Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
I dont currently have access to a SQL server cluster for development (one using cluster services with a SAN behind it) but will do so in the future for the live environment.
When programming this from a .NET environment what should I expect to have to do in my code if the node I am connected to dies? Is the fail-over to the other node completely transparent? Do I need to catch some sort of exception and try to reconnect myself? Is there a code pattern for this?
Failover is a big fat disconnect that will cut all connection to the server. You will be able to reconnect in a few seconds to minutes usually, after the standby server has started and recovered the databases from the shared drive(s). Any transaction in-flight at the moment of the failover will be rolled back.
In your application a failover will be a lights-out event, you’ll loose all your connections and any attempt to reconnect will fail until the recovery of the standby server is complete. If you use proper transaction semantics, the data will be correct and an well written application will resume without problems. If you don’t use transactions properly, you will have all sort of inconsistencies from logical half-writes not properly guarded by a transaction boundary.
You can always test failover on a couple of virtual machines. Both Virtual Server and HyperV can simulate clusters on virtual hardware (not Virtual PC though), and I expect VMWare can as well (although I never tried it).