I'm not having any serious issues I need help with. I just want to understand how the following works:

 

1) I used reiserfsck to recover data from a disk13

2) reiserfsck recovered 2/3 of the files and created a lost+found folder.

3) I followed the 'Trust My Array' procedure to reset my entire array.

4) I reassigned all the disks to the same slots minus the parity drive. I matched all serials to the old slot and triple checked against a screenshot. Yes, i'm unprotected.

5) I started the array.

6) Now my disk13 shows the original file structure with 100% of the files. I tested a few files that were missing after the reiserfsck rebuild and they work fine.

 

My assumption was that disk13 would still show the reiserfsck rebuilt folder/file structure even after the 'Trust My Array' procedure.

 

How did 100% my original disk13 folder/file structure return after the "reiserfsck --rebuild-tree /dev/md13" rebuilt and created 2/3 the original folder/file structure?

 

Thanks,

Christopher

 

The following is a guess.

 

At the time you ran the reiserfsck was disk13 red-balled?  If so you were not running against the physical drive (unRAID does not write to a red-balled disk) but one being emulated from the combination of parity and the other data disks.  How accurately it was being emulated would depend on the state of the parity disk and the other data disks..  If parity was bad or any of the other data disks was bad then the emulated disk would have bad contents.

 

The next step you describe would discard the emulation of disk13 and revert to trying to use the physical disks (including disk13). This is not the normal way to try and recover disk13 and clear a red-ball state if the emulated version was the one with good data because it would lead to data loss.  However in this case it looks as if the physical disk was in a better state than the emulated one.

 

Is the above guess about disk13 being red-balled and there being problems with either the parity disk or another data disk valid?

Is the above guess about disk13 being red-balled and there being problems with either the parity disk or another data disk valid?

 

Your right on both points. Thank you very much for your explanation!

 

-Christopher