AWS Certified CloudOps Engineer Associate SOA-C03 Practice Questions with Explanations
Questions
An admin needs memory and disk-space utilization metrics for EC2 instances. These metrics aren't available by default. What should be done?
- Install and configure the CloudWatch agent
- Enable detailed monitoring on the instances
- Turn on VPC Flow Logs for the subnets
- Create a CloudTrail trail for EC2 API calls
Where is the recommended place to store a CloudWatch agent configuration so many instances can use it?
- Systems Manager Parameter Store
- An EBS snapshot shared across accounts
- Each instance's user data script only
- A public S3 bucket
A team wants CPU, memory, and network metrics for every task and service in its ECS clusters. What should it enable?
- CloudWatch Container Insights
- VPC Flow Logs
- CloudTrail data events for the ECS API
- AWS Config recorder
Application logs in CloudWatch Logs contain the word ERROR. The team wants an alarm when errors exceed 50 in 5 minutes. What should be created first?
- A metric filter on the log group
- A subscription filter to Kinesis
- A CloudTrail trail
- A new log stream per instance
An on-call team gets too many alerts. They want a page only when both high CPU and high latency happen together. What should be used?
- A CloudWatch composite alarm
- Two separate alarms with SNS email targets
- A metric filter that sums both metrics
- A CloudWatch dashboard with both widgets
Which action can a CloudWatch alarm on an EC2 metric take directly, without EventBridge?
- Stop, terminate, reboot, or recover the instance
- Resize the instance's root EBS volume to add more space
- Change the instance's subnet
- Create a new AMI of the instance
A new alarm stays in INSUFFICIENT_DATA. What is the most likely reason?
- The metric hasn't reported enough data points yet
- The alarm threshold is set too high
- The SNS topic has no subscribers
- The alarm is attached to a composite alarm that is already in ALARM
Leadership wants one dashboard showing alarms and metrics from 12 accounts in three Regions. What should be set up?
- Cross-account observability with one shared dashboard
- A separate dashboard in each account, emailed weekly
- An AWS Config aggregator for all accounts
- A CloudTrail organization trail
An alarm must email the ops team when it changes state. What should its action target?
- An Amazon SNS topic with email subscriptions
- An Amazon SES email template
- A CloudWatch Logs log group
- An S3 bucket with event notifications sent to the team
A team runs Kubernetes workloads and wants to store and query Prometheus metrics without running Prometheus servers. Which service fits?
- Amazon Managed Service for Prometheus
- Amazon Managed Grafana
- Self-managed Prometheus on EC2
- CloudWatch Logs Insights with saved queries
By default, EC2 sends basic monitoring metrics to CloudWatch every 5 minutes.
An EC2 instance fails its system status check because of a problem with the underlying host. What is the usual fix?
- Stop and start the instance
- Reboot the operating system from inside the instance
- Increase the instance's EBS volume size
- Detach and reattach the security group
An EC2 instance fails its instance status check, but its system status check passes. Where is the problem?
- Inside the instance, such as its OS
- In the AWS physical host hardware
- In the AWS Region's network backbone
- In the account's service quotas
An EventBridge rule should invoke a Lambda function, but nothing happens. The event pattern is correct. What should be checked next?
- The function's resource-based policy
- Whether the Lambda function has enough memory
- Whether CloudTrail is enabled in another Region
- Whether the event bus has a CloudWatch dashboard
An EventBridge rule should pass only the instance ID and state to a target, in a custom format. Which feature fits?
- An input transformer
- A dead-letter queue
- An archive and replay
- A schema registry
An admin wants to restart an EC2 instance automatically, using a predefined AWS procedure, whenever a specific alarm fires. Which tool fits?
- A Systems Manager Automation runbook
- A CloudFormation change set run against the instance
- An AWS Config conformance pack for EC2
- A CloudTrail Lake query on the instance ID
Some EventBridge events fail to reach their target because of temporary errors. How can the team keep them for later investigation?
- Add a dead-letter queue to the target
- Increase the rule's event pattern size
- Create a second event bus with no rules
- Turn on CloudWatch detailed monitoring
Which AWS service uses AI to automatically investigate incidents when an alarm fires and suggest root causes and fixes?
- AWS DevOps Agent
- AWS Security Agent
- AWS Config
- Amazon Inspector
An engineer wants an AI tool in the IDE to help write and fix the scripts used in remediation runbooks. Which AWS tool fits?
- Kiro
- AWS Service Catalog
- Amazon Macie
- AWS Global Accelerator
Which can be used to automatically remediate an issue when a CloudWatch alarm fires? (Select TWO.)
- An EventBridge rule that starts a Systems Manager Automation runbook
- An alarm action that invokes a Lambda function
- A CloudWatch dashboard widget
- An IAM password policy
A gp2 volume runs slowly once its BurstBalance metric drops to zero. What is the best fix?
- Change the volume to gp3 and set the IOPS it needs
- Take a snapshot and restore it to a new volume of the same gp2 type
- Attach the volume to a second instance
- Turn on EBS encryption by default
Can an EBS volume's type and size be changed while it's attached and in use?
- Yes, with Elastic Volumes, usually without downtime
- No, it must be detached and recreated from a snapshot
- Only if the instance is stopped first
- Only for volumes smaller than 1 TB
A volume restored from a snapshot is slow the first time each block is read. What removes this first-read delay?
- Enable Fast Snapshot Restore on the snapshot
- Switch the volume to st1
- Enable S3 Transfer Acceleration for the snapshot data
- Add more security groups
Large uploads to S3 from distant offices are slow. What can speed them up?
- S3 Transfer Acceleration with multipart uploads
- S3 Object Lock in governance mode on the bucket
- Moving the bucket to S3 Glacier
- S3 Requester Pays
An app makes very high request rates to one S3 prefix and gets 503 Slow Down errors. What helps?
- Spread objects across more prefixes
- Turn on S3 Versioning
- Use a single larger object
- Move the bucket to another Region closer to the app
Many Linux instances need a shared file system whose cold files should move to cheaper storage automatically. Which fits?
- Amazon EFS with a lifecycle policy
- Amazon EBS gp3 with Multi-Attach
- EC2 instance store volumes
- Amazon FSx for Windows File Server
A DBA must find which SQL queries cause most of the load on an RDS database. Which feature fits?
- RDS Performance Insights
- RDS Enhanced Monitoring only
- AWS Config
- VPC Flow Logs
Hundreds of Lambda functions open new connections to RDS and exhaust its connection limit. What fixes this?
- Amazon RDS Proxy
- A larger EBS volume
- RDS Multi-AZ
- An RDS read replica in another Region
Which service recommends better-sized EC2 instances, EBS volumes, and Lambda memory settings based on usage?
- AWS Compute Optimizer
- AWS Health Dashboard
- AWS Config rules and conformance packs
- Amazon Inspector
Tightly coupled HPC nodes need the lowest network latency between them. What should be used?
- A cluster placement group
- A spread placement group
- Instances in different Regions
- Instances in three different AZs
Which EBS metric suggests a volume can't keep up with the I/O requests being sent to it?
- A high VolumeQueueLength
- A low VolumeIdleTime
- A high BurstBalance
- A low VolumeReadOps count
Developers need to search and aggregate log data across several log groups with a query language. Which tool fits?
- CloudWatch Logs Insights
- CloudWatch metric math
- AWS Config advanced queries
- VPC Reachability Analyzer
A team uses Windows apps that need SMB file shares integrated with Active Directory. Which shared storage fits?
- Amazon FSx for Windows File Server
- Amazon EFS
- Amazon FSx for Lustre
- EC2 instance store
An Auto Scaling group should keep average CPU near 60% with the least setup. Which policy fits?
- Target tracking
- Simple scaling
- Step scaling with five steps
- Scheduled scaling
New instances in an Auto Scaling group take 10 minutes to boot and warm up. Scale-out is too slow during spikes. What helps most?
- A warm pool of pre-initialized instances
- A shorter health check grace period for new instances
- A smaller instance type
- Scheduled scale-in every night
Before an instance is terminated during scale-in, a script must copy its logs to S3. What enables this?
- An Auto Scaling lifecycle hook
- A longer default cooldown period for the group
- A target tracking policy
- A spread placement group
Admins updated the launch template with a new AMI. How can they replace all instances in the group in a controlled, rolling way?
- Start an instance refresh
- Delete the Auto Scaling group and recreate it
- Reboot every instance at once
- Change the group's maximum size to zero
Newly launched instances are being marked unhealthy and replaced before the app finishes starting. What should be adjusted?
- Increase the health check grace period
- Lower the desired capacity
- Turn off the load balancer
- Shorten the group's cooldown period to zero seconds
A read-heavy RDS MySQL database is near its CPU limit because of read queries. What scales reads with the least app change?
- Add read replicas and send read traffic to them
- Enable Multi-AZ and point the read queries at the standby
- Turn on storage autoscaling
- Increase the backup retention period
An RDS database keeps running low on storage. What prevents outages from a full disk?
- Enable RDS storage autoscaling
- Add a read replica in the same AZ
- Lower the backup retention to 1 day
- Change the instance class to burstable
A DynamoDB table with provisioned capacity is throttled at peak hours. Which option adjusts capacity automatically with demand?
- DynamoDB auto scaling
- DynamoDB Streams
- Point-in-time recovery
- A local secondary index
A relational workload is idle overnight and spikes unpredictably by day. Which option scales database capacity automatically in fine steps?
- Aurora Serverless v2
- RDS on a fixed db.r6g.large
- A single Aurora provisioned writer
- RDS with Provisioned IOPS only
A web app keeps querying a database for the same product details. What reduces database load and improves scalability?
- Cache results in Amazon ElastiCache
- Add more NAT gateways
- Increase the RDS backup window and retention period
- Enable VPC Flow Logs
A global site's static assets are overloading its origin servers. What should be added?
- Amazon CloudFront
- AWS Transit Gateway
- An internet gateway
- AWS DataSync
A scheduled scaling action can raise an Auto Scaling group's capacity before a known daily traffic spike.
An ALB marks every target unhealthy, though the app works when tested directly. What should be checked first?
- The health check settings
- The ALB's access log bucket
- The account's Service Quotas
- The Region's CloudTrail settings
Route 53 must fail over based on the health of a resource with only a private IP address. Public health checkers can't reach it. What should be used?
- A health check based on a CloudWatch alarm
- A standard health check aimed at the resource's private IP
- A latency-based routing policy
- A Route 53 Resolver inbound endpoint
A production RDS database must survive the loss of an Availability Zone with automatic failover. What should be enabled?
- RDS Multi-AZ deployment
- Automated backups only
- A read replica in the same AZ
- Provisioned IOPS storage
An ElastiCache cluster must keep working if its primary node fails. What should be enabled?
- Multi-AZ with automatic failover to a replica
- A larger node type with no replicas in other AZs
- Backups to S3 every hour
- Encryption in transit
An operator needs to quickly move traffic away from one impaired Availability Zone for an application behind a load balancer. Which feature fits?
- ARC zonal shift
- Route 53 geolocation routing
- An S3 lifecycle rule
- AWS Config conformance packs