MSI K8N Diamond - Am i going to be flogging a dead horse?

[Kode]

This is the message i sent to them:

Hi, Last week i bought 3 1.5TB Harddrives from you, They are to be used as data drives, so before using them i have put them through a preclear stage that gets an initial SMART report, pre-reads the entire disk, writes zeros to the entire disk, post-reads the entire disk and then gets a final SMART report, 2 of the disks passed this with no issues, but one of them seems to be randomly struggling, it is randomly unable to re-read what it has previously written to the platters.  Also each smart report indicates that sectors are pending re-allocation, but when i do another preclear on it, the pending count changes, but the re-allocated amount has not changed, indicating that on this pass, it could write to these sectors, but not a new section.  As i said, the other 2 drives have not exhibited this behaviour.  I have the drives attached to a 700watt FSP EPSILON FX700GLN PSU so having just 3 drives shouldnt be overloading the rails, plus the drive in question is attached in the middle, so if there were problems with the power, it should be affecting the last one as well.

If you have any suggestions please let me know, i am going to double check the connections and redo the preclear process, but the process takes about 27hours to complete, and if it still is having issues i will probably have to RMA the disk as i cant trust data to this disk currently.  (I have run 3 preclears on this disk so far as usually any bad sectors should be re-allocated during this process and when it stops re-allocating them the drive should be fairly safe to use.

 

I'll see what they say, in the mean time, i have attached it to a different power connector and have started another preclear

[Kode]

This is so frustrating! No sectors have been reallocated, and the pending has dropped to 2! BUT the syslog shows errors, and i'm fairly sure that if i add it to the array again i'm just going to get a load of read errors, but if i RMA it, when they check it, it will just say its passed the S.M.A.R.T test:

 

May  7 15:56:39 Tower preclear_disk-finish[774]: === START OF READ SMART DATA SECTION ===

May  7 15:56:39 Tower preclear_disk-finish[774]: SMART overall-health self-assessment test result: PASSED

 

though it does say:

 

May  7 15:56:39 Tower preclear_disk-finish[774]: Self-test execution status:      ( 121)^IThe previous self-test completed having

May  7 15:56:39 Tower preclear_disk-finish[774]: ^I^I^I^I^Ithe read element of the test failed.

 

*Edit:*

Is there a way to run some kind of read test on the disk so at least i have a metric i can use when RMA-ing the disk as unreliable?

syslog-2010-05-07.txt