Unstable network on EC2 WIndows server after scheduled reboot

Nothing has fixed this, nothing was changed to break this either. An AWS scheduled reboot was the last action before this failed

Based on the symptoms you've described, it appears you're experiencing a network connectivity issue with your EC2 Windows instance after a scheduled reboot. This problem could be related to several factors, but a few possibilities stand out:

1. Time synchronization issues: If your instance's time is not correctly synchronized with UTC (Coordinated Universal Time), it can cause network connectivity problems. Windows instances use UTC by default, and changing the time zone without proper configuration can lead to temporary IP address loss after a reboot.

2. PV (Paravirtual) driver issues: Older Windows Server AMIs, particularly those from before September 2014, can experience network and storage connectivity loss after a reboot due to the Plug and Play Cleanup feature incorrectly identifying the EC2 network device as inactive.

3. Network adapter driver problems: If you've recently changed the instance type, especially from a Xen-based instance to a Nitro instance, you might be experiencing issues with the network adapter driver.

To address these potential causes, consider the following steps:

1. Ensure your instance is using the correct time settings. Set the RealTimeIsUniversal registry key to 1 to use a time zone other than UTC persistently.

2. Check and update your PV drivers. You may need to perform an in-place driver upgrade or run a helper script to fix PV driver-related issues.

3. Verify that you have the correct network adapter drivers installed, especially if you've changed instance types recently. You may need to install or update the Enhanced Networking (ENA) and AWS NVMe drivers.

4. Review your security group settings to ensure they allow inbound RDP traffic (port 3389) from your IP address.

5. Check the Windows Firewall settings on the instance, as they might be interfering with RDP connections.

If the problem persists after trying these steps, you may want to use AWS Systems Manager to troubleshoot further or consider reaching out to AWS support for more in-depth assistance. Additionally, creating a snapshot of your instance before making any changes is always a good practice to ensure you can recover if needed.

Sources
EC2 instance losing connection | AWS re:Post
Troubleshoot issues connecting to your Amazon EC2 Windows instance - Amazon Elastic Compute Cloud
Troubleshoot PV drivers on Windows instances - Amazon Elastic Compute Cloud
Troubleshoot issues with Amazon EC2 Windows instances - Amazon Elastic Compute Cloud
Troubleshoot an RDP connection on Windows Server after EC2 change | AWS re:Post