Restore "Raid created Image" to IDE (SATA)

Discussion in 'Acronis True Image Product Line' started by ieichens, Oct 16, 2005.

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

    ieichens Registered Member

    Joined:
    Oct 16, 2005
    Posts:
    5
    I'm using an Intel SRCS14L Raid-Controller with one Raid-5 Array.
    I'm able to create an image unsing TI9.2289.
    I would like to use this image on a backup PC without this
    Raid-Controller in case of hardware failure.

    What i tested was: Removed Raid-Controller and activated
    on-board SATA-Promise controller. I've choosen IDE-Mode for this
    controller and plugged in a brand new harddisk.

    I was able to restore the image to an attached disk using
    the recovery CD.

    System booted and prompted me te select a boot mode (safe, normal etc.)
    I guess that is because i created the image from a running XP (right?).
    After any selection the System reboots within less than a second.
    The imaged XP never saw the on-board controller, sinced it
    was deactivated in bios (may this bee a problem).

    Should it bee possible to restore an image using a different controller ?

    Ingo
     
  2. mareke

    mareke Registered Member

    Joined:
    Oct 19, 2004
    Posts:
    200
    Location:
    Sydney Australia
    The problem is that the image contains the wrong drivers for the Sata hard disks and to install the drivers for the SATA-Promise controller you would most likely have to reinstall Windows choosing the repair option after the image has been restored by Acronis. As long as you are given the repair option as a choice this should work but you would lose some of the Windows updates and some programs may have to be reinstalled.
     
  3. noonie

    noonie Registered Member

    Joined:
    Jun 7, 2004
    Posts:
    299
    The purpose of image software is to reimage to the same hardware, specifically hard drive failure.
    It is not meant to become a cheap deployment software package. These can cost thousands of dollars. There are workarounds, one such as mareke suggested.
    Another would be to prep the drives with ms sys prep first/ then image.
    Another would be to have and consider the raid controller part of the image backup (keep a spare), if you are more concerned with the rest of your system failing aside from the raid5 array.
     
  4. ieichens

    ieichens Registered Member

    Joined:
    Oct 16, 2005
    Posts:
    5
    The aim is not to do software-deployment. Two weeks ago my mainboard failed and my hardware-vendor needed a week to repair that.
    I had good luck and was able to use my notebook.
    But now i'm wondering what would be the minimal backup-pc,
    based on the assumption that only the image is available
    in the worst case. I would like to be able to continue work
    within let's say 2 hours.

    I think imaging is here a solution, because you do not have to
    keep the backup-pc in sync which is sometimes not possible
    due to activation methods. It takes for example up to a day
    to get InstallShield running after a new installation,
    just because activation is lost with a failing system and you
    have to phone around to get your legally purchased software
    working again. (By the way: this is not acceptable in my opinion).

    Because the intel raid-controller is the most expensive part and
    in the meanwhile discontinued i was wondering how to replace it.

    Ingo
     
  5. noonie

    noonie Registered Member

    Joined:
    Jun 7, 2004
    Posts:
    299
    Software-deployment is just the term used for transferring a working os to new hardware, whether to 1 box or 100. Sorry, didn't mean to imply anything else.
    Nonetheless I got the distinct impression you were having hardware problems other than hdrives. They can be devastating.
    Other than the options I previously mentioned, the simplest backup would be a pc with the identical mb as the one you are using. the os only cares about the bios, chipset and controllers, so your drives would still work.
    If you went to raid 1 or 10 the drives can be read separately on an ide channel and put on a different board with the methods above.
    There are many very expensive hardware methods for redundancy, but being in this forum, I doubt you would want to consider them.
    Not knowing your hardware, I would suggest using server quality mb's such as Tyan or Supermicro with raid and just keeping a spare mb. That would be relatively cheap and in your time frame. You can get a Tyan with raid etc for around $150.00. Probably cheaper than your separate raid card. Activation probably would not be an issue.
    I'm typing this on a Tyan and I have customers still using boards with 286's in them. That's 10 years. Frankly, I have never built a Tyan that went bad.
    Good mb, good memory, good psu, good ups, adequate cooling>reliable system.
    Hope this helps.
     
  6. ieichens

    ieichens Registered Member

    Joined:
    Oct 16, 2005
    Posts:
    5
    Thanks for your comments. You're right i'm not only thinking of hdrives,
    my fear is that i might end with completley intact drives but cannot
    access them due to other hardware damage, especially the raid-controller.

    What about the data if i cannot get an identical controller.
    Currently i'm doing some backups on a second (non-raid) drive on the
    same controller, seems to be no good idea.

    In IDE times things where simple. I always could plug my harddrive
    into another machine and read the data, even boot if the hardware
    was similar enough.

    Now i see that don't understand enough of raid-controllers to develop
    a suitable backup-strategy. My hope was to become a little bit
    controller-independet making images.
     
  7. noonie

    noonie Registered Member

    Joined:
    Jun 7, 2004
    Posts:
    299
    TI is really not meant to be a data backup solution. It is supposed to be a imager that will restore the os to an imaged state easily, without having to install the os, progs etc to a new hdrive. It happens to bring alll data with it, so some people have wrongly decided to use it for a data backup method.
    The same holds true for raid 1,5,10, it is also not meant for backup, but as an easy way to replace a hdrive that fails without having to reinstall everything.
    For your data, backup software is still the most reliable, and the best softwares are the most expensive. Veritas makes professional solutuions.
    Duplicate all your hardware.
    All 4 methods should be used in conjunction, if you want to recover to the point of failure in minimal time.

    The most secure is to have an ti image, a mirrored raid hdisk, data only backups on removable media, and a complete duplicate set of hardware (whether it be in a completely assembled computer or just parts) if you want to recover in the least amount of time.
    If you don't do all of this for whatever reason, you will not be up and running in 2 hrs in a worst case scenario, simply because you cannot predict what will go wrong.
    eg. You may just simply get a nasty trojan that will trash your raid, trash you image because it doesn't have the latest data and without a regular software data backup, you won't be at your crash point in 2 hrs.
    If that ever happens, believe me, a $300.00 raid card or mb will seem like a drop in a big bucket.
    I just had a customer that didn't believe in a totally redundant system, lost his server in a small sign shop wth 7 boxes, and not only spent 2 weeks getting part of the data back, but also lost future jobs etc. The results can be far reaching.
    If you want to go with a separate raid controller, Adaptec and Promise are good.
    It all depends on what your data and even more so what your time is worth.
    Only you can decide that.
     
Thread Status:
Not open for further replies.