Technical lineage known issues and limitations

The following table shows known issues and limitations in the current lineage harvester version.

Important The success rate of a technical lineage, as shown in the Sources tab page, gives a good indication of the processing success. A success rate less than 100%, however, does not mean processing was unsuccessful. A parsing error, for example, which negatively affects the success rate, does not always negatively affect the completeness of the lineage.

Known issue

Description

Stitching results of BI source have a gray background.

Usually, data objects that Collibra Data Lineage stitches to assets in Data Catalog have a yellow background in the technical lineage graph. However, assets of BI sources, for example Power BI, that are stitched to other assets in Data Catalog currently have a gray background. This does not indicate that stitching failed. You can see which assets are stitched on the Stitching tab page.

The lineage harvester currently does not support Java version 16.

If you have Java version 16 and run the lineage harvester with the full-sync command, the harvester fails during the API key retrieval process.

As a workaround, we recommend the following:

  1. Set the JAVA_OPTS to the following: 
    JAVA_OPTS='--illegal-access=deny'
  2. Run the lineage harvester in the same command line window.
Collibra Data Lineage does not reuse the database model or DDL statements from other sources in the lineage harvester configuration file.

Currently, all sources in the lineage harvester configuration file are analyzed separately. As a result, the database model and DDL statements that are used for one source are not taken into account when analyzing another source.

As a workaround, we recommend that you make sure that each source has all DDL statements that it needs to be processed properly.

Tip Saving the DDL statements in separate files and adding the preview "_" before their names might speed up the analysis of the DDL statements.
Harvesting an Amazon Redshift data source fails when using a CDATA JDBC driver. If you use a CDATA JDBC driver to harvest metadata from an Amazon Redshift data source, you have to set the QueryPassthrough property in the connection configuration to true, otherwise the driver fails to execute the query.