"Rollback RX" luvs "Image For Windows" - REDUX

Discussion in 'backup, imaging & disk mgmt' started by TheRollbackFrog, Apr 26, 2015.

  1. jwcca

    jwcca Registered Member

    Joined:
    Dec 6, 2003
    Posts:
    772
    Location:
    Toronto
    The worst utility would be one (like Privazer) that has an option to "Clean free space".
     
  2. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    Actually I was just talking about a regular defrag. Sure it does take a bit longer, and sure the next incremental is larger, but there is no danger of overwriting sectors windows doesn't know about that Rollback is using. That is the issue to which I was referring.

    Pete
     
  3. pvsurfer

    pvsurfer Registered Member

    Joined:
    Sep 1, 2004
    Posts:
    1,617
    Location:
    USA
    I never heard of any such problem. Not saying it ain't so; just that I never came across it. Would you have any references?
     
  4. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    Okay a test for you. Assuming you have Rollback installed, download a trial of Perfect Disk from Raxco, install it and run a defrag. Are all your Rollback snapshots okay?
     
  5. pvsurfer

    pvsurfer Registered Member

    Joined:
    Sep 1, 2004
    Posts:
    1,617
    Location:
    USA
    No thanks, like I've said many times I don't run file defrags on a RBrx system. It sounds like you once experienced this problem with Perfect Disk and RBrx; is that the situation?
     
  6. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    HI pvsurfer

    Well you are wise not to run file defrags on an RBrx system. That was my point. It would be a problem because all the defragers run using windows API's and I am sure they wouldn't account for the non windows sectors.
     
  7. manolito

    manolito Registered Member

    Joined:
    Apr 23, 2013
    Posts:
    407
    What will get you in trouble under RBrx is a boot-time defrag. A regular defrag within Windows is under the control of the Rollback driver and will not cause any damage. It will cause a huge next snapshot, though. And it will be completely useless, because it cannot touch the "real" file system.

    Cheers
    manolito


    //EDIT//
    Wrong. The Rollback driver intercepts all calls to the Windows API for HDD access.
     
    Last edited: Apr 27, 2015
  8. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    Thansk manolito, I stand corrected.
     
  9. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,945
    Location:
    The Pond - USA
    From the FWIW Dept...

    There are basically 3-places in a Windows BOOT stream that things can happen... pre-BOOT, the Native App API, and the Windows API. An example of pre-BOOT is the DOS looking BLACK screen that BOOTmgr offers you when selecting other BOOT options. An example of the Native App API is a place where something like CheckDsk usually runs... here there is a native NT4 API of about 200+ different calls where Windows has never publicly documented more than about 25. At the Native App API, Windows (and most of its drivers) is not yet running on your system. Leaving this level starts up Windows and loads all the good stuff we're most familiar with. The main reason the Native Mode API is not well documented is that Windows would like to keep that level to itself (other then what certain 3rd party apps really need)... that way they can make changes to the undocumented parts and not worry about the effect on 3rd party apps (since it was never offered as a documented feature anyway).

    The special Rollback disk masking driver is running when you arrive at the Native App API (as well as some required device drivers <disk, some basic hardware interfaces, etc.>). That's why CheckDsk will not wreck a RBrx system if running under the protected volume (it will complete destroy your system if any anomalies are detected while running CheckDsk from OUTSIDE of the protected system <WinPE/RE, etc.>). Also, all BOOT time defrags that I'm aware of (I don't know them all) run either at the Native Mode API or under Windows itself... they should not break or destroy a RBrx protected system. If the BOOT time defrag is designed to run at the pre-BOOT level, say goodbye to your RBrx system... the required RBrx driver is NOT THERE at that time.

    The reason Pete had trouble with the "Perfect Defrag" BOOT time process and AX64 was that the BOOT time defrag ran at the Native Mode API level and AX64's tracking file runs under full Windows... a recipe for disaster.
     
    Last edited: Apr 28, 2015
  10. Peter2150

    Peter2150 Global Moderator

    Joined:
    Sep 20, 2003
    Posts:
    20,590
    There still is a problem there with PD and AX64, but it is manageable if one just realizes there are a few simple rules. I was able to restore across Boot time defrags, but just found the next image was going to be a full once. The imagers that use tracking files are not the ideal solution.

    Pete
     
  11. pvsurfer

    pvsurfer Registered Member

    Joined:
    Sep 1, 2004
    Posts:
    1,617
    Location:
    USA
    I say again (to RBrx users) for the umpteenth time, do not use any file defragger (or other system utilities) while RBrx is installed.

    Fact: When in a RBrx environment, file defragging will make the subsequent RBrx snapshot extraordinarily large.
    Fact: When in a RBrx environment, the use of a file defragger serves no (constructive) purpose whatsoever

    I recommend uninstalling RBrx (to your current snapshot) periodically, then run chkdsk/f, run a safe file cleaning tool (eg.,CCleaner), now run your file defragger, make an image-backup of your system disk/partition, then reinstall RBrx.

    pv
     
    Last edited: Apr 28, 2015
  12. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,945
    Location:
    The Pond - USA
    A quick report on the project...

    Using our "specially" installed IMAGE FOR WINDOWS, we (Thanks, Brian!) are now doing successful HOT, DIFFERENTIAL imaging of a Rollback RX protected system incl. ALL SNAPSHOTs. Although the image time is extensive (almost a FULL due to a complete comparison of the partition surface called for in the "All Sector" mode), the size is exactly as expected... minimal size. I will be testing INCREMENTAL imaging as well, and based on the above success for DIFFERENTIAL imaging, I expect the same for the INCREMENTAL.
     
  13. pvsurfer

    pvsurfer Registered Member

    Joined:
    Sep 1, 2004
    Posts:
    1,617
    Location:
    USA
    Froggie, I realize the above is being accomplished by using IFW 2.95 (with PhyLock). A while back you reported that your IFW hot imaging method no longer worked for RBrx 10.x, so I was wondering if any version of IFW prior to 2.95 can also be used successfully to make hot-raw images (full/diff) in an RBrx environment?
     
    Last edited: Apr 28, 2015
  14. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,945
    Location:
    The Pond - USA
    I probably won't be able to test them all but if you have any specific version you're interested in, I can make a quick check.

    And remember... anyone licensed to a v2 version at any level can use v2.95 for FREE...
     
  15. Brian K

    Brian K Imaging Specialist

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

    I can't test when IFW warm imaging and Rollback Rx became compatible as my test computer died. I had restored an image of Win10, installed Rollback Rx and clicked Restart. The computer didn't POST and gave 5 short beeps which according to the code book is a Processor error. Strange! I tried but I couldn't fix the computer. As the test computer had a LGA1155 socket I've ordered a LGA1150 Motherboard and processor. My test computer will be better than before.
     
  16. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,945
    Location:
    The Pond - USA
    Yea... see, Rollback eats your hardware, causes your processors to go up in smoke, and makes your mainboards melt.

    You really have to stay away from that application... :argh: (Yea... I'm just kiddin')
     
    Last edited: Apr 29, 2015
  17. pvsurfer

    pvsurfer Registered Member

    Joined:
    Sep 1, 2004
    Posts:
    1,617
    Location:
    USA
    @Brian, sorry to hear that and I hope that your unfortunate incident doesn't precipitate any more HDS 'hate mail'.

    @Froggie, please add 'you're just kiddin' to the above, as some around here are sure to take it (as is) and run with it.

    I don't have an IFW license. I'm just trying to understand what happened to turn-around your prior report that the IFW hot-raw method no longer worked with v10.x?
     
    Last edited: Apr 28, 2015
  18. pandlouk

    pandlouk Registered Member

    Joined:
    Jul 15, 2007
    Posts:
    2,976
    LoL:argh:
    Brian, sorry for the pc (never in my life had a processor go bad).
    ps1. You were determined to take RollbackRX restore capabilities to another level that even HDS hasn't dared to!!!:argh:
    ps2. submit a ticket support and then send it to Kurtis... Who knows... Maybe you'll be able to rollback the problem...:argh:

    Panagiotis
     
  19. pandlouk

    pandlouk Registered Member

    Joined:
    Jul 15, 2007
    Posts:
    2,976
    All versions from 2.70 upwards should work.(they all use version 8 of Phylock driver).

    Panagiotis
     
  20. pvsurfer

    pvsurfer Registered Member

    Joined:
    Sep 1, 2004
    Posts:
    1,617
    Location:
    USA
    Panagiotis, that doesn't help my confusion as to what changed? A while back, Froggie's hot-raw backups of RBrx v10.x failed (I believe he was using IFW 2.77 at the time). Now it's working (with IFW 2.95). :confused:
     
  21. Brian K

    Brian K Imaging Specialist

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

    Same here.

    But I didn't want to take the chance and buy a LGA1155 processor and find it was the motherboard that had failed.
     
  22. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,945
    Location:
    The Pond - USA
    PV, without going all the way back to v2.6x I don't have documented the exact point at which it started to fail... what I've spoken of was from a distant memory. I also have a slight recollection of a failure while still using v9.x so the memory may not be so good.

    When I saw the failure it was as a result of another user reporting the failure... I retested at that point and also saw the failure. I then started to warn users about the documented workaround (and put the WARNING in the document Forum post) and stopped using IFW for that purpose.
     
  23. TheRollbackFrog

    TheRollbackFrog Imaging Specialist

    Joined:
    Mar 1, 2011
    Posts:
    4,945
    Location:
    The Pond - USA
    I thought that's what the ROTFLMAO emoji was for... :confused:
     
  24. pandlouk

    pandlouk Registered Member

    Joined:
    Jul 15, 2007
    Posts:
    2,976
    I doubt that it had to do with a change in IFW. More probably a change in RollbackRX's drivers.

    Panagiotis
     
  25. Brian K

    Brian K Imaging Specialist

    Joined:
    Jan 28, 2005
    Posts:
    12,113
    Location:
    NSW, Australia
    Try the one month trial. It is fully functional.
     
  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.