SQL Server Performance

Desing Applications to manage Failover.

Discussion in 'SQL Server 2005 Clustering' started by tomasnelson, May 22, 2007.

  1. tomasnelson New Member

    HI.....
    Please i need some help... I hope someone knows which is the syntax necessary to develop an applications that have to connect to instance off SQL Server, and can handle the state of failover and they are reconecten automatically to the instance, with no need to restart the application again.
    Some time ago i read a link, but i don't find yet again.
    Please if someone knows or find the link letme know please.
  2. MohammedU New Member

    In failover environment SQL server name will not be chaged when it failover from one node another because it uses VIRTUAL servername which is different than the physical/machine name...

    http://msdn2.microsoft.com/en-us/library/ms143182.aspx

    MohammedU.
    Moderator
    SQL-Server-Performance.com

    All postings are provided “AS IS” with no warranties for accuracy.
  3. tomasnelson New Member

    I read a document some time ago where it said that the application had to handle the connection when this detected a failover state. So that the application does not report an error.

    tomás nelson
  4. satya Moderator

    Fyihttp://www.microsoft.com/technet/technetmag/issues/2007/03/HighAvailability/ too for more information on high availability.

    Satya SKJ
    Microsoft SQL Server MVP
    Writer, Contributing Editor & Moderator
    http://www.SQL-Server-Performance.Com
    This posting is provided AS IS with no rights for the sake of knowledge sharing. Knowledge is of two kinds. We know a subject ourselves or we know where we can find information on it.
  5. Argyle New Member

    What the application need to handle is to reconnect after a failover. The SQL Service is stopped and then started during a failover so the application need to handle the reconnect if it had a constant connection to the SQL Server (like some VB applications for example). Data might also be rolled backed if there where any open transactions. Those need to be recommitted.

    For a state less application like a web site it will often work by just refreshing the specific web page. It depends on the application though.
  6. tomasnelson New Member

    Thanks a lot SatYa, exist a similiar code for sql server 2000 in a cluster enviroment? Argyle do you know where can i find an example for your comment?

    tomás nelson
  7. satya Moderator

    Problem with SQL 2000 you have to go for Clustering only, as log shipping is not a complete high availability solution. Another linkhttp://www.microsoft.com/technet/prodtechnol/sql/2000/deploy/harag05.mspx about HA series.

    Satya SKJ
    Microsoft SQL Server MVP
    Writer, Contributing Editor & Moderator
    http://www.SQL-Server-Performance.Com
    This posting is provided AS IS with no rights for the sake of knowledge sharing. Knowledge is of two kinds. We know a subject ourselves or we know where we can find information on it.

Share This Page