asked on 9/9/2011

SQL NATIVE

in the application layer we are giving the failover db instance name.
I like to understand if we failover db to the other instance rather which was mention in the application code will SQL NATIVE CLIENT will redirect the client to failed over db which was not mentioned in the application code?

2.In the application layer if i dont mention the failover db name, and due some bad failure in the server ,server went down. And there is mirroring setup and it has been manual failover.
Now how will application will connect to failured over server?since sql native is down on old priniciple server?

Microsoft SQL Server 2008 Microsoft SQL Server Microsoft SQL Server 2005

Last Comment

Kevin Cross

9/15/2011

jogos

9/10/2011

If you have cluster with 2 nodes (servers) your db-instance can work on both. Your cluster has a name that has to be mentioned for the connections, not the name of the active node. Failover means that the instance with sitll the same name changed from the one node to the other node but beacuse the cluster and instance name don't change there is no 'connection problem'.

DB-mirroring <> failover cluster.

VIVEKANANDHAN_PERIASAMY

9/10/2011

ASKER

no i am using cluster technology here. I just having 2 servers where principle is on one server and mirrior is on another server.In this case for question above which i have posted, how it will work?

ASKER CERTIFIED SOLUTION

Kevin Cross

9/11/2011

THIS SOLUTION IS ONLY AVAILABLE TO MEMBERS.

View this solution by signing up for a free trial.

Members can start a 7-Day free trial and enjoy unlimited access to the platform.

See Pricing Options

Start Free Trial

VIVEKANANDHAN_PERIASAMY

9/15/2011

ASKER

So if I am not mentioning in appdomain fileor in application layer then application or users will not be able to redirect to appropriate sql box Am i right?
so how does SQL native client is useful?

Kevin Cross

9/15/2011

Correct, if you are not using a name that resolves to the new server once you have a failure on the production one, then the clients won't be able to automatically switch. I am not sure I understand the second question. The issue is that the Native Client connects to a DNS name or IP address for example and said entity has failed. Without a change either on client to a new entity OR binding of the old information to a new system, the Native Client can't connect. Sounds like you want a cluster whether physical or virtually managed by an appliance that directs a virtual IP address to the appropriate physical IP node based on status.