Replication for redundancy

  • I am trying to setup replication for following scenario.

    We have four SQL server for example A,B,C and D.

    A and B are on the same network collecting online data. Both machine are collecting same data. I need to replicate this data to C and D which are also on same network but different network than A and B. I need to design transactional push replication. I need design ideas from you guys so If server A goes down, server B can replicate to C and D. similarly if server B goes down, server A can replicate to C and D.

    Thanks.

  • Sorry, your setup is unclear to me.

    A and B are on the same network collecting online data. Both machine are collecting same data.

    By "Both machine are collecting same data." do you mean they are replicated to each other? Are they in sync with each other?

    I need to replicate this data to C and D which are also on same network but different network than A and B.

    If A and B are on a different network than C and D, then how will you get data across from A to C and D, or B to C and D in a disaster recovery scenario?

    There are no special teachers of virtue, because virtue is taught by the whole community.
    --Plato

  • No, A and B servers are not replicating. They are both connected to same interface.

    Firewall setup allows communication between A,B and C,D server.

  • OK, from your original post:

    If server A goes down, server B can replicate to C and D. similarly if server B goes down, server A can replicate to C and D.

    Are you saying you want all databases on A to be available on both C and D? And in addition, are you saying you also want all databases on B to be available on both C and D as well?

    It may help if you write out the use-case of what would happen if server A went down, but all other stayed online, where would your clients connect to reach databases that were previously hosted on server A? Same questions for server B.

    There are no special teachers of virtue, because virtue is taught by the whole community.
    --Plato

  • Yes, I want all databases to replicate from A to C And D, Same with B server.

    We are trying to provide data for clients which are also connected to both server. Applications need to point to different server. In case A goes down Apps need to be connected to B, Which I believe happens automatically through Apps.

    I am replicating data from A to C and D, working fine. (I have setup push replication. Distributer and publisher are on same local machine.)

    Now I need to setup replication on B incase A goes down.

    Thanks...

  • As long as database names on A are not overlapping with any database names on B then you can use continuous Transaction Replication to achieve your goal of having all databases on A and B on both C and D.

    To do this you would make C and D subscribers of all databases on A. And similarly, also make C and D subscribers of all databases on B.

    In your application that uses databases on A, make either C or D the secondary option when A goes down. And similarly, in your application that uses databases on B, make either C or D the secondary option when B goes down.

    There are no special teachers of virtue, because virtue is taught by the whole community.
    --Plato

  • Thanks for your help. Database name are same in A and B both server. I will try to change it at subsciber level if that works.

  • 2cool222 (3/26/2012)


    Thanks for your help. Database name are same in A and B both server. I will try to change it at subsciber level if that works.

    Anytime, happy to assist.

    You will have the option to set it up to replicate A and B into different database names in C and D.

    There are no special teachers of virtue, because virtue is taught by the whole community.
    --Plato

Viewing 8 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic. Login to reply