›

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In with Google

Sign In Register
Our Sites

covecube.com 
community.covecube.com 
blog.covecube.com 
wiki.covecube.com 
bitflock.com 
stablebit.com 

Categories

Poll

No poll attached to this discussion.

In order to better support our growing community we've set up a new more powerful forum.


The new forum is at: http://community.covecube.com


The new forum is running IP.Board and will be our primary forum from now on.


This forum is being retired, but will remain online indefinitely in order to preserve its contents. This forum is now read only.


Thank you,


Failing drive unable to remove from pool

RimBlockRimBlock
edited April 2012 in DrivePool
I am just trying out drivepool and have hit a problem with a failing drive.

I added 4 drives to a new pool and copied on my data.  I have since noticed one drive (WD Green 1.5TB) is failing with bad block errors and the number of read errors are rapidly increasing.  I need to remove the drive from the pool (there is about 1.8GB storage free with all drives added) but cannot as it fails trying to copy the data off the drive due to CRC checks (result of the drive failing).

What I would like to see;
  • Information on which files are on which physical drives so I can see which files will be physically affected on drive failure.
  • Option to ignore CRC failure and continue migrating data off the drive being removed but to log which file was affected so I can replace it with a known good backup.
  • A waring about the pool state.  Currently it is listed as healthy although I have over 7,000 bad block errors in the event viewer for the failing drive.
I have the majority of files backed up so can recover fairly easily but how to take the faulty drive out without loosing all the data in the pool.  Presumably just pulling the drive would not be a good solution.

Ultimately, how do I get my drive out of the pool whilst migrating the non-corrupted data and  noting which data need replacing.

Thanks
RB
Share on Twitter

Comments

  • ShaneShane
    edited April 2012 Resident Guru
    Hi RimBlock.

     * The "Everything" program from www.voidtools.com can list every file on every NTFS volume; if you select "Match Case" and "Match Path" from its "Search" menu and then enter "\PoolPart." into the search field, it will display every file and folder in the pool's physical volumes (in other words, a complete list of which pooled drives contain which files). You can then choose to "Export" the entire list from the "File" menu to a text file.

     * CRC failures are generated by Windows, not by DrivePool. I would suggest tools such as TeraCopy or Unstoppable Copier which have the ability to ignore and/or log such errors.

     * As far as I know the pool state runs at the file, not disk block, level; I'm presuming that you aren't using duplication, otherwise I would expect that the pool would have noticed that duplicated files on the bad disk aren't matching their twins on the other disk(s).

     * If you pull the bad drive, the pool will continue to operate; the files on your other disks will remain available.

    "Ultimately, how do I get my drive out of the pool whilst migrating the non-corrupted data and noting which data need replacing."

    If your pool has duplication on for everything, simply pull the bad disk and THEN remove it from the pool. DrivePool will re-duplicate your files and nothing should be lost. Otherwise, if your pool does not have duplication on for everything:

     * If possible, make sure nothing besides the Dashboard is accessing the pool or the bad disk.
     * Use "Everything" to export a list of all files in "\PoolPart." folders, as described above.
     * Open an explorer window on the server, and turn on show hiddden folders if not already on.
     * Rename the bad disk's PoolPart folder (to anything you like, e.g. stick "bad" in front of it).
     * DrivePool will notice the disk has gone missing from the pool (it may take a few seconds).
     * Remove the "missing" disk from the pool. This should be quick if you are not using duplication.
     * Use TeraCopy or Unstoppable Copier (or similar) to copy only the still-good files off the bad disk to a good disk.
     * Once this is done, you can move the still-good files from their new location back to their original location in the pool (note: skip existing files).
     * Use "Everything" again to export a new list. Compare the new and old lists to find your lost files.

    (apologies if you were spammed with notifications, I had some trouble posting)
    Share on Twitter
  • ShaneShane
    Resident Guru
    Hmm. You could also simply use the log from TeraCopy or Unstoppable Copier as it should show which files it couldn't copy.
    Share on Twitter
  • ShaneShane
    Resident Guru
    I think I might have missed a step. Just in case, reboot WHS after renaming the bad disk's PoolPart folder but before removing the "missing" disk from the pool.
    Share on Twitter
  • jbenfieldjbenfield
    Member

    I'm just wondering how the drive removal went?

    This past weekend I had a drive go bad on my old WHS and it simply couldn't be removed from the pool. DE is in a hosed state and I ended up having to move 6TB of data onto non-DE drives. So, with that done, I'm finally moving to 2011. Is Drivepool going to give me the same removal/replacement headaches as DE has or can I do a pull and replace and just refresh the duplication?

    Share on Twitter
  • ShaneShane
    Resident Guru
    A few people have had trouble with removing bad drives, but (a) usually a pull and replace works just fine if you've got duplication on, (b) when it has got stuck, Alex has been excellent with support (and has added fixes to subsequent versions, q.v. drivepool's changelog).
    Share on Twitter
Sign In or Register to comment.