Request Help

Request Help

We are happy to help! Please fill out this form or call us
Call: 866-438-6932

Determining a Stale Hard Drive in Most RAIDs

  In my last installment I covered two of the three reasons why we do a parity check.  First we want to make sure that we do not have a stale hard drive in the array.  Although I did not cover how one determines if in fact a stale hard drive exists, I will mention it now. 

  In reference to stale drives there is a trick to identifying if in fact the parity is failing because of a stale hard drive or if there is another reason.  When a drive goes stale it will have a mix of current data, and old data.  As time passes the disparity between the current data and old data grows larger.  However, the fact remains that there is some data on a clients system that no matter how long the drive has been stale it will never change.  For example, all the operating system files, those hardly ever get updated.  Another example is if a client has many picture files, or music files, these files almost never change because they are put on the media once, only to be used for viewing, not for editing.  There are other examples, however these are the most prevalent.

  In conjunction with this fact, most file system handlers will try to fill the media from sector zero to sector X contiguously.  Like filling a glass of water.  So the operating system files and older data that has not been deleted will remain at the beginning of the drive and the more recent data towards the end of the drive.  There are factors that effect this, the main one being deleted data.  If data has been deleted then the void left by the deleted file will quickly be filled by another file or files.  Internet cache is notorious for this as they are usually smaller files that can be stuffed into areas vacated by deleted data.  So if your clients RAID has gone down and they have been deleting and adding files it will affect the way the parity check looks when scanning with a stale drive.

  Next installment I will finish my parity check example and how to see if there is in fact a stale drive, and hopefully explain the staggered offsets which plagued me in one of my most recent recovery attempts.

  Until next time…

Links to Previous RAID Data Recovery Articles In This Series

  • Check RAID Hard Drive Consistency BEFORE a Rebuild – the three most important steps to take prior to rebuilding a RAID array.
  • Analyzing RAID Parity – How to use our free RAID Diagnostic Toolkit to analyze RAID parity.
  • Hard Drive Recovery – More information about hard drive data recovery services by


5 Responses to “Determining a Stale Hard Drive in Most RAIDs”

  1. Joshua Dibble January 15, 2009 3:29 pm #

    I was wondering about the consistency of the hard drive. My e:/ is showing inconsistency when starting up windows xp. It continuely wants to run a disk check so I really don’t where to start. Any one have any solutions? I don’t know where to find info on the net that is safe for me to diagnose the problem. Thank you for the feed back. Josh…

  2. DTI Data Recovery January 16, 2009 3:52 pm #


    Is your E: drive part of a RAID? Is it a stand alone drive or a partition? We can help you, but need more info

  3. Dick Correa January 16, 2009 5:14 pm #


    The reason you are having problems is because you have bad sectors on the drive and they are getting worse. This accounts for the long defrag as well as the scan hanging in the same place.
    It is not good for the drive and ultimately your data to continue in this vein of recovery as it will exacerbate the problem to the point where you will destroy all data on your drive.
    If you wish to try this yourself you must isolate the bad drive, image it onto another drive using a smart cloner so as not to belabor multiple retries on a single sector and then recover your data from the newly imaged drive using a data recovery software product. The cloner can run from $99.00 to several hundred dollars and the data recovery software is usually around $50.00. If you use a cheaper cloner and not a smart cloner you will have to hand jump the bad areas of the drive.

    This is a long and involved process and could take many hours. You must sit with the drive and watch this process, as you do not want the drive to overheat or hang in a particular place.

    I hope this helps. We do this kind of work at the shop all the time. We have special tools specifically designed for your problem and can retrieve data that most normal cloners cannot. The price is extremely reasonable.

    Dick Correa

  4. Joshua Dibble January 16, 2009 4:16 pm #

    I don’t really know. My hard drive C: is a stand alone 80gig and the drive e: is internal. It is not an NTFS file system it is a fat32. I really don’t know the difference between the two. Basically when the cpu starts up it goes to scan disk automatically Checks drive e: for consistency. Then it gets stuck at 16% and will repeat the process for infinety. repeatingly turning cpu on and going through the scan untill it gets to 16% and then shuts down. I can click a key within 10 secs or so and skip the process but I would like to figure out this error. I thought that maybe the disk partition was lost or damaged but it shows that a partion is still there: E: = recovery. This problem started when I installed Verizon internet security. I had performed a 3 hour long defrag using the verizon security suite. This was way longer then usual. I then tried to restore to a point which i did but since this scan disk stuff started it slowed me right down to about three times slower. Sell i will try to find info to copy to this page and let you see. I don’t know if I answered your question or not. If you have more questions please ask me.

    Microsoft Windows XP
    Home Edition
    Version 2002
    Service Pack 3
    Cleron(R) D CPU 3.20GHz
    3.20Ghz, 448 Mb of Ram

  5. Joshua Dibble January 19, 2009 2:54 pm #

    Well thank you so much for the info. I really am just a newbe to this stuff. I kinda understand the what i have to do but i am sure it is harder than it looks. Do you think that if I just leave my cpu on and just go into standby that it will prolonge the inevitable? See this cpu in all cost me 300 dollars and I might be able to save the files to CD and SD cards and then transfer them onto a newer cpu.
    What stuff will I lose if the bad sector fails?
    Will this sector crash?
    How can I get an idea on how fast it will crash?
    If I don’t add or remove programs as frequentely will it slow the problems down?
    If I leave the cpu on and leave it in stand by when not using it will it be ok?
    What strains a sector the most?
    Is there anything I can avoid doing that will ease the sector?
    How can I find the bad sector?
    What, if any, programs will I have to buy/run to find the bad sector on the E;/?

    Well I understand that I would normally have to pay for the answers in one way or another so first thing first, where are you guys located? If I wanted to have you fix these problems, how much are we talking (roughly)? What would I have to send you as far as the cpu goes?

    Well again thank you so much for the reply.

Leave a Reply