Unreachable Instances

Instance Reboot

If an instance is unreachable, try rebooting it using the AWS CLI or AWS Management Console.

AWS CLI:

aws ec2 reboot-instances --instance-ids i-1234567890abcdef0

Console:

1. Go to the **EC2 Dashboard**.
2. Select the unreachable instance.
3. Click **Instance State** → **Reboot Instance**.

Instance Console Output

Check instance logs using the EC2 Serial Console or AWS CLI:

aws ec2 get-console-output --instance-id i-1234567890abcdef0

Or use the EC2 console:

1. Go to **EC2 Dashboard**.
2. Select the instance.
3. Click **Actions** → **Monitor and Troubleshoot** → **Get Instance Screenshot**.

Capture a Screenshot of an Unreachable Instance

If the instance is running but unresponsive, capture a screenshot to diagnose issues.

aws ec2 get-instance-screenshot --instance-id i-1234567890abcdef0

Common Screenshots for Windows Instances

• **Blue screen (BSOD)** – Possible system crash.
• **Blank black screen** – Could indicate GPU or driver failure.
• **Windows login screen** – Instance is running but has network issues.

Instance Recovery When a Host Computer Fails

If an EC2 instance becomes unreachable due to a host failure, AWS automatically moves it to another host.

To manually recover:

aws ec2 stop-instances --instance-ids i-1234567890abcdef0
aws ec2 start-instances --instance-ids i-1234567890abcdef0

Instance Appeared Offline and Unexpectedly Rebooted

Check if your instance rebooted unexpectedly due to AWS maintenance.

Use AWS CLI to check instance status:

aws ec2 describe-instance-status --instance-id i-1234567890abcdef0

Or review **CloudTrail** logs:

aws cloudtrail lookup-events --lookup-attributes AttributeKey=EventName,AttributeValue=RebootInstances