Post-installation Configuration and Verification Steps

Verify that all installation and configuration steps in the Instructions.html file have been completed. The Instructions.html file is located in \SAS-configuration-directory\Levn\Documents\. Follow the instructions that are provided in this file.

If you used the SAS Deployment Manager to update passwords for the SAS Administrator and SAS Trusted User, then you must also update the users’ passwords for SAS Model Manager 14.2 Local Services before you sign in to the SAS Decision Manager web application. For more information, see Update User Passwords for Local User Service.

Create application users and assign permissions.

Verify that the WebDAV folder permissions are set appropriately to allow authorized users to create and edit user-defined templates.

(Optional) SAS Decision Manager supports the use of multiple SAS Application Servers to be used as part of a performance definition, a scoring test, or a model retrain definition. To use a SAS Application Server context other than the default SASApp, you must add a new workspace server and create a new SAS Application Server context using the SAS Deployment Wizard. For more information, see Add a New Workspace Server and Create a SAS Application Server Context.

If you are using a locale where national characters or if multi-byte characters are used in table names or variable names, you must extend naming support for double-byte characters. For more information, see Extend Support for Double-Byte Characters in Table and Variable Names.

(Optional) Run the scripts to create Oracle synonyms in the database if you do not want to use the default Oracle schema.

If you cleared the Automatically create tables and load data check box during installation, then you must manually create and load the Oracle database tables for business rules data and modeling project metadata, including history, job definitions, and job logs. For more information, see Create Oracle Database Tables.

(Optional) Configure your deployment for single sign-on web authentication.

In order to publish models from SAS Decision Manager to a database for scoring, additional configuration steps are required to prepare the database. For more information, see Preparing a Data Management System for Use with SAS Model Manager in SAS In-Database Products: Administrator’s Guide.

If your system is configured for Kerberos authentication for Hadoop or Teradata, each user must have a valid Kerberos ticket. You must also complete post-installation configuration steps to enable users to publish models from the SAS Decision Manager web application. For more information, see Configure Users Authenticated by Kerberos for Publishing Models.

Verify that the Certificate Authority certificate is available.

(Optional) Configure your deployment to use HTTPS.

(Optional) Add the Visual Analytics: Data Building and Data Management: Lineage roles to the Decision Manager Users group. See Administering Group and Role Membership for more information. These roles enable users to run SAS Visual Data Builder and view lineage information for rule flows and models.

(Optional) Configure SAS Workflow.

(Optional) Configure SAS Micro Analytic Service to support Python code. If your decisions run code that is written in Python, you must configure SAS Micro Analytic Service in order to be able to test the decisions in Decision Builder. See SAS Micro Analytic Service: Programming and Administration Guide for information about configuring a SAS application server to support the DS2 pymas package.

(Optional) Configure the limit for the number of observations for a scoring result set.

(Optional) If you have a license for SAS Factory Miner and want to integrate with SAS Decision Manager to be able to register models to the model repository, see SAS Factory Miner: Administration and Configuration.

(Optional) If you are using Oracle for the SAS Decision Manager database, you can stop the SAS Decision Manager Common Data Server.