A fleet of Linux EC2 instances runs in one account and uses a Systems Manager Automation task for patching. During the latest patch cycle some instances failed because they ran out of disk space. What combination of actions ensures instances have enough disk space during future patching? (Choose two.)

To monitor disk space, the CloudWatch agent must be installed on the EC2 instances. By default, CloudWatch only collects basic metrics like CPU utilization and network I/O; detailed metrics like disk space require the agent. Once the agent is installed and configured to send disk space metrics, a CloudWatch alarm can be created to monitor these metrics. This alarm can then be integrated as a safety control into the Systems Manager Automation task. If the alarm state changes (e.g., disk space falls below a threshold), the automation can be stopped or paused, preventing patching failures due to insufficient disk space. A cron job to delete temporary files is a reactive measure and doesn't provide proactive monitoring or integration with the patching process. Writing disk space to CloudWatch Logs is logging, not monitoring, and requires additional processing to be useful. A Lambda function checking log streams is an indirect and less efficient way to monitor compared to direct CloudWatch metrics and alarms.