Amazon Web Services SOA-C03 Practice Test Questions Answers

Amazon Aurora backtracking allows a DB cluster to be rewound to a specific point in time without creating a new DB cluster. This feature is designed for fast recovery from logical errors, such as accidental data changes, within a configured backtrack window. Because backtracking operates directly on the existing cluster, it satisfies the requirement that the restore occur in the same production DB cluster.

Point-in-time recovery (Option D) restores data by creating a new DB cluster, which violates the requirement. Option A involves promoting a replica, which does not allow rolling back to an arbitrary historical point. Option B introduces unnecessary complexity and is not supported for restoring directly into the same cluster.

Backtracking provides near-instant rollback and minimal operational disruption, making it the correct solution.

S3 Lifecycle policies are the correct managed mechanism for automatically expiring objects after a defined age. Because the company never accesses data older than 30 days, an expiration rule can delete objects after 30 days and reduce storage growth without custom code. This has the least operational overhead because S3 performs the expiration automatically. Option A introduces Lambda code, scheduling, permissions, retries, and edge-case handling, which is unnecessary. S3 Storage Lens can identify usage trends and aging objects, but it does not itself delete objects or enforce retention. Option C is vague and not an AWS-managed configuration as written. For CloudOps cost optimization, lifecycle expiration is the standard approach when objects have a predictable retention period and no long-term access requirement.

===============

Per the AWS Cloud Operations and S3 Data Management documentation, Cross-Region Replication (CRR) automatically replicates new objects between S3 buckets across Regions. However, CRR alone does not retroactively replicate existing objects created before replication configuration. To include such objects, AWS introduced S3 Batch Replication.

S3 Batch Replication scans the source bucket and replicates all existing objects that were not copied previously. Additionally, it can retry failed replication tasks automatically, ensuring regulatory compliance for complete dataset replication.

S3 Replication Time Control (S3 RTC) guarantees predictable replication times for new objects only—it does not cover previously stored data. S3 Lifecycle rules (Option D) move or transition objects between storage classes or buckets, but not in a replication context.

Therefore, the correct solution is to use S3 Cross-Region Replication (CRR) combined with S3 Batch Replication to ensure all current and future data is synchronized across Regions with retry capability.

As per AWS Cloud Operations, Bedrock, and Governance documentation, AWS CloudTrail is the authoritative service for capturing API activity and audit trails across AWS accounts. For Amazon Bedrock, CloudTrail records all user-initiated API calls, including interactions with agents, knowledge bases, and generative AI model parameters.

Using CloudTrail Lake, organizations can store, query, and analyze CloudTrail events directly without needing to export data. CloudTrail Lake supports SQL-like queries for generating audit and compliance reports, enabling the company to retrieve information such as user identity, API usage, timestamp, model or agent ID, and invocation parameters.

In contrast, CloudWatch focuses on operational metrics and log streaming, not API-level identity data. OpenSearch or Flink would add unnecessary complexity and cost for this use case.

Thus, the AWS-recommended CloudOps best practice is to leverage CloudTrail with CloudTrail Lake to maintain auditable, queryable API activity for Bedrock workloads, fulfilling governance and compliance requirements.

Amazon S3 event notifications are the native mechanism for invoking processing when objects are uploaded to an S3 bucket. The CloudOps engineer can configure the bucket to send ObjectCreated events to an AWS Lambda function, and the Lambda function can process each new log file by using the bucket name and object key from the event payload. This is serverless, event-driven, and has the least operational overhead. CodePipeline is intended for software delivery workflows, not generic log-file processing. Step Functions is useful for orchestrating multi-step workflows, but polling or waiting for S3 uploads is unnecessary here. EventBridge can receive some S3 events, but the direct S3 event notification to Lambda is simpler and more targeted for this requirement.