This how-to will teach you how to set up a new SQL Server database.
2 Set Up a New Database
When setting up a new database for Mendix, most of the settings can be left to the default configuration. When looking at the general settings, you only need to set up the database name. Set up the database files according to the Microsoft SQL Server best practices (for details, see SQL Server Installation Best Practices).
In the database options, the default properties need to be evaluated. When choosing a collation, pay attention to the type of collation you are going to use. Mendix uses UTF-8 for all data evaluation. Depending on your exact locale, you will most likely want to choose one of the
SQL_Latin1_General_ collations. The exact encoding will depend on your OS. For an en_US installation, this will be
The last two options identify how sorting and uniqueness is interpreted. For example, the collation option
_CS_ indicates that the collation sorting style will be case-sensitive. For more information on collations, see Windows Collation Sorting Style.
As a recovery model, Mendix only requires the Simple mode. The functionality offered in the Full recovery model option is not used by Mendix; usage of this recovery model will not hurt Mendix, but it could increase the data usage of all the transactions and might slow down any rollbacks in case of an error.
After the database is created, the Mendix Platform can initiate the initial setup and prepare all the tables and functions for usage by the platform. When starting the platform for the first time, there are two queries that require elevated privileges from a
sysadmin role. The
sysadmin role can be temporarily assigned to the user, or these queries can be executed by the administrator as well.
There are two more queries that are only executed once, but these privileges are implicitly assigned to the
db_owner role. As long as the user is
db_owner, the function and procedure can be created without problems.
3 Configuring the Read Committed Snapshot Isolation Level
The database schema needs to be configured so that the Read Committed Snapshot feature is enabled. This can be achieved by executing the following command on the database:
ALTER DATABASE [MySchema] SET READ_COMMITTED_SNAPSHOT ON;
You need to replace
MySchema with the name of your schema.
4 Configuring the necessary SQL Server extensions
For the correct functioning of Mendix, some SQL Server extensions need to be installed. This can be achieved via the following command:
CREATE ASSEMBLY [Mendix.SqlServerExtensions] FROM "D:\MyFolder\Mendix\server\runtime\lib\Mendix.SqlServerExtensions.dll" WITH PERMISSION_SET = SAFE;
Of course, the file path needs to be set to the corresponding file which is part of your Mendix installation.
The permission above requires CLR to be enabled on the SQL Server instance. CLR can be enabled using this query:
EXEC sp_configure 'clr enabled' , '1'; RECONFIGURE;
4.1 Queries to Be Executed with the
db_ddladmin Database Role
The function and assembly need to be created for timezone handling. Also for these functions, Common Language Runtime (CLR) needs to be enabled on the SQL Server instance. If your are not working on the same timezone as UTC, you need to enable CLR and this function. Without time zone support in the platform, development of your functionality will be much more difficult. This can be achieved using the following commands:
CREATE FUNCTION [dbo].[mx_toLocalDateTime] (@utcDateTime datetime, @dstTimeZone nvarchar(50)) RETURNS datetime AS EXTERNAL NAME [Mendix.SqlServerExtensions].[Mendix.SqlServerExtensions.DateTimeLocalizer].[ConvertToLocalDateTime];
CREATE PROCEDURE [dbo].[usp_nextsequencevalue] @SeqName nvarchar(128) AS BEGIN DECLARE @NewSeqVal bigintSET NOCOUNT ON UPDATE [mendixsystem$sequence] SET @NewSeqVal = [current_value] = [current_value] + 1 WHERE [name] = @SeqName RETURN @NewSeqVal END;