locked
SQL Cluster date integrity RRS feed

  • Question

  • Hi there,

    We are evaluating database solutions for our upcoming eCommerce project, and trying to finalize if we use MS SQL cluster or MySQL. Besides many aspects to be considered, one of elements has been argued is how to ensure unique user ids across the system. Does an MS SQL cluster automatically support the data integrity, which can generate and manage unique ids across the entire system or we have to come up with a mechanism ourselves to generate the user ids and allocate them to each server to ensure the uniqueness? Could you please explain it in details?

    Thursday, August 6, 2020 6:27 AM

All replies

  •  is how to ensure unique user ids across the system

    Not clear, what for user ids you mean here? SQL Logins or user data? Anyway, in a Failover cluster the database is one database which ist hosted in one node, and on a failover the same database is hosted on the other node; it's always exact the same database.

    Olaf Helper

    [ Blog] [ Xing] [ MVP]

    Thursday, August 6, 2020 7:16 AM
    Answerer
  • I have a feeling that we're talking about user data here and that Grace is under the impression that "clustering" means that we can modify data on both nodes at the same time - which we can't. I agree, clarification of the question is needed.

    Tibor Karaszi, SQL Server MVP (Web Blog)

    Thursday, August 6, 2020 2:09 PM