System Administration Guide: IP Services

Chapter 2 Planning an IPv4 Addressing Scheme (Tasks)

This chapter describes the issues you must resolve in order to create your network in an organized, cost-effective manner. After you resolve these issues, you can devise a network plan as you configure and administer your network in the future.

This chapter contains the following information:

For tasks for configuring a network, refer to Chapter 5, Configuring TCP/IP Network Services and IPv4 Addressing (Tasks).

Network Planning (Task Map)

Task	Description	For Information
1. Plan your hardware requirements and network topology	Determine the types of equipment that you need and the layout of this equipment at your site.	For general network topology questions, refer to Determining the Network Hardware. For IPv6 topology planning, refer to Preparing the Network Topology for IPv6 Support. For information about a specific type of equipment, refer to the equipment manufacturer's documentation.
2. Obtain a registered IP address for your network	Your network must have a unique IP address if you plan to communicate outside your local network, for example, over the Internet.	Refer to Obtaining Your Network's IP Number.
3. Devise an IP addressing scheme for your systems, based on your IPv4 network prefix or IPv6 site prefix.	Determine how addresses are to be deployed at your site.	Refer to Designing an IPv4 Addressing Scheme or refer to Preparing an IPv6 Addressing Plan.
4. Create a list that contains the IP addresses and host names of all machines on your network.	Use the list to build network databases	Refer to Network Databases
5. Determine which name service to use on your network.	Decide whether to use NIS, LDAP, DNS, or the network databases in the local `/etc` directory.	Refer to Selecting a Name Service and Directory Service
6. Establish administrative subdivisions, if appropriate for your network	Decide if your site requires that you divide your network into administrative subdivisions	Refer to Administrative Subdivisions
7. Determine where to place routers in the network design.	If your network is large enough to require routers, create a network topology that supports them.	Refer to Planning for Routers on Your Network
8. If required, design a strategy for subnets.	You might need to create subnets for administering your IP address space or to make more IP addresses available for users.	For IPv4 subnet planning, refer to What Is Subnetting? For IPv6 subnet planning, refer to Creating a Numbering Scheme for Subnets

Determining the Network Hardware

When you design your network, you must decide what type of network best meets the needs of your organization. Some of the planning decisions you must make involve the following network hardware:

The network topology, the layout, and connections of the network hardware
The number of host systems your network can support
The types of hosts that the network supports
The types of servers that you might need
The type of network media to use: Ethernet, Token Ring, FDDI, and so on
Whether you need bridges or routers extend this media or connect the local network to external networks
Whether some systems need separately purchased interfaces in addition to their built in interfaces

Based on these factors, you can determine the size of your local area network.

Note –

How you plan the network hardware is outside the scope of this manual. For assistance, refer to the manuals that come with your hardware.

Deciding on an IP Addressing Format for Your Network

The number of systems that you expect to support affects how you configure your network. Your organization might require a small network of several dozen standalone systems that are located on one floor of a single building. Alternatively, you might need to set up a network with more than 1,000 systems in several buildings. This setup can require you to further divide your network into subdivisions that are called subnets.

When you plan your network addressing scheme, consider the following factors:

The type of IP address that you want to use: IPv4 or IPv6
The number of potential systems on your network
The number of systems that are multihomed or routers, which require an IP address for each interface
Whether to use private addresses on your network
Whether to have a DHCP server that manages pools of IPv4 addresses

The worldwide growth of the Internet since 1990 has resulted in a shortage of available IP addresses. To remedy this situation, the Internet Engineering Task Force (IETF) has developed a number of IP addressing alternatives. Types of IP addresses in use today include the following:

If your organization has been assigned more than one IP address for your network or uses subnets, appoint a centralized authority within your organization to assign network IP addresses. That authority should maintain control of a pool of assigned network IP addresses, and assign network, subnet, and host addresses as required. To prevent problems, ensure that duplicate or random network numbers do not exist in your organization.

IPv4 Addresses

These 32-bit addresses are the original IP addressing format that was designed for TCP/IP. Originally, IP networks have three classes, A, B, and C. The network number that is assigned to a network reflects this class designation plus 8 or more bits to represent a host. Class-based IPv4 addresses require you to configure a netmask for the network number. Furthermore, to make more addresses available for systems on the local network, these addresses were often divided into subnets.

Today, IP addresses are referred to as IPv4 addresses. Although you can no longer obtain class-based IPv4 network numbers from an ISP, many existing networks still have them. For more information about administering IPv4 addresses, refer to Designing Your IPv4 Addressing Scheme.

IPv4 Addresses in CIDR Format

The IETF has developed Classless Inter-Domain Routing (CIDR) addresses as a short to medium term fix for the shortage of IPv4 addresses. In addition, CIDR format was designed as a remedy to the lack of capacity of the global Internet routing tables. An IPv4 address with CIDR notation is 32 bits in length and has the same dotted decimal format. However, CIDR adds a prefix designation after the rightmost byte to define the network portion of the IPv4 address. For more information, refer to Designing Your CIDR IPv4 Addressing Scheme.

DHCP Addresses

The Dynamic Host Configuration Protocol (DHCP) protocol enables a system to receive configuration information from a DHCP server, including an IP address, as part of the booting process. DHCP servers maintain pools of IP address from which to assign addresses to DHCP clients. A site that uses DHCP can use a smaller pool of IP addresses than would be needed if all clients were assigned a permanent IP address. You can set up the Solaris DHCP service to manage your site's IP addresses, or a portion of the addresses. For more information, refer to Chapter 11, About Solaris DHCP (Overview).

IPv6 Addresses

The IETF has deployed 128–bit IPv6 addresses as the long term solution to the shortage of available IPv4 addresses. IPv6 addresses provide greater address space than is available with IPv4. The Solaris OS supports IPv4 and IPv6 addressing on the same host, through the use of dual-stack TCP/IP. As with IPv4 addresses in CIDR format, IPv6 addresses have no notion of network classes or netmasks. As in CIDR, IPv6 addresses use prefixes to designate the portion of the address that defines the site's network. For an introduction to IPv6, refer to IPv6 Addressing Overview.

Private Addresses and Documentation Prefixes

The IANA has reserved a block of IPv4 addresses and an IPv6 site prefix for use on private networks. You can deploy these addresses on systems within an enterprise network but be aware that packets with private addresses cannot be routed across the Internet. For more information on private addresses, refer to Using Private IPv4 Addresses.

Note –

Private IPv4 addresses are also reserved for documentation purposes. The examples in this book use private IPv4 addresses and the reserved IPv6 documentation prefix.

Obtaining Your Network's IP Number

An IPv4 network is defined by a combination of an IPv4 network number plus a network mask, or netmask. An IPv6 network is defined by its site prefix, and, if subnetted, its subnet prefix.

Unless your network plans to be private in perpetuity, your local users most likely need to communicate beyond the local network. Therefore, you must obtain a registered IP number for your network from the appropriate organization before your network can communicate externally. This address becomes the network number for your IPv4 addressing scheme or the site prefix for your IPv6 addressing scheme.

Internet Service Providers provide IP addresses for networks with pricing that is based on different levels of service. Investigate with various ISPs to determine which provides the best service for your network. ISP's typically offer dynamically allocated addresses or static IP addresses to businesses. Some ISPs offer both IPv4 and IPv6 addresses.

If your site is an ISP, you obtain IP address blocks for your customers from the Internet Registry (IR) for your locale. The Internet Assigned Numbers Authority (IANA) is ultimately responsible for delegating registered IP addresses to IRs around the world. Each IR has registration information and templates for the locale that the IR services. For information about the IANA and its IRs, refer to the IANA's IP Address Service page.

Note –

Do not arbitrarily assign IP addresses to your network, even if you are not currently attaching the network to external TCP/IP networks. Instead, use private addresses as described in Using Private IPv4 Addresses.

Designing an IPv4 Addressing Scheme

Note –

For IPv6 address planning information, refer to Preparing an IPv6 Addressing Plan.

This section gives an overview IPv4 addressing to aid you in designing an IPv4 addressing plan. For information on IPv6 addresses, see IPv6 Addressing Overview. For information on DHCP addresses, see Chapter 11, About Solaris DHCP (Overview).

Each IPv4-based network must have the following:

A unique network number that is assigned by either an ISP, an IR, or, for older networks, registered by the IANA. If you plan to use private addresses, the network numbers you devise must be unique within your organization.
Unique IPv4 addresses for the interfaces of every system on the network.
A network mask.

The IPv4 address is a 32-bit number that uniquely identifies a network interface on a system, as explained in How IP Addresses Apply to Network Interfaces. An IPv4 address is written in decimal digits, divided into four 8-bit fields that are separated by periods. Each 8-bit field represents a byte of the IPv4 address. This form of representing the bytes of an IPv4 address is often referred to as the dotted-decimal format.

The following figure shows the component parts of an IPv4 address, 172.16.50.56.

Figure 2–1 IPv4 Address Format

The figure divides the IPv4 address into two parts, network
part and network host, which are described in the next context.

172.16: Registered IPv4 network number. In class-based IPv4 notation, this number also defines the IP network class, Class B in this example, that would have been registered by the IANA.
50.56: Host part of the IPv4 address. The host part uniquely identifies an interface on a system on a network. Note that for each interface on a local network, the network part of the address is the same, but the host part must be different.

If you plan to subnet a class-based IPv4 network, you need to define a subnet mask, or netmask, as explained in netmasks Database.

The next example shows of the CIDR format address 192.168.3.56/22

Figure 2–2 CIDR Format IPv4 Address

The figure shows the three parts of the CIDR address,
network part, host part, and network prefix, which are described in the next
context.

192.168.3: Network part, which consists of the IPv4 network number that is received from an ISP or IR.
56: Host part, which you assign to an interface on a system.
/22: Network prefix, which defines how many bits of the address comprise the network number. The network prefix also provides the subnet mask for the IP address. Network prefixes are also assigned by the ISP or IR.

A Solaris-based network can combine standard IPv4 addresses, CIDR format IPv4 addresses, DHCP addresses, IPv6 addresses, and private IPv4 addresses.

Designing Your IPv4 Addressing Scheme

This section describes the classes into which standard IPv4 address are organized. Though the IANA no longer gives out class-based network numbers, these network numbers are still in use on many networks. You might need to administer the address space for a site with class-based network numbers. For a complete discussion of IPv4 network classes, refer to Network Classes.

The following table shows the division of the standard IPv4 address into network and host address spaces. For each class, “Range” specifies the range of decimal values for the first byte of the network number. “Network Address” indicates the number of bytes of the IPv4 address that are dedicated to the network part of the address. Each byte is represented by xxx. “Host Address” indicates the number of bytes that are dedicated to the host part of the address. For example, in a class A network address, the first byte is dedicated to the network, and the last three bytes are dedicated to the host. The opposite designation is true for a class C network.

Table 2–1 Division of the IPv4 Classes


Class	Byte Range	Network Number	Host Address
A	0–127	`xxx`	`xxx`.`xxx`.`xxx`
B	128–191	`xxx`.`xxx`	`xxx`.`xxx`
C	192–223	`xxx`.`xxx`.`xxx`	`xxx`

The numbers in the first byte of the IPv4 address define whether the network is class A, B, or C. The remaining three bytes have a range from 0–255. The two numbers 0 and 255 are reserved. You can assign the numbers 1–254 to each byte, depending on the network class that was assigned to your network by the IANA.

The following table shows which bytes of the IPv4 address are assigned to you. The table also shows the range of numbers within each byte that are available for you to assign to your hosts.

Table 2–2 Range of Available IPv4 Classes


Network Class	Byte 1 Range	Byte 2 Range	Byte 3 Range	Byte 4 Range
A	0–127	1–254	1–254	1–254
B	128–191	Preassigned by IANA	1–254	1–254
C	192–223	Preassigned by IANA	Preassigned by IANA	1–254

IPv4 Subnet Number

Local networks with large numbers of hosts are sometimes divided into subnets. If you divide your IPv4 network number into subnets, you need to assign a network identifier to each subnet. You can maximize the efficiency of the IPv4 address space by using some of the bits from the host part of the IPv4 address as a network identifier. When used as a network identifier, the specified part of the address becomes the subnet number. You create a subnet number by using a netmask, which is a bitmask that selects the network and subnet parts of an IPv4 address. Refer to Creating the Network Mask for IPv4 Addresses for details.

Designing Your CIDR IPv4 Addressing Scheme

The network classes that originally constituted IPv4 are no longer in use on the global Internet. Today, the IANA distributes classless CIDR format addresses to its registries around the world. Any IPv4 address that you obtain from an ISP is in CIDR format, as shown in Figure 2–2.

The network prefix of the CIDR address indicates how many IPv4 addresses are available for hosts on your network. Note that these host addresses are assigned to interfaces on a host. If a host has more than one physical interface, you need to assign a host address for every physical interface that is in use.

The network prefix of a CIDR address also defines the length of the subnet mask. Most Solaris 10 commands recognize the CIDR prefix designation of a network's subnet mask. However, the Solaris installation program and /etc/netmask file require you to set the subnet mask by using dotted decimal representation. In these two cases, use the dotted decimal representation of the CIDR network prefix, as shown in the next table.

Table 2–3 CIDR Prefixes and Their Decimal Equivalent


CIDR Network Prefix	Available IP Addresses	Dotted Decimal Subnet Equivalent
/19	8,192	255.255.224.0
/20	4,096	255.255.240.0
/21	2,048	255.255.248.0
/22	1024	255.255.252.0
/23	512	255.255.254.0
/24	256	255.255.255.0
/25	128	255.255.255.128
/26	64	255.255.255.192
/27	32	255.255.255.224

For more information on CIDR addresses, refer to the following sources:

For technical details on CIDR, refer to RFC 1519, Classless Inter-Domain Routing (CIDR): an Address Assignment and Aggregation Strategy.
More general information about CIDR is available from Pacific Bell Internet at Classless Inter-Domain Routing (CIDR) Overview.
Another CIDR overview can be found in the Wikipedia article,"Classless inter-domain routing".

Using Private IPv4 Addresses

The IANA has reserved three blocks of IPv4 addresses for companies to use on their private networks. These addresses are defined in RFC 1918, Address Allocation for Private Internets. You can use these private addresses, also known as 1918 addresses, for systems on local networks within a corporate intranet. However, private addresses are not valid on the Internet. Do not use them on systems that must communicate outside the local network.

IPv4 Address Range	netmask
10.0.0.0 - 10.255.255.255	10.0.0.0
172.16.0.0 - 172.31.255.255	172.16.0.0
192.168.0.0 - 192.168.255.255	192.168.0.0

How IP Addresses Apply to Network Interfaces

To connect to the network, a system must have at least one physical network interface. Each network interface must have its own unique IP address. During Solaris installation, you must supply the IP address for the first interface that the installation program finds. Usually that interface has the name device-name0, for example eri0 or hme0. This interface is considered the primary network interface.

If you add a second network interface to a host, that interface also must have its own unique IP address. When you add the second network interface, the host then becomes multihomed. By contrast, when you add a second network interface to a host and enable IP forwarding, that host becomes a router. See Configuring an IPv4 Router for an explanation.

Each network interface has a device name, a device driver, and an associated device file in the /devices directory. The network interface might have a device name such as eri or smc0, which are device names for two commonly used Ethernet interfaces.

For information and tasks related to interfaces, refer to Part I, Administering Single Interfaces, in System Administration Guide: Network Interfaces and Network Virtualization.

Note –

This book assumes that your systems have Ethernet network interfaces. If you plan to use different network media, refer to the manuals that come with the network interface for configuration information.

Naming Entities on Your Network

After you receive your assigned network IP address and you have given the IP addresses to your systems, the next task is to assign names to the hosts. Then you must determine how to handle name services on your network. You use these names initially when you set up your network and later when you expand your network through routers, bridges, or PPP.

The TCP/IP protocols locate a system on a network by using its IP address. However, if you use a recognizable name, then you can easily identify the system. Therefore, the TCP/IP protocols (and the Solaris OS) require both the IP address and the host name to uniquely identify a system.

From a TCP/IP perspective, a network is a set of named entities. A host is an entity with a name. A router is an entity with a name. The network is an entity with a name. A group or department in which the network is installed can also be given a name, as can a division, a region, or a company. In theory, the hierarchy of names that can be used to identify a network has virtually no limit. The domain name identifies a domain.

Administering Host Names

Many sites let users pick host names for their machines. Servers also require at least one host name, which is associated with the IP address of its primary network interface.

As a system administrator, you must ensure that each host name in your domain is unique. In other words, no two machines on your network can both have the name “fred.” However, the machine “fred” might have multiple IP addresses.

When planning your network, make a list of IP addresses and their associated host names for easy access during the setup process. The list can help you verify that all host names are unique.

Selecting a Name Service and Directory Service

The Solaris OS enables you to use three types of name services: local files, NIS, and DNS. Name services maintain critical information about the machines on a network, such as the host names, IP addresses, Ethernet addresses, and so forth. The Solaris OS also gives you the option of using the LDAP directory service in addition to or instead of a name service. For an introduction to name services on Solaris, refer to Part I, About Naming and Directory Services, in System Administration Guide: Naming and Directory Services (DNS, NIS, and LDAP).

Network Databases

When you install the operating system, you supply the host name and IP address of your server, clients, or standalone system as part of the procedure. The Solaris installation program adds this information into the hosts This database is part of a set of network databases that contain information necessary for TCP/IP operation on your network. The name service that you select for your network reads these databases.

The configuration of the network databases is critical. Therefore, you need to decide which name service to use as part of the network planning process. Moreover, the decision to use name services also affects whether you organize your network into an administrative domain. Network Databases and the nsswitch.conf File has detailed information on the set of network databases.

Using NIS or DNS as the Name Service

The NIS and DNS name services maintain network databases on several servers on the network. System Administration Guide: Naming and Directory Services (DNS, NIS, and LDAP) describes these name services and explains how to configure the databases. In addition, the guide explain the “namespace” and “administrative domain” concepts in detail.

Using Local Files as the Name Service

If you do not implement NIS, LDAP, or DNS, the network uses local files to provide the name service. The term “local files” refers to the series of files in the /etc directory that the network databases use. The procedures in this book assume you are using local files for your name service, unless otherwise indicated.

Note –

If you decide to use local files as the name service for your network, you can set up another name service at a later date.

Domain Names

Many networks organize their hosts and routers into a hierarchy of administrative domains. If you are using the NIS or DNS name service, you must select a domain name for your organization that is unique worldwide. To ensure that your domain name is unique, you should register the domain name with the InterNIC. If you plan to use DNS, you also need to register your domain name with the InterNIC.

The domain name structure is hierarchical. A new domain typically is located below an existing, related domain. For example, the domain name for a subsidiary company can be located below the domain of the parent company. If the domain name has no other relationship, an organization can place its domain name directly under one of the existing top-level domains.

The following are a few examples of top-level domains:

.com – Commercial companies (international in scope)
.edu – Educational institutions (international in scope)
.gov – U.S. government agencies
.fr – France

You select the name that identifies your organization, with the provision that the name must be unique.

Administrative Subdivisions

The question of administrative subdivisions deals with matters of size and control. The more hosts and servers that you have in a network, the more complex your management task. You might want to handle such situations by setting up additional administrative divisions. Add networks of a particular class. Divide existing networks into subnets. The decision about setting up administrative subdivisions for your network is determined by the following factors:

How large is the network?

A single administrative division can handle a single network of several hundred hosts, all in the same physical location and requiring the same administrative services. However, sometimes you should establish several administrative subdivisions. Subdivisions are particularly useful if you have a small network with subnets and the network is scattered over an extensive geographical area.

Do users on the network have similar needs?

For example, you might have a network that is confined to a single building and supports a relatively small number of machines. These machines are divided among a number of subnetworks. Each subnetwork supports groups of users with different needs. In this example, you might use an administrative subdivision for each subnet.

Planning for Routers on Your Network

Recall that in TCP/IP, two types of entities exist on a network: hosts and routers. All networks must have hosts, while not all networks require routers. The physical topology of the network determines if you need routers. This section introduces the concepts of network topology and routing. These concepts are important when you decide to add another network to your existing network environment.

Note –

For complete details and tasks for router configuration on IPv4 networks, refer to Packet Forwarding and Routing on IPv4 Networks. For complete details and tasks for router configuration on IPv6 networks, refer to Configuring an IPv6 Router.

Network Topology Overview

Network topology describes how networks fit together. Routers are the entities that connect networks to each other. A router is any machine that has two or more network interfaces and implements IP forwarding. However, the system cannot function as a router until properly configured, as described in Configuring an IPv4 Router.

Routers connect two or more networks to form larger internetworks. The routers must be configured to pass packets between two adjacent networks. The routers also should be able to pass packets to networks that lie beyond the adjacent networks.

The following figure shows the basic parts of a network topology. The first illustration shows a simple configuration of two networks that are connected by a single router. The second illustration shows a configuration of three networks, interconnected by two routers. In the first example, Router R joins Network 1 and Network 2 into a larger internetwork. In the second example, Router R1 connects Networks 1 and 2. Router R2 connects Networks 2 and 3. The connections form a network that includes Networks 1, 2, and 3.

Figure 2–3 Basic Network Topology

Diagram shows the topology of two networks that are connected
by a single router.

In addition to joining networks into internetworks, routers route packets between networks that are based on the addresses of the destination network. As internetworks grow more complex, each router must make more and more decisions about the packet destinations.

The following figure shows a more complex case. Router R3 directly connects networks 1 and 3. The redundancy improves reliability. If network 2 goes down, router R3 still provides a route between networks 1 and 3. You can interconnect many networks. However, the networks must use the same network protocols.

Figure 2–4 A Network Topology That Provides an Additional Path Between Networks

Diagram shows the topology of three networks that are
connected by two routers.

How Routers Transfer Packets

The IP address of the recipient, which is a part of the packet header, determines how the packet is routed. If this address includes the network number of the local network, the packet goes directly to the host with that IP address. If the network number is not the local network, the packet goes to the router on the local network.

Routers maintain routing information in routing tables. These tables contain the IP address of the hosts and routers on the networks to which the router is connected. The tables also contain pointers to these networks. When a router receives a packet, the router checks its routing table to determine if the table lists the destination address in the header. If the table does not contain the destination address, the router forwards the packet to another router that is listed in its routing table. Refer to Configuring an IPv4 Router for detailed information on routers.

The following figure shows a network topology with three networks that are connected by two routers.

Figure 2–5 A Network Topology With Three Interconnected Networks

Diagram shows a sample of three networks that are connected
by two routers.

Router R1 connects networks 192.9.200 and 192.9.201. Router R2 connects networks 192.9.201 and 192.9.202. If Host A on network 192.9.200 sends a message to Host B on network 192.9.202, the following events occur:

Host A sends a packet out over network 192.9.200. The packet header contains the IPv4 address of the recipient Host B, 192.9.202.10.
None of the machines on network 192.9.200 has the IPv4 address 192.9.202.10. Therefore, Router R1 accepts the packet.
Router R1 examines its routing tables. No machine on network 192.9.201 has the address 192.9.202.10. However, the routing tables do list Router R2.
R1 then selects R2 as the “next hop” Router. R1 sends the packet to R2.
Because R2 connects network 192.9.201 to 192.9.202, R2 has routing information for Host B. Router R2 then forwards the packet to network 192.9.202, where Host B accepts the packet.