AWS DevOps High Availability 4 — Questions and Answers
Question 1: A company uses AWS Systems Manager Run Command to patch EC2 instances. How should the DevOps team structure patching to maintain high availability?
- Patch all instances simultaneously during a maintenance window
- Use Patch Manager with maintenance window rate controls to patch a percentage of instances at a time (Correct answer)
- Terminate and relaunch instances from a patched AMI all at once
- Manually patch one instance and create a new AMI from it
Correct answer: Use Patch Manager with maintenance window rate controls to patch a percentage of instances at a time
Patch Manager's rate controls allow patching a defined percentage or count of targets concurrently, maintaining service capacity during patching.
Question 2: An application running on EC2 uses an Elastic IP address for high availability. What is a significant limitation of this approach compared to using a load balancer?
- Elastic IPs cannot be associated with instances in a VPC
- Failover requires re-associating the Elastic IP, causing brief downtime and requiring automation (Correct answer)
- Elastic IPs are not supported in multiple Availability Zones
- Elastic IPs do not support IPv4 traffic from the internet
Correct answer: Failover requires re-associating the Elastic IP, causing brief downtime and requiring automation
Elastic IP failover requires disassociation from the failed instance and association to a replacement, introducing downtime and complexity compared to load balancer health-check-based routing.
Question 3: Which AWS service provides a fully managed, highly available message broker that supports industry-standard protocols like AMQP and MQTT for migrating on-premises messaging?
- Amazon SQS
- Amazon SNS
- Amazon MQ (Correct answer)
- Amazon Kinesis Data Streams
Correct answer: Amazon MQ
Amazon MQ is a managed Apache ActiveMQ and RabbitMQ service that supports AMQP, MQTT, and other standard protocols, easing migration from on-premises brokers.
Question 4: A DevOps team is designing a pipeline to deploy a containerized microservice. Which approach best supports zero-downtime deployments in Amazon EKS?
- Delete existing pods before deploying new ones
- Use Kubernetes rolling update strategy with maxUnavailable=0 and maxSurge=1 (Correct answer)
- Scale the cluster to zero nodes then redeploy
- Rebuild the node group and redeploy all pods simultaneously
Correct answer: Use Kubernetes rolling update strategy with maxUnavailable=0 and maxSurge=1
Setting maxUnavailable=0 ensures no pods are terminated before replacement pods are ready, and maxSurge=1 allows one extra pod during the rollout.
Question 5: AWS CloudWatch composite alarms can improve high-availability alerting. What is the primary advantage over individual alarms?
- Composite alarms can trigger Auto Scaling policies directly
- They reduce alarm noise by only alerting when multiple conditions are simultaneously true, indicating a real issue (Correct answer)
- Composite alarms support cross-account metric aggregation
- They automatically remediate issues using Systems Manager automation
Correct answer: They reduce alarm noise by only alerting when multiple conditions are simultaneously true, indicating a real issue
Composite alarms combine multiple alarms with AND/OR logic, reducing false positives by alerting only when a combination of conditions indicates a genuine incident.
Question 6: A company needs its API Gateway to remain available even during a regional AWS outage. Which architecture achieves this?
- Enable API Gateway caching with a 3600-second TTL
- Deploy identical APIs in multiple regions with Route 53 health-check-based failover routing (Correct answer)
- Use API Gateway usage plans to throttle traffic during outages
- Enable API Gateway access logging to CloudWatch for faster incident detection
Correct answer: Deploy identical APIs in multiple regions with Route 53 health-check-based failover routing
Deploying API Gateway endpoints in multiple regions with Route 53 health checks provides automatic DNS-based failover if an entire region becomes unavailable.
Question 7: What does the RPO (Recovery Point Objective) measure in a high-availability and disaster recovery strategy?
- The maximum time allowed to restore service after a failure
- The maximum acceptable amount of data loss measured in time from the last backup to the failure (Correct answer)
- The percentage of traffic that must remain available during an incident
- The number of Availability Zones required for compliance
Correct answer: The maximum acceptable amount of data loss measured in time from the last backup to the failure
RPO defines the maximum tolerable data loss, expressed as the time window between the last recovery point and the failure event.
AWS DevOps High Availability 4
A company uses AWS Systems Manager Run Command to patch EC2 instances.
How should the DevOps team structure patching to maintain high availability?