{"slug": "copy-job-cdc-with-sql-estate-is-now-ga-in-microsoft-fabric", "title": "Copy Job CDC with SQL estate is now GA in Microsoft Fabric", "summary": "Microsoft Fabric Data Factory has made Change Data Capture (CDC) support for SQL estate in Copy Job generally available, enabling teams to keep analytical data aligned with operational systems as business changes occur. The GA release supports CDC for SQL estate sources and destinations, with additional connectors for Fabric Lakehouse, Google BigQuery, Snowflake, Oracle, and Fabric SQL database moving through preview. The update also extends SCD Type 2 support for Fabric Warehouse and Synapse SQL Pool, allowing teams to preserve historical record versions with effective dating and soft delete handling.", "body_md": "Originally published at [https://shai-kr.github.io/data-ninja-ai-lab/blog/2026-05-26-copy-job-cdc-sql-estate-ga.html](https://shai-kr.github.io/data-ninja-ai-lab/blog/2026-05-26-copy-job-cdc-sql-estate-ga.html).\n\nCopy Job CDC with SQL estate is now generally available in Microsoft Fabric Data Factory.\n\nThat sounds like a data movement update.\n\nIt is more useful than that.\n\nFor many BI and data engineering teams, the hard part is not copying data once. The hard part is keeping analytical data aligned with operational systems after the business changes.\n\nA customer address is updated.\n\nAn order status changes.\n\nA subscription is canceled.\n\nInventory moves.\n\nA record is deleted in the source system, but the reporting layer still needs to explain what happened.\n\nThose are not edge cases. They are normal business behavior. If the analytical platform cannot handle them clearly, trust starts to leak out of the reporting layer.\n\nThat is why CDC matters.\n\nMicrosoft announced that **Change Data Capture support for SQL estate in Copy Job is now generally available**.\n\nThe generally available SQL estate sources include:\n\nThe generally available destinations include:\n\nMicrosoft is also moving more CDC connectors through preview, including Fabric Lakehouse table, Google BigQuery, Snowflake, Oracle, SQL database in Fabric, and Fabric Data Warehouse scenarios.\n\nThe important part is not the connector list by itself. The important part is that CDC is becoming a normal Data Factory pattern inside Fabric, not a side script that each team has to invent again.\n\nSource: [Microsoft Fabric Updates Blog: Simplify your data movement with Copy job: CDC with SQL estate](https://community.fabric.microsoft.com/t5/Fabric-Updates-Blog/Simplify-your-data-movement-with-Copy-job-CDC-with-SQL-estate/ba-p/5184211)\n\nFull refresh is simple until it is not.\n\nIt works when the tables are small, the source systems can handle the load, and nobody cares about latency. That changes quickly with operational SQL systems.\n\nCDC lets the pipeline focus on what changed. That can reduce load on the source system, reduce movement volume, and make analytical updates closer to operational reality.\n\nThis is especially useful for tables such as orders, customers, products, subscriptions, transactions, inventory, service tickets, and account status history.\n\nA lot of organizations already have replication logic around their SQL estate. Some of it is mature. Some of it is a set of custom jobs nobody wants to touch.\n\nCopy Job CDC gives Fabric teams a cleaner option for the right workloads.\n\nInstead of maintaining another custom replication layer, a team can move more of the pattern into Fabric Data Factory, where the data movement is visible as part of the platform.\n\nThat does not mean every existing pipeline should be replaced tomorrow. It does mean new Fabric architecture decisions should consider CDC as a first-class option.\n\nFor reporting, the latest value is often not enough.\n\nIf a customer changed region last month, some reports need the current region. Other reports need the region that was true when the order happened.\n\nThat is where slowly changing dimension Type 2 patterns matter.\n\nMicrosoft also highlighted extended SCD Type 2 support for Fabric Warehouse and Synapse SQL Pool. With native SCD Type 2 in Copy Job, teams can preserve historical versions of records with effective dating and soft delete handling.\n\nThat is not just a data warehouse modeling detail. It is the difference between a report that shows the current answer and a report that can explain the historical answer.\n\nDeletes are dangerous in analytics.\n\nIf a source record disappears and the destination simply removes it, the reporting layer may lose the ability to explain prior results.\n\nSoft delete handling is useful because the destination can mark a record as inactive instead of physically deleting it. That keeps the history visible for audit, reconciliation, and operational reporting.\n\nFor finance, subscriptions, customer lifecycle, compliance, and operational analytics, that distinction matters.\n\nThe real value is not that Fabric can copy data from A to B.\n\nTeams have had ways to copy data for years.\n\nThe value is that Fabric is making change capture, history, deletes, latency, and ownership easier to discuss as platform concerns.\n\nThat changes the conversation from:\n\nHow do we move the data?\n\nTo:\n\nHow do we trust the data after it changes?\n\nThat is a better architecture question.\n\nI would look for workloads with these signals:\n\nGood candidates are customer dimension sync, order status tracking, subscription lifecycle reporting, inventory movement, financial transaction replication, and support or case management analytics.\n\nGA does not remove design responsibility.\n\nBefore moving a production workload, I would still define:\n\nCDC makes the movement pattern easier. It does not automatically make the architecture clean.\n\nCopy Job CDC with SQL estate becoming GA is a practical Fabric milestone.\n\nIt gives BI and data engineering teams a stronger native option for moving operational SQL changes into analytical systems, while preserving history and making deletes more traceable.\n\nThe best use of this feature is not to treat it as another ETL checkbox.\n\nUse it to make change history, auditability, and trust explicit in the Fabric architecture.\n\nThat is where the feature starts to matter.\n\n**Shai Karmani**\n\n[Let’s connect on LinkedIn](https://www.linkedin.com/in/shai-kr)", "url": "https://wpnews.pro/news/copy-job-cdc-with-sql-estate-is-now-ga-in-microsoft-fabric", "canonical_source": "https://dev.to/shai_karmani_2521c2f8e837/copy-job-cdc-with-sql-estate-is-now-ga-in-microsoft-fabric-ggb", "published_at": "2026-05-26 23:00:53+00:00", "updated_at": "2026-05-26 23:03:06.993775+00:00", "lang": "en", "topics": ["ai-products", "ai-tools", "ai-infrastructure"], "entities": ["Microsoft Fabric", "Microsoft", "Data Factory", "SQL", "Google BigQuery", "Snowflake", "Oracle"], "alternates": {"html": "https://wpnews.pro/news/copy-job-cdc-with-sql-estate-is-now-ga-in-microsoft-fabric", "markdown": "https://wpnews.pro/news/copy-job-cdc-with-sql-estate-is-now-ga-in-microsoft-fabric.md", "text": "https://wpnews.pro/news/copy-job-cdc-with-sql-estate-is-now-ga-in-microsoft-fabric.txt", "jsonld": "https://wpnews.pro/news/copy-job-cdc-with-sql-estate-is-now-ga-in-microsoft-fabric.jsonld"}}