Rebooted on drive removal process

kenergy007

I have a 15 disk pool. All are 2tb or higher. 2 of the 15 disks started to show smart failure.

This is what I did exactly:

- I removed disk 1 failure (N Drive) from the pool using the fast removal option.
- The pool started to balance again.
- Before the balance completed I removed disk 2 failure (P Drive) from the pool using the migrate option.
- During removal of failure 2 drive (P Drive) the server froze up and I was forced to reboot.

After the reboot N drive was not part of the pool anymore and P drive was still part of it. The pool started to duplicate but every time it did it ended up with a pool condition of %50 with warnings. Also the total sizes of the folders is showing 3tb less then before. I checked random files but all the data seems to be there.

I tried adding the N drive back into the pool but after duplication it still shows %50 condition and 3tb missing.

Is my whole pool fubar'ed or can I fix this?

Shane

Pool condition of 50% means there are unduplicated files that should be duplicated (or some other problem involving duplication).

Try to run a consistency check on your pool: go to the "Pool" tab of the Dashboard's "Server Folders and Hard Drives" section, select your pool drive, choose "Check duplication consistency" from the task list on the right.

If this does not work, what are the warnings?

kenergy007

Ok I tried that didn't work. Just tried to duplicate again.

This is the warning:

Duplication Warnings
There were problems checking, duplicating or cleaning up one or more files on the pool.

Files:

\ServerFolders\Videos\... etc

Just tells me to duplicate again, but it doesn't do anything.

kenergy007

A lot of the drives have "Other" as files in the pie graph instead of "Duplicated files"

Shane

It's possible DrivePool could be having trouble migrating files off the damaged P drive.

Can you remove (via the slower migrate option, NOT the quick remove) the P drive now, or does it still freeze or does DrivePool refuse due to the duplication problem?

And do you have enough room on your good drives in the pool to take all the files from your two bad drives?

kenergy007

Everytime I try to remove P drive it fails or the server reboots on its own.

I do have enough space. I'm going to shutdown the server, physically remove the P drive and chkdsk it on another computer to see if I can flag the bad sectors. Then pop it back in to the server and try the slow drive removal process with it.

I guess worst case is to fast removal with data loss and manually transfer the files back to the server from the drive off a sata-dock. Am I right?

Also thanks so much for your help Shane!

Shane

Hmm. If you're at all worried that the drive might imminently die, and you don't have backups you can restore of the missing pool files, then my suggested priority would be "get data first, check drive second".

In such a case, I'd put it in the other machine's dock and copy all I could (beginning my efforts with something like Unstoppable Copier, initial setting Auto Skip Damaged Files tick, Overwrite unticked, rest as default, and proceeding from there as needed), and only then worry about "chkdsk /r". Afterwards, tell DP to forget the missing disk and manually merge the copied files back into the pool.