AWS Certification CloudOps Engineer Associate (SysOps) 2022 past questions and answers

1. Enable CloudTrail data events for each EC2 instance
2. Install and configure the Amazon CloudWatch agent to collect memory metrics
3. Enable detailed monitoring for each EC2 instance
4. Create a VPC Flow Logs subscription for the instance subnets

Answer: B. Install and configure the Amazon CloudWatch agent to collect memory metrics

EC2 does not publish guest operating-system memory metrics by default; the CloudWatch agent can collect and publish them. Detailed monitoring increases the frequency of supported EC2 metrics but does not add memory metrics.

1. A CloudTrail trail with management events delivered every five minutes
2. A CloudWatch dashboard widget with a text search query
3. An EC2 status check alarm for the application instances
4. A CloudWatch Logs metric filter for TIMEOUT and a CloudWatch alarm on its metric

Answer: D. A CloudWatch Logs metric filter for TIMEOUT and a CloudWatch alarm on its metric

A metric filter turns matching log events into a CloudWatch metric, which can be evaluated by an alarm over the requested period. A dashboard or EC2 status check does not count application log patterns.

1. Configure a CloudWatch alarm on the system status check metric with the EC2 recover action
2. Configure an alarm to terminate the instance and launch an unrelated instance manually
3. Configure an alarm to stop the instance and wait for an administrator to start it
4. Configure an alarm to reboot the instance whenever its CPU utilization exceeds 80 percent

Answer: A. Configure a CloudWatch alarm on the system status check metric with the EC2 recover action

The EC2 recover action responds to eligible system status check failures by recovering the instance on healthy hardware while retaining its identity and supported network attributes. A reboot is not the appropriate response to an underlying host failure, and termination loses the original instance identity.

1. Increase the number of subnets in the DB subnet group
2. Enable deletion protection on the existing DB instance
3. Create one or more read replicas and direct appropriate read traffic to them
4. Increase the backup retention period for the Multi-AZ standby

Answer: C. Create one or more read replicas and direct appropriate read traffic to them

A Multi-AZ standby is intended for availability and failover, not for serving read traffic. Read replicas can serve read-only workloads; changing subnet count or backup retention does not increase query-serving capacity.

1. Place one EC2 instance and one single-AZ RDS instance in separate subnets in the same Availability Zone
2. Place an Application Load Balancer in multiple Availability Zones, run an Auto Scaling group across those zones, and use an RDS Multi-AZ deployment
3. Create an EC2 AMI once per month and use a single-AZ RDS read replica as the only database
4. Place two EC2 instances in one subnet and rely on an RDS automated backup to fail over the application

Answer: B. Place an Application Load Balancer in multiple Availability Zones, run an Auto Scaling group across those zones, and use an RDS Multi-AZ deployment

A load balancer and Auto Scaling group spanning Availability Zones can continue serving from healthy instances, while RDS Multi-AZ provides managed database failover. An AMI or automated backup alone requires restoration and does not provide automatic in-region application and database failover.