locked
Exchange 2010 without a traditional backup? RRS feed

  • Question

  • Has anyone actually implemented an Exchange 2010 environment using a combination of single instance recovery and a lagged DB copy in a DAG to completely replace traditional backups?  I keep reading how this could be done and the cost savings has the CTO very interested in the concept so we are looking at attempting it in a relatively small environment of 5000 mailboxes broken out into six DBs, with four DB copies spread across two AD sites, with one being lagged.  I was just wondering if anyone has actually tried this and, if so, how is that working out?
    TBrennan
    Friday, July 29, 2011 8:46 PM

Answers

  • The technology works.

    Question is more of monitoring and reaction if monitoring solution triggers event.
    I think you need to have really good monitoring solution and a clear plan on what/when/who do in different situations.

     


    lasse at humandata dot se, http://anewmessagehasarrived.blogspot.com
    • Marked as answer by emma.yoyo Thursday, August 4, 2011 1:57 AM
    Sunday, July 31, 2011 9:28 AM
  • Yep. What Lasse said. What I dont like about lagged copies is the uncertainty of technology.

    And by that I mean in a DR situation, all heck will be breaking loose. You decide to activate the lagged copy. The Exchange Admin better know all the necessary steps to bring that copy online and its a one-way trip. Once you activate the lagged databases, the other copies are invalid.

    The assumption is also that the CAS and HT servers/roles are still intact. If not, the lagged copy doesnt do you much good if clients cant connect to it.

    So that goes back to the original problem: If this a DR scenario, in what situation would only the mailbox servers be completely down and not the other Exhcange servers? Probably in the case of logical corruption. If so, then the lagged copy isnt of much use unless the Exchange Admin knows at what point the logical corruption occured and then only replays the transaction logs on the lagged copy to that point.

    If you want true DR/HA , then use DAC Mode with DAGs and not lagged copies.

    Just my opinion.

    • Marked as answer by emma.yoyo Thursday, August 4, 2011 1:57 AM
    Sunday, July 31, 2011 1:56 PM

All replies

  • Yes , have tried it.

    What do you want to know?

     


    lasse at humandata dot se, http://anewmessagehasarrived.blogspot.com
    Friday, July 29, 2011 8:53 PM
  • What do you want to know?


    Sukh
    Friday, July 29, 2011 9:41 PM
  • I would avoid the lagged copy thing personally. I dont see the value. Do what Microsoft IT has done. Have 4 copies of each database and forget about the lagged copy stuff. Circular Logging, 30 days of delete item retentenion and single item recovery and you are good to go.

     

     


    Saturday, July 30, 2011 2:12 AM
  • I was just wondering if it worked as well as it sounds.  It's great to read about the theory of changing everything you knew about backing up email; its' another for it to work in real life.  Have you, your users, and your management been satisified with results and the level of protection?

     

    It sounds as if AndyD does not think much of the lagged DB copies.  Has anyone actually tried to use one to recover a "lost" DB?

     

     


    TBrennan
    Sunday, July 31, 2011 4:28 AM
  • The technology works.

    Question is more of monitoring and reaction if monitoring solution triggers event.
    I think you need to have really good monitoring solution and a clear plan on what/when/who do in different situations.

     


    lasse at humandata dot se, http://anewmessagehasarrived.blogspot.com
    • Marked as answer by emma.yoyo Thursday, August 4, 2011 1:57 AM
    Sunday, July 31, 2011 9:28 AM
  • Yep. What Lasse said. What I dont like about lagged copies is the uncertainty of technology.

    And by that I mean in a DR situation, all heck will be breaking loose. You decide to activate the lagged copy. The Exchange Admin better know all the necessary steps to bring that copy online and its a one-way trip. Once you activate the lagged databases, the other copies are invalid.

    The assumption is also that the CAS and HT servers/roles are still intact. If not, the lagged copy doesnt do you much good if clients cant connect to it.

    So that goes back to the original problem: If this a DR scenario, in what situation would only the mailbox servers be completely down and not the other Exhcange servers? Probably in the case of logical corruption. If so, then the lagged copy isnt of much use unless the Exchange Admin knows at what point the logical corruption occured and then only replays the transaction logs on the lagged copy to that point.

    If you want true DR/HA , then use DAC Mode with DAGs and not lagged copies.

    Just my opinion.

    • Marked as answer by emma.yoyo Thursday, August 4, 2011 1:57 AM
    Sunday, July 31, 2011 1:56 PM