COF-C03 Question 194
Single answerData replication and failoverA company runs a business-critical analytics application in Snowflake on AWS us-east-1. To meet disaster recovery requirements, the team has configured cross-region replication to a secondary account in AWS us-west-2. During a regional outage affecting us-east-1, analysts must resume access to the replicated data in the secondary account as quickly as possible with minimal administrative effort. Which action should the Snowflake administrator take?
- A
Refresh the replication schedule in the primary account so the secondary account can automatically become read/write
- B
Promote the replicated database in the secondary account so it becomes the primary, writable database
- C
Create database clones in the secondary account from the replicated database and point users to the clones
- D
Enable Time Travel in the secondary account to convert the replicated database into an active database
Show answer and explanation
Correct answer: B
Explanation
Snowflake supports replication and failover for disaster recovery across regions and, where supported, across cloud platforms. In normal operation, the secondary replicated database is read-only. If the primary region becomes unavailable, the correct operational step is to promote or fail over the secondary database so it becomes the new primary and accepts read/write workloads. Candidates should distinguish replication refresh from failover: replication keeps the secondary copy up to date, while failover/promotion changes roles so business operations can continue. This aligns with Snowflake documentation on database replication and failover/failback, which emphasizes that secondaries remain read-only until explicitly promoted.
- A. Incorrect.
Incorrect. Refreshing replication updates the replicated objects with the latest available changes from the primary, but it does not automatically convert a secondary replicated database into a writable primary database. During an outage, the administrator must explicitly perform failover or promotion actions on the replicated object in the target account/region.
- B. Correct.
Correct. In a disaster recovery scenario, the administrator should promote the replicated database in the secondary account so it becomes the primary, writable database. This is the key failover action that allows applications and users to resume work against the secondary environment. Snowflake replication and failover are designed for exactly this use case: maintaining a read-only secondary copy until promotion is needed.
- C. Incorrect.
Incorrect. A replicated secondary database is read-only, and while cloning is a common Snowflake feature, it is not the standard failover mechanism for turning a replicated disaster recovery copy into the active production database. The supported DR workflow is to promote/fail over the replicated database, not to rely on manual clone-based workarounds during an outage.
- D. Incorrect.
Incorrect. Time Travel provides access to historical data states within retention limits, but it does not change the replication role or convert a secondary replicated database into a writable primary database. This option confuses data recovery capabilities with disaster recovery/failover capabilities.