A manufacturing company's website uses an Amazon Aurora PostgreSQL DB cluster. Which configurations will result in the LEAST application downtime during a failover? (Choose three.)

The correct answer is A. Use the provided read and write Aurora endpoints to establish a connection to the Aurora DB C. Edit and enable Aurora DB cluster cache management in parameter groups. E. Set JDBC connection string timeout variables to a low value.. Aurora PostgreSQL Failover Minimization Using Aurora-provided endpoints (A) ensures automatic DNS rerouting to the new primary instance during failover without manual intervention. Aurora DB cluster cache management (C) warms the cache on the replica before promotion, so the new

Submitted by suresh_in· Mar 6, 2026Management and Operations

Question

Options

AUse the provided read and write Aurora endpoints to establish a connection to the Aurora DB
BCreate an Amazon CloudWatch alert triggering a restore in another Availability Zone when the
CEdit and enable Aurora DB cluster cache management in parameter groups.
DSet TCP keepalive parameters to a high value.
ESet JDBC connection string timeout variables to a low value.
FSet Java DNS caching timeouts to a high value.

How the community answered

(50 responses)

A
78% (39)
B
6% (3)
D
12% (6)
F
4% (2)

Explanation

Aurora PostgreSQL Failover Minimization

Using Aurora-provided endpoints (A) ensures automatic DNS rerouting to the new primary instance during failover without manual intervention. Aurora DB cluster cache management (C) warms the cache on the replica before promotion, so the new primary doesn't suffer a "cold cache" performance penalty that would cause application slowdowns. Setting JDBC timeout variables to a low value (E) ensures the application quickly detects and abandons failed connections, reconnecting to the new primary faster rather than waiting through long timeout periods.

Why the distractors are wrong:

B is incorrect because CloudWatch alerts triggering manual restores introduce significant delay - Aurora handles failover automatically without this
D is wrong because high TCP keepalive values mean the system waits longer to detect dead connections, increasing downtime
F is wrong because high Java DNS caching (TTL) means the application caches the old IP address longer, delaying discovery of the new primary endpoint

Memory Tip

Think "Fast Fail, Warm Cache, Smart Endpoint" - you want Aurora to route smartly (A), stay warm (C), and fail fast (E). For timeouts and caching, remember the rule: DNS TTL and TCP keepalives should be LOW to detect failures quickly, not high.

Topics

#Aurora Failover#High Availability#Connection Management#Cluster Cache Management

Community Discussion

No community discussion yet for this question.

Full DBS-C01 Practice