Server automatically shut down

When an EC2 instance shuts down unexpectedly without a corresponding CloudTrail event, it typically means the shutdown was initiated from within the operating system rather than through AWS API calls.

Based on your event log showing "LiteAgent.exe has initiated the shutdown" with reason code 0x8000000c, this appears to be an OS-level shutdown. Here are several places you should look to identify the cause:

1. Check the AWS Health Dashboard for any notifications about degraded hardware or scheduled maintenance that might have affected your instance.

2. Review other Windows Event Logs around the time of the shutdown for additional clues - look for system errors, application crashes, or resource constraints that might have triggered the shutdown.

3. Monitor your instance's resource utilization (CPU, memory, disk space) through CloudWatch metrics to identify if resource exhaustion might have caused the issue.

4. Examine your application logs for any errors or crashes that could have led to the system shutdown.

5. If this is a recurring issue, consider enabling more detailed logging or implementing monitoring tools to capture the system state before shutdowns occur.

6. Check if there were any automated scripts, scheduled tasks, or services running on the instance that might have triggered the shutdown.

The "LiteAgent.exe" reference suggests this might be related to AWS Xen virtualization tools, which could indicate an underlying hardware or virtualization issue. If the problem persists, you might want to stop and start the instance (not just reboot) to move it to new underlying hardware, which can resolve issues related to the physical host.
Sources
Windows server automatically poweroff issue | AWS re:Post
'Power key pressed.' reported in server logs | AWS re:Post
EC2 instance shutdown without alert & CloudTrail event. | AWS re:Post
EC2 Instance Shutdown by SystemD event "Power key pressed" but no CloudTrail event were logged | AWS re:Post

Just ensure you checked the correct region for cloudtrail where your EC2 resource resides

Confirmed the cloudtrail region but no entries there for a shutdown. Resource usage at the time of the shutdown was also low and no application error entries near the time of the shut down.

Are there any other logs to check?