COF-C03 Question 312
Select 2Data replication and failoverA company uses Snowflake Business Critical Edition in AWS us-east-1 for its production account. To meet disaster recovery requirements, the team has configured a secondary account in AWS us-west-2. They need a solution that minimizes recovery time if the primary region becomes unavailable and ensures that important reporting data and access control metadata can be made available in the secondary account. Which TWO actions should the architect implement?
- A
Create a failover group in the primary account that includes the required databases and account objects, replicate it to the secondary account, and plan to promote the secondary account during a regional outage.
- B
Configure database replication only for the reporting databases because users, roles, and warehouses are automatically recreated in the secondary account during failover.
- C
Use a reader account in us-west-2 as the disaster recovery target because reader accounts can be promoted to full standalone accounts during failover.
- D
Create a replication group for the required databases and account objects, replicate it to the secondary account, and use database failover/failback operations as needed.
- E
Schedule replication refreshes for the configured replication or failover group so the secondary account has a recent copy when a failover event occurs.
Show answer and explanation
Correct answers: A, E
Explanation
For disaster recovery across regions/accounts in Snowflake, the architect should use a failover group when the requirement includes both replicated data and supported account objects needed to resume operations in another account. Failover groups are intended for business continuity scenarios and support failover/failback workflows. By contrast, database replication alone protects database contents but does not provide full account-level recovery for users, roles, and other supported account objects. Another key operational requirement is regularly refreshing the replicated objects so the secondary remains current enough to meet the organization's RPO. These recommendations align with Snowflake documentation on replication groups, failover groups, cross-region/cross-cloud replication, and account failover best practices.
- A. Correct.
Correct. A failover group is the Snowflake object designed for cross-region and cross-account business continuity. It can include databases and certain account-level objects such as users, roles, and other supported account objects, which is important when access control metadata must also be available in the secondary account. In a disaster event, the secondary can be promoted so workloads can resume there with minimal recovery time compared with rebuilding metadata manually.
- B. Incorrect.
Incorrect. Database replication alone does not automatically recreate all account-level security and governance objects such as users and roles in the target account. This option reflects a common misconception that database replication is sufficient for full disaster recovery. It may protect data, but it does not by itself provide complete account-level continuity for access management.
- C. Incorrect.
Incorrect. Reader accounts are intended to let non-Snowflake customers consume shared data. They are not the standard disaster recovery target for account failover, and they cannot be promoted into a normal standalone production account for regional failover. This distractor targets confusion between data sharing and disaster recovery features.
- D. Incorrect.
Incorrect. Replication groups are used for replication of databases and supported account objects, but failover groups are the specific construct used to support failover and failback scenarios between accounts. Saying to use a replication group and then perform failover/failback operations mixes two related but distinct features and is not the best answer for minimizing recovery time during a regional outage.
- E. Correct.
Correct. Replication in Snowflake is not continuous streaming replication by default; the secondary copy is updated through refresh operations. Scheduling refreshes for the failover group ensures the secondary account has a reasonably current copy of the data and supported metadata, which directly affects recovery point objectives during an outage.