Image For Windows

Discussion in 'backup, imaging & disk mgmt' started by Stigg, Apr 15, 2014.

  1. Brian K

    Brian K Imaging Specialist

    Joined:
    Jan 28, 2005
    Posts:
    12,113
    Location:
    NSW, Australia
    Restoring an Entire Drive Backup with AUTO First Track Sectors restores the boot code, partition table and Disk Signature. The partition that was Active at the time of backup will be Active after the restore. With a GPT disk, as with a MBR disk, it restores the appropriate GUID sectors in the First track.

    With a partition image backup you would restore the SRP first with these options...

    Set Active
    Restore First Track
    Log Results to File

    Then you would restore the OS partition with these options...

    Update Boot Partition
    Log Results to File

    Validates are optional.

    Edit... The Disk Signature is restored with "Restore First Track" with both Entire Drive and Partition restores.

    Edit... If you were doing the restore (not to an empty HD) for an OS issue you only need to restore the OS image with these options...

    Update Boot Partition
    Log Results to File
     
    Last edited: Jun 25, 2015
  2. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    Brian
    a question , image for windows has Validate or VALIDATE byte for byte option
    what do exactly do these 2 options?
    i mean if i restore an image with Write Changed Sectors to a drive (that doesn't support) DRAT , validate or validate byte for byte should compare the restore image data with the data written ?
    so it should tell me if the images was restore correctly right?
     
  3. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,944
    Location:
    The Pond - USA
    There are 3-types of TRIM in use by SSDs...
    • Non-deterministic Trim: each read command to the Logical block address (LBA) after a Trim may return different data.
    • Deterministic Trim (DRAT): all read commands to the LBA after a Trim shall return the same data, or become determinate.
    • Deterministic Read Zero after Trim (DZAT): all read commands to the LBA after a Trim shall return zero.
    You can use the TRIMcheck tool and give each a quick check. The drives that support DRAT (or DZAT) will always TRIM to the same value, the non-Deterministic drives will TRIM to varying values, different from the original DATA but usually non-ZERO and non-ALL ONEs. Drives that do not support TRIM or an OS that does not have TRIM enabled will always return TRIMcheck's DATA value only. When using TRIMcheck you need to give your OS/SSD a chance to do its thing... it may take 10s of seconds to perform this operation depending on the OS' I/O load and the SSD's controller design

    If attempting TRIMcheck under a Rollback RX-enabled OS (which will be TRIM disabled), the result will be false and look like a non-Deterministic TRIMming SSD most of the time due to Rollback's obfuscation of your System's Logical Blocks.
     
    Last edited: Jun 25, 2015
  4. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    hi TheRollbackFrog
    i tried trimcheck but it tell me only the trim appears to be working
    thanks
     
  5. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,944
    Location:
    The Pond - USA
    Mantra, if it says it's working, your SSD is either DRAT or DZAT (the TRIM data following the operation will always be the same but not the same as the original). When you run TRIMcheck with a non-Deterministic SSD, TRIMcheck will tell you it cannot determine the state of TRIM due to the different data its receiving back which is also different than the original data written.

    It sounds like all your SSDs are DRAT or DZAT which should be fine for your IFW "changed sectors" scenario.
     
  6. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    hi
    today i restored an image to a friend's pc with ifw "changed sectors" , it boots but with mft errors ,reboot i restore a full image without changed secotors and it work perfectly
    have 't tried trimcheck on my pc's friends
     
  7. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    Brian , i guess i have restored many times entire drive without automatic boot partition update
    what could be caused in the worst scenario restoring an entire drive image without automatic boot partition update?
     
  8. Brian K

    Brian K Imaging Specialist

    Joined:
    Jan 28, 2005
    Posts:
    12,113
    Location:
    NSW, Australia
    In the restore situation, Validate checks the image before the restore and Validate Byte-for-Byte checks the image after the restore. But they are different....

    http://www.terabyteunlimited.com/kb/article.php?id=350

    In the situation of restoring your images to an empty HD you usually restore them to the same LBA offset as the original HD. Update Boot Partition isn't really necessary unless you are using the Scaling options. In my GPT disk tests I've been restoring images to different LBA offsets from the original and often in reverse order from the original. Update Boot Partition is needed otherwise the OS doesn't boot. BCD issues. So I use Update Boot Partition with each partition restore other than the ESP. The Boot Partition (SRP or ESP) needs to know where the other partitions are situated.

    When you have Automatic Boot Partition Update selected in Settings, the Update Boot Partition option doesn't appear when you are doing an Entire Drive restore. But it is present when you are doing a Partition Restore.
     
  9. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    Hi TheRollbackFrog
    may i ask you which ssd do you use?
    is there another utility to know if the ssd support drat , dzat or Non-deterministic Trim ?
    thanks
     
  10. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,944
    Location:
    The Pond - USA
    There are three SSDs I've been mostly using... OCZ Agility 3, OCZ ARC-100 and a Transcend model (can't remember the model #). The OCZ devices (Agility and ARC-100) all demonstrate the DZAT style, while the Trancend shows up to be a non-Deterministic type. Most likely this is determined by the SSD Controller chipset in use in the devices.

    When the OS issues an "ATA Identify Device" command during its discovery process, two words in the returned data block (word #69 and word #169) contain the information letting the OS know what type of TRIMming device it is (DRAT, DZAT or "other"). I know of no current utility application that reports that particular status to the user... but there may be one out there somewhere.
     
  11. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    hi TheRollbackFrog
    do you find out that
    &
    using the tool trimcheck?
     
  12. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,944
    Location:
    The Pond - USA
    Yes. When TRIMcheck-ing the OCZ devices, a quick re-check after the pattern is set returned the original DATA saying that TRIM wasn't active yet. About 10-12 sec. later a re-check showed the pattern DATA to be ZEROs. With the Trancend, the same execution showed the original DATA after about 3-sec then a re-check later showed a completely different data pattern than the original and each consecutive re-check show yet another different DATA pattern. When TRIMcheck sees this it tells you that it cannot determine the status of TRIM, it does not say that TRIM is inactive.

    I also ran this on another (don't remember the brand) SSD and it always returned an ALL ONEs pattern after TRIM kicked in (very common also)... clearly a DRAT product.
     
  13. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    thanks
    might you pleae upload a screenshot to understand if the drive support drat or not please?
     
  14. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,944
    Location:
    The Pond - USA
    If you're familiar with TRIMcheck you really don't need a screenshot. When you run TRIMcheck, it sets up its DATA pattern, locates all the file pieces (location on disk) it just created, picks the piece it will do its checking with, DELETEs the file and asks you to EXIT. You then run it again quickly, and most likely it will show you its original DATA pattern it's checking for and tell you TRIM is not active and ask you to check again in a few seconds. If TRIM is operational, eventually it will show a different data pattern (usually ZEROs with DZAT and ALL ONEs with DRAT <but it doesn't have to be>. If it sees the same "different" data pattern (but not ZEROs) it assumes DRAT and proclaims TRIM is working... why, 'cause the pattern is always the same but not its TEST DATA. If it keeps seeing different DATA patterns but not its original TEST DATA, it assumes non-deterministic and tells you it can't determine the status of TRIM.

    There's an outside chance, if you're running a very busy DATA generating system of some sort, TRIMcheck will not be able to determine the status of TRIM due to the fact that the OS may be re-using TRIMcheck's TEST DATA block quickly after it releases it and before it can really check for TRIM. This would be a very unusual situation based on the way Windows allocates additional needed storage.
     
  15. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    hi thanks
    but how many times should i run trimcheck? :(
     
  16. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,944
    Location:
    The Pond - USA
    When TRIMcheck runs the 1st time and sets up its check for DATA retention or TRIMming, it leaves a special AUDIT file (.JSON) in the same place it runs in. That audit file will remain until TRIMcheck decides TRIM is active, then it deletes it so that the next time it runs it'll do yet another DATA write/read/delete followed by a check.

    From the 2nd run on it will always check its old data (via the .JSON file) to see if TRIM has been applied to its DATA. Once it detects TRIM active, that audit file will go away. Based on that, TRIM will continue to do checks, not re-writes (each time it runs) until it sees TRIM active. It will continue to do the check if it sees its original TEST DATA (and tells you TRIM is not active), OR if it sees continuously different DATA than its CHECK DATA at which point it'll tell you it cannot determine TRIM's status. This is the condition caused by non-Deterministic TRIM data or a fast re-use by the OS of the data area.
     
  17. puff-m-d

    puff-m-d Registered Member

    Joined:
    Feb 13, 2002
    Posts:
    5,703
    Location:
    North Carolina, USA
    Hello,

    An update was released today to version 2.97a:
    Homepage: http://www.terabyteunlimited.com/index.htm
    Downloads: http://www.terabyteunlimited.com/downloads-image-for-windows.htm
    Upgrade History: http://www.terabyteunlimited.com/upgradehist-image-for-windows.htm
     
  18. Hadron

    Hadron Registered Member

    Joined:
    Apr 1, 2014
    Posts:
    2,137
    I wonder why it had such a minor version increment.
     
  19. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    thanks
    because Write Changed Sectors Only could be a serius problem for some drives
    but i don't know how the program can know if the drive support it or not
     
  20. Brian K

    Brian K Imaging Specialist

    Joined:
    Jan 28, 2005
    Posts:
    12,113
    Location:
    NSW, Australia
    mantra,

    Does IFW ver 297a detect that your SSD doesn't support Write Changed Sectors Only?
     
  21. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    i don't know , i haven't created the winpe iso
    does your 297a detect your ssd?
    is the feature "write only..." enbled ?
     
    Last edited: Jul 5, 2015
  22. Brian K

    Brian K Imaging Specialist

    Joined:
    Jan 28, 2005
    Posts:
    12,113
    Location:
    NSW, Australia
    Yes, but all versions detected my SSD. My SSDs are Intel 520.

    The Options list has been reordered in 297a. Write Changed Sectors Only is now towards the top.
     
  23. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    hi
    i mean , the program should check the drive and disable Write Changed Sectors Only if it doesn't support drat
     
  24. Brian K

    Brian K Imaging Specialist

    Joined:
    Jan 28, 2005
    Posts:
    12,113
    Location:
    NSW, Australia
    That's my understanding too. If the drive is OK then Write Changed Sectors Only will be present.
     
  25. mantra

    mantra Registered Member

    Joined:
    Jan 25, 2005
    Posts:
    6,167
    hi
    i hope the winpe will check it everytime it boot
    because if you install and create winpe with a drive that doesn't support drat , you could use the feature "write only" even with drives that support it
     
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.