So I have a Drive that has been marked as failed by the array due to the ' UltraDMA CRC Error Count ' number being high. Think is this drive is coneected back to an IBM branded LSI 9211-8i. On that card I have 8 drive attached, the other three drives on that cable ALSO have controller errors. This leads me to believe it's the cable, not the drive as I doubt all 4 drives would have the same problem at the same time, given that two are SSD's of different sizes and batches and the two disk drives were purchased a year apart. 

 

So, how do I pull the drive back into the array, I'd hate to waste the drive if I don't have to. Yes I could warranty it out, but that seems so....wasteful.

43 minutes ago, eagle470 said:

So I have a Drive that has been marked as failed by the array due to the ' UltraDMA CRC Error Count ' number being high. Think is this drive is coneected back to an IBM branded LSI 9211-8i. On that card I have 8 drive attached, the other three drives on that cable ALSO have controller errors. This leads me to believe it's the cable, not the drive as I doubt all 4 drives would have the same problem at the same time, given that two are SSD's of different sizes and batches and the two disk drives were purchased a year apart. 

 

So, how do I pull the drive back into the array, I'd hate to waste the drive if I don't have to. Yes I could warranty it out, but that seems so....wasteful.

Is the disk actually disabled?

 

Tools - Diagnostics, post complete zip

It's shut down until I replace the cable later today, I don't want to risk any more disks failing right now. That would be a huge PITA.

Just now, eagle470 said:

It's shut down until I replace the cable later today, I don't want to risk any more disks failing right now. That would be a huge PITA.

The concept of a "disabled disk" has a very specific meaning in unRAID, and unfortunately your description of the problem doesn't give enough information to recommend a course of action.

 

If the disk is disabled, replacing the cable will not re-enable the disk, though it might fix the cause of the problem.

 

It's too late now, but you should always get a diagnostics before shutting down if you have an issue you want to get support on. Since you did shut down, the diagnostics will not allow us to see the actual event that caused the issue.

 

Nevertheless, please post your diagnostics when you are ready to proceed with support for your problem.

Will do. Now I know for next time. 

 

Just a quick question, why doesn't the system do a dump to a DIAG share or something when a disk fails out?

7 minutes ago, eagle470 said:

Will do. Now I know for next time. 

 

Just a quick question, why doesn't the system do a dump to a DIAG share or something when a disk fails out?

I don't know exactly what you mean by "when a disk fails out", which is why I asked if the disk was disabled.

Here is a screen shot of the failed drive and it's smart report. And the total DIAG report.

unraid1-smart-20170411-1518.zip

unraid1-diagnostics-20170411-1519.zip

Not necessary to post SMART since Diagnostics includes SMART for all disks as well as syslog(s) and other useful things.

 

The red X is what I was trying to get at when I asked if the disk was disabled. unRAID disables a disk when a write to it fails. But, that failed write and any subsequent writes to the disk are emulated by the parity array. So the array has the correct data, but the disk does not.

 

SMART for all disks looks OK. Check all connections. You can rebuild the disk to itself as follows:

 

Stop array (looks like it is already stopped in your screenshot)

Unassign disk

Start array

Stop array

Reassign disk

Start array to rebuild

 

(That was the TL;DR version of this wiki:

https://lime-technology.com/wiki/index.php/Troubleshooting#What_do_I_do_if_I_get_a_red_X_next_to_a_hard_disk.3F 

)