Home › AWS SysOps Associate › cloudcomputing › sysopsmonitoring › Which automation pattern is BEST for self-healin…
Which automation pattern is BEST for self-healing a stuck EC2 instance — for example, restart on a specific log message?
AMigrating the workload to AWS Snowball
BCloudWatch alarm or Logs metric filter → EventBridge rule → Systems Manager Automation runbook
CManually paging an engineer every 5 minutes
DDisabling all CloudWatch alarms
Answer & Solution
Correct answer: B. CloudWatch alarm or Logs metric filter → EventBridge rule → Systems Manager Automation runbook
Alarm/log-filter → EventBridge → SSM Automation runbook is the canonical self-healing pattern. Manual paging is the anti-pattern; Snowball isn't compute; disabling alarms removes the signal.
Related questions
An EC2 Auto Scaling group is set to launch from an AMI baked weekly. After the latest AMI A team needs disaster recovery with an RTO ≤ 15 minutes and RPO ≤ 5 minutes across two AWSWhich service detects unexpected drift between an AWS resource's actual configuration and A multi-AZ RDS deployment for a write-heavy database experiences a primary AZ failure. WhaAn RDS database needs point-in-time-restore for 14 days plus a guaranteed daily backup. WhAn EC2 instance's CPU steady-state is fine but bursts to 100% during scheduled jobs. The tWhich Route 53 routing policy is used to fail traffic over from a primary endpoint to a stAn EC2 fleet behind an ALB needs to scale out when average CPU goes above 70% for 5 minute