Version 26.11.0

Date: 14th November 2026

Data Reliability

• Score Contribution Indicator in Policy Execution History: In the policy execution history, runs that are excluded from the Overall Reliability Score calculation now show a Not Scored badge, and the summary of each run shows Yes or No for Include in Overall Reliability Score. Inclusion depends on the Include in Overall Reliability Score calculation? setting of the policy, and a change to this setting applies from the next execution. For more information, see View Policy Execution and Results.
• List View for Rules in Policy Execution Summary: The execution summary of a policy now lets you view rules in a list view in addition to the existing card view. Use the list view to find rules faster in policies that contain many rules. For more information, see View Policy Execution and Results.
• Filter Policies by Execution Date:The date filter on the Manage Policies page now lets you filter policies by Executed Between in addition to Modified Between. Use Executed Between to find the policies that ran in a selected period, regardless of when their configuration was last modified. For more information, see Manage Policies.
• Spark App Name in Policy Jobs: The Jobs page now includes a Spark App Name column for Data Quality, Data Freshness, and Reconciliation policy jobs. Copy the application name to locate the Spark driver and executor pods of a job in Kubernetes when you troubleshoot it. For more information, see Monitoring and Managing Jobs.

Data Sources

• Revamped Data Source Pages: ADOC introduces a redesigned data source listing page and a new data source details page. The details page brings together the configuration and schedules of a data source in one place.

Discover Assets

• Asset Disambiguation in Search Results: The Discover Assets page now displays the hierarchical context of each asset next to its name in search results, so you can distinguish assets that share the same name without opening each one. For more information, see Discover Assets.
• Column Tooltips on the Discover Assets Page: The 30-Day Update Count and 30-Day Profile Anomalies columns on the Discover Assets page now display tooltips that describe what each column measures. For more information, see Discover Assets.
• Owner and Team Columns on the Discover Assets Page: You can now add the Owner and Team columns to the asset table on the Discover Assets page. Use these columns to review asset ownership across the list without opening the details page of each asset. For more information, see Discover Assets.

Incident Management and Notifications

• Incident Management Enhancements: ADOC now lets you filter incidents by asset tags and policy tags, group incidents by incident policy, and record external reference tickets, such as ServiceNow or Jira tickets, on an incident. You can also define what an Incident Policy monitors by tag instead of selecting each asset or policy individually. For more information, see Incident Management.
• Custom Email Headers in Notification Templates: You can now add custom headers to email notification templates, such as X-ICCategory, so that your mail system can categorize, route, and filter ADOC notification emails. Notification emails don’t include custom headers unless you add them. For more information, see Working With Notification Templates.

Pipeline Observability

• Generic OpenLineage Connector: ADOC now accepts OpenLineage run events from any OpenLineage-compliant producer, including orchestration tools that ADOC doesn't integrate with natively and custom applications. ADOC creates pipelines, runs, jobs, datasets, and lineage from the events it receives, and correlates the datasets with your existing assets. For more information, see OpenLineage Integration.

Data Products

• Insights Tab for Data Products: ADOC now lets you add a dashboard to the Insights tab of a Data Product. Users who open the Data Product can view the dashboard on the Data Product details page instead of opening it separately. For more information, see Data Products Detail Page.

Agentic Data Management (ADM)

• File Uploads in Conversations and Knowledge Base Improvements: You can now attach files to an ADM conversation, up to 10 files of 20 MB each, and choose whether ADM searches all Knowledge Base files, selected files, or none. ADM can now filter, total, and compare rows in tables inside documents, and its answers list the files they used, which you can preview. Uploading and deleting Knowledge Base files now requires the ADM Configuration Modify permission. For more information, see Knowledge Base.
• Workspace Memory: ADM can now remember facts about your organization, such as pipeline owners, business terms, or escalation channels, and apply them in every conversation in your tenant. You save a memory from the lightbulb icon under an answer, by asking ADM, or from Settings > ADM Settings > Workspace Memory, and you confirm each memory before ADM saves it. Workspace Memory is off by default; contact Acceldata Support to turn it on.  For more information, see Core Concepts.
• Generative Dashboards: ADM now builds dashboards from a plain-language request, such as a data quality board for the last 30 days, and saves each change that you ask for as a version. Published dashboards appear under Dashboards for everyone in the workspace, load data live, and show each viewer only the data they have access to. Generative dashboards are off by default; contact Acceldata Support to turn them on. For more information, see Generative Dashboards.
• ADM Actions: ADM can now make changes in ADOC, such as creating, scheduling, or archiving policies, running profiling, applying tags, and updating data products. ADM prepares each change as a card that you review and run, and every change runs with your own permissions. Actions is off by default; contact Acceldata Support to turn it on. For more information, see ADM Actions.
• Turn Individual MCP Tools On or Off: Administrators can now turn individual tools of an MCP server on or off in Settings > ADM Settings > MCP Servers, instead of only the whole server. ADM doesn’t use a turned-off tool for any user in the tenant. For more information, see Understanding MCP Server.
• Policy Settings in Data Quality Policies Created from Chat: Data quality policies that you create in an ADM conversation now apply the settings in your request, including SQL-based checks that join or aggregate across tables, the engine and SQL dialect, and where to store good or bad records. ADM validates SQL queries before it saves the policy. For more information, see Data Quality Management.
• Knowledge Base Search in AI Analyze Copilots: The AI Analyze copilots now accept file attachments and search the Knowledge Base when they answer.

APIs

• OpenAPI Specifications for ADOC Public APIs: The ADOC API reference now includes a Notifications API for managing notification channel groups. The Catalog API adds endpoints for bulk policies (rule sets), exporting and importing catalog entities, and managing data reliability policies in bulk. You can download each API’s OpenAPI specification in JSON format and import it into OpenAPI-compatible tools, such as API clients, code generators, and mock servers.

Pipelines & Lineage

• Resolved an issue where inferred lineage links between duplicate nodes were incorrectly resolved across pipeline runs.
• Resolved an issue where the scheduleInterval parameter was omitted from the Pipeline API response, preventing start-time monitoring policy configuration.
• Resolved an issue where scheduled pipeline execution failure email summaries sent at 6:00 AM UTC reported inaccurate pipeline failure counts and statuses.

Data Reliability (Data Quality & Profiling)

• Resolved an issue where Pushdown Data Quality failed-record filtering returned zero records due to Parquet column-name casing mismatches.
• Resolved an issue where Auto Anomaly policies failed with a JDBC timestamp conversion error when partition configuration was applied to timestamp columns in Databricks.
• Resolved an issue where policy execution status was displayed inconsistently between the Manage Policies view and the Policy Execution Details view.
• Resolved an issue where data profiling jobs failed on BigQuery tables containing wide schemas exceeding 900 columns.
• Resolved an issue where columns in tables, such as the Policies and Rules tables on the Manage Policies page, disappeared or were hard to read when you scrolled horizontally. The Name column now stays fixed while you scroll to view other columns.

Dashboards & Reporting

• Resolved an issue where the Policy Alert Profile widget failed to load on policy execution dashboards.
• Resolved an issue where data quality reliability scores were omitted from dashboard widgets when using Rule-Based Row-Weighted Averaging.

Connectors & Crawlers

• Resolved high latency and timeouts in the catalog service that caused SDK API calls and Databricks crawler jobs to fail.
• Resolved an AWS S3 crawler failure where keys containing slash (/) characters caused a Jedis null-pointer exception on V3 Dataplanes.
• Resolved an issue where Databricks crawler lineage queries ignored configured asset inclusion and exclusion filters and scanned the entire catalog lineage history.
• Resolved a connection failure when testing Redshift Serverless datasources with IAM/IRSA authentication due to a missing AWS SDK client dependency.

Platform, Catalog & UI

• Fixed an issue where Collibra showed N/A or outdated policy scores after a policy run, because ADOC notified data-catalog integrations before the run’s results were saved.
• Resolved an issue in Reliability Explorer where the tag filter dropdown excluded newly created tags that were not yet linked to assets.
• Resolved an issue where the Tags API returned only the first 1,000 records, omitting subsequent tags from selection menus.
• Added descriptive error messaging when large filter selections exceed HTTP request header thresholds.
• Added Kubernetes liveness and readiness health probes to YuniKorn scheduler pods to prevent unrecovered scheduling outages.

ADM

• Resolved an issue where ADM included runs still in progress when it calculated average pipeline run durations. Average durations now include only completed runs.
• Resolved an issue where ADM listed the same pipeline dependency more than once. Each dependency now appears once.
• Resolved an issue where ADM answers about data quality incidents didn’t show the policy, run, and quality score. These details now appear in the answer.
• Resolved an issue where ADM showed an internal user ID as the creator of a policy. ADM now shows the creator’s name.
• Resolved an issue where ADM answers, such as lists of schema drift policies, contained broken links. Links in answers now open the correct page.
• Resolved an issue where Pipeline AI Analyze didn’t work when ADM was turned off and the pipeline copilot was turned on. Pipeline AI Analyze now works in this configuration.