HCVA0-003 practice question 9 of 35

During a service outage, you must ensure all current tokens and leases are copied to another Vault cluster for failover so applications don’t need to authenticate. How can you accomplish this?

Full set: 285 questions
Answer options
A: Have Vault write all the tokens and leases to a file so you have a second copy of them
B: Configure all applications to use the auto-auth feature of the Vault Agent
C: Configure Disaster Recovery replication and promote the secondary cluster during an outage Correct answer
D: Replicate to another cluster using Performance Replication and promote the secondary cluster during an outage
Correct Answer:
C. Configure Disaster Recovery replication and promote the secondary cluster during an outage
Explanation

Disaster Recovery (DR) replication is the Vault Enterprise feature designed for "hot standby" failover scenarios. It replicates nearly all data from a primary to a secondary cluster, including authentication tokens, leases, and policies. During an outage, the secondary DR cluster can be promoted to become the new primary. Because the tokens and leases have been synchronized, applications can continue their operations against the newly promoted cluster without needing to re-authenticate. This provides a seamless failover experience, which is the core requirement of the question.

Why Incorrect

A. Writing tokens and leases to a file is an insecure manual process, not a built-in, automated high-availability or disaster recovery solution provided by Vault.
B. The Vault Agent's auto-auth feature simplifies re-authentication for applications, but it does not replicate existing tokens or leases to a failover cluster.
D. Performance Replication is designed for scaling read-heavy workloads and explicitly does not replicate transient data like tokens and leases to secondary clusters.

References

1. HashiCorp Vault Official Documentation, "Disaster Recovery Replication": "Disaster Recovery (DR) Replication is designed for a "hot standby" scenario. It replicates all data from a primary to a secondary cluster. If the primary cluster fails, the secondary can be promoted to take its place." This confirms that DR replication is the intended mechanism for this scenario.

2. HashiCorp Vault Official Documentation, "Performance Replication": "Performance replication does not replicate ephemeral data such as tokens and leases. This means that clients cannot use a token generated on a primary cluster with a performance secondary cluster." This directly invalidates option D.

3. HashiCorp Vault Official Documentation, "Replication" (Model Comparison Table): The table comparing replication models clearly states that for "Tokens and Leases," Disaster Recovery replication provides "Full Sync," while Performance Replication provides "None." This highlights the fundamental difference and supports the choice of C over D.

Scroll to Top