locked
How do I solve this corrupt file problem? RRS feed

  • Question

  • I keep getting these messages in the WHS tray icon saying there is a file conflict. When I check it out, there will be a single file (usually a .jpeg, but once it was a thumbs.db file) listed and the conflict description is "The file or directory is corrupt or unreadable". Even though it says this, I can still open the file fine. Perhaps this is because I have file duplication turned on for all of my data. Rebooting the server gets rid of the error message. It will go through a chkdsk routine and "restore" the file. All will be well for the next 5 or 6 hours, until I get the next file conflict alert message.

    How do I resolve this issue? Is this a sign that a hard drive is failing? If so, how do I determine which drive (I have 8 2TB Western Digital hard drives) it is so that I can remove it? I appreciate your help.

    Monday, March 22, 2010 1:51 PM

Answers

  • Yes, it's probably a sign that you have a failing drive. I would suggest you check all the drives in your server for errors. Do you have any USB drives in the storage pool?

     

    In any computer with many drives, identifying the specific drive that's failing can be a challenge. Assuming chkdsk shows you that one of your storage pool drives is having problems, you should be able to identify it by noting which mount point chkdsk identifies as having issues, then disconnect drives one at a time and reboot until you get the one associated with that mount point. Leave it disconnected and remove the "missing" drive using the console. You may be warned that you will lose some backups; I would recommend that you accept this and continue. (If this occurs, afterward you should use the Repair button in the console.) Since you have duplication turned on for everything, I wouldn't expect any warnings about losing files in your shares.


    I'm not on the WHS team, I just post a lot. :)
    • Proposed as answer by kariya21Moderator Tuesday, March 23, 2010 2:31 AM
    • Marked as answer by PullnPray Monday, July 12, 2010 4:23 AM
    Monday, March 22, 2010 2:25 PM
    Moderator

All replies

  • Yes, it's probably a sign that you have a failing drive. I would suggest you check all the drives in your server for errors. Do you have any USB drives in the storage pool?

     

    In any computer with many drives, identifying the specific drive that's failing can be a challenge. Assuming chkdsk shows you that one of your storage pool drives is having problems, you should be able to identify it by noting which mount point chkdsk identifies as having issues, then disconnect drives one at a time and reboot until you get the one associated with that mount point. Leave it disconnected and remove the "missing" drive using the console. You may be warned that you will lose some backups; I would recommend that you accept this and continue. (If this occurs, afterward you should use the Repair button in the console.) Since you have duplication turned on for everything, I wouldn't expect any warnings about losing files in your shares.


    I'm not on the WHS team, I just post a lot. :)
    • Proposed as answer by kariya21Moderator Tuesday, March 23, 2010 2:31 AM
    • Marked as answer by PullnPray Monday, July 12, 2010 4:23 AM
    Monday, March 22, 2010 2:25 PM
    Moderator
  • Also you might be interested to know that there is an Add-In called Disk Management that will give your serial number information from the drive.  Also I think WHS Smart will give you information about the serial numbers as well as useful health information that may indicate if other drives have a problem.

    http://www.wegotserved.com/2010/01/23/new-whs-add-in-home-server-smart/


    --

    • Proposed as answer by sdisk Tuesday, June 8, 2010 5:57 AM
    Monday, March 22, 2010 4:34 PM
  • First, I would like to apologize for not responding to this thread for so long. Life problems got in the way, and I put my Windows Home Server project on hold for a couple of months. Ken, back in March I ran the chkdsk script that you linked to, and this alone solved my file corruption issue; haven't had an error since. But it also drastically reduced my available storage space. The "system" is now using up 2.5 terabytes of storage. I looked for the chkdsk log in the event viewer, hoping it might explain what happened, but I couldn't find any such log. Recently, I took the time to investigate this further. I ran chkdsk on all the drives again, but this time only one at a time, copying the results from the DOS screen for each disk.  Below are the bad sector numbers for each of my 8 2TB disks.

    • D: 879 GB
    • G: 41 GB
    • I: 2.4 GB
    • K: 16 GB
    • M: 1.8 GB
    • Q: 879 GB
    • W: 0
    • Y: 879 GB

    So I guess this is where the 2.5 of "system" storage comes from. What percentage of bad sectors is considered acceptable in a hard drive? I ordered all of my hard drives at the same time, and so I guess I just got a bad batch. The one hard drive that has zero bad sectors was a replacement hard drive for one that was DOA. I think that at least for the 3 that have 879 GB of bad sectors, I should try to return them to WD. The problem is, one of those is the system drive. What's the best way to go about doing this? Also, I guess I'll need to buy additional hard drives so that I have enough space to remove the bad ones without turning off file duplication.

    I've never RMA'd a hard drive before that I had already extensively used (just DOAs). Should I be concerned with privacy/security issues?

    Thanks for the help.

    • Proposed as answer by sdisk Tuesday, June 8, 2010 5:57 AM
    Tuesday, June 8, 2010 12:03 AM