> For the complete documentation index, see [llms.txt](https://docs.collibra.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.collibra.com/data-quality-self-hosted/release-notes/data-quality-observability-self-hosted-release-archive/release-2026-04.md). # Release 2026.04 * Release date of Data Quality & Observability (self-hosted) 2026.04: May 1, 2026 * Release notes publication date: April 16, 2026 ## New and improved ### Platform {% hint style="danger" %} As of Data Quality & Observability (self-hosted) 2026.03, all Data Quality & Observability (self-hosted) repo keys used to pull images for self-hosted Kubernetes installations will expire on April 30, 2026. You must update your workflows before this date to avoid disruptions.As of Data Quality & Observability (self-hosted) 2026.02, a self-service option to download temporary repo keys from the "Docker Images" section of the [Downloads page](https://productresources.collibra.com/downloads/data-quality-observability-classic-2026-02/) (login required) is available. Temporary repo keys are valid for 5 days from the download date. The process of requesting long-lived keys from Collibra Support will no longer be supported.To ensure deployment stability, we recommend using the temporary keys to download images, uploading them to your internal Artifactory or registry, and generating internal credentials for production deployments. There is no impact to self-hosted Spark Standalone installations. For instructions on pulling images from the Collibra registry, go to [Install on self-hosted Kubernetes](/data-quality-self-hosted/installation/installing-data-quality-observability-self-hosted-on-self-hosted-kubernetes/install-on-self-hosted-kubernetes.md) . {% endhint %} * You can now configure access to scorecards and reports using the "Require INSIGHTS\_VIEWER role to view scorecards and reports" setting in Security Settings > General. When enabled, only users with the ROLE\_INSIGHTS\_VIEWER role can access scorecards and reports, including sub-pages, ensuring stricter access control. Users without the role are redirected to the Access Denied page when attempting to access restricted content. * You can now manage the WASBS storage account key securely through Azure Key Vault (AKV). This eliminates the need to store the key directly in the Helm values file, enhancing security. * We upgraded the Databricks driver version to 2.8.0. * Databricks Pushdown is now ANSI-compliant. * We upgraded the OWASP ESAPI (Enterprise Security API) library to enhance application security and address known vulnerabilities. While no functional changes are expected, edge cases may behave differently. * Data Quality & Observability (self-hosted) now supports the SQL Assistant for Data Quality feature when you have a Data Quality & Observability (self-hosted) to Collibra Platform OAuth-based integration. Previously, this capability was available only with Basic Authentication. This enhancement ensures broader integration options for leveraging SQL Assistant for Data Quality. * You can now propagate labels and annotations to the Kubernetes metastore secret created by the DQ Agent service at runtime. This is configurable via Helm chart values, allowing you to customize the secret used by Pullup jobs. * You can now customize the Kubernetes pod template for Spark driver and executor pods launched by the DQ Agent. * You can now enable the PUSH\_NA\_TO\_DIP application configuration setting to push adaptive rules with "Not Applicable" status to Collibra Platform. This setting is disabled by default and can be enabled in the Admin Console when the “Create Metric Assets” option is set to true in the integration configuration. * You can now grant non-admin users access to the Admin > Integrations page based on their existing connection role mappings. When you enable the new "Enable Integration Management for Connection Users" setting in Security Settings, users with roles mapped to specific connections can view and configure integration settings for only those connections they have access to. * You can now manage standardized tags using the new "Strict Tagging" feature. This feature ensures consistent metadata application by allowing users to select only approved tags from a dropdown. Admins can manage tags directly in the Admin Console under "Metatags," with validation to prevent duplicates and empty values. Additionally, new API endpoints are available for creating, updating, and deleting tags. * As a one-time conversion process, the values in the t1, t2, t3, and t4 columns of the owl\_catalog table will be used to populate the new defined\_tags table, with each unique value stored as a tag and assigned a unique identifier. ### Jobs * Databricks Pushdown jobs now run successfully on tables with special characters in column names. * Jobs now support the -metadataEnhancedBreakRecords command-line option. This allows archived break records to include additional metadata columns, providing more detailed insights. * You can now enable metadata-enhanced break records for Pullup jobs. This option allows you to include additional data in pullup break records, aiding in remediation and reporting. * The replay functionality no longer executes unnecessary SQL queries when you select "REPLAY" from the Run dropdown menu. This improvement significantly reduces wait times when working with complex SQL queries, as the system now only updates the timeframe without re-running the underlying query logic. ### Findings * You can now control user access to export functionality on the Findings page "Export" tab through role-based permissions. When the "Require DATA\_EXPORT role to export rule breaks" security setting is enabled, users without the DATA\_EXPORT role cannot access export checkboxes or buttons, while users with the role retain full export capabilities. ### Rules * You can now control export permissions for the "Copy Results" and "Download Results" buttons in Rule Workbench's Result Preview section. When the "Require DATA\_EXPORT role to export rule breaks" setting is enabled, users without the DATA\_EXPORT role will see these buttons disabled with a tooltip explaining they don't have permission to copy or download data. * You can now configure the Purpose field as mandatory for custom rules to meet integration requirements. When the "MANDATORY\_RULE\_PURPOSE" configuration flag is enabled, the Purpose field is required for all rule types in Rule Workbench. The "Save" and "Submit" buttons are disabled until you enter a purpose. * You can now add metadata columns to archived rule break records in Pullup jobs. ### Alerts * Email alerts now use 6-digit hexadecimal color codes to ensure the "Job Status" field displays correctly across all email clients. ### Dataset Manager * The "Copy to Clipboard" and "Download Results" options now respect the global exportlimit configuration setting, ensuring data extraction controls are uniformly enforced while maintaining full visibility of preview data on screen. ### Collibra Platform integration * The Data Quality & Observability (self-hosted) and Collibra Platform integration now uses a dedicated thread pool to increase integration job throughput without affecting other scheduler tasks. New properties provide granular control over job concurrency and thread queuing. For most customers, no changes are needed. However, if integration job throughput exceeds 50 jobs per hour, you may need to adjust the following properties: * **dq.dgc.integration.job.thread.pool.size**: Maximum concurrent jobs based on available threads (default: 10). * **dq.dgc.integration.job.thread.pool.queue.capacity**: Threads (jobs) concurrent + queued for the pool (default: 256). * **dq.dgc.max.concurrent.integration.job**: Batch size for integration jobs (default: 10). * The metastore mapping table for integrating Data Quality & Observability (self-hosted) with Collibra Platform now includes a new column that shows the time stamp of when the table was mapped. This time stamp is added when you use the auto-map feature for a schema. * You can now delete all mapping records at the schema level by selecting "Delete Mapping" from the three-dot menu. This action removes existing mappings for a specific schema, allowing you to clear outdated mappings before running the auto-map job. The delete mapping option works alongside the existing auto-map feature to help you maintain accurate schema-level mappings. * When you delete a dataset from Data Quality & Observability (self-hosted), the system now automatically removes the corresponding integration assets (job asset, score domain, and rules domain) from Collibra Platforminstances when the [Application configuration](/data-quality-self-hosted/admin-configuration/admin-console/configuration-settings/application-configuration-settings.md) property, ENABLE\_DELETE\_INTEGRATION, is set to true. This eliminates the need to manually identify and delete orphaned assets, reducing cleanup time and potential errors. * You can now control which assets are created during the integration setup process. When setting up an integration, you can choose to exclude "Metrics" assets from being published to the Data Catalog while still ingesting the underlying data. This option helps reduce catalog clutter and makes it easier to find relevant datasets. The "Metrics" asset creation is enabled by default to maintain existing behavior. * You can now automatically sync tolerance values from Data Quality & Observability (self-hosted) rules to Collibra Platform as threshold parameters when rules are promoted or updated. The tolerance values become visible on the rule asset in the platform. Each sync action is logged with timestamps, rule IDs, and user information for complete traceability. If syncing fails, the system logs detailed error information to help with troubleshooting. ## Fixes ### Platform * Data Quality & Observability (self-hosted) now starts successfully when the SAML tenant manager is enabled. Previously, the web application failed to initialize after upgrading to version 2026.03 if SAML\_TENANT\_MANAGER\_ENABLED was set to true. * We added support for additional types of SAML envelope configurations. * The Helm chart value global.web.security.jwt.token.secretkey now defaults to a sentinel value that causes a random JWT secret key to be generated at deploy time. This random key is regenerated on every application restart, which means short-term tokens issued from previous restarts will be invalidated. To maintain a consistent key across restarts, set global.web.security.jwt.token.secretkey to a custom value of at least 32 characters. Custom values shorter than 32 characters will now cause chart rendering to fail with an error. * Users with the ROLE\_ADMIN\_VIEWER role can now access the Schedule Restriction page without encountering the "Failed to get schedule restrictions" error. This resolves an issue where business administrators were unable to view schedule restrictions due to insufficient permissions. The role maintains appropriate read-only access while continuing to restrict edit and delete capabilities for these users. ### Jobs * Databricks Pushdown jobs now run successfully on tables with special characters in column names. ### Findings * When exporting findings with Link IDs, the export limit now applies to each rule individually rather than to all rules collectively. Previously, if you set an export limit, the system would take the first N records from all rules combined, which could result in some rules being excluded entirely from the export if earlier rules filled the limit. ### Rules * Rule syntax validation for rules no longer fails when datasets sourced from Edge Information Core (EIC) return zero preview rows. You can now create and modify rules without encountering "No data preview available" errors. The validation process now evaluates SQL logic independently of dataset row count, ensuring consistent rule development across all data sources. * The "Preview Breaks" dialog box now displays columns in the same order as specified in your rule query, rather than alphabetical order. This improvement applies to Databricks, Redshift, and Snowflake connections for both Pushdown and Pullup jobs, including complex queries with CTEs, window functions, and column aliases. * The Rule Builder data preview now loads only when you select the "Data Preview" tab. * The purpose field in data quality rules now behaves consistently across all rule types. Previously, quick rules returned an empty string ("") for the purpose field when no value was provided, while manually created rules returned null. Both rule types now handle empty purpose fields uniformly. ## Patch release ### 2026.04.1 * We improved the security of our Java libraries and PostgreSQL packages. * The DQ Agent pod no longer fails to start in Kubernetes deployments after upgrading to 2026.04.