locked
Mirroring and the safety mode settings during an outage on the principal. RRS feed

  • Question

  • Hi Mirroring Guru's!

    Scenario: 2 Servers with 2 Instances each.
    There is mirroring in High Safety Mode between the instances on Server A and Server B.

    An outage is required on Server 1 for 12-14 hrs.

    I have a two questions relating to mirroring and the safety mode.

    Q1: If I was to take the principal server offline for a hardware related outage (12-14hrs) is it best to: 
    A.) failover the mirrored DB's from Server A Instance 1 to Server B Instance 1 during this outage? Why?
    B.) OR pause the mirroring on the principal then shut it down? Why?
    (note: the DB's on Instance 1 are not used by the business during this time)

    Q2: DB's on Server 1 Instance 2 is required to be online for continued business processing and therefore a failover to Server 2 Instance 2 is required. Hence, is it best to:
    A.) keep the DB's in high safety mode while the principal? Does it matter - Why?
    B.) OR place the DB's into high performance mode? Why?

    Thanks for your input!
    • Edited by altrstar Sunday, November 4, 2012 11:38 PM typo
    Sunday, November 4, 2012 11:37 PM

Answers

  • Hello,

    Q1: If I was to take the principal server offline for a hardware related outage (12-14hrs) is it best to:  A.) failover the mirrored DB's from Server A Instance 1 to Server B Instance 1 during this outage? Why? B.) OR pause the mirroring on the principal then shut it down? Why? (note: the DB's on Instance 1 are not used by the business during this time)

    My personal preference is to fail the mirror databases over to the secondary making it now the principal.Even though they might not be used by the business, you never know what may or may not happen during an outage if something unplanned occurs. I always like to take the safe road when it comes to these types of things. Even though it won't be used, it'll be available should the outage window be broken or extended, etc. It also makes it easy to fail back and make sure everything is running normally.

    Q2: DB's on Server 1 Instance 2 is required to be online for continued business processing and therefore a failover to Server 2 Instance 2 is required. Hence, is it best to: A.) keep the DB's in high safety mode while the principal? Does it matter - Why? B.) OR place the DB's into high performance mode? Why?

    High safety mode is synchronous with no witness, high performance is asynchronous. In this instance the secondary will not be available so this shouldn't have any impact initially... When the other server comes back up this will start to make an impact but I can't say how much, you know your systems. Depending on the system it might be worth changing but only because of the amount of work done while down.

    -Sean


    Sean Gallardy, MCC | Blog | Twitter

    Monday, November 5, 2012 12:31 AM
    Answerer

All replies

  • Hello,

    Q1: If I was to take the principal server offline for a hardware related outage (12-14hrs) is it best to:  A.) failover the mirrored DB's from Server A Instance 1 to Server B Instance 1 during this outage? Why? B.) OR pause the mirroring on the principal then shut it down? Why? (note: the DB's on Instance 1 are not used by the business during this time)

    My personal preference is to fail the mirror databases over to the secondary making it now the principal.Even though they might not be used by the business, you never know what may or may not happen during an outage if something unplanned occurs. I always like to take the safe road when it comes to these types of things. Even though it won't be used, it'll be available should the outage window be broken or extended, etc. It also makes it easy to fail back and make sure everything is running normally.

    Q2: DB's on Server 1 Instance 2 is required to be online for continued business processing and therefore a failover to Server 2 Instance 2 is required. Hence, is it best to: A.) keep the DB's in high safety mode while the principal? Does it matter - Why? B.) OR place the DB's into high performance mode? Why?

    High safety mode is synchronous with no witness, high performance is asynchronous. In this instance the secondary will not be available so this shouldn't have any impact initially... When the other server comes back up this will start to make an impact but I can't say how much, you know your systems. Depending on the system it might be worth changing but only because of the amount of work done while down.

    -Sean


    Sean Gallardy, MCC | Blog | Twitter

    Monday, November 5, 2012 12:31 AM
    Answerer
  • Thanks Sean!

    Regarding High Safety vs High Performance in the second part of your response - during the catchup phase (i.e. Server 1 Instance 2 is started) do you know how it will be applying the missing transactions? i.e., would it be in High Safety mode but apply the old transactions in High Performance mode and then once it's caught up, then it adheres to the double commit requirement?

    Cheers!

    Monday, November 5, 2012 12:40 AM
  • If Mirror is unavailable, then we run database in exposed state.
    http://msdn.microsoft.com/en-us/library/ms189284.aspx

    Balmukund Lakhani | Please mark solved if I've answered your question, vote for it as helpful to help other user's find a solution quicker
    --------------------------------------------------------------------------------
    This posting is provided "AS IS" with no warranties, and confers no rights.
    --------------------------------------------------------------------------------
    My Blog | Team Blog | @Twitter

    Monday, November 5, 2012 12:51 AM