RAW VALUE
1     Raw Read Error Rate                    0x002F     190     189     051     Pre-fail     Always      Never     179871
3     Spin Up Time                           0x0027     253     144     021     Pre-fail     Always      Never     3291
4     Start/Stop Count                       0x0032     100     100     000     Old age      Always      Never     121
5     Reallocated Sector Count               0x0033     200     200     140     Pre-fail     Always      Never     0
7     Seek Error Rate                        0x002E     200     200     000     Old age      Always      Never     0
9     Power On Hours                         0x0032     091     091     000     Old age      Always      Never     7153
10    Spin Retry Count                       0x0032     100     100     000     Old age      Always      Never     0
11    Calibration Retry Count                0x0032     100     253     000     Old age      Always      Never     0
12    Power Cycle Count                      0x0032     100     100     000     Old age      Always      Never     69
192   Power Off Retract Count                0x0032     200     200     000     Old age      Always      Never     53
193   Load Cycle Count                       0x0032     160     160     000     Old age      Always      Never     122743
194   Temperature Celsius                    0x0022     111     090     000     Old age      Always      Never     41
196   Reallocated Event Count                0x0032     200     200     000     Old age      Always      Never     0
197   Current Pending Sector Count           0x0032     200     195     000     Old age      Always      Never     0
198   Offline Uncorrectable Sector Count     0x0030     198     198     000     Old age      Offline     Never     707
199   UDMA CRC Error Count                   0x0032     200     200     000     Old age      Always      Never     0
200   Multi Zone Error Rate                  0x0008     001     001     000     Old age      Offline     Never     199970
Error Log Version: 1
Error Count: 0
Error Log: 
No errors logged.
Self Test Log Revision: 1
Self Test Log Count: 0
Self Test Log:

No self tests logged.

 

[Rifter]

I slam my 2TB EADS drive full, like to 95%+ capacity full every week doing backups of my media servers RAID storage array and it gets a consistent 60-80MB/sec transfers(the limit of my Gigabit network).

Has been performing fine for the last year or so, i thought the weekly max capacity backups would kill it fast but its been very reliable so far.

 

[AstroGuardian]

Post SMART info here; including raw values for ALL SMART variables; i do not believe your SMART data is fine, you probably have Current Pending Sector or UDMA CRC Error Count. You can RMA the drive, but ultimately this likely is lack of HDD maintenance, rather than HDD failure.

What HDD maintenance? You mean like oil change?
There is definitely something wrong with the drive. Maybe corrupted firmware.