Hi,

 

I currently have 6 x 2TB drives in my array including parity.

1 of my data drives has died.

It will take 1 week before I will get a replacement from my supplier.

Today I bought a 4TB for my other PC.

 

Instead of putting the 4TB in my PC, will be work if I stick that 4TB in my NAS and rebuild parity etc until my replacement 2TB drive arrives?

I want it to only use half of the 4TB, but I just want to run protected again so I don't lose everything.

Then when my replacement 2TB drive arrives I will pull out the 4TB and put in the 2TB.

 

Thanks!

 

No, that won't work without altering the parameters of the new drive.  What you COULD do is use HDAT2's SetMax function to make the 4TB drive "look" like a 2TB drive, and then do the rebuild onto that drive.  http://www.hdat2.com/

 

But I wouldn't recommend that.  I'd just turn off the UnRAID server until the replacement drive arrives.  If any other drive fails before you do the rebuild, you'll lose all the data from both the current failed drive and the 2nd failed drive.

 

If you want the rest of the array parity protected, you could also just do a New Config without the failed drive; and then let the parity drive be updated with the new parity computations.  Then, when you later get the other drive, repeat the process and then copy all of the data that was on the failed drive from your backups.    But this requires that (a) you have a current directory of the failed drive, so you know what files you need to copy; and (b) that your backups are all current.  If either of those isn't true you can't do this !!

 

 

 

 

 

A suggestion would be to move all the data from the failed drive eg. /mnt/disk2 onto the other working drives if there is capacity, and then start a new config with only 5 disks an allow parity to rebuild, just ensure you select the same disk you used for parity again else you'll loose data, and once your new drive arrives just add it and expand the array to 6 disks again.

 

It would be awesome if there was an evacuate disk option, which would allow you to evacuated a disk from the array and then remove it, basically doing exactly what I have just explained above but supported from the gui.

 

Install 4TB as a cache disk.  Copy files from simulated drive to your 4TB cache drive.  Do a new Config and don't assign the bad drive to the array and keep the 4TB as your cache drive. MAKE SURE YOU GET THE CORRECT DRIVE AS PARITY. Rebuild parity for the reduced array.  Copy the files from the cache drive back to your now reduced size array or leave them on the cache drive until you get your replacement.  Add the replacement to the array as a new drive. Basically what the others were saying but using the 4TB as a cache drive so that you could still use the array while you are doing these operations.

Bob's idea is a good one.  Hadn't thought about doing that, but I assume it will work okay (as long as you're running v5 UnRAID -- v4.7 doesn't have > 2TB support).

 

Another thing you can do, which may be preferable, since that's what you plan for the 4TB drive anyway, is to go ahead and install the 4TB drive in your PC;  then create a folder called "FailedUnRAIDDrive" (or whatever you want to call it);  and copy ALL of the data from the failed drive to that folder.    UnRAID will do this by reconstructing the data from the other drives (that's what fault-tolerance is all about).

 

THEN just do a New Config on UnRAID, and do NOT include the failed drive in the configuration.  It will rebuild parity ... and you'll have a parity protected array.

 

When you get your replacement drive, pre-clear it;  then add it to your array (the array will remain protected throughout this process) ==> and then copy the data from your PC back to the UnRAID server.