Release 2025.03

Collibra Data Lineage Service

• Asset names that include "\G" are now shown correctly when creating technical lineage.
• When ingesting Azure Data Factory data sources, SQL statements of the type “expression” no longer result in analyze errors.
• When ingesting dbt sources, Collibra Data Lineage now improved the support for Postgres.
• When ingesting SQL Server Integration Services (SSIS) data sources, Collibra Data Lineage now extracts lineage from SQL statements, even if the statements don't start with "SELECT" or "WITH."
• When ingesting Informatica PowerCenter sources, Collibra Data Lineage now handles undefined objects gracefully.
• We fixed an issue in the Power BI integration that was causing a “UNIQUE constraint failed” error during metadata processing.
• Collibra Data Lineage now correctly handles column name casing when ingesting SQL Server data sources via the Power BI integration.
• When ingesting Microsoft SQL Server data sources, Collibra Data Lineage now correctly parses OpenQuery within block code and stored procedures. Unsupported uses of OpenQuery, such as executing stored procedures, result in UNKNOWN Query errors.

Metadata integrations

• The source-driven sampling feature for Oracle now continues correctly when an "OracleDatabaseException" occurs.

Lineage Harvester (CLI and Edge)

• When you create technical lineage for Databricks Unity Catalog, you can now use the new Ingest SQL transformations option on the synchronization page to determine whether to include SQL transformations in the technical lineage viewer.
• When creating technical lineage for Google Dataplex, Collibra Data Lineage now supports the "Technical Lineage Admin" connection.
• We fixed an issue affecting technical lineage for Databricks Unity Catalog, which was caused by a schema change introduced by Databricks.

Collibra Data Lineage Service

• When ingesting BigQuery data sources, Collibra Data Lineage now supports column options for view statements.
• When ingesting DB2 data sources, Collibra Data Lineage now supports CURRENT TIMESTAMP(integer) statements.
• Improved lineage results when integrating Power BI with DAX analysis enabled.
• When ingesting Oracle data sources, Collibra Data Lineage now correctly sets the database and schema for packages and shows the correct package name.

Infrastructure

• We now support HashiCorp Vault version 1.19.x.

Security

• We improved the security of Edge.

Metadata integrations

• When you add custom AI metric mappings for an Azure AI Foundry synchronization, the "Metric" field is now a dropdown list of available and supported Azure AI Foundry metrics.
• When including SAP Analytics Cloud assets as part of the unified SAP Datasphere Catalog integration, the “Owner In Source” attribute on SAC Story and SAC Data Model asset pages is now populated. It is populated with the value of the “Created By” field from the API response, for the corresponding SAP Analytics Cloud assets.
• The "GCP Service Account" field in the GCP connection is now renamed to "Service Account / Workload Identity Federation (WIF)."

Lineage Harvester (CLI and Edge)

• When harvesting JDBC data sources for technical lineage via Edge, the relevant technical lineage Edge capability templates now include a “Use Default value” checkbox next to the default SQL queries. By default, the checkbox is selected, indicating that you want to use the default queries. To use custom queries, clear the checkbox and enter your custom queries.
  
  Note 

  • If you use custom queries, ensure that you use only supported SQL syntax.
  • Collibra Support does not provide support for custom SQL queries.

Collibra Data Lineage Service

• When using the "Dependent On Sources" feature, which enables sharing database models across multiple data sources, Collibra Data Lineage now handles large volumes of synonyms more efficiently. This reduces the risk of out-of-memory errors during metadata processing.
• When ingesting OpenLineage sources, Collibra Data Lineage automatically recognizes the database and schema if they are not provided in the source configuration file.
  

Protect

• Protect capabilities now report more detailed information on how policies have been applied.

Security

• We improved the security of Edge.

Metadata integrations

• The scalable ingestion flow now replaces the UTF-8 NULL character with an empty character, allowing the synchronization to complete correctly.

  Note If the UTF-8 NULL character appears in the name of the asset, the synchronization will still fail.

Collibra Data Lineage Service

• When ingesting Azure Data Factory (ADF) data sources, Collibra Data Lineage now supports
  • Lineage for parametrized pipeline runs invoked by other pipelines.
  • Empty column names in datasets gracefully.
• When ingesting OpenLineage data sources, Collibra Data Lineage now provides table-level lineage for jobs.
• When ingesting Oracle data sources, Collibra Data Lineage now supports the “^=” operator.
• When ingesting BigQuery data sources, Collibra Data Lineage now supports:
  • CONSTRAINT and PRIMARY keywords.
  • SWITCH/EXCHANGE statements.
  • CREATE EXTERNAL TABLE statements.
• When integrating any of the supported BI or ETL tools, Collibra Data Lineage now supports UNION/SELECT queries for dialects handled by our analyzer v1.
• When ingesting SQL Server data sources, Collibra Data Lineage now supports OPENROWSET clauses.
  
  Note The OPENROWSET function is not yet supported.

Metadata integrations

You can now use Workload Identity Federation (WIF) authentication for Dataplex and Dataplex Catalog integrations, as well as for creating technical lineage for Dataplex.

Lineage Harvester (CLI and Edge)

• Collibra Data Lineage now harvests lineage from all projects specified by the "Project ID" field on the synchronization page.
• You can now use a custom Catalog to ingest metadata from Databricks Unity Catalog and create technical lineage.
• When integrating Power BI, you can now benefit from new filtering options, which we refer to as “filtering v2”. In addition to filtering on capacities and workspaces, v2 allows you to also filter on dashboards and reports, including reports that are published in Power BI apps. For complete information, go to Prepare a <source ID> configuration file and Power BI workspaces.
• When integrating Looker, Collibra Data Lineage can now access different API endpoints, depending on your Looker user permissions. Therefore, your Looker user no longer requires an Admin role. You can now configure a custom role with a specific set of permissions. For complete details, go to Set up Looker.
• We updated the Snowflake JDBC driver that is bundled with the CLI lineage harvester to version 3.23.1.

Collibra Data Lineage Service

• When integrating Power BI, Collibra Data Lineage now correctly processes filters that include multiple domains.
• In the technical lineage viewer, CREATE TABLE AS and CREATE VIEW statements are now given the correct highlighting and the query name is correctly shown.
• When ingesting SQL Server data sources, Collibra Data Lineage can now parse OPENDATASOURCE statements.
• Collibra Data Lineage now handles synonyms and other user-provided information more effectively when using the "Dependent On Sources" feature, which allows you to share database models across data sources.
• OAuth authentication from the technical lineage platform to Collibra Platform now works correctly when used via the technical lineage admin connection.
• You can now create technical lineage without encountering issues in particular external ID mapping setups.
• When ingesting Azure Data Factory (ADF) data sources, Collibra Data Lineage now only processes pipelines that are entry points.

Security

• We improved the security of Edge and Protect capabilities via Edge.

Collibra Data Lineage Service

• When ingesting any of the supported JDBC data sources, Collibra Data Lineage now provides improved analysis of SQL statements when a required database model is not available.
• When ingesting SQL Server Integration Services (SSIS) data sources, Collibra Data Lineage now:
  • Supports SSIS connections referenced by connection name.
  • Handles incorrect connection strings gracefully.
• When ingesting DataStage data sources, Collibra Data Lineage now provides a suggestion for the Jobs property in the source configuration file when not provided.
• When ingesting Azure Data Factory data sources, Collibra Data Lineage now provides lineage for manual parameterized pipeline runs.
• SQL files with WHERE clauses that contain a huge number of conditions no longer cause out-of-memory errors.
• The last sync time is now correctly populated for custom technical lineage sources, even when only the default source code is present.

Metadata integrations

• The Azure AI Foundry integration via Edge is now available in public beta. When you integrate Azure AI Foundry, you integrate the metadata of AI models and AI agents from Azure AI Foundry to the Collibra Platform. The resulting assets represent the Azure AI Foundry model and agent.
  For information, go to About the Azure AI Foundry integration via Edge.
• With the release of SAP version 2025.5, when integrating SAP Datasphere Catalog, the user who owns the system in SAP Datasphere Catalog must have access to all of the spaces that they want to synchronize in SAP Datasphere Catalog. To own a system:
  • You need the “Catalog Admin” permission in SAP Datasphere Catalog.
  • You can’t be the System Owner.

Lineage Harvester (CLI and Edge)

• Collibra Data Lineage now ingests lineage from file assets for Google Dataplex.
• Collibra Data Lineage now supports OpenLineage, AWS Glue (via OpenLineage), and Apache Airflow (via OpenLineage) in public beta. This functionality is available on Edge.
  For information, go to Create a technical lineage via Edge.
• When creating technical lineage for Google Dataplex, column-level lineage support is generally available.
• When integrating MicroStrategy, Collibra Data Lineage now correctly processes data source mapping configurations. You no longer encounter "java.lang.CharSequence.length()" errors.

Collibra Data Lineage Service

• We fixed an issue that prevented choosing the correct dialect while processing lineage.

Collibra Data Lineage Service

• When ingesting SAP HANA data sources with calculated views, synonyms no longer generate analysis errors.
• When ingesting IBM Db2 data sources, Collibra Data Lineage now supports the "CURRENT USER" keyword.
• When integrating Power BI, Collibra Data Lineage now correctly constructs URLs for embedded Power BI reports in Power BI apps. These URLs are included on the respective asset pages in Collibra.
• When ingesting Azure Data Factory,Collibra Data Lineage now "parses pipeline().RunId" correctly.
• When ingesting Informatica PowerCenter, Collibra Data Lineage has improved support for SQL statement analysis.
• When ingesting data sources, Collibra Data Lineage now generates an analyze error for unsupported JDBC connectors.

Metadata integrations

• When ingesting SAP Datasphere metadata as part of the SAP Datasphere Catalog integration, we now ingest the URLs for the following asset types. The URLs point to the instances of the assets the data builder.
  • SAP Datasphere Analytic Model
  • SAP Datasphere Intelligent Lookup
  • Table
  • Database View
• We improved the clarity of Azure ML capability logs.

Lineage Harvester (CLI and Edge)

• Collibra Data Lineage now includes transformation source code in the technical lineage viewer for Databricks Unity Catalog.
• Collibra Data Lineage now ingests table-level lineage for Databricks Unity Catalog.

Collibra Data Lineage Service

• When ingesting Informatica Intelligent Cloud Services (IICS) data sources, Collibra Data Lineage now uses the source or output file name from "sessionProperties", if available, as the file or table name.

Infrastructure

• When you install an Edge site on a managed Kubernetes cluster, you can now:
  • Configure custom priority classes.
  • Skip the configuration of priority classes.
    Warning  We don't recommend configuring or skipping priority classes unless you have an experienced Kubernetes engineer in your organization, as doing so may result in Edge site failures.
• When you run a JDBC Commercial offering capability from an Edge site, the log files are stored when the capability fails or succeeds.

Metadata integrations

• When you ingest Excel files from a local system via an Edge capability, the capability now successfully ingests the files metadata.

Lineage Harvester (CLI and Edge)

• When ingesting Azure Synapse Analytics data sources, you can now choose between two Edge capabilities, depending on whether you want to ingest metadata from one database or multiple databases.
  • Use the existing “Technical Lineage for Azure” capability to harvest metadata from a single database (dedicated SQL pool).
  • Use the new “Technical Lineage for Azure (multi-DB)” capability to harvest metadata from multiple databases (serverless SQL pool).
• When ingesting Azure Data Factory data sources, Collibra Data Lineage now supports pipeline run data.
• When harvesting Azure synapse data sources, Collibra Data Lineage now supports multiple databases in one capability.

Collibra Data Lineage Service

• When ingesting DataStage, Collibra Data Lineage now removes leading and trailing whitespace from the "TableName" tag in the XML definition when extracting schema and table information.
• When integrating MicroStrategy, if project permissions are missing, the integration no longer fails. Instead, an analyze error is generated.
• We fixed an issue so that when integrating MicroStrategy, Collibra Data Lineage now creates relations of the type “Data Attribute used in / uses Report” between MicroStrategy Data Attribute assets and MicroStrategy Report assets, when the data attribute is a Metric.
• When ingesting Snowflake data sources, Collibra Data Lineage now supports:
  • Table functions in CALL parameters.
  • EXPLAIN statements.
• When integrating PowerCenter, the integration no longer fails if Collibra Data Lineage doesn't receive a value for the shared attribute in the Folder.
• When integrating SAP Hana calculated views, Collibra Data Lineage now supports columns coming from synonyms used in joins.

Security

• We've improved the security of capabilities via Edge.

Metadata integrations

• The Unified Data Classification process no longer collects sample data if the data classes don’t include any "regular expression for data" or "list of values for data" classification rules. It also skips sample data collection if classifications can be based on "regular expression for column names" or "data type" classification rules. This update allows you to classify data without collecting sample data.
• When adding the Edge data classification capability for a data source, you can now define separate timeout parameters. For example, in "Other settings", you can use the "table-sampling-queries-timeout" and "table-sampling-queries-timeout-unit" parameters to specify the timeout for querying samples from the data source.
• For schemas integrated using an earlier version of the Databricks Unity Catalog integration, domains with spaces in their names will remain unchanged. Duplicate empty domains without spaces will no longer be created in future integrations.
• When you synchronize with Azure ML, access tokens are now refreshed to reduce failures if the synchronization takes longer than an hour to complete.

Lineage Harvester (CLI and Edge)

• Collibra Data Lineage now supports the OAuth authentication type, via the Technical Lineage Edge capabilities, to authenticate connections to the Collibra Data Lineage service instances.

Collibra Data Lineage Service

• When ingesting Azure Data Factory data sources, Collibra Data Lineage now improved the parsing for expressions.
• When integrating DataStage, Collibra Data Lineage now supports virtual database connections.
• When integrating SSRS, integration no longer fails if Collibra Data Lineage can't find a referenced data source. Instead, an analyze error is generated.
• When ingesting JDBC data sources, Collibra Data Lineage no longer creates duplicate transformations when indirect lineage is involved.
• When ingesting Teradata data sources, Collibra Data Lineage now supports:
  • PIVOT statements with string literals.
  • BEGIN statements.
  • EXPAND ON statements.
• When integrating MicroStrategy, Collibra Data Lineage now correctly creates stitching for dossiers.
• After integrating Tableau, if you hover your pointer over a “TDM” node in the technical lineage graph, the correct label “Tableau Data Model” is now shown.
• When ingesting Snowflake data sources, Collibra Data Lineage now supports chained LATERAL clauses.
• We improved the performance of the SQL statement ordering algorithm, to mitigate the risk of out-of-memory errors during batch analysis.
• When integrating Power BI with DAX analysis enabled, Collibra Data Lineage now correctly identifies the implicit current table in DAX queries.

Infrastructure

• All Edge site pods now have static labels, which allow you to better track your Edge site capabilities.
• The Edge CLI is now automatically installed with new Edge site installations.

Security

• We've improved the security of Edge and capabilities via Edge.

Metadata integrations

• We improved the "Catalog JDBC ingestion" capability to address issues that could occur when synchronizing an empty schema.
• When creating technical lineage for Databricks Unity Catalog, the SQL transformation code is no longer ingested. We plan to reintroduce this support in a future release.
• Foreign keys with the same name but belonging to different tables are now synchronized correctly via the scalable ingestion flow. The foreign key naming convention has been updated to: "edgeConnectionName>jdbccatalog>schemaName>tableName>foreignKeyName (foreign key)."

Lineage Harvester (CLI and Edge)

• When ingesting dbt Cloud data sources, Collibra Data Lineage now supports aliases.
• When ingesting data source types that allow for multiple databases (meaning HiveQL, Microsoft SQL Server, MySQL, Netezza, Spark SQL, Sybase and Teradata), you can now use the optional “Database name JSON” field in your technical lineage Edge capabilities, to provide multiple database names in a list format.
• Collibra Data Lineage now supports the OAuth authentication type, via the Technical Lineage Edge capabilities, to authenticate connections to the Collibra Data Lineage service instances.

Collibra Data Lineage Service

• When ingesting dbt data sources, Collibra Data Lineage now supports SQL statements that include CTEs and the "LOCKING" keyword.
• We optimized the creation of lineage relations for large technical lineage graphs.
• When ingesting IICS data sources, Collibra Data Lineage now applies SQL queries defined in mapping configuration task files only to transformations which are assigned to the same connection.
• Collibra Data Lineage now includes transformation source code in the technical lineage viewer for Databricks.
• When ingesting Oracle data sources, Collibra Data Lineage now:
  • Generates a single analysis error if the required database link mapping configuration is not provided. The analyze error includes detailed information about the missing configuration.
  • Supports database link mapping with multiple links under the same scope.
• We fixed an issue that occurred during Tableau integration, whereby Collibra Data Lineage was using incorrect database columns as source columns.
• When integrating Looker, Collibra Data Lineage now creates a relation of the type "Data Element sources / targets Data Element" between Looker Report Attribute assets and Column assets. (idea #LINA-I-39)

Infrastructure

• Edge now supports using proxy certificates that have a negative serial number. To get this feature support, you must download the new Edge site installer.
  Note This support is offered to customers with older proxy certificates, which may still have negative serial numbers. However, we don't recommend using a proxy certificate with a negative serial number, as this is no longer best practice and is considered invalid according to the X.509 standard (RFC 5280). As such, Edge will remove this support by the end of 2025. If your proxy certificates have negative serial numbers, update your certificates before the end of 2025.

Security

• We've improved the security of integrations via Edge.

Metadata integrations

• The MLflow integration via Edge is now available in public beta. When you integrate MLflow, you integrate the metadata of ML models from MLflow to Collibra Data Intelligence Platform. The resulting assets represent the MLflow model.
• Collibra Data Lineage now includes transformation source code in the technical lineage viewer for Databricks.
• You can now select "us" and "eu" from the "Location" field when you configure the synchronization of the Dataplex Catalog ingestion.
• In the Databricks Unity Catalog synchronization configuration, you can now define how the compute resource reacts after source tags are extracted. If the "Stop Compute Resource" field is set to "Yes," the compute resource is stopped by the integration. When set to "No," the compute resource remains active.

Protect

• Protect for BigQuery now sets retry parameters for all Google APIs. This ensures that operations such as setting or clearing IAM members on policy tags are retried if a service is temporarily unavailable.

Collibra Data Lineage Service

• When integrating Power BI with DAX analysis enabled, an improvement to the caching mechanism allows for faster processing.
• When ingesting Oracle data sources, Collibra Data Lineage now:
  • Supports the SYSTIMESTAMP function.
  • Supports FETCH statements in CURSORS.
• When ingesting SQL Server data sources, Collibra Data Lineage now supports "LOAD" as a column name.
• When ingesting BigQuery data sources, Collibra Data Lineage now correctly handles SQL statements that contain a lot of common table expressions and wildcards.
• When ingesting Snowflake data sources using the SQL-API mode, Collibra Data Lineage now ignores object types that are not columns in the modified objects returned by Snowflake.
• We added the new "aws-me" Collibra Data Lineage service instance.

Collibra Data Lineage Service

• We fixed an issue that was causing telemetry data to not be exported properly. Telemetry data are traces and metrics that our Support teams use to help resolve issues.

Collibra Data Lineage Service

• When creating a custom technical lineage:
  • Collibra Data Lineage now raises an analysis error when the "props" attribute is defined for a column, in a database > schema > table hierarchy.
  • Collibra Data Lineage now raises an analysis error if the “useCollibraSystemName” property is set to “false”, but a system is specified in the JSON file.
• When ingesting Teradata data sources, Collibra Data Lineage now supports UPDATE statements inside TRANSACTIONS.
• When ingesting Teradata data sources via shared storage connection (if via Edge) or folder connection (if via CLI harvester), Collibra Data Lineage now offers better support for scenarios involving auto-generated DDL in combination with user-provided DDL statements.
• When integrating Tableau, Collibra Data Lineage now correctly handles database mapping for databases with multiple schemas.
• When ingesting Snowflake data sources using SQL-API mode, the Collibra Data Lineage does not create lineage anymore between base object and direct object.
• When ingesting Azure Data Factory data sources, Collibra Data Lineage now groups the analysis errors that are related to activity output not being supported.
• When integrating Informatica Intelligent Cloud Services (IICS), Collibra Data Lineage now also creates lineage when the "createTaret" attribute is set to "False" in a transformation definition.
• Collibra Data Lineage now fetches information from the Dataplex API differently to prevent failures.

Security

• We have improved the security of capabilities via Edge.

Metadata integrations

• For assets in SAP Datasphere Catalog that have a "summary" and a "description", if the API returns both of these values, they are now both shown in the "Description from Source System" attribute on the respective asset pages in Collibra Data Catalog. To differentiate between the two, the value of the "summary" attribute in SAP Datasphere Catalog is shown with the prefix "summary:".
• You can now configure which initial status you want your SAP assets to have when they are ingested via our SAP Datasphere Catalog integration. This enhancement gives you the same control over SAP-related asset types as exists for all other asset types.
• Google Dataplex Catalog integration now ingests columns for views.
• The ADLS with Purview integration now synchronizes correctly when the "Max Schema Level" field in the capability is set to a value other than 1.
• You can again synchronize Snowflake data sources when the "Include Source Tags" option is selected in synchronization rules.