Version 4.8.0

Date: 29 September 2025

• Export-Import Hub Enhancements: The Export-Import Hub now supports policies, rule sets, segments, tags, and UDT variables. Users can selectively import/export entities with dependency validation, download a sample mapping file, and simplify tenant-to-tenant migrations.
• Profile Metrics for Pushdown Data Engine: Profiling in the pushdown engine (Snowflake, BigQuery, Databricks) now achieves parity with Spark. Users gain richer insights such as patterns, arrays, keys, and exact distinct counts, with tenant-level controls for governance.
• Service User Management V1: Service accounts replace personal API keys for integrations and system operations. This ensures stable, auditable, and secure credentials that are not tied to individual user lifecycles.
• Backward Lineage from Power BI to Redshift: Backward lineage is now supported for Power BI, providing end-to-end traceability from dashboards to underlying sources. This extends parity with Snowflake, BigQuery, Databricks, and Redshift.
• Custom Notifications V2: Custom Notifications now integrate with Slack, Teams, Google Chat, Jira, and ServiceNow. Teams can receive alerts directly in their collaboration and ITSM tools alongside existing email and webhook channels.
• Detailed Column-Level Data Quality Reporting: Column-level data quality scores are now surfaced in Discover Assets page, enabling deeper visibility into field-level health, faster root cause analysis, and greater trust in data assets.
• Hive Cadence Support: Hive external tables now support cadence-based updates using native SQL commands, ensuring freshness and freshness policies run against the latest Hive Metastore metrics.
• Kafka Reconciliation Jobs – Start/End Value Support: Kafka reconciliation and profiling jobs now include start and end offsets/timestamps, aligning with Oracle and other sources for full auditability and traceability.
• Tenant-Level Configuration for Profile Metrics: Tenant admins can now set profiling defaults across all assets or new assets only. This ensures consistent governance, optimized performance, and parity across Spark and pushdown engines.

Connectors

• Oracle → Databricks: Fixed reconciliation failures on select Oracle tables with ORA-01858.
• Databricks: Fixed profiling and data quality failures after enabling partition configuration.
• Databricks pushdown: Resolved Pattern Match (Regex) rule failures on TIMESTAMP columns.
• SAP → Snowflake: Fixed row-count reconciliation job failures caused by “Spark Applications Already Exists.”

Data Quality & Policies

• Fixed intermittent DQ policy execution errors (No Request Received in Dataplane).
• Corrected alerts that were triggered incorrectly for failed reconciliation jobs.
• Fixed import errors for reconciliation policies created from different sources.
• Resolved issue where policies were archived after crawling despite no metadata changes.
• Fixed Schema Drift policy import failures.
• Reconciliation policy exports now capture timezone correctly, preventing schedule mismatches after import.
• Fixed policy errors when column names contained spaces or dots.
• Improved reliability in retrieving asset column metadata for programmatic DQ policy creation using custom SQL.
• Pushdown parity: Data Violations tab now displays rows correctly when columns contain NULL values (aligned with Spark behavior).
• Resolved failures migrating data quality policies from Torch to ADOC.

Scheduling & Execution

• Fixed DQ scans/jobs that were stuck in Waiting state indefinitely.
• Fixed reconciliation jobs stuck in Running state after upgrade to 4.6.1.
• Autosys: Corrected case-mismatch handling for job names that blocked status pickup.

Reporting & UI

• Navigator: Fixed over-provisioning of permissions when viewing datasources.
• Profile Settings Migration: Notification group selections are now preserved.
• Fixed incremental profiling and sample data views for date-partitioned Parquet files.
• UI Config: KUBERNETES_SPARK_NODE_SELECTOR settings now apply correctly at pod level.

Other Fixes

• Corrected permissions to allow deletion of profiling resources (previously missing cluster role).
• Fixed engine type migration so left/right engine settings carry forward between tenants.
• Resolved policy failures with error: “Dataplane not running, please contact your administrator.”