Transcript DNS Name

Module 2: Designing an
Active Directory Naming
Strategy
Overview

Identifying Business Needs

DNS and Active Directory

Planning Active Directory Domain Names

Designing a DNS Naming Strategy for Active Directory

Resolution of unique names is the cornerstone of
identifying and accessing objects in Microsoft®
Windows® 2000 Active Directory™ directory service.
Active Directory uses the Domain Name System (DNS)
as a basis for naming domains. The hierarchical
structure of Active Directory is derived from the root
domain, which is the first domain created. Carefully
selecting an inclusive DNS name for the root domain is
crucial because an inclusive name may make it easier
for users to access the network over the Internet and
also enable network flexibility.
At the end of this module, you will be able to:

Identify business needs that impact the selection of
Active Directory names.

Describe how Active Directory is integrated with DNS.

Plan Active Directory names within the Active Directory
hierarchy.

Design a DNS naming strategy for Active Directory root
domains.
Identifying Business Needs

Main Business Needs that Impact a Naming Strategy:

Intended Scope of Active Directory

Internet Presence

The initial root domain name will influence the structure
of the Active Directory hierarchy. A properly selected
name should accommodate the current and future
planned business needs of an organization. The two
primary business considerations that affect the naming
of an Active Directory structure are how much of the
organization Active Directory should include, and
whether or not the organization plans to make some or
all of its resources available on the Internet.
Intended Scope of Active Directory

When assessing business needs, you need to determine
the scope of the planned Active Directory structure.
Before you implement Active Directory, you must first
determine how the Active Directory structure will meet
the business requirements of the organization. Thus,
the design of the Active Directory structure should
accommodate one or more of the following possibilities,
depending on the business requirements:



Will the Active Directory structure include the entire
organization, including subsidiaries?
Will the Active Directory incorporate partners or
customers in the future?
Are you anticipating any mergers or acquisitions in the
next two to five years?
Internet Presence

You must consider whether or not the organization's
Active Directory will ever be available on the Internet. If
so, you must choose a name for the Active Directory
root that adheres to Internet standards. You must also
choose a DNS strategy to support the Active Directory.
DNS and Active Directory

Distinguishing Between DNS and Active Directory

Interoperability with BIND

Active Directory follows DNS standards for naming
domains, servers, and services. Active Directory also
uses DNS as the domain locator service. You can use
DNS for name resolution of both intranet (internal) and
Internet (external) resources in your organization. There
are special considerations you must take into account if
your organization uses a Berkeley Internet Name
Domain (BIND) DNS server and insists on maintaining it.
In this lesson you will learn about the following topics:

Distinguishing between DNS and Active Directory

Interoperability with BIND
Distinguishing Between DNS and Active Directory

DNS Servers Store Resource Records

Active Directory Servers Store Domain Objects
Domain Name System
(DNS)
contoso.msft

Active Directory can consist of one or more domains.
You identify Active Directory domains by the DNS
names you assign them.

AD Concepts

The Active Directory domain and the corresponding
DNS domain have the same name, yet each has a
distinct role. These two domains store different
information and manage different objects.
DNS servers store and manage resource records within
a zone database file. A DNS zone database file contains
all resource records for a single DNS domain, or a
discreet portion of a DNS domain tree.

Active Directory stores and manages domain objects.
Objects in the Active Directory include users,
computers, printers, servers, workstations, services and
shares. All objects are stored within Active Directory
and managed either by scripting, or by tools within
Microsoft Management Console (MMC).
Because Active Directory and DNS domain names are
identical and DNS is the mechanism for performing
name resolution, each Active Directory domain requires
a corresponding DNS domain. However, each DNS
domain does not require a corresponding Active
Directory domain.
Interoperability with BIND

Windows 2000 DNS Server Service Offers Maximum
Compatibility

BIND DNS Servers Can Be Integrated with Active
Directory

BIND 8.2.1 or later recommended

Windows 2000 DNS Server service is fully compatible
with Active Directory. However, some organizations may
use BIND DNS servers and insist on maintaining their
use.

Certain versions of BIND can be integrated with Active Directory. If
a business need requires that the existing BIND servers be kept in
place, you can make one of four choices:




Use BIND for external access and Windows 2000 DNS for internal
access.
Use BIND for both Internet (external) and intranet (internal) access.
Use BIND for internal and external access, but place the Active
Directory domain on the Windows 2000 DNS server.
Use Windows 2000 DNS for both external and internal access.

If a BIND server is used for external resources and a
Windows 2000 DNS server for Active Directory, you may
need to delegate the Active Directory subdomains.
These are the subdomains that are used by client
computers to gain access to services in Active
Directory.
While some earlier versions of BIND will function with
Active Directory, BIND 8.2.1 is the minimum
recommended version. BIND 8.2.1 supports the SRV
(service) resource records, dynamic updates, negative
caching, and incremental zone transfers.

The following table compares the features supported in Windows
2000 DNS and BIND:

Note: For more information on SRV resource records and dynamic
DNS, see RFC 2782 and RFC 2136.
Feature
Windows 2000
BIND 8.2.1
BIND 4.9.7
SRV records
Yes
Yes
Yes
Dynamic Update
Yes
Yes
No
Secure Dynamic Update
Yes
No
No
WINS & WINS-R records
Yes
No
No
Fast zone transfer
Yes
Yes
Yes
Incremental zone transfer
Yes
Yes
No
UTF-8 character encoding
Yes
No
No
Planning Active Directory Domain Names

Determining the Scope of Active Directory

Designing the Naming Hierarchy

Choosing Active Directory Domain Names

Because Active Directory is tightly integrated with DNS,
you should adhere to DNS standards when planning the
naming strategy for Active Directory. Your Active
Directory design should include:



Determining the scope of Active Directory within your
organization.
Designing a hierarchical DNS name.
Choosing Active Directory domain names by using a
naming strategy.
Determining the Scope of Active Directory


DNS Name Should Represent Entire Organization

Headquarters

Branch Locations

Business Partners
Active Directory Name Can Be Internet Name

Register Name with ICANN

When you determine the scope of Active Directory,
make it as broad as possible to avoid having to
restructure it later. When considering the scope for
naming, you should consider not just the internal
organization, but whether the organization will have an
Internet presence.
Your DNS Name Should Represent Your Entire Organization

Because Active Directory can accommodate only one
DNS root name, you should choose a name that
represents the entire organization. This name will allow
all users within the organization to access all
information and resources within the organization.
There are instances, however, when the organization
may require more than one root name. For example, if
the organization has acquired a company that already
has a well-established identity on the Internet, you
might want to retain the DNS name of the newly
acquired company.
The Active Directory Name Can Be Your Internet Name

Active Directory can exist within the scope of the
Internet DNS structure. In this case, you must register
the DNS root name with the Internet Corporation for
Assigned Names and Numbers (ICANN). Registration
ensures the global uniqueness of all DNS names. You
will have the authority to manage your own hierarchy of
child domains, zones, and hosts within the root domain.
Designing the Naming Hierarchy
Root
DNS Name: contoso.msft
contoso.msft
Child
namerica.contoso.msft
DNS Name: namerica.contoso.msft
Child
europe.contoso.msft
DNS Name: europe.contoso.msft

Domains in Active Directory are arranged in a
hierarchical structure that reflects their relationships to
each other. Within the network, each domain must have
a unique DNS name that matches the Active Directory
domain name. Client computers running Active
Directory use these names to identify and locate domain
controllers for logon purposes. The goal in designing
your naming hierarchy is to reflect the organization
logically.
The First Domain Is the Root Domain

The first domain created in Active Directory is the
starting point, or root, of Active Directory. All other
domains are derived from the root domain. Only one
name can be used for the root domain. If you plan an
Internet presence using this root name, the name must
be unique on the Internet.
Important: You cannot change the name of the root
domain in your Active Directory forest without removing
Active Directory and creating a new forest
Domains Derived from the Root Domain Form a
Hierarchical Tree

If you have multiple domains in Active Directory, a
naming hierarchy is formed. For example, a network
with a single domain might have the single domain
name, contoso.msft. If business needs require you to
divide your network into three domains, North America,
Europe, and Africa, your domain names could be
namerica.contoso.msft, europe.contoso.msft, and
africa.contoso.msft. Each domain name is separate, but
maintained within the same Active Directory tree.
Choosing Active Directory Domain Names

Choose a Root Domain Name Unique to the Internet

Conform to DNS Naming Regulations

Register Your DNS Domain Name

Choose Meaningful, Stable, Scalable Names

Use An Existing DNS Domain Name

Because Active Directory naming follows the standard
DNS format, DNS must be installed on your network,
and Active Directory names must conform to DNS
naming guidelines. When choosing Active Directory
domain names, consider the following strategies:

The Active Directory root domain name must be unique
in the DNS hierarchy. If your network connects to the
Internet, choose a name that is unique on the Internet.

If your network connects to the Internet, the DNS name
of each Active Directory domain must conform to
Internet domain naming rules.
Note: For more information on DNS character
standards, see RFC 1034, RFC 1035, and RFC 1123.

You will need to register the DNS domain name you
designate as the root Active Directory domain with
ICANN if you plan to use the name for both external and
internal access.

Choose domain names that are recognizable,
meaningful, and stable, such as geographical names.
Choose a name that can last three to five years, and that
can accommodate the addition of child domains, which
might occur in the case of a reorganization or
acquisition.

Use an existing DNS domain name within the registered
DNS hierarchy for the organization. Create a new
domain name as a child domain of the existing DNS
domain namespace, such as corp.contoso.msft.
Note: The draft that reserves .local for local domain
names has expired. For more information, see
www.ietf.org/proceedings/99jul/I-D/draft-ietf-dnsindlocal-names-07.txt.
Designing a DNS Naming Strategy for Active
Directory

Making Initial Naming Decisions

Using a Delegated Subdomain Name for the Internal
Network

Using a Single DNS Name for Public and Private
Networks

Using a Different DNS Name for Public and Private
Networks

Designing a DNS Solution to Integrate with BIND

Design Guidelines
Designing a DNS Naming Strategy for Active Directory

There are several ways that you can design a DNS naming strategy
for the Active Directory root domain. Most of these choices will
reflect whether the organization has, or plans to have, an Internet
presence. If the organization chooses to have an Internet presence
you must decide whether to separate your internally accessible
resources from your externally accessible resources. Depending on
your Internet strategy and existing DNS implementations, you can
use one or more of the following guidelines.



Use a subdomain of a registered DNS domain name as the Active
Directory root.
Use the same DNS root domain name for your Active Directory
structure and Internet presence.
Use a different DNS domain name for your Active Directory root to
maintain separation between your Active Directory structure and
your Internet presence.
In this lesson you will learn about the following topics:

Making initial naming decisions.

Using a delegated subdomain name for the internal
network.

Using a single DNS name for public and private
networks.

Using a different DNS name for public and private
networks.

Designing a DNS solution to integrate with BIND.

Design guidelines.
Making Initial Naming Decisions

Registering the DNS Root Name

Designing with an Existing DNS Implementation

Determining Internal and External Naming Strategies

Meeting Requirements of the DNS Design

Assuring Client Name Resolution

There are important initial decisions you need to make
when designing a DNS naming strategy. The impact of
not considering these items could necessitate a reinstallation of Active Directory in the future.
Registering the DNS Root Name

If you choose to have an Internet presence, you must
obtain a registered DNS domain name. Even if you do
not anticipate an Internet presence, it is recommended
that you still register your chosen root name with ICANN
so that a future move to the Internet will not require
renaming your root.
Designing with an Existing DNS Implementation

Windows 2000 DNS server service is recommended
because it supports full functionality with Active
Directory. If you use the BIND DNS server, you will need
to make design choices based on your naming strategy.
You may need to upgrade your BIND servers, or opt to
convert to Windows 2000 DNS.
Determining Internal and External Naming Strategies

If your organization has an Internet presence, securing
the internal resources is a priority. How you choose to
separate your internally accessible resources from your
externally accessible resources will impact the design.
Meeting Requirements of the DNS Design

When determining the naming approach for your
organization, make sure the design meets the following
requirements:



Expose only the public part of the organization's
namespace to the Internet.
Enable all Active Directory client computers to resolve all
of the organization's internal and external names.
Make sure that client computers requiring access to the
Internet can resolve any names from the Internet.
Assuring Client Name Resolution to the Organization's
External Resources

The external DNS server should only have resource
records for the external hosts and services that will be
exposed to users of the Internet. To allow internal
clients to gain access to your organization's Internet
data, you can place servers with duplicate content on
the internal network, using the same DNS names as the
servers on the Internet. Therefore, the external DNS
servers will resolve the organization's DNS names to an
external IP address, and the internal DNS servers will
resolve the organization's DNS name to an internal IP
address.
Assuring Client Name Resolution of Internet Hosts

Additional configuration may be required for the internal
hosts to access public Internet resources. You can
configure the internal DNS server to forward all
irresolvable DNS requests to an external DNS server. If a
proxy server is used, create an exclusion list for the
clients. The proxy server will forward all requests
received to an external DNS server.
Considerable planning will be required for the firewall to
specify exactly what types of traffic from internal clients
will be allowed through the firewall to the external
network. This can be accomplished by using either a
firewall or a combination of a firewall and a proxy
server.
Using a Delegated Subdomain Name for the Internal
Network



Create a New DNS Zone in New
Domain
Configure Authoritative DNS
Server in Existing DNS Domain
to Delegate to New Domain
Create Active Directory Forest
Root in New Domain
contoso.msft
Zone 1
Firewall
ad.contoso.msft
Zone 2

Instead of using the organization's registered DNS
domain name as the root domain of Active Directory,
you can use a DNS child domain. The child domain
exists in a separate zone, and is designated as the
Active Directory root domain. You should only expose
the zone containing the DNS root domain to the Internet.

For example, the company called Contoso, Ltd. decides
to use the domain ad.contoso.msft for its Active
Directory root domain. The Contoso, Ltd. Information
Technology (IT) team must first do the following:



Create a new DNS zone on a separate DNS server that
encompasses the ad.contoso.msft domain.
Configure the DNS server that is authoritative for
contoso.msft with a delegation record to the other DNS
server for the ad.contoso.msft domain.
Create the root of the Active Directory forest in the
ad.contoso.msft domain. The name of the Active
Directory root domain will be ad.contoso.msft.

The implications of using a Delegated Subdomain Name
for the Internal Network are as follows:




This solution allows isolation of all Active Directory data
in its domain or domain tree.
This solution provides a contiguous namespace, creating
less confusion for the administrative staff and clients.
The delegated Active Directory root domain requires its
own DNS server. However, it does not require upgrading
the DNS servers that currently serve the existing
registered DNS domain.
Active Directory name structure is longer because
naming starts at a third-level domain.
Using a Single DNS Domain Name for Public and
Private Networks
Zone for contoso.msft
without internal servers
Public Internet
Firewall
Zone for contoso.msft
with internal servers
Private Internal
Network

You may want to retain a single DNS domain name for
both internal and external use. However, you want to
ensure that the internal network is kept separate from
the Internet.
Separating DNS Zones by Using a Firewall

You can create two DNS zones with the same name on
either side of a firewall, and then manage the contents
of those zones so that the appropriate records are
present in each. In this scenario, the DNS server on the
public network would maintain records only for hosts to
be accessed from the Internet. The DNS server on the
internal network would have the resource records that
would be exposed to all internal hosts, including all
resource records related to Active Directory.
Using the firewall to separate DNS zones requires
administration of the SRV resource records. You must
ensure that only the appropriate resource records are
exposed to external requests.

The implications of using a single DNS domain name for Public and
Private Networks are as follows:





Users can use a single domain name whether accessing resources
from the internal network or from the Internet.
Additional administration by the DNS administrators is required to
ensure the appropriate records are on the internal and external DNS
servers.
Configuration of a firewall can be more complex when protecting the
internal network from the Internet.
No additional names need to be registered with the ICANN.
If external resources are mirrored on the internal network,
synchronizing the data can be challenging.
Using a Different DNS Name for Public and Private
Networks
Zone for contoso.msft
without internal servers
Public Internet
Firewall
Zone for contosoltd.msft
Private Internal
Network

An alternative to creating separate views of a single
domain for public and internal clients is to use different
domain names for the public and private networks. For
example, Contoso, Ltd. could use contoso.msft on the
public network, and use contosoltd.msft for the internal
network. This solution makes the division of public and
private resources very simple. All public resources use
one domain name, and all internal resources on the
private network use another domain name.
 Assuring Client Name Resolution

Remember that you can allow clients to resolve external
names for both your organization and other Internet
hosts by configuring the internal DNS server to forward
all requests it is unable to resolve to the external DNS
server. For proxy clients, create an exclusion list.
Considerable planning will be required for the firewall to
specify exactly what types of traffic from internal clients
will be allowed through the firewall to the external
network. This can be accomplished by using either a
firewall or a combination of a firewall and a proxy
server.

The implications of using a different DNS name for public and private
networks are as follows:

Resources are more manageable and secure because there is a clear
distinction between public and private resources.

The internal naming hierarchy is not exposed on the Internet.

Internal resources are inaccessible from the Internet by using the external
domain name.

You will no longer need to replicate external server content to internal
servers.

When accessing a company resource from the Internet, some users may be
confused by the different name.

It may not be necessary to register additional names with ICANN.

You may need to upgrade DNS servers to provide support for SRV resource
records.

Existing DNS infrastructure and host names can remain unchanged and will
match the Active Directory domain name.

Existing DNS zones and DNS topology can remain unchanged.
Designing a DNS Solution to Integrate with BIND
To Integrate BIND and Microsoft DNS You Can

Use Existing DNS Strategy as the Root of Active
Directory

Create a Subdomain of the Existing DNS Strategy as the
Root of Active Directory

Keep the Existing BIND DNS Strategy, and Register
Another Domain Name for the Root of Active Directory

If your organization mandates that existing BIND servers version
8.2.1 or higher be maintained, you must decide how to integrate
them with Microsoft DNS servers.
Strategy
To Integrate:
Use existing DNS
strategy as the root of
Active Directory
On the BIND DNS server, create the following subdomains:
_msdcs.domainname
_sites.domainname
_tcp. domainname
_udp. domainname
Create a subdomain of
the existing DNS strategy
as the root of Active
Directory
On the BIND DNS server, delegate the subdomain to the Microsoft DNS server.
On the Microsoft DNS server, create the subdomain.
On the Microsoft DNS server, enable dynamic updates.
Configure the necessary Microsoft DNS Secondary servers, or integrate the
zones with Active Directory.
Keep the existing BIND
DNS strategy, and
register another domain
name for the root of
Active Directory
Create the new zone on the Microsoft DNS server as an Active Directory
integrated zone.
Create a secondary zone for the new domain on the BIND server.
Create a secondary zone for the existing DNS name on the Microsoft DNS
server.
Have DNS clients point to either the existing BIND DNS server or the MS DNS
server for the DNS name resolution.

The table summarizes DNS naming strategy design choices and
their implications.
Design Choice
Implications
Delegated
subdomain for the
internal network
Isolates all Active Directory data from the public resources in
its domain or domain tree.
Contiguous namespace.
The delegated Active Directory root domain requires its own
DNS server.
Does not require upgrading any existing DNS servers.
Fully qualified domain names of hosts will be longer.
Users can use a single domain name.
Additional administration is required.
Configuration of a firewall can be more complex.
No additional names need to be registered.
Must synchronize with mirrored external resources.
Management and security are easier due to a clear distinction
between public and private resources.
The internal naming hierarchy is not exposed on the Internet.
Single DNS name
for public and
private networks
Different DNS name
for public and
private networks
Design Guidelines
Naming Strategies Include:

Delegated Subdomain for the Internal Network

Single DNS Name for Public and Private Networks

Different DNS Name for Public and Private Networks

The following flow chart represents a decision tree that
is useful for determining the appropriate naming
strategy for an organization.
Lab A: Planning an Active Directory Naming Strategy
Review

Identifying Business Needs

DNS and Active Directory

Planning Active Directory Domain Names

Designing a DNS Naming Strategy for Active Directory