ODI Scanner Guide

Follow these steps to configure a connection to the Oracle Data Integrator (ODI) scanner.

Step 1: Configure the Connection

Create a new connection in Admin UI http://localhost:8281/manta-admin-gui/app/index.html?#/platform/connections/ for each ODI instance that you would like to have the lineage covered by IBM Manta Data Lineage. You must create one ODI connection for each unique combination of ODI repository and ODI context used for analysis. For example, when you use two contexts in one repository, for example context Prod1 for ODI jobs A and B, and context Prod2 for ODI jobs C and D, then create two ODI connections against the same repository. Then, configure the context and jobs to extract for each connection. The necessary connection requirements, prerequisites, and privileges can be found in ODI Integration Requirements.

Mandatory Configuration Fields

Optional Properties

Please review ODI Resource Configuration for the full list plus a detailed explanation of the properties that can be configured for the ODI scanner.

Optional/Additional Steps

Connection Definition Settings

The connection settings for the ODI Repository are automatically extracted by Manta Data Lineage during lineage analysis. However, in some cases (e.g., Manta Data Lineage connects to a non-prod environment for ODI but prod for the database, or when ODI jobs are provided manually rather than via automated extraction), it may be necessary to override them or define them manually as follows.

Variable and Expression Override Files

ODI jobs are usually parameterized, and the jobs cannot be analyzed on their own without the variable/expression values. The analysis may not be accurate if the values are constructed dynamically or if there are no defaults. The override files are currently the only source of these values. Configured values are currently only used during the analysis of ODI procedures.

Values of variables and ODI expressions can be configured manually using override files. To configure variable values, create the file referenced by the property odi.variable.override.file.

Example of the file structure:

[global]
SOME.VARIABLE=value

[PROCEDURE/procedure_name]
OTHER.VARIABLE=other value

Values configured under the global heading will be used for all procedures, while values under PROCEDURE/<procedure name> will only be used for the given procedure. Configurations for specific procedures override global configurations.

Configuring expression values works the same. As an example, assume we want to configure the value of <%= odiRef.getOption("OPT001") %> to be TRUE for all procedures except PROC_1, where it should be FALSE. In such cases, the file referenced by the property odi.expression.override.file should be created with the following content. 

[global]
odiRef.getOption("OPT001")=TRUE

[PROCEDURE/PROC_1]
odiRef.getOption("OPT001")=FALSE

Connecting to Databases Not Supported by ODI by Default

The ODI scanner supports connections to Snowflake and BigQuery. These databases are not supported by ODI by default, but the user can add them as new technologies.

To make the connections to, for example, Snowflake work properly, all Snowflake data sources must be placed under one dedicated technology. Manta Data Lineage tries to recognize these additional technologies heuristically by their codes. If some of these technologies are not resolved correctly, it may help to change the code of the corresponding technology to: