Deployment Considerations for DirectAccess on Amazon Web Services (AWS)

Organizations are rapidly deploying Windows server infrastructure with public cloud providers such as Amazon Web Services (AWS) and Microsoft Azure. With traditional on-premises infrastructure now hosted in the cloud, DirectAccess is also being deployed there more commonly.

Supportability

Interestingly, Microsoft has expressly stated that DirectAccess is not formally supported on their own public cloud platform, Azure. However, there is no formal statement of non-support for DirectAccess hosted on other non-Microsoft public cloud platforms. With supportability for DirectAccess on AWS unclear, many companies are taking the approach that if it isn’t unsupported, then it must be supported. I’d suggest proceeding with caution, as Microsoft could issue formal guidance to the contrary in the future.

DirectAccess on AWS

Deploying DirectAccess on AWS is similar to deploying on premises, with a few notable exceptions, outlined below.

IP Addressing

It is recommended that an IP address be exclusively assigned to the DirectAccess server in AWS, as shown here.

Deployment Considerations for DirectAccess on Amazon Web Services (AWS)

Prerequisites Check

When first configuring DirectAccess, the administrator will encounter the following warning message.

“The server does not comply with some DirectAccess prerequisites. Resolve all issues before proceed with DirectAccess deployment.”

The warning message itself states that “One or more network adapters should be configured with a static IP address. Obtain a static address and assign it to the adapter.

Deployment Considerations for DirectAccess on Amazon Web Services (AWS)

IP addressing for virtual machines are managed entirely by AWS. This means the DirectAccess server will have a DHCP-assigned address, even when an IP address is specified in AWS. Assigning static IP addresses in the guest virtual machine itself is also not supported. However, this warning message can safely be ignored.

No Support for Load Balancing

It is not possible to create load-balanced clusters of DirectAccess servers for redundancy or scalability on AWS. This is because enabling load balancing for DirectAccess requires the IP address of the DirectAccess server be changed in the operating system, which is not supported on AWS. To eliminate single points of failure in the DirectAccess architecture or to add additional capacity, multisite must be enabled. Each additional DirectAccess server must be provisioned as an individual entry point.

Network Topology

DirectAccess servers on AWS can be provisioned with one or two network interfaces. Using two network interfaces is recommended, with the external network interface of the DirectAccess server residing in a dedicated perimeter/DMZ network. The external network interface must use either the Public or Private Windows firewall profile. DirectAccess will not work if the external interface uses the Domain profile. For the Public and Private profile to be enabled, domain controllers must not be reachable from the perimeter/DMZ network. Ensure the perimeter/DMZ network cannot access the internal network by restricting network access in EC2 using a Security Group, or on the VPC using a Network Access Control List (ACL) or custom route table settings.

External Connectivity

A public IPv4 address must be associated with the DirectAccess server in AWS. There are several ways to accomplish this. The simplest way is to assign a public IPv4 address to the virtual machine (VM). However, a public IP address can only be assigned to the VM when it is deployed initially and cannot be added later. Alternatively, an Elastic IP can be provisioned and assigned to the DirectAccess server at any time.

An ACL must also be configured for the public IP that restricts access from the Internet to only inbound TCP port 443. To provide additional protection, consider deploying an Application Delivery Controller (ADC) appliance like the Citrix NetScaler or F5 BIG-IP to enforce client certificate authentication for DirectAccess clients.

Network Location Server (NLS)

If an organization is hosting all of its Windows infrastructure in AWS and all clients will be remote, Network Location Server (NLS) availability becomes much less critical than with traditional on-premises deployments. For cloud-only deployments, hosting the NLS on the DirectAccess server is a viable option. It eliminates the need for dedicated NLS, reducing costs and administrative overhead. If multisite is configured, ensure that the NLS is not using a self-signed certificate, as this is unsupported.

Deployment Considerations for DirectAccess on Amazon Web Services (AWS)

However, for hybrid cloud deployments where on-premises DirectAccess clients share the same internal network with cloud-hosted DirectAccess servers, it is recommended that the NLS be deployed on dedicated, highly available servers following the guidance outlined here and here.

Client Provisioning

All supported DirectAccess clients will work with DirectAccess on AWS. If the domain infrastructure is hosted exclusively in AWS, provisioning clients can be performed using Offline Domain Join (ODJ). Provisioning DirectAccess clients using ODJ is only supported in Windows 8.x/10. Windows 7 clients cannot be provisioned using ODJ and must be joined to the domain using another form of remote network connectivity such as VPN.

Additional Resources

DirectAccess No Longer Supported in Microsoft Azure

Microsoft Server Software Support for Azure Virtual Machines

DirectAccess Network Location Server (NLS) Guidance

DirectAccess Network Location Server (NLS) Deployment Considerations for Large Enterprises

Provisioning DirectAccess Clients using Offline Domain Join (ODJ)

DirectAccess SSL Offload and IP-HTTPS Preauthentication with Citrix NetScaler

DirectAccess SSL Offload and IP-HTTPS Preauthentication with F5 BIG-IP

Planning and Implementing DirectAccess with Windows Server 2016 Video Training Course

Implementing DirectAccess with Windows Server 2016 Book

Configure F5 BIG-IP for DirectAccess NLS

Recently I wrote about the Network Location Server (NLS) and its importance for DirectAccess deployments. As I described previously, the NLS is nothing more than a web server with an SSL certificate installed. It should also be made highly available to prevent potential service disruption caused by planned or unplanned NLS server downtime. Any web server can serve as the NLS. In addition, if you have the F5 BIG-IP Local Traffic Manager (LTM) in your environment, you can easily configure the LTM to serve as the NLS.

To accomplish this, import the SSL certificate for the NLS and create an SSL client profile using its certificate and private key. Next, create a new iRule that contains the following code.

when HTTP_REQUEST {
HTTP::respond 200 
}

Configure F5 BGIP for DirectAccess NLS

Finally, create a new virtual server listening on TCP port 443 and assign this iRule as a resource for the virtual server. Once NLS reachability has been verified, update the DirectAccess configuration using the Remote Access Management console or the Set-DANetworkLocationServer PowerShell cmdlet.

DirectAccess Network Location Server Guidance

Introduction

The Network Location Server (NLS) is a critical component in a DirectAccess deployment. The NLS is used by DirectAccess clients to determine if they are inside or outside of the corporate network. If a DirectAccess client can connect to the NLS, it must be inside the corporate network. If it cannot, it must be outside of the corporate network. It is for this reason that the NLS must not be reachable from the public Internet. A client configured for DirectAccess will probe the NLS when it first starts, and on subsequent network interface status changes.

What is the NLS?

The NLS itself is nothing more than a web server with an SSL certificate installed. Beginning with Windows Server 2012, the NLS can be collocated on the DirectAccess server itself. Although there may be scenarios in which this is acceptable, it is generally recommended that NLS be configured on a server dedicated to this role.

NLS Configuration

Any web server can be used, including IIS, Apache, Nginx, Lighttpd, and others. You can also use an Application Delivery Controller (ADC) like the F5 BIG-IP Local Traffic Manager (LTM), as described here. The web server must have a valid SSL certificate installed that includes a subject name that matches the NLS FQDN (e.g. nls.corp.example.com). The DNS record for the NLS must configured using an A host record. A CNAME DNS entry will not work. In addition, the NLS must also respond to ICMP echo requests.

DirectAccess Network Location Server Guidance

DirectAccess Network Location Server Guidance

The certificate can be issued by an internal PKI or a public third-party Certificate Authority (CA). A self-signed certificate can be used if the certificate is distributed to all DirectAccess clients and servers, but this is not advisable. To avoid possible service disruptions, the NLS should be made highly available by deploying at least two NLS in a load balanced configuration.

What Happens if the NLS is Offline?

If the NLS is offline for any reason, remote DirectAccess clients will be unaffected. However, DirectAccess clients on the internal network will mistakenly believe they are outside of the corporate network and attempt to establish a DirectAccess connection. If the DirectAccess server is not accessible from the internal network, the client will be unable to connect to any local network resources by name until the NLS is brought online or other actions are taken.

Collocation Issues

As mentioned previously, it is possible in some scenarios to collocate the NLS on the DirectAccess server. This is probably acceptable for proof-of-concept deployments, but any production deployment should have the NLS configured on a server dedicated to this role. If the NLS is located on the DirectAccess server and the server is offline for any reason, DirectAccess clients on the internal network will be unable to access local resources by name until the DirectAccess server is back online.

Don’t Use Existing Web Application Servers

Occasionally I will encounter a scenario in which an administrator who wants to avoid implementing additional infrastructure will use an existing internal web application server for the NLS, such as a SharePoint server. Although this will work, it quickly becomes an issue when remote DirectAccess clients need to access the server. Since the NLS is not resolvable or reachable externally, connectivity will fail, preventing DirectAccess clients from reaching the internal application.

Summary

The NLS is a vitally important piece of the DirectAccess architecture. DirectAccess clients use the NLS to determine their location, and if the service is unavailable for any reason (planned or unplanned) internal DirectAccess clients will be negatively affected. The NLS isn’t necessarily complicated, as it is nothing more than a web server with an SSL certificate installed. However, don’t overlook the importance of this service, and make sure it is highly available to avoid any potential network connectivity issues.

Additional Resources

DirectAccess Network Location Server (NLS) Deployment Considerations for Large Enterprises

Configure KEMP LoadMaster Load Balancer for DirectAccess Network Location Server (NLS)

Configure Citrix NetScaler for DirectAccess Network Location Server (NLS)

Configure F5 BIG-IP for DirectAccess Network Location Server (NLS) 

SSL Offload for IP-HTTPS DirectAccess Traffic from Windows 7 Clients using F5 BIG-IP

From a client perspective, DirectAccess is an IPv6 only solution. It requires IPv6 connectivity from end-to-end to provide seamless, transparent, always-on remote access. DirectAccess clients are most commonly connected to the IPv4 Internet, so to overcome the limitations imposed by the exclusive use of IPv6 for transport, DirectAccess leverages IPv6 transition technologies such as 6to4, Teredo, or IP-HTTPS to tunnel IPv6 DirectAccess client communication over the IPv4 Internet. These transition protocols are favored by the operating system in the order in which I have listed them here. 6to4 uses IP protocol 41 for transport and requires that the client have a public IPv4 address, so if the DirectAccess client is behind a firewall that does not allow outbound IP protocol 41, or is located behind a NAT and has a private IPv4 address, it will fall back to Teredo. Teredo uses UDP for transport on port 3544, and if this communication is blocked by a firewall the DirectAccess client will then fall back to IP-HTTPS. IP-HTTPS, as its name implies, tunnels DirectAccess IPv6 traffic in HTTP, which is authenticated and encrypted using SSL or TLS.

Historically the challenge with the IP-HTTPS IPv6 transition protocol is that it encrypts DirectAccess communication which is already encrypted using IPsec. This double encryption places significant demands on CPU and memory resources on the DirectAccess server, resulting in poor throughput and performance and limiting the overall scalability of the solution. To address these shortcomings, Windows Server 2012 DirectAccess introduced support for IP-HTTPS NULL encryption. SSL/TLS is still used for authentication, but the IPsec traffic is no longer double encrypted. This dramatically reduces resource consumption on the DirectAccess server, resulting in improved performance and allowing many more DirectAccess clients to be handled by a single server. The only drawback is that IP-HTTPS NULL encryption is only supported with Windows 8 clients. When Windows 7 clients connect to a Windows Server 2012 DirectAccess server using IP-HTTPS, they will continue to use encrypted IP-HTTPS.

An ideal solution would be to terminate SSL off box using a dedicated hardware appliance like the F5 BIG-IP Local Traffic Manager (LTM). Unfortunately there is no provision in Windows Server 2012 DirectAccess to enable SSL termination for IP-HTTPS traffic. However, using some of the advanced features of the LTM, we can effectively offload SSL on the F5 by configuring LTM to emulate Windows 8 DirectAccess client behavior. This is accomplished by having the F5 LTM exclusively negotiate the use of a NULL encryption cipher suite with the Windows Server 2012 DirectAccess server on behalf of Windows 7 DirectAccess clients.

Note: This post assumes that you are familiar with the configuration and management of the F5 BIG-IP LTM solution, and that you’ve already imported your SSL certificates and configured nodes, pools, and virtual servers for your Windows Server 2012 DirectAccess server.

To configure the F5 LTM to provide SSL offload for Windows 7 DirectAccess clients, we’ll need to create SSL profiles to allow the use of specific cipher suites for our IP-HTTPS traffic. In its default configuration, the BIG-IP LTM does not support the use of NULL encryption cipher suites. Since Windows 8 DirectAccess clients use NULL cipher suites exclusively, we need to explicitly enable these on the LTM to support our Windows 8 clients. Since our Windows 7 clients will use only encrypted cipher suites, we’ll be sure to include those as well. To do this, open the F5 management console, expand Local Traffic, Profiles, SSL, and then click the green icon next to Client.

f5_directaccess_iphttps_offload_01

Provide a name for the new Client SSL Profile, select Advanced configuration, check the Custom box and specify DEFAULT:NULL for Ciphers. Be sure to select the appropriate SSL certificate and key. Click Finished at the bottom of the screen to save these settings. This change allows NULL cipher suites in addition to encrypted cipher suites, allowing us to support both Windows 8 and Windows 7 DirectAccess clients.

f5_directaccess_iphttps_offload_02

Next we need to configure the LTM to use only NULL cipher suites when communicating with the Windows Server 2012 DirectAccess server. To do this, expand Profiles, SSL, and then click the green icon next to Server.

f5_directaccess_iphttps_offload_03

Provide a name for the new Server SSL Profile, select Advanced configuration, check the Custom box and specify NULL-SHA for Ciphers. Click Finished at the bottom of the screen to save these settings. The end result here will be to force the exclusive use NULL encryption cipher suites for all IP-HTTPS traffic, regardless if it is a Windows 8 or Windows 7 client.

f5_directaccess_iphttps_offload_04

Once you’ve completed the client and server SSL profiles, it will be necessary to assign these profiles to the virtual servers that represent your Windows Server 2012 DirectAccess server. Navigate to Virtual Servers and click on Virtual Server List. Click the virtual server that corresponds to your DirectAccess server, and then scroll down to the bottom of the page. For SSL Profile (Client), select DA_IPHTTPS_CLIENT and add that to the list. Repeat this step for the SSL Profile (Server), this time selecting DA_IPHTTPS_SERVER. Click Update to apply these changes.

f5_directaccess_iphttps_offload_05

Once complete, the F5 BIG-IP LTM will now effectively be offloading SSL traffic on behalf of Windows 7 DirectAccess clients by emulating the Windows 8 DirectAccess client behavior and using only NULL encryption for IP-HTTPS sessions established with the Windows Server 2012 DirectAccess server. Although I can see no issues with this deployment model, be advised that this configuration may not be supported by Microsoft, so make these changes at your own risk. I’ll be working with Microsoft and F5 to get this solution reviewed and tested and I will provide clarification on supportability here once I have that information.

Special thanks to Jeff Bellamy, Ryan Korock, and John Wagnon at F5 for their assistance with this developing solution.

DirectAccess and NAT

One of the more common barriers to adoption for DirectAccess in Windows Server 2008 R2 and Forefront Unified Access Gateway (UAG) 2010 is the strict requirement for two consecutive public IPv4 addresses to be assigned to the external network interface of the DirectAccess server. Many small and mid-sized businesses have only a single public IPv4 address, or have a very small range of public IPv4 addresses that are already in use. For large organizations, corporate security policies often dictate that Windows-based systems cannot be internet facing, and many object to having a domain-joined Windows system exposed directly to the Internet. Further complicating matters is the fact that deploying a Window Server 2008 R2 or Forefront UAG 2010 DirectAccess server behind a border router or edge firewall performing Network Address Translation (NAT) is explicitly not supported.

Beginning with Windows Server 2012, deploying the DirectAccess server behind a border router or edge firewall performing NAT is now fully supported. No longer is there a requirement to have public IPv4 addresses assigned to the DirectAccess server’s external network interface. In fact, DirectAccess in Windows Server 2012 can be deployed with a single network adapter, allowing the DirectAccess server to be completely isolated in a perimeter or DMZ network.

Windows Server 2012 DirectAccess Network Topology

Be advised that deploying a Windows Server 2012 DirectAccess server behind a NAT device will result in all DirectAccess client communication being delivered to the server exclusively using the IP-HTTPS IPv6 transition protocol. If you are using Windows 8 clients, there’s nothing to worry about in terms of performance and scalability because Windows 8 clients leverage NULL encryption for IP-HTTPS traffic. However, Windows 7 clients cannot utilize NULL encryption and will instead encrypt all DirectAccess client communication using SSL/TLS. DirectAccess communication is already encrypted using IPsec, so this presents a problem. Double encryption places high demands on the DirectAccess server’s CPU and memory and will significantly impact performance on the client and the server. It will also impede the scalability of the solution by dramatically reducing the number of DirectAccess clients supported on a single DirectAccess server.

So, if you’re planning to deploy a Windows Server 2012 DirectAccess server behind a NAT, and you are also planning to support a lot of Windows 7 clients, please proceed cautiously. Monitor the DirectAccess server performance closely during your pilot and, if at all possible, offload SSL/TLS off box using F5 BIG-IP Local Traffic Manager (LTM) or equivalent device.

%d bloggers like this: