Am i right in saying this disk did not even last 2 days!!!

[abs0lut.zer0]

pls check my log

 

p.s. am running a pre-clear as i post this

sdasmart.txt

[Johnm]

that my friend looks like a failing drive

 

I would also check your power cables and size of your power supply unless you really did power it on and off 31 times in its 43 hour  life time.

 

The high error rate "could" be attributed to a lose power plug,

 

but the high pending sector counts looks grim. I would guess you in fact that you got a DOA...

that or you dropped your server while running ...more likely the first one.

 

RMA it. try and return to your vendor first to save time and expense

 

[abs0lut.zer0]

here is a single pre-clear pass any thoughts?

preclear_rpt__WD-WMAZA4313074_2011-12-04.txt

preclear_start__WD-WMAZA4313074_2011-12-04.txt

preclear_finish__WD-WMAZA4313074_2011-12-04.txt

[SSD]

I'd return the drive and get a new one (hopefully you're within your return period from the seller to do an exchange and get a new one, and would not need to RMA it and get a refurb).  It is not aweful, but a brand new drive should not have these reallocated or pending sectors.

 

My experience is that once sectors start to reallocate, that the drive is on a fast or slow spiral to a premature death.  If you have only 1 or 2 reallocated sectors, and they don't increase, it is more common for it to not get worse and all if fine.  But it seems to have started life with 6 reallocated sectors, and now it is up to 19.  

 

If you want to keep testing it, you could run 3 more preclears, and monitor the reallocated sectors.  If they hold constant at 19, the drive may have settled down and work well.  But if each preclear is seeing the number increase, that will be a sure sign to return the disk.

 

In addition to the reallocated sectors, I am seeing some other disk errors in the log.  95% of the time, these types of errors indicate a problem with the cabling.  But rarely a bad drive can cause similar errors.  But even if there is a loose cable, it would NEVER generate reallocated sectors, so correcting this cabling issue would not make your reallocated sectors stop happening.

 

There have already been almost 1500 of these errors (only the most recent 5 are shown in the log, here is the most recent one):

 

Error 1445 occurred at disk power-on lifetime: 43 hours (1 days + 19 hours)
 When the command that caused the error occurred, the device was active or idle.

 After command completion occurred, registers were:
 ER ST SC SN CL CH DH
 -- -- -- -- -- -- --
 40 51 08 18 a3 61 e6  Error: UNC 8 sectors at LBA = 0x0661a318 = 107062040

 Commands leading to the command that caused the error were:
 CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
 -- -- -- -- -- -- -- --  ----------------  --------------------
 c8 00 08 18 a3 61 e6 08      01:48:53.235  READ DMA
 c8 00 08 10 a3 61 e6 08      01:48:53.235  READ DMA
 c8 00 08 08 a3 61 e6 08      01:48:53.235  READ DMA
 c8 00 08 00 a3 61 e6 08      01:48:53.235  READ DMA
 c8 00 08 f8 a2 61 e6 08      01:48:53.235  READ DMA

 

[lionelhutz]

FI, there was one poster who had a drive showing pending sectors. He replaced some cabling and maybe the PS and the drive cleared the pending sectors without reallocating them and began to work as if it had never had any issues.

 

[WeeboTech]

The DMA errors could be a cable, power, interference on the cable or the drive timing out when it tests a sector.

 

I would suggest doing a scrub using badblocks.

 

The preclear is a great tool, but it's not as extensive as bad blocks for exercising sectors.

 

badblocks will do 4 passes each of a specific pattern in "write" mode.

 

If you have data on the drive you can do a non destructive read/write/read mode.

This uses a random pattern for each block of sectors.

 

These tests will take a long time.

 

I did a non destructive read/write/read test on 2TB drive and it took about 130 hours.

I have not done a 4 pass write mode test on a drive that large, but plan to before putting any drives in service.

 

A full/write read badlocks test took a little over 24 hours to run on a 500gb drive. So I would assume 4x that for a 2TB drive.

 

The output of badblocks can be used as input to mkrieserfs so that it avoids them, but I really believe if there are any pending sectors, then a future parity check/sync/create will fail unless they get reallocated.

 

In addition, you could use this thread to assist with possibly forcing the bad sectors to get reallocated.

http://lime-technology.com/forum/index.php?topic=14563.0

 

Possibly running badblocks in a confined location to exercise the blocks.