TI-8: what are incrementals based on?

Discussion in 'Acronis True Image Product Line' started by RikL, Mar 21, 2006.

Thread Status:
Not open for further replies.
  1. RikL

    RikL Registered Member

    Joined:
    Jan 26, 2005
    Posts:
    8
    Using TI-8, suppose I'm paranoid and I want to keep two completely separate sets of backups on different network devices.

    I want to do a full backup to A, then a full backup to B, then incremental to A, incremental to B, incremental to A, to B, to A, etc.

    Does this get me two separate sets of valid backups?

    Another way of asking the question:

    Is incremental backup determined solely by the current state of the disk being backed up, compared to what is on the target backup sequence,
    --- OR ---
    is incremental backup determined by some sort of flags maintained on the disk being backed up?

    If it's the former, then I would expect the A/B sequence to work fine; if it's the latter then my backups would be splattered over A and B combined.

    The disk being backed up is NTFS, if that matters.

    Please advise.

    Thanks,
    --Rik
     
  2. bVolk

    bVolk Registered Member

    Joined:
    Dec 22, 2005
    Posts:
    954
    In my opinion the answer is yes, it does.

    But I think a greater level of security would be achieved by creating one set only and copying the image files onto DVDs or on an external HD. See, I'm paranoid too. :D
     
  3. RikL

    RikL Registered Member

    Joined:
    Jan 26, 2005
    Posts:
    8
    Hhmm... bVolk, thanks for the reply, but I was hoping for something stronger than opinion. If the answer is definitely yes, then I can save a substantial amount of separate checking to make sure that my second copies really are, uh, second copies and not just unreadable trash.

    Can somebody at Acronis who knows the data structures and algorithms please tell me if interleaving two sets of backups does or does not generate separate independent sets?

    Thanks very much,
    --Rik
     
  4. paultwang

    paultwang Registered Member

    Joined:
    Mar 13, 2006
    Posts:
    36
    Incremental backup is based on the current state of your source drive and all existing backups in the same series (1 full + many increments) in the same directory. Nothing is written to the source drive to tag stuff.
     
  5. Acronis Support

    Acronis Support Acronis Support Staff

    Joined:
    Apr 28, 2004
    Posts:
    25,885
    Hello RikL,

    Thank you for choosing Acronis Disk Backup Software.

    As paultwang has already pointed out, each new incremental image is based on the current state of the source hard drive compared to the already existing set of images, i.e. the initial full plus a number of incremental images, so the answer is definitely yes. This will give you two separate sets of images.

    You can find more information on how to use Acronis True Image in the respective User's Guide.

    If you have any further questions concerning Acronis software, please feel free to submit a request for technical support or post any of them on this forum. We will certainly try to help you in resolving any issues.

    Thank you.
    --
    Alexey Popov
     
  6. RikL

    RikL Registered Member

    Joined:
    Jan 26, 2005
    Posts:
    8
    Excellent -- now a quick followup question...

    Background... Once in while, something fails in a way that leaves me wondering whether my last incremental really did finish correctly.

    I suspect, from the discussion above, that it's OK to simply delete the last incremental, or even the last several, and then move forward with more incrementals, using the truncated set of backups as a base.

    Is that correct?

    Thanks very much,
    --Rik
     
  7. RikL

    RikL Registered Member

    Joined:
    Jan 26, 2005
    Posts:
    8
    And a follow-up to my follow-up...

    If it's OK to truncate the backup set and then move forward, this allows an interesting scheme for more efficient backups under some common circumstances.

    The scheme is to do a full backup, then a series of incrementals, call them A1, A2, etc. When the A series consumes too much space, move them off to a holding directory, leaving behind just the full. Then start another series of incrementals, call them B1, B2, etc. B1 actually acts more like a differential since it captures in one place all of the differences between the full backup and the current state. Subsequent B's act like ordinary incrementals. After B1 has been generated and verified, all of the A's can be deleted because they are redundant.

    Is there some flaw to this scheme? Is it described someplace that I have overlooked?

    Thanks,
    --Rik
     
  8. paultwang

    paultwang Registered Member

    Joined:
    Mar 13, 2006
    Posts:
    36
    TI does not know that those incrementals ever existed if you chose to delete them, as long as you delete from the tail.
     
  9. Menorcaman

    Menorcaman Retired Moderator

    Joined:
    Aug 19, 2004
    Posts:
    4,661
    Location:
    Menorca (Balearic Islands) Spain
    Hello Rick,

    No flaw. It'll work as you envisage.

    Regards
     
  10. RikL

    RikL Registered Member

    Joined:
    Jan 26, 2005
    Posts:
    8
    Menorcaman and paultwang, thanks very much for the confirmation. This should save me a lot of time & trouble.
    --Rik
     
Thread Status:
Not open for further replies.