Results 1 to 6 of 6

Thread: What do you do?

  1. #1
    Senior Member GregE's Avatar
    Join Date
    May 2006
    Posts
    2,100
    Thanks
    9
    Thanked 17 Times in 16 Posts

    Default What do you do?

    Many of you have done DR testing at many levels and we are nearing our first DR test.

    My manager asked me a question I had thought about, but really could only answer with......"we just won't have that data."

    The question is, what do you do if you get your tapes to the DR site and discover that a necessary file (or many files) are missing? Bad tape, broken tape, etc.

    What we don't want is for something crucial to kill our expensive and time consuming DR test. But my answer to my manager is what it is, short of having a primary volume rushed to the site.

    What do you guys think about such a situation?

  2. #2
    Senior Member
    Join Date
    Jul 2008
    Location
    Dallas, Texas
    Posts
    207
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Default

    We usually have our primary site send the tape we need to complete the DR. In the real situation, the primary site is destroyed, I would explain the issue and that we were not able to recover the data.

  3. #3
    Moderator mikeymac's Avatar
    Join Date
    Jun 2003
    Location
    Syracuse, NY
    Posts
    896
    Thanks
    9
    Thanked 12 Times in 12 Posts

    Default

    Yup. It's all about probability. In a real DR situation, assuming one offsite copy of each file, there is a very real chance that you will experience some data loss. If this is an unacceptable risk to your company, then they will have to create redundancy by generating a second set of offsite tapes, or by backing up stgpools to offsite disk.

    Another option is to identify mission-critical data (which is probably a relatively small portion of your total backup load), and provide it with VIP treatment. Offsite a second copy of it each day. Reduce those odds.

  4. #4
    Moderator chad_small's Avatar
    Join Date
    Dec 2002
    Location
    Gilbert, AZ
    Posts
    2,191
    Thanks
    1
    Thanked 25 Times in 24 Posts

    Default

    If its business critical to the point of do or die then the company should establish a remote mirror of that data. Otherwise you would probably roll the data back to a previous version. Just because one "version" was damaged doesn't mean you couldn't reconstruct the server from previous versions. Not perfect or wanted, but if it gets you back up and running you take what you can get.
    Chad Small
    IBM Certified Deployment Professional
    chadsmal@gmail.com
    http://www.tsmadmin.com

  5. #5
    Member Mikey D's Avatar
    Join Date
    Mar 2007
    Location
    Southeast Michigan
    Posts
    259
    Thanks
    0
    Thanked 4 Times in 4 Posts

    Default

    Sometimes when we've encountered that we just restore back to a day before that damaged file. Depending on the level of media failure that sometimes can get things back up but like everyone else mentioned. It's a very real possibility and has happened to use more than once that a tape gets damaged in shipping or what not.

  6. #6
    Senior Member GregE's Avatar
    Join Date
    May 2006
    Posts
    2,100
    Thanks
    9
    Thanked 17 Times in 16 Posts

    Default

    Thanks everyone.

    It's always a possibility of course, and when the other side of the house (iSeries guys running BRMS backups) are taking two different days of backups, a primary recovery point, and secondary recovery point in case the primary is fouled up, they are in that mindset of full+incremental and a tape-centric backup system, where TSM is neither. Hard to explain to someone that we will NOT be taking two "sets of TSM tapes, 10 days apart" because that's not how TSM works. We'll take several days of TSM dbbackup, but that's different.
    Last edited by GregE; 02-09-2009 at 01:29 PM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •