Hi all,

 

After changing the disks to spin down after 15 mins recently (v4.7), after a couple days one of my newer drives showed some errors, the icon was red and the drive was inaccessible. After a soft reboot the drive showed as unavailable and had been removed from the array.

 

I have today down a shutdown and power on - the drive now shows as added to the array (same disk) and I get the following message:

 

Start will bring the array on-line, start Data-Rebuild, and then expand the file system (if possible).

 

Is there any way to re-add this disk to the array without rebuilding? I would think the data is still intact and the error may have been caused by a faulty connection (reseated all the cables and drive bays when shutdown).

 

There wasn't much used space on the drive (probably about 90% free on a 2TB drive).

 

I just want to make sure a rebuild is the only solution and that everything will function as before should I continue with this option. I've never had to rebuild a drive before, so just a bit nervous at the moment.

 

Any help would be great.

It it's asking to rebuild, unraid doesn't recognize the drive as your old array drive. That probably means there is data missing... After it came up red balled... I'd replace it and rebuild on to a good disk.

 

Before all that check your SATA and power connections!

The red-ball usually means that a write to it has failed. Could have been a loose cable, but can also be a failing drive.

 

You can 'trust parity' if you think the disk and data on it is ok.

 

But I would at least get a s.m.a.r.t report first from that red-balled drive.

 

 

I don't believe v4.7 has a "trust parity" option.

 

Your best option is to simply let the rebuild proceed.  But note that if the drive is actually bad, that will fail => you'll then need to replace it with a new drive and let it rebuild again.    Don't, however, procrastinate, as you're "running at risk" until the rebuild is done -- i.e. any other disk failures = lost data.

 

The red-ball usually means that a write to it has failed.

No, it ALWAYS means a write to it failed.

Could have been a loose cable, but can also be a failing drive.

True.

 

You can 'trust parity' if you think the disk and data on it is ok.

But since a WRITE to it FAILED, it is almost always incorrect.

But I would at least get a s.m.a.r.t report first from that red-balled drive.

I would too.  But it has nothing to do with the fact that it is almost certainly incorrect in some way, either missing data, or file-system damage of some kind exists.  a "reiserfsck --check" is certainly in order.

Thank you garycase and Joe L. for the corrections!

 

Always learning...