​

Improvements

• Removed beta label from useDynamicAssertion in AutomatorTestCaseField component
• Added paginated assets API endpoints for Domain, Data Product, Glossary Term, Tag, Team, and User entities for improved scalability
• Added support for MF4 (Measurement Data Format version 4) file reader for automotive and industrial measurement data
• Added support for BigQuery clone queries (CREATE TABLE … CLONE) in lineage tracking
• Improved handling of nested column structures (subfields) in Datalake sources

​

Fixes

• Fixed German language UI translations for login page
• Fixed UC Ingestion failing due to non-selected tables
• Fixed MF4 file reader import errors with proper lazy loading of dependencies
• Fixed domain assets count mismatch between API and UI responses
• Fixed Bulk Import to preserve existing metadata when fields are not included in the CSV
• Pinned pydantic dependency to version less than 2.12.0 for compatibility

​

Deprecation Notice

• The inline assets field in Domain and Data Product entities has been deprecated and will be removed in a future release.
• New paginated API endpoints have been introduced for better performance with large asset collections:
  • GET /v1/domains//assets - Paginated access to domain assets by ID
  • GET /v1/domains/name//assets - Paginated access to domain assets by name
  • GET /v1/dataProducts//assets - Paginated access to data product assets by ID
  • GET /v1/dataProducts/name//assets - Paginated access to data product assets by name

• Update your integrations to use the new paginated endpoints instead of relying on the inline assets field. Python SDK users should migrate to ometa.get_domain_assets(domain_id) and ometa.get_data_product_assets(data_product_id).

​

What’s New

​

Improvements

• Fetched entities for import should exclude only fields sourced from the CSV.

​

Improvements

• Remove unnecessary braces in additional_config for cleaner code.
• Implemented comprehensive Security Service with schema changes, UI integration, and OSS visibility controls.
• Added backward-compatible support for documented owner configuration in dbt.
• Implemented configurable sink and reverse metadata support for Ranger, along with cleanup, bug fixes, and UI enhancements. (Collate)
• Added lineage propagation enhancements in Automator with stop conditions and propagation depth, along with related UI fixes, configuration support, and test improvements. (Collate)

​

Fixes

• Fixed Incorrect or missing security labels.

​

Improvements

• Added missing database migrations for searchSettings changes introduced in version 1.8.7.
• Introduced support for the SSAS (SQL Server Analysis Services) connector.
• Enhanced memory management and ensured proper resource cleanup in the profiler.
• Updated OMeta logging for improved clarity and easier debugging.
• Enabled support for service accounts in the Synapse Connector. $
• Implemented chunked reading of archived logs in Argo Workflows to improve performance and memory efficiency. $
• Added external table reverse metadata support for Databricks and Unity Catalog. $
• Introduced Collate SaaS activity metrics integrated with OpenSearch telemetry. $

​

Fixes

• Added Tableau CA certificate authentication to ensure secure connectivity.
• Skipped creation of redundant indexes for foreign keys referencing the same column multiple times.
• Fixed DataLake ingestion to handle larger files without failures or timeouts.
• Resolved Tableau ingestion issues by properly handling None entities.
• Ignored non-current columns for Iceberg tables in Glue and Athena environments.
• Fixed JSON processing issues related to Jakarta in the MCP Patch Tool.
• Addressed clusterAlias errors in the MCP SearchMetadataTool by implementing proper alias resolution for cross-environment compatibility.
• Fixed cyclic lineage node collapsing to prevent incorrect root node removal, with added unit and Playwright tests for validation.
• Ensured Airflow ingestion compatibility with older Airflow versions.
• Resolved Trino column validation errors for highly complex fields to ensure proper handling.
• Fixed alert mechanism to correctly display test case names and URLs upon failure.

​

Improvements

• SAdd Permission Debugger to understand roles & permissions for a user
• Smart Reindex functionality for improved performance
• Pass GX description on test case creation
• Add Actual Case on the displayName to aggregate with right bucket

​

Fixes

• Add support for editing “Compute row count” from parameter screen
• Explore page count mismatch
• Fix lineage search issue in UI
• Fix domain custom property persistence improvements
• Fix table query pagination showing incorrect pageSize
• Fix stored procedure temp table processing and lineage filtering with db & schema

​

Improvements

• Search RBAC enhancements for better performance and accuracy.
• Revamped metrics system with detailed request latencies and breakdown of database/search operations.
• Added Virtual Threads and Semaphore control for improved performance.
• Enhanced browser language support with fixes for Chinese (zh) language search indexing.
• Improved German UI translations.
• Dashboard service prefix support for better organization.
• Added columns.description in search settings for enhanced discoverability.
• Improved handling of appPrivateConfig with empty parameters.
• Enhanced JVM parameters optimization for Java 21.
• Added missing supportsMetadataExtraction flag in connectors.
• Implement Freshness Test on Pandas. (Collate)

​

Fixes

• Handle text overflow in CommonEntitySummaryInfo and improved loader in search bar.
• Fixed soft-delete and restore handling for charts linked to dashboards.
• Fixed null columns handling in various scenarios.
• Fixed Postgres query column name for execution time.
• Fixed pagination alignment issues in UI.
• Resolved race condition in bulk import between websocket and REST API.
• Fixed column lineage validation and update/delete operations.
• Fixed activity feed not showing column-level metadata changes.
• Fixed issues with Chinese (zh) language search index mapping.
• Corrected parent FQN extraction for column FQN.
• Fixed NPE when reading appPrivateConfig with empty parameters.
• Fixed recursive suggestion application for deeply nested columns.
• Fixed auto pilot trigger button enable state.
• Fixed custom logo failure causing repetitive API calls.
• Fixed unaligned charts display.
• Removed “Default: null” from columns for better JSON to POJO handling.
• Updated button names for consistency.
• Set Presidio logger to ERROR level to reduce noise.
• Reverse Metadata - Illegal result error. (Collate)

​

Improvements

• Enhanced tab labels for better readability.
• Improved German UI translations.

​

Fixes

• Resolved issue with Databricks default schema.
• Fixed bugs in ElasticSearchSink and OpenSearchSink.
• Corrected problem with Prometheus endpoint.
• Removed debug test code.

​

Improvements

• Implemented Prefix For Dashboard Service.
• Automator - Domain Lineage Propagation Support.
• Added Session Age for Cookies.
• Add maxRequestHeaderSize to server.applicationConnectors section in OpenMetaData default config file.
• Pbi display table name from source.
• Removed McpIntegrationTest.java.
• Added Virtual Threads and Semaphore to control.
• Browser language support.
• Add logs for all api calls.
• Add more debug logs; improve JVM params to be JVM 21.
• Show toast error on duplicate domain.
• Add filterJsonTree to the automatorAppConfig.
• Primary color customization.
• Add support for DBX system metrics.
• Improve the alert destination selection warning message.
• Supported rendering all suggestion on user avatar click.
• Add mcp preview.

​

Fixes

• Relevant fields are pulled for bulk import.
• Snowflake map key type error.
• Data quality tab table pagination issue.
• Domain inheritance issue after team removal.
• Okta multi tab refresh issue.
• Api version endpoint page breaking due to STRING changeDescription updated.
• Fixed TotalDataAssetsWidget a stacked graph.
• Cron validations in Ingestion.
• Text Overflow from Widget and Table.
• Fixed column selection not persisting for all action in dropdown.
• Fixed mentions formatting in block editor.
• Fixed the token expiry options order.
• Column lineages are not getting updated/deleted when columns are updated/deleted.
• Custom logo failure leads to repetitive api calls for that image url.
• Fix psql migration data freshness.
• ActivityFeedProvider context not available in KnowledgeCenter page.
• Automator conditions not showing properly for the filters for custom properties.

​

Improvements

• REST service process nested objects inside array dtype in schema.
• Throw an error on query filter parsing and an error on column test case.
• Display object column type.
• Add support for adding tags in test case form.
• Add Postgres SP and UDF descriptions.
• Enhance AutomatorTestCaseField with tag and glossary support.

​

Fixes

• Cache to have immutable entities.
• Remove edit permission from version page.
• Table pagination count issue in the Schema page.
• Version link and fetch logic with 1 hour wait time.
• Tableau Lineage in multiple schema model.
• Soft deleted entities are not removed rom system, so relations are still valid if there is entry in DB.
• Custom properties display issue in advanced search.
• Image was not rendering correctly in description editor.
• Handle sampler SQA sessions.
• Typo on query cost record index.
• Reply count being shown as 0 in task feed card.
• Pagination state lost after multiple back navigations in table list view.
• query_cost_record_search_index Search exception for elasticsearch instance.
• Search functionality for domain edit in user profile.
• Username issue in comments card.
• Quickcloud get script tables.
• Added Databricks pipeline Lineage.
• Add parquet alternate filetypes in table fileformat.
• Snowflake reverse ingestion.
• Typo for Chart preview.
• Update version link for Collate.
• Sorting issue for charts.

​

OpenMetadata MCP Server — Generative-AI-Ready Metadata with Rich Context

​

Key highlights:

• One graph, one endpoint: The MCP server exposes every entity, relationship, data quality metric, lineage detail, and governance rule you’ve curated in OpenMetadata.
• LLM-optimized responses: JSON schemas are tailored for semantic search and RAG workflows, enabling chatbots and copilots to ground their answers in accurate, up-to-date metadata.
• Enterprise-grade insights: Gain real-time KPIs on asset distribution, metadata coverage (descriptions, ownership), tiering, and PII tagging to strengthen data governance.
• Zero-friction adoption: It comes bundled with OpenMetadata—just enable the service, generate an API key, and start querying from Claude, Cursor, ChatGPT, and more.

​

SCIM Provisioning for Okta & Azure AD — Hands-Free User & Group Management (Collate Only)

​

Key benefits:

• Automated onboarding & offboarding: The Tier Agent continuously analyzes usage patterns and lineage to highlight your most critical data assets.
• Consistent governance: The Documentation Agent generates accurate asset descriptions and powers seamless Text2SQL chat experiences.
• Standards-based interoperability: Built on the SCIM spec, making it easy to extend to other IdPs like JumpCloud, OneLogin, and more.

​

Data Contracts - API & Specifications

​

Features:

• Schema & Quality Specifications: Define schemas, semantic tags, quality checks, and SLAs explicitly.
• API and UI-driven Management: Easy creation, versioning, and management through both intuitive UI and REST APIs.