help with disk errors

[wgstarks]

Thanks. I'll add this to my homework. Anyone have any ideas what I should do with the failed disk. Should I just trash it and get a new one? Is it just a bad sector that needs to be repaired and I can put it back into service? I'm fairly sure I can use Disk Utility to repair the disk. Just not sure if I should.

Post a SMART report.

Finally got the smart report (attached). Maybe someone can take a look who understands them and let me know what I should do with the disk?

smart.txt

[BobPhoenix]

The drive is old 22465 (2.5+ years) power on hours but it doesn't have any serious smart errors that I can see.  It does have some "High_Fly_Writes" but I've always seen posted that they are nothing to really worry about.  However when a drive gets to be 2.5 to 3 years old and I start having problems (note plural) with it I usually replace it with a newer drive and use it for offline backups.  So if this is the first time you have had problems with it I would continue to use it but watch it closely and if it does this again then replace it.  That's what I did with my bad drive but it dropped a 2nd time now so I'm going to RMA it if it is still under warranty.  It took my drive about 1-2 months before it did it again so it really didn't last very long.  Maybe you would have better luck but in any case I would plan on replacing it in the near future so I would have a drive on hand precleared and ready to go.  Not that a preclear is required before you rebuild onto it but that way you can have it tested before you would need to use it.

[wgstarks]

The drive is old 22465 (2.5+ years) power on hours but it doesn't have any serious smart errors that I can see.  It does have some "High_Fly_Writes" but I've always seen posted that they are nothing to really worry about.  However when a drive gets to be 2.5 to 3 years old and I start having problems (note plural) with it I usually replace it with a newer drive and use it for offline backups.  So if this is the first time you have had problems with it I would continue to use it but watch it closely and if it does this again then replace it.  That's what I did with my bad drive but it dropped a 2nd time now so I'm going to RMA it if it is still under warranty.  It took my drive about 1-2 months before it did it again so it really didn't last very long.  Maybe you would have better luck but in any case I would plan on replacing it in the near future so I would have a drive on hand precleared and ready to go.  Not that a preclear is required before you rebuild onto it but that way you can have it tested before you would need to use it.

Thanks. Think I'll preclear it and see how it handles that. If there aren't any problems I'll re-install it. Newegg has the 4TB WD Reds on sale again. I've already ordered one to setup as a hot spare.

 

Thanks to everyone who has helped me with this. It is much appreciated.

[SSD]

The attribute called hardware_ecc_recovered is a bit concerning having been as low as 4 on a normalized scale but currently 20. I believe this is measuring the level of error correction being applied to the data. But it has no reallocated or pending sectors. I would not be feeling the need to immediately replace this drive.

[wgstarks]

I have finished pre clearing this disk and I'm wondering if I can place it back into service?

 

Preclear Results

916
No SMART attributes are FAILING_NOW

0 sectors were pending re-allocation before the start of the preclear.
0 sectors were pending re-allocation after pre-read in cycle 1 of 4.
0 sectors were pending re-allocation after zero of disk in cycle 1 of 4.
0 sectors were pending re-allocation after post-read in cycle 1 of 4.
0 sectors were pending re-allocation after zero of disk in cycle 2 of 4.
0 sectors were pending re-allocation after post-read in cycle 2 of 4.
0 sectors were pending re-allocation after zero of disk in cycle 3 of 4.
0 sectors were pending re-allocation after post-read in cycle 3 of 4.
0 sectors were pending re-allocation after zero of disk in cycle 4 of 4.
0 sectors are pending re-allocation at the end of the preclear,
    the number of sectors pending re-allocation did not change.
0 sectors had been re-allocated before the start of the preclear.
0 sectors are re-allocated at the end of the preclear,
    the number of sectors re-allocated did not change. 

 

pre clear reports attached.

preclear_rpt_9XK0FBJV_2014-05-23.txt

preclear_start_9XK0FBJV_2014-05-23.txt

preclear_finish_9XK0FBJV_2014-05-23.txt

[SSD]

I'd say yes. Just keep an eye on that hardware_ecc_recovered attribute.

[wgstarks]

This disk red balled again. Tried to run a SMART report but it failed (IO error). I've got a WD Red 4TB preclearing. Should I go ahead and just replace this one?

[SSD]

Have you verified the cables are secure? If so you might try swapping out the sata cable for a fresh one.

 

If it continues to be non-responsive to even smart requests, switch to the new drive.

 

This drive seems ok based on the SMART attributes, so it certainly is acting like a cabling issue. But drives can and do die in ways that you can't detect using SMART.

 

Loose cables are the #1 reason for drives red-balling here. I recommend getting 5in3s, 4in3s, DS380, Norco, or similar that lets you easily exchange drives without opening the case and disturbing the cabling. I used to think this was a luxury but now feel it is a requirement once you get over 3-4 drives. My fav is the SuperMicro CSE-M35T-1B. Not so quiet with the stock fan but they work great and keep drives cool. The horseshoe shaped aluminum cages allow optimal airflow to the bottom of the drive where it gets its hottest.

[wgstarks]

Thanks. It's going to be a couple of days before I can shutdown the array. The new drive is warming up in an empty slot. Loose cables may be a possibility, I just replaced the fans. The sata cables are locking but I'm sure that doesn't always guarantee anything. I'll try replacing the cable as soon as I can shut down the array.

[DaleWilliams]

Loose power connectors can happen also, when you open up your case and move wires around.

The other, rarer issue, is that folks sometime try to be 'neat and tidy' with their SATA cables and bundle them together or route them next to each other...this creates 'crosstalk'...better to let them be random routed in the case.