SAP-C02 Continuous Improvement for Existing Solutions Practice Test 19 — 15 Questions

Question 1

A company runs a production database on an Amazon RDS for MySQL instance with a single Availability Zone. The database is critical and experiences occasional read-heavy workloads. The company wants to improve read scalability and ensure high availability with minimal application changes. Which solution meets these requirements?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Enable Multi-AZ deployment and create a read replica in another Availability Zone, then update the application to use the reader endpoint.. Enabling Multi-AZ provides automatic failover for high availability, while a read replica in another AZ offloads read queries and improves scalability. Updating the application to use the reader endpoint is a minimal change. Other options either lack high availability, require major migration, or do not address read scaling.

Answer

Migrate the database to Amazon Aurora MySQL with a single writer and multiple readers, and update the application to use the reader endpoint.

Answer

Increase the instance size of the RDS instance to a larger instance type to handle read traffic.

Answer

Create a read replica in a different Availability Zone and configure the application to direct read queries to the replica endpoint.

Question 2

A company runs a legacy three-tier web application on a set of Amazon EC2 instances behind an Application Load Balancer. The application writes session state to local disk on each instance, so users are logged out whenever an instance is replaced during a scaling event. The company wants to remove this limitation with the least operational overhead and no application code changes. Which solution meets these requirements?

Accepted Answer

Mount an Amazon EFS file system on all instances and relocate the session directory to the mounted path with no code changes.. The application writes sessions to a local directory, so the lowest-effort fix is to make that directory a shared, durable file system that every instance can mount. Amazon EFS delivers a managed, multi-AZ POSIX file system, so relocating the session path preserves existing application behavior while surviving instance replacement. Sticky sessions do not persist data, S3 is not a POSIX store, and Redis would require application changes.

Answer

Copy the session directory to an Amazon S3 bucket on a schedule and restore it when a new instance launches.

Answer

Store session state in an Amazon ElastiCache for Redis cluster and point the application at it through a configuration change.

Answer

Enable stickiness on the Application Load Balancer target group so each user is pinned to the instance that created their session.

Question 3

A company has a legacy application running on a single Amazon EC2 instance with an attached Amazon EBS volume. The application writes logs to a local directory on the EBS volume. The company wants to improve the durability and availability of the logs by shipping them to Amazon S3 in near-real-time. The logs are continuously written and can grow up to 100 GB per day. The solution must minimize operational overhead and ensure that logs are delivered even if the EC2 instance is terminated. Which approach should a solutions architect recommend?

Accepted Answer

Install and configure the Amazon Kinesis Agent on the EC2 instance to stream logs to Amazon Kinesis Data Firehose, which delivers to Amazon S3.. The Amazon Kinesis Agent streams log data to Kinesis Data Firehose, which reliably delivers to Amazon S3 with automatic scaling and buffering. This managed service minimizes operational overhead and ensures logs are delivered even if the EC2 instance is terminated, because Firehose retains data until delivery. It supports near-real-time ingestion and is ideal for continuous log data without application changes.

Answer

Use AWS DataSync to replicate the log directory from the EC2 instance to an Amazon S3 bucket on a scheduled basis.

Answer

Configure the application to write logs directly to Amazon CloudWatch Logs, and then use a CloudWatch Logs subscription to stream to Amazon S3.

Answer

Create a cron job that runs every 5 minutes to copy new log files to Amazon S3 using the AWS CLI.

Question 4

A company is migrating a legacy on-premises application to AWS. The application writes log files to a local directory and requires a shared file system that supports the NFSv4 protocol and Windows ACLs. The company wants to minimize operational overhead and ensure the file system is highly available across multiple Availability Zones. Which two solutions should a solutions architect recommend? (Choose two.)

Accepted Answer

Create an Amazon FSx for Windows File Server file system and mount it from the application instances.. The application requires a shared file system with NFSv4 and Windows ACLs, minimal operational overhead, and high availability across AZs. Amazon EFS provides a fully managed NFSv4 file system with Multi-AZ durability. Amazon FSx for Windows File Server provides a fully managed file system that supports NFSv4 and native Windows ACLs with Multi-AZ deployment. Both eliminate the need to manage file servers.

Answer

Set up a Linux NFS server on an EC2 instance in an Auto Scaling group and mount it from the application instances.

Answer

Use Amazon S3 with the Mountpoint for Amazon S3 client to mount the bucket as a file system.

Answer

Deploy an AWS Storage Gateway File Gateway and present an SMB share to the application.

Question 5

A company has a microservices application running on Amazon ECS with the Fargate launch type. The application uses an Application Load Balancer (ALB) to distribute traffic. The operations team notices that during deployments, some requests fail with HTTP 503 errors. The deployment configuration is set to rolling update with a minimum healthy percent of 100 and a maximum percent of 200. The tasks have a health check endpoint that returns 200 only after the application has fully initialized, which takes about 30 seconds. What should a solutions architect recommend to eliminate the 503 errors during deployments?

Accepted Answer

Enable the ECS task health check and set the health check grace period appropriately.. The 503 errors occur because new tasks are registered with the ALB before the application is fully initialized. ECS task health checks, when configured, ensure that a task is only considered healthy and registered with the load balancer after it passes the health check. Setting an appropriate health check grace period allows the application time to start. This prevents traffic from being routed to unready tasks, eliminating 503 errors.

Answer

Configure a longer deregistration delay on the target group.

Answer

Increase the health check interval and timeout on the target group.

Answer

Use the ECS deployment circuit breaker with rollback.

Question 6

A company runs a critical application on Amazon EC2 instances in an Auto Scaling group. The application writes logs to local disk and also sends custom metrics to Amazon CloudWatch. The operations team wants to detect and automatically remediate instances that are not sending metrics for more than 5 minutes. The remediation should terminate the instance and allow the Auto Scaling group to replace it. Which solution meets these requirements with the LEAST operational overhead?

Accepted Answer

Create a CloudWatch alarm based on the metric's missing data points, set the alarm to treat missing data as breaching, and configure the alarm to trigger an EC2 action that terminates the instance.. CloudWatch alarms can monitor custom metrics and trigger EC2 actions. By treating missing data as breaching, the alarm will fire when no metrics are received for the specified period. Configuring the alarm to terminate the instance automates remediation with minimal overhead. Other options either do not detect missing metrics or require manual intervention or custom code.

Answer

Create an Amazon EventBridge rule that matches the 'EC2 Instance State-change Notification' event for stopped instances and triggers a Lambda function to terminate the instance.

Answer

Configure a CloudWatch Logs subscription filter to detect the absence of log entries and trigger a Lambda function to terminate the instance.

Answer

Use AWS Systems Manager OpsCenter to create an OpsItem for each instance that fails to send metrics, and manually terminate the instance from the OpsCenter console.

Question 7

A company runs a stateful legacy application on a single Amazon EC2 instance with an attached Amazon EBS volume. The application writes transaction logs continuously. The CIO wants to improve the recovery time objective (RTO) and recovery point objective (RPO) without re-architecting the application. The solutions architect proposes using Amazon EBS snapshots and a scripted recovery process. Which two actions should the architect take to meet a 15-minute RPO and a 30-minute RTO? (Choose two.)

Accepted Answer

Schedule EBS snapshots every 15 minutes using Amazon Data Lifecycle Manager or a cron job with the AWS CLI.. Frequent EBS snapshots every 15 minutes satisfy the RPO, and a pre-provisioned launch template plus restoring the latest snapshot enables a fast, scripted recovery within 30 minutes. AMI creation per snapshot, Multi-Attach, and cross-Region copies do not address the required recovery objectives efficiently or safely.

Answer

Create an Amazon Machine Image (AMI) of the instance after each snapshot and use it to launch a replacement instance.

Answer

Use AWS Backup to copy snapshots to a vault in another Region and restore from that vault during a disaster.

Answer

Enable EBS Multi-Attach on the volume and attach it to a second instance in another Availability Zone for automatic failover.

Question 8

A company runs a latency-sensitive gRPC service on an Auto Scaling group of EC2 instances behind a Network Load Balancer. The service uses a custom TCP health check on port 8080. During deployments, instances are terminated while they still have long-lived gRPC streams, causing client errors. The team wants zero-downtime deployments and faster scale-in. Which solution best meets these requirements?

Accepted Answer

Use an Application Load Balancer with gRPC support, enable connection draining, and implement a graceful shutdown handler in the service that stops accepting new RPCs, finishes active streams, and then exits after the ALB deregistration delay.. gRPC requires HTTP/2, which Network Load Balancers do not terminate at Layer 7, so they cannot manage gRPC streams gracefully. An Application Load Balancer supports gRPC and provides a deregistration delay that keeps targets in service until in-flight requests finish. Pairing that with an application-level graceful shutdown handler ensures active streams complete before the process exits, eliminating client errors during deployments and allowing faster scale-in.

Answer

Configure the NLB target group to use HTTP health checks on a /health endpoint that reports unhealthy when the instance is marked for termination, and rely on the load balancer to drain connections.

Answer

Attach an Elastic Load Balancing connection idle timeout of 60 seconds on the NLB and set the Auto Scaling group termination policy to OldestInstance.

Answer

Enable connection draining on the NLB with a 300-second deregistration delay, and add an EC2 Auto Scaling lifecycle hook on termination that waits for the same duration.

Question 9

A company uses AWS Lambda functions to process messages from an Amazon SQS queue. The functions occasionally fail due to downstream service throttling, and messages are being lost after the maximum number of receives. The company wants to ensure that failed messages are not lost and can be reprocessed later, while also reducing the number of retries for non-retryable errors. Which solution meets these requirements with the LEAST development effort?

Accepted Answer

Configure a dead-letter queue (DLQ) on the source SQS queue and set the maximumReceiveCount to a low value.. A dead-letter queue on the source SQS queue automatically captures messages that exceed the maximum receive count, ensuring they are not lost and can be reprocessed. Adjusting the maximumReceiveCount to a lower value reduces the number of retries for messages that are likely non-retryable, such as those causing persistent errors. This solution requires no code changes and leverages native SQS features.

Answer

Modify the Lambda function code to catch exceptions and send failed messages to a separate SQS queue.

Answer

Use an SQS FIFO queue instead of a standard queue to ensure exactly-once processing.

Answer

Increase the Lambda function's reserved concurrency and timeout to prevent throttling.

Question 10

A solutions architect is optimizing a batch processing workload that runs on AWS Lambda functions. The functions process large JSON files from Amazon S3, each up to 5 GB, and currently use the AWS SDK for Java to download the entire file into memory before parsing. The functions have a memory allocation of 1024 MB and often time out after 15 minutes. The architect needs to reduce execution time and cost without changing the output. Which approach is MOST effective?

Accepted Answer

Use S3 Select to filter and retrieve only the necessary data from the JSON files, and process the smaller result set.. S3 Select pushes down filtering to the S3 storage layer, so the Lambda function retrieves only the data it needs instead of the entire 5 GB object. This reduces network transfer, memory pressure, and processing time. The other options either increase cost, do not address the root cause, or require significant migration effort without solving the inefficiency.

Answer

Convert the Lambda function to run on AWS Fargate with a larger task size and use the same code.

Answer

Enable Lambda provisioned concurrency to reduce cold start latency and improve throughput.

Answer

Increase the Lambda function memory to 3008 MB and extend the timeout to 15 minutes.

Question 11

A company runs a microservices application on Amazon ECS using the Fargate launch type. The services communicate over HTTP and are deployed across multiple Availability Zones. The company wants to improve the existing solution to reduce inter-service latency and improve fault tolerance. The services are registered with AWS Cloud Map. Which two actions should be taken to meet these requirements? (Choose two.)

Accepted Answer

Enable ECS service discovery by configuring the services to use AWS Cloud Map for service registration and DNS-based discovery.. To reduce inter-service latency and improve fault tolerance, ECS services should use awsvpc network mode and be deployed across multiple Availability Zones. AWS Cloud Map service discovery enables direct task-to-task communication via DNS, eliminating the need for load balancers and reducing latency. Together, these actions provide a resilient, low-latency microservices architecture. Other options either add latency, are unsupported, or do not address the requirements.

Answer

Use Amazon CloudFront with the ECS services as origins to cache responses and reduce latency.

Answer

Place an Application Load Balancer in front of each ECS service to distribute traffic and improve fault tolerance.

Answer

Configure the ECS tasks to use the host network mode to avoid network address translation overhead.

Question 12

A company operates a REST API on Amazon API Gateway backed by AWS Lambda functions. During monthly financial close, request volume increases tenfold for several hours. The Lambda functions currently use the default concurrency settings, and the operations team observes throttling errors returned to clients. The company wants to absorb the burst without over-provisioning capacity year-round and without changing the API clients. Which action should a solutions architect recommend?

Accepted Answer

Configure provisioned concurrency on the Lambda functions for the financial close window using a scheduled scaling schedule.. The throttling stems from Lambda concurrency limits being reached when tenfold traffic arrives. Provisioned concurrency reserves initialized execution environments for a specified period, so requests are served without cold starts and without exhausting the on-demand pool. Applying a scheduled scaling configuration to the alias during the close window aligns capacity with the temporary spike, and API Gateway continues to invoke the same alias, so no client changes are required.

Answer

Increase the Lambda function timeout and memory allocation so each invocation completes faster during the close period.

Answer

Raise the account-level Lambda concurrency quota and remove any reserved concurrency so the functions can scale to the surge.

Answer

Enable API Gateway caching on the stage so repeated requests are served from the cache instead of invoking Lambda.

Question 13

A company has a two-tier application with a web tier on Amazon EC2 and a database tier on Amazon RDS for MySQL. The application experiences performance degradation during peak hours, and monitoring shows high read IOPS on the database. The database is a Single-AZ deployment with 500 GB of General Purpose SSD (gp2) storage. The company wants to improve read performance and reduce latency without changing the application code. What should a solutions architect recommend?

Accepted Answer

Change the storage type to Provisioned IOPS SSD (io1) with a higher IOPS value.. The database is experiencing high read IOPS, and the current gp2 storage may not provide sufficient IOPS. Switching to io1 with a higher provisioned IOPS value can deliver the needed performance without application changes. Other options either require code changes, do not improve read performance, or are more complex migrations.

Answer

Create a read replica of the RDS instance and configure the application to use it for read queries.

Answer

Enable Multi-AZ for the RDS instance.

Answer

Migrate the database to Amazon Aurora MySQL with a reader endpoint.

Question 14

A company runs a stateful web application on Amazon EC2 instances in an Auto Scaling group. The application stores session data locally on each instance. During scaling events, users are logged out because their sessions are lost. The company wants to make the application stateless without modifying the application code, and needs to minimize latency for session access. Which solution should a solutions architect recommend?

Accepted Answer

Use an Application Load Balancer with a target group that routes to a fleet of EC2 instances, and configure the instances to use Amazon ElastiCache for Redis as the session store via the application's existing session handler configuration.. To make a stateful application stateless without code changes, session data must be externalized to a shared store. ElastiCache for Redis offers sub-millisecond latency and can be configured via standard session handler settings (e.g., PHP's session.save_handler and session.save_path) without altering application logic. This allows any instance in the Auto Scaling group to access the session, eliminating user logouts during scaling.

Answer

Enable sticky sessions on the Application Load Balancer and increase the session timeout.

Answer

Store session data in an Amazon ElastiCache for Redis cluster and configure the application to use it via a configuration file.

Answer

Use an Application Load Balancer with a target group that routes to a fleet of EC2 instances, and configure the instances to store sessions in Amazon DynamoDB via the AWS SDK for PHP session handler.

Question 15

A company runs a stateless web application on an Auto Scaling group of EC2 instances behind an Application Load Balancer. The application retrieves secrets from AWS Secrets Manager at startup. The security team wants to reduce the blast radius of compromised credentials and improve the reliability of secret retrieval. Which two actions should a solutions architect take? (Choose two.)

Accepted Answer

Enable automatic rotation for the Secrets Manager secrets using a Lambda rotation function, and configure the application to retrieve the latest secret version on each connection.. Automatic rotation with Lambda shrinks the window in which a leaked secret is valid, and fetching the latest version per connection keeps the application working after rotation. Scoping the instance profile to only the required secret ARNs enforces least privilege so a compromised instance cannot read unrelated secrets. Together these actions reduce blast radius and improve retrieval reliability without introducing custom secret management.

Answer

Store the secrets in AWS Systems Manager Parameter Store as SecureString parameters and grant the instances access via an IAM role.

Answer

Hardcode the secrets in the application's environment variables within the EC2 user data script to avoid runtime API calls.

Answer

Store the secrets in an encrypted Amazon S3 bucket and have the application download them at startup using the instance role.