Network traffic within a VPC

Question

What would cause intermittent network disruptions between servers in the same VPC? 
 
For testing purposes I setup a Windows Active Directory server (10.0.0.190) and Web Server (10.0.0.133) in the same VPC (for testing purposes).  The web server has joined the AD domain.  I real all internal traffic is by default disabled in a VPN and so I allowed all inbound traffic on the intranet (10.0.0.0/16) with this security group rule:

**IP version = IPv4;
Type = All Traffic;
Protocol = All;
Port Range = All;
Source = 10.0.0.0/16**

Windows Firewall is turned off on both servers.

DNS server is installed on the AD server.  Web server has its DNS set to the IP of the AD Server.  It is set manually in the network adapter for IPv4.  IPv6 is disabled on both servers.

Sometimes the web server can not ping the AD server by name or by IP address.  Sometimes the web server can ping by name (in domain's DNS) and by IP address.

What am I missing?

Thanks,
Mike

Tushar_J · Answer

+1 to what @Brettski mentioned. Additionally, to rule out OS (windows) specific issues you can spin up Linux EC2 instances in the same exact subnet and see if exhibits same ping/packet loss also? This will help you narrow down the troubleshooting scope. It sounds like both your source and destination servers are windows currently.

Secondly, refer this [KB article](https://aws.amazon.com/premiumsupport/knowledge-center/vpn-packet-loss/) I know this is not VPN related issue but there is 1 Windows related setting (See the section: Explicit Congestion Notification (ECN)) that is suggested to toggle. You may want to give it a try, if it does not resolve the issue you can simply toggle it back.

Brettski-AWS · Answer

I can't think of a good reason why you'd be seeing intermittent network connectivity here. I'd strongly recommend that you raise a support ticket to investigate as the support team can delve into the instances and network services that you're using in order to troubleshoot. On re:Post we have no visibility into your account, services or anything so the support team is best positioned to help.

That said: You haven't mentioned which region you're running in nor which instance types you're using. The first is just for interest; but the second might have an impact - smaller instances have less bandwidth available to them - if there is a lot of other traffic between the two instances then that might cause some issues.

Network traffic within a VPC

Relevant content