https://community.databricks.com/t5/data-engineering/cross-region-dr-in-azure-databricks-24h-rpo-rto/m-p/166654#M55609 <P>Your overall design is sensible, but Databricks’ current recommendation is to use <STRONG>Managed Disaster Recovery</STRONG> when your account is eligible. If Managed DR is unavailable, use an <STRONG>active-passive warm standby</STRONG> with Terraform/DABs, cross-region data replication, and a tested failover runbook. Your separate North Europe and West Europe workspaces/metastores are a normal pattern, and both regions are listed as supported for Managed DR.</P><P>The main caution is that a nightly copy alone does not guarantee a 24-hour RPO/RTO. You must also recover pipelines, streaming state, dashboards, models, sharing, networking, and other dependencies that are not automatically replicated</P> Thu, 27 Aug 2026 21:06:44 GMT pradeep_singh 2026-08-27T21:06:44Z https://community.databricks.com/t5/data-engineering/cross-region-dr-in-azure-databricks-24h-rpo-rto/m-p/166285#M55553 <DIV><P>Hi,</P><P>We're designing a DR strategy for an Azure Databricks platform and would appreciate guidance on current best practices for achieving approximately <STRONG>24-hour RPO and RTO</STRONG> across Azure regions.</P><P>Our platform includes Unity Catalog, DAB, Jobs, Notebooks, SQL Warehouses, Lakeflow Declarative Pipelines, Materialized Views, Streaming Tables, Delta Sharing, and related security/governance components.</P><P>Our current approach is:</P><UL><LI>Primary workspace in North Europe</LI><LI>Standby workspace in West Europe</LI><LI>Separate metastores per region</LI><LI>DAB deployment to both workspaces</LI><LI>DR resources kept paused until failover</LI><LI>Nightly replication of business-critical data</LI><LI>Recovery through a combination of replication, replay, and recomputation</LI></UL><P>A few questions:</P><OL><LI>Is a warm-standby workspace the recommended pattern today?</LI><LI>What is the preferred approach for Unity Catalog and metadata across regions?</LI><LI>How are customers handling Lakeflow Pipelines, Streaming Tables, and Materialized Views given the lack of native cross-region replication?</LI><LI>Is replay/recomputation the recommended recovery model for pipeline outputs?</LI><LI>What are common approaches for SQL Warehouses, dashboards, alerts, and region-specific IDs?</LI><LI>Are there reference architectures or proven customer patterns for 24h RPO/RTO on Azure Databricks?</LI><LI>What DR limitations should we account for regarding Managed Volumes, MLflow models, Delta Shares, Vector Search, Genie assets, and similar services?</LI></OL><P>We're interested in both official Databricks guidance and real-world implementations from other customers.</P><P>Thanks in advance.</P></DIV> Mon, 24 Aug 2026 08:53:47 GMT https://community.databricks.com/t5/data-engineering/cross-region-dr-in-azure-databricks-24h-rpo-rto/m-p/166285#M55553 Deny1 2026-08-24T08:53:47Z https://community.databricks.com/t5/data-engineering/cross-region-dr-in-azure-databricks-24h-rpo-rto/m-p/166654#M55609 <P>Your overall design is sensible, but Databricks’ current recommendation is to use <STRONG>Managed Disaster Recovery</STRONG> when your account is eligible. If Managed DR is unavailable, use an <STRONG>active-passive warm standby</STRONG> with Terraform/DABs, cross-region data replication, and a tested failover runbook. Your separate North Europe and West Europe workspaces/metastores are a normal pattern, and both regions are listed as supported for Managed DR.</P><P>The main caution is that a nightly copy alone does not guarantee a 24-hour RPO/RTO. You must also recover pipelines, streaming state, dashboards, models, sharing, networking, and other dependencies that are not automatically replicated</P> Thu, 27 Aug 2026 21:06:44 GMT https://community.databricks.com/t5/data-engineering/cross-region-dr-in-azure-databricks-24h-rpo-rto/m-p/166654#M55609 pradeep_singh 2026-08-27T21:06:44Z