This chapter contains tasks for configuring IPv6 on a network. The following major topics are covered:Configuring an IPv6 Interface Enabling IPv6 on an Interface (Task Map) Configuring an IPv6 Router Modifying an IPv6 Interface Configuration for Hosts and Servers Modifying an IPv6 Interface Configuration (Task Map) Configuring Name Service Support for IPv6 For an overview of IPv6 concepts, refer to Chapter 3, Planning an IPv6 Addressing Scheme (Overview). For IPv6 planning tasks, refer to Chapter 4, Planning an IPv6 Network (Tasks). To find reference information about the tasks in this chapter, refer to Chapter 11, IPv6 in Depth (Reference). The initial step in IPv6 configuration is enabling IPv6 on an interface. You can enable IPv6 support during the Solaris 10 installation process or by configuring IPv6 on the interfaces of an installed system. During the Solaris 10 installation process, you can enable IPv6 on one or more of a system's interfaces. After installation, the following IPv6-related files and tables are in place:Each interface that was enabled for IPv6 now has an associated /etc/hostname6.interface file, such as hostname6.dmfe0. The /etc/nsswitch.conf file has been modified to accommodate lookups using IPv6 addresses. The IPv6 address selection policy table is created. This table prioritizes the IP address format to use for transmissions over an IPv6-enabled interface. This section describes how to enable IPv6 on the interfaces of an installed system. Task Description For Instructions Enable IPv6 on an interface on a system that has already been installed with the Solaris 10 OS. Use this task for enabling IPv6 on an interface after the Solaris 10 OS has been installed. How to Enable an IPv6 Interface for the Current Session Make the IPv6-enabled interface persist across reboots. Use this task to make the IPv6 address of the interface permanent. How to Enable Persistent IPv6 Interfaces Turn off IPv6 address autoconfiguration. Use this task if you need to manually configure the interface ID portion of the IPv6 address. How to Turn Off IPv6 Address Autoconfiguration Begin your IPv6 configuration process by enabling IPv6 on the interfaces of all systems that will become IPv6 nodes. Initially, the interface obtains its IPv6 address through the autoconfiguration process, as described in IPv6 Address Autoconfiguration. You then can tailor the node's configuration based on its function in the IPv6 network, either as a host, server, or router. If the interface is on the same link as a router that currently advertises an IPv6 prefix, the interface obtains that site prefix as part of its autoconfigured addresses. For more information, refer to How to Configure an IPv6-Enabled Router. The following procedure explains how to enable IPv6 for an interface that was added after Solaris 10 installation. Complete the planning tasks for the IPv6 network, such as upgrading hardware and software, and preparing an addressing plan. For more information, see IPv6 Planning (Task Maps). Log in to the prospective IPv6 node as Primary Administrator or as superuser.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. Enable IPv6 on an interface.# ifconfig inet6 interface plumb up Start the IPv6 daemonin.ndpd.# /usr/lib/inet/in.ndpdYou can display the status of a node's IPv6-enabled interfaces by using the ifconfiga6 command. This example shows how to enable IPv6 on the qfe0 interface. Before you begin, check the status of all interfaces configured on the system.# ifconfig -a lo0: flags=1000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 qfe0: flags=1000863 <UP,BROADCAST,NOTRAILERS,RUNNING,MULTICAST,IPv4> mtu 1500 index 2 inet 172.16.27.74 netmask ffffff00 broadcast 172.16.27.255 ether 0:3:ba:13:14:e1 Only the qfe0 interface is currently configured for this system. Enable IPv6 on this interface as follows:# ifconfig inet6 qfe0 plumb up # /usr/lib/inet/in.ndpd # ifconfig -a6 lo0: flags=2000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv6> mtu 8252 index 1 inet6 ::1/128 qfe0: flags=2000841 <UP,RUNNING,MULTICAST,IPv6> mtu 1500 index 2 ether 0:3:ba:13:14:e1 inet6 fe80::203:baff:fe13:14e1/10The example shows the status of the system's interface before and after qfe0becomes IPv6-enabled. The a6 option of ifconfig shows just the IPv6 information for qfe0 and the loopback interface. Note that the output indicates that only a link-local address was configured for qfe0, fe80::203:baff:fe13:14e1/10. This address indicates that as of yet no router on the node's local link advertises a site prefix.After IPv6 is enabled, you can use the ifconfig a command to display both IPv4 and IPv6 addresses for all interfaces on a system. To configure the IPv6 node as a router, go to Configuring an IPv6 Router. To maintain the IPv6 interface configuration across reboots, see How to Enable Persistent IPv6 Interfaces. To disable address autoconfiguration on the node, see How to Turn Off IPv6 Address Autoconfiguration. To tailor the node as a server, see the suggestions in Administering IPv6-Enabled Interfaces on Servers. This procedure explains how to enable IPv6 interfaces with autoconfigured IPv6 addresses that persist across subsequent reboots.If the interface is on the same link as a router that currently advertises an IPv6 prefix, the interface obtains that site prefix as part of its autoconfigured addresses. For more information, refer to How to Configure an IPv6-Enabled Router. Log in to the IPv6 node as Primary Administrator or as superuser.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. Create IPv6 addresses for interfaces that were added after installation.# touch /etc/hostname6.interface (Optional) Create an /etc/inet/ndpd.conf file that defines parameters for interface variables on the node.If you need to create temporary addresses for the host's interface, refer to Using Temporary Addresses for an Interface. For details about /etc/inet/ndpd.conf, refer to the ndpd.conf4 man page and ndpd.conf Configuration File. Reboot the node.# reboot -- -rThe reboot process sends router discovery packets. If a router responds with a site prefix, the node can configure any interface with a corresponding /etc/hostname6.interface file with a global IPv6 address. Otherwise, the IPv6-enabled interfaces are configured solely with link-local addresses. Rebooting also restarts in.ndpd and other network daemons in IPv6 mode. This example shows how to make the IPv6 configuration for the qfe0 interface persist across reboots. In this example, a router on the local link advertises the site prefix and subnet ID 2001:db8:3c4d:15/64.First, check the status of the system's interfaces.# ifconfig -a lo0: flags=1000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 qfe0: flags=1000863 <UP,BROADCAST,NOTRAILERS,RUNNING,MULTICAST,IPv4> mtu 1500 index 2 inet 172.16.27.74 netmask ffffff00 broadcast 172.16.27.255 ether 0:3:ba:13:14:e1 # touch /etc/hostname6.qfe0 # reboot -- -rVerify that the IPv6 address you configured is still applied to the qfe0 interface.# ifconfig -a6 qfe0: flags=2000841 <UP,RUNNING,MULTICAST,IPv6> mtu 1500 index 2 ether 0:3:ba:13:14:e1 inet6 fe80::203:baff:fe13:14e1/10 qfe0:1: flags=2180841 <UP,RUNNING,MULTICAST,ADDRCONF,IPv6> mtu 1500 index 2 inet6 2001:db8:3c4d:15:203:baff:fe13:14e1/64The output of ifconfig a6 shows two entries for qfe0. The standard qfe0 entry includes the MAC address and the link-local address. A second entry, qfe0:1, indicates that a pseudo-interface was created for the additional IPv6 address on the qfe0 interface. The new, global IPv6 address, 2001:db8:3c4d:15:203:baff:fe13:14e1/64, includes the site prefix and subnet ID advertised by the local router. To configure the new IPv6 node as a router, go to Configuring an IPv6 Router. To disable address autoconfiguration on the node, see How to Turn Off IPv6 Address Autoconfiguration. To tailor the new node as a server, see the suggestions in Administering IPv6-Enabled Interfaces on Servers. You normally should use address autoconfiguration to generate the IPv6 addresses for the interfaces of hosts and servers. However, sometimes you might want to turn off address autoconfiguration, especially if you want to manually configure a token, as explained in Configuring an IPv6 Token. Log in to the IPv6 node as Primary Administrator or as superuser.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. Create an /etc/inet/ndpd.conf file for the node.The /etc/inet/ndpd.conf file defines interface variables for the particular node. This file should have the following contents in order to turn off address autoconfiguration for all of the server's interfaces:if-variable-name StatelessAddrConf falseFor details about /etc/inet/ndpd.conf, refer to the ndpd.conf4 man page and ndpd.conf Configuration File. Update the IPv6 daemon with your changes.# pkill -HUP in.ndpd The first step in configuring IPv6 on a network is configuring IPv6 on a router. Router configuration involves a number of discrete tasks, which are described in this section. You might perform some or all of the tasks, depending on your site requirements.Perform the next tasks in the order that is shown in the table.Task Description For Instructions 1. Ensure that you have completed the required prerequisites before you begin IPv6 configuration. You must complete planning tasks and Solaris installation with IPv6 enabled interfaces before you configure an IPv6-enabled router. Chapter 4, Planning an IPv6 Network (Tasks) and Configuring an IPv6 Interface. 2. Configure a router. Define the site prefix for the network. How to Configure an IPv6-Enabled Router 3. Configure tunnel interfaces on the router. Set up a manual tunnel or a 6to4 tunnel interface on the router. The local IPv6 network needs tunnels to communicate with other, isolated IPv6 networks. How to Configure a 6to4 Tunnel How to Manually Configure IPv6 Over IPv4 Tunnels How to Manually Configure IPv6 Over IPv6 Tunnels How to Configure IPv4 Over IPv6 Tunnels 4. Configure the switches on the network. If your network configuration includes switches, configure them for IPv6 at this point in the configuration process. Refer to switch manufacturer's documentation. 5. Configure any hubs on your network. If your network configuration includes hubs, configure them for IPv6 at this point in the configuration process. Refer to hub manufacturer's documentation. 6. Configure the network name service for IPv6. Configure your primary name service (DNS, NIS, or LDAP) to recognize IPv6 addresses after the router is configured for IPv6. How to Add IPv6 Addresses to DNS and How to Display IPv6 Information Through NIS 7. (Optional) Modify the addresses for the IPv6-enabled interfaces on hosts and servers. After IPv6 router configuration, make further modifications on IPv6-enabled hosts and servers. Modifying an IPv6 Interface Configuration for Hosts and Servers Configure applications to support IPv6 Different applications might require different actions in order to support IPv6. Refer to applications' documentation This procedure assumes that all interfaces of the router were configured for IPv6 during Solaris installation. On the system that will become the IPv6 router, assume the Primary Administrator role or become superuser.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. Review which interfaces on the router were configured for IPv6 during installation.# ifconfig -aCheck the output to ensure that the interfaces that you wanted to configure for IPv6 are now plumbed with link-local addresses. The following sample command output of ifconfig -a shows the IPv4 and IPv6 addresses that were configured for the router's interfaces. o0: flags=1000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 dmfe0: flags=1000843 <UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2 inet 172.16.26.232 netmask ffffff00 broadcast 172.16.26.255 ether 0:3:ba:11:b1:15 dmfe1: flags=1000843 <UP,BROADCAST,RUNNING,MULTICAST,IPv4 mtu 1500 index 3 inet 172.16.26.220 netmask ffffff00 broadcast 172.16.26.255 ether 0:3:ba:11:b1:16 lo0: flags=2000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv6> mtu 8252 index 1 inet6 ::1/128 dmfe0: flags=2000841 <UP,RUNNING,MULTICAST,IPv6> mtu 1500 index 2 ether 0:3:ba:11:b1:15 inet6 fe80::203:baff:fe11:b115/10 dmfe1: flags=2000841 <UP,RUNNING,MULTICAST,IPv6> mtu 1500 index 3 ether 0:3:ba:11:b1:16 inet6 fe80::203:baff:fe11:b116/10 The output also shows that the primary network interface dmfe0 and the additional interface dmfe1 were configured during installation with the IPv6 link–local addresses fe80::203:baff:fe11:b115/10 and fe80::203:baff:fe11:b116/10. Configure IPv6 packet forwarding on all interfaces of the router.Use either of the following to enable packet forwarding:Use the routeadm command, as follows:# routeadm -e ipv6-forwarding -u Use the following Service Management Facility (SMF) command, as follows:.# svcadm enable ipv6-forwarding Start the routing daemon.The in.ripngd daemon handles IPv6 routing. Turn on IPv6 routing in either of the following ways:Use the routeadm command as follows:# routeadm -e ipv6-routing -u Use SMF to enable IPv6 routing:# svcadm enable ripng:default For syntax information on the routeadm command, see the routeadm1M man page. Create the /etc/inet/ndpd.conf file. You specify the site prefix to be advertised by the router and other configuration information in /etc/inet/ndpd.conf. This file is read by the in.ndpd daemon, which implements the IPv6 Neighbor Discovery protocol.For a list of variables and allowable values, refer to ndpd.conf Configuration File and the ndpd.conf4man page. Type the following text into the /etc/inet/ndpd.conf file:ifdefault AdvSendAdvertisements true prefixdefault AdvOnLinkFlag on AdvAutonomousFlag onThis text tells the in.ndpd daemon to send out router advertisements over all interfaces of the router that are configured for IPv6. Add additional text to the /etc/inet/ndpd.conf file to configure the site prefix on the various interfaces of the router.The text should have the following format:prefix global-routing-prefix:subnet ID/64 interfaceThe following sample /etc/inet/ndpd.conf file configures the router to advertise the site prefix 2001:0db8:3c4d::/48 over the interfaces dmfe0 and dmfe1.ifdefault AdvSendAdvertisements true prefixdefault AdvOnLinkFlag on AdvAutonomousFlag on if dmfe0 AdvSendAdvertisements 1 prefix 2001:0db8:3c4d:15::0/64 dmfe0 if dmfe1 AdvSendAdvertisements 1 prefix 2001:0db8:3c4d:16::0/64 dmfe1 Reboot the system.The IPv6 router begins advertising on the local link any site prefix that is in the ndpd.conf file. The following example shows output from the ifconfig a command such as you would receive after you finish the Configuring an IPv6 Router procedure.lo0: flags=1000849 <UP LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1 inet 127.0.0.1 netmask ff000000 dmfe0: flags=1000843 <UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2 inet 172.16.15.232 netmask ffffff00 broadcast 172.16.26.255 ether 0:3:ba:11:b1:15 dmfe1: flags=1000843 <UP,BROADCAST,RUNNING,MULTICAST,IPv4 mtu 1500 index 3 inet 172.16.16.220 netmask ffffff00 broadcast 172.16.26.255 ether 0:3:ba:11:b1:16 lo0: flags=2000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv6> mtu 8252 index 1 inet6 ::1/128 dmfe0: flags=2100841 <UP,RUNNING,MULTICAST,ROUTER,IPv6> mtu 1500 index 2 ether 0:3:ba:11:b1:15 inet6 fe80::203:baff:fe11:b115/10 dmfe0:1: flags=2180841 <UP,RUNNING,MULTICAST,ADDRCONF,ROUTER,IPv6> mtu 1500 index 2 inet6 2001:db8:3c4d:15:203:baff:fe11:b115/64 dmfe1: flags=2100841 <UP,RUNNING,MULTICAST,ROUTER,IPv6> mtu 1500 index 3 ether 0:3:ba:11:b1:16 inet6 fe80::203:baff:fe11:b116/10 dmfe1:1: flags=2180841 <UP,RUNNING,MULTICAST,ADDRCONF,ROUTER,IPv6> mtu 1500 index 3 inet6 2001:db8:3c4d:16:203:baff:fe11:b116/64In this example, each interface that was configured for IPv6 now has two addresses. The entry with the name of the interface, such as dmfe0, shows the link-local address for that interface. The entry with the form interface:n, such as dmfe0:1, shows a global IPv6 address. This address includes the site prefix that you configured in the /etc/ndpd.conf file, in addition to the interface ID. To configure any tunnels from the routers that you have identified in your IPv6 network topology, refer to Configuring Tunnels for IPv6 Support. For information about configuring switches and hubs on your network, refer to the manufacturer's documentation. To configure IPv6 hosts, refer to Modifying an IPv6 Interface Configuration for Hosts and Servers. To improve IPv6 support on servers, refer to Administering IPv6-Enabled Interfaces on Servers. For detailed information about IPv6 commands, files, and daemons, refer to Solaris 10 IPv6 Implementation. This section explains how to modify the configuration of IPv6-enabled interfaces on nodes that are hosts or servers. In most instances, you should use address autoconfiguration for IPv6-enabled interfaces, as explained in Stateless Autoconfiguration Overview. However, you can modify the IPv6 address of an interface, if necessary, as explained in the tasks of this section. Task Description For Instructions Turn off IPv6 address autoconfiguration. Use this task if you need to manually configure the interface ID portion of the IPv6 address. How to Turn Off IPv6 Address Autoconfiguration Create a temporary address for a host. Hide a host's interface ID by configuring a randomly created temporary address that is used as the lower 64 bits of the address. How to Configure a Temporary Address Configure a token for the interface ID of a system. Create a 64-bit token to be used as the interface ID in an IPv6 address. How to Configure a User-Specified IPv6 Token An IPv6 temporary address includes a randomly generated 64-bit number as the interface ID, instead of an interface's MAC address. You can use temporary addresses for any interfaces on an IPv6 node that you want to keep anonymous. For example, you might want to use temporary addresses for the interfaces of a host that needs to access public web servers. Temporary addresses implement IPv6 privacy enhancements. These enhancements are described in RFC 3041, available at “Privacy Extensions for Stateless Address Autoconfiguration in IPv6”.You enable a temporary address in the /etc/inet/ndpd.conf file for one or more interfaces, if needed. However, unlike standard, autoconfigured IPv6 addresses, a temporary address consists of the 64-bit subnet prefix and a randomly generated 64-bit number. This random number becomes the interface ID segment of the IPv6 address. A link-local address is not generated with the temporary address as the interface ID.Be aware that temporary addresses have a default preferred lifetime of one day. When you enable temporary address generation, you may also configure the following variables in the /etc/inet/ndpd.conf file:valid lifetime TmpValidLifetimeTime span in which the temporary address exists, after which the address is deleted from the host. preferred lifetime TmpPreferredLifetimeElapsed time before the temporary address is deprecated. This time span should be shorter than the valid lifetime. address regenerationDuration of time before the expiration of the preferred lifetime, during which the host should generate a new temporary address. You express the duration of time for temporary addresses as follows:nn number of seconds, which is the default n hn number of hours (h) n dn number of days (d) Log in to the IPv6 host as Primary Administrator or as superuser.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. If necessary, enable IPv6 on the host's interfacesRefer to How to Enable an IPv6 Interface for the Current Session. Edit the /etc/inet/ndpd.conf file to turn on temporary address generation.To configure temporary addresses on all interfaces of a host, add the following line to /etc/inet/ndpd.conf:ifdefault TmpAddrsEnabled true To configure a temporary address for a specific interface, add the following line to /etc/inet/ndpd.conf:if interface TmpAddrsEnabled true (Optional) Specify the valid lifetime for the temporary address. ifdefault TmpValidLifetime durationThis syntax specifies the valid lifetime for all interfaces on a host. The value for duration should be in seconds, hours, or days. The default valid lifetime is 7 days. You can also use TmpValidLifetime with the if interface keywords to specify the valid lifetime for a temporary address of a particular interface. (Optional) Specify a preferred lifetime for the temporary address, after which the address is deprecated.if interface TmpPreferredLifetime durationThis syntax specifies the preferred lifetime for the temporary address of a particular interface. The default preferred lifetime is one day. You can also use TmpPreferredLifetime with the ifdefault keyword to specify the preferred lifetime for the temporary addresses on all interfaces of a host.Default address selection gives a lower priority to IPv6 addresses that have been deprecated. If an IPv6 temporary address is deprecated, default address selection chooses a nondeprecated address as the source address of a packet. A nondeprecated address could be the automatically generated IPv6 address, or possibly, the interface's IPv4 address. For more information about default address selection, see Administering Default Address Selection. (Optional) Specify the lead time in advance of address deprecation, during which the host should generate a new temporary address.ifdefault TmpRegenAdvance durationThis syntax specifies the lead time in advance of address deprecation for the temporary addresses of all interfaces on a host. The default is 5 seconds. Change the configuration of the in.ndpd daemon.# pkill -HUP in.ndpd # /usr/lib/inet/in.ndpd Verify that temporary addresses have been created by running the ifconfig a6 command, as shown in Example 7–5.The output from ifconfig should have the word TEMPORARY in the same line as the interface definition. The following example shows a segment of an /etc/inet/ndpd.conf file with temporary addresses enabled for the primary network interface.ifdefault TmpAddrsEnabled true ifdefault TmpValidLifetime 14d ifdefault TmpPreferredLifetime 7d ifdefault TmpRegenAdvance 6s This example shows the output of the ifconfig command after temporary addresses are created.# ifconfig -a6 lo0: flags=2000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv6> mtu 8252 index 1 inet6 ::1/128 hme0: flags=2000841 <UP,RUNNING,MULTICAST,IPv6> mtu 1500 index 2 ether 8:0:20:b9:4c:54 inet6 fe80::a00:20ff:feb9:4c54/10 hme0:1: flags=2080841 <UP,RUNNING,MULTICAST,ADDRCONF,IPv6> mtu 1500 index 2 inet6 2001:db8:3c4d:15:a00:20ff:feb9:4c54/64 hme0:2: flags=802080841<UP,RUNNING,MULTICAST,ADDRCONF,IPv6,TEMPORARY> mtu 1500 index 2 inet6 2001:db8:3c4d:15:7c37:e7d1:fc9c:d2cb/64Note that the line following interface hme0:2 includes the word TEMPORARY. This designation indicates that the address 2001:db8:3c4d:15:7c37:e7d1:fc9c:d2cb/64 has a temporary interface ID. To set up name service support for IPv6 addresses, see Configuring Name Service Support for IPv6. To configure IPv6 addresses for a server, see How to Configure a User-Specified IPv6 Token. To monitor activities on IPv6 nodes, see Chapter 8, Administering a TCP/IP Network (Tasks). The 64-bit interface ID of an IPv6 address is also referred to as a token, as introduced in IPv6 Addressing Overview. During address autoconfiguration, the token is associated with the interface's MAC address. In most cases, nonrouting nodes, that is IPv6 hosts and servers, should use their autoconfigured tokens.However, using autoconfigured tokens can be a problem for servers whose interfaces are routinely swapped as part of system maintenance. When the interface card is changed, the MAC address is also changed. Servers that depend on having stable IP addresses can experience problems as a result. Various parts of the network infrastructure, such as DNS or NIS, might have stored specific IPv6 addresses for the interfaces of the server.To avoid address change problems, you can manually configure a token to be used as the interface ID in an IPv6 address. To create the token, you specify a hexadecimal number of 64 bits or less to occupy the interface ID portion of the IPv6 address. During subsequent address autoconfiguration, Neighbor Discovery does not create an interface ID that is based on the interface's MAC address. Instead, the manually created token becomes the interface ID. This token remains assigned to the interface, even when a card is replaced.The difference between user-specified tokens and temporary addresses is that temporary addresses are randomly generated, rather than explicitly created by a user. The next instructions are particularly useful for servers whose interfaces are routinely replaced. They also are valid for configuring user-specified tokens on any IPv6 node. Assume the Primary Administrator role or become superuser on the node.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. Verify that the interface you want to configure with a token is plumbed.An interface must be plumbed before you can configure a token for its IPv6 address.# ifconfig a6qfe0: flags=2000841 <UP,RUNNING,MULTICAST,IPv6> mtu 1500 index 2 ether 0:3:ba:13:14:e1 inet6 fe80::203:baff:fe13:14e1/10This output shows that the network interface qfe0 is plumbed and has the link-local address fe80::203:baff:fe13:14e1/10. This address was automatically configured during installation. Create one or more 64-bit hexadecimal numbers to be used as tokens for the node's interfaces. For examples of tokens, refer to Link-Local Unicast Address. Configure each interface with a token.Use the following form of the ifconfig command for each interface to have a user-specified interface ID (token):ifconfig interface inet6 token address/64For example, you would use the following command to configure interface qfe0 with a token:# ifconfig qfe0 inet6 token ::1a:2b:3c:4d/64 Repeat this step for every interface that will have a user-specified token. (Optional) Make the new IPv6 address persist across reboots.Edit or create an /etc/hostname6.interface file for each interface you configured with a token. Add the following text at the bottom of each /etc/hostname.6interface file:token ::token-name/64For example, you might add the following text to the bottom of an/etc/hostname6.interface file:token ::1a:2b:3c:4d/64 After the system reboots, the token that you configured in an /etc/hostname6.interface file is applied to the interface's IPv6 address. This IPv6 address remains persistent across subsequent reboots. Update the IPv6 daemon with your changes.# pkill -HUP -in.ndpd In the following example, the interface bge0:1 has an autoconfigured IPv6 address. The subnet prefix 2001:db8:3c4d:152:/64 is advertised by a router on the node's local link. The interface ID 2c0:9fff:fe56:8255 is generated from bge0:1's MAC address.# ifconfig -a6 lo0: flags=2002000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv6,VIRTUAL> mtu 8252 index 1 inet6 ::1/128 bge0: flags=2100801 <UP,MULTICAST,IPv6> mtu 1500 index 5 inet6 fe80::2c0:9fff:fe56:8255/10 ether 0:c0:9f:56:82:55 bge0:1: flags=2180801 <UP, MULTICAST,ADDRCONF,IPv6>mtu 1500 index 5 inet6 2001:db8:3c4d:152:c0:9fff:fe56:8255/64 # ifconfig bge0 inet6 token ::1a:2b:3c:4d/64 # vi /etc/hostname6.bge0 token ::1a:2b:3c:4d/64 # pkill -HUP -in.ndpd # ifconfig -a6 lo0: flags=2002000849 <UP,LOOPBACK,RUNNING,MULTICAST,IPv6,VIRTUAL> mtu 8252 index 1 inet6 ::1/128 bge0: flags=2100801 <UP,MULTICAST,IPv6> mtu 1500 index 5 inet6 fe80::2c0:9fff:fe56:8255/10 ether 0:c0:9f:56:82:55 bge0:1: flags=2180801 <UP, MULTICAST,ADDRCONF,IPv6>mtu 1500 index 5 inet6 2001:db8:3c4d:152:1a:2b:3c:4d/64After the token is configured, the global address on the second status line of bge0:1 now has 1a:2b:3c:4dconfigured for its interface ID. To update the name services with the IPv6 addresses of the server, see Configuring Name Service Support for IPv6. To monitor server performance, see Chapter 8, Administering a TCP/IP Network (Tasks). When you plan for IPv6 on a server, you must make a few decisions as you enable IPv6 on the server's interfaces. Your decisions affect the strategy to use for configuring the interface IDs, also known as tokens, of an interface's IPv6 address.The next procedure assumes the following:Solaris 10 OS is already installed on the server. You enabled IPv6 on the server's interfaces either during Solaris OS installation or later, using the procedures in Configuring an IPv6 Interface. If applicable, upgrade the application software to support IPv6. Note that many applications that run on the IPv4 protocol stack also successfully run on IPv6. For more information, refer to How to Prepare Network Services for IPv6 Support. On the server, assume the Primary Administrator role or become superuser.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. Ensure that an IPv6 subnet prefix is configured on a router on the same link as the server. For more information, refer to Configuring an IPv6 Router. Use the appropriate strategy for the interface ID for the server's IPv6-enabled interfaces.By default, IPv6 address autoconfiguration uses the MAC address of an interface when creating the interface ID portion of the IPv6 address. If the IPv6 address of the interface is well known, swapping one interface for another interface can cause problems. The MAC address of the new interface will be different. During address autoconfiguration, a new interface ID is generated.For an IPv6-enabled interface that you do not plan to replace, use the autoconfigured IPv6 address, as introduced in IPv6 Address Autoconfiguration. For IPv6-enabled interfaces that must appear anonymous outside the local network, consider using a randomly generated token for the interface ID. For instructions and an example, refer to How to Configure a Temporary Address. For IPv6-enabled interfaces that you plan to swap on a regular basis, create tokens for the interface IDs. For instructions and an example, refer to How to Configure a User-Specified IPv6 Token. &ipv6-config-tunnels;This section describes how to configure the DNS and NIS name services to support IPv6 services.LDAP supports IPv6 without requiring IPv6-specific configuration tasks. For full details for administering DNS, NIS, and LDAP, refer to the System Administration Guide: Naming and Directory Services (DNS, NIS, and LDAP).Log in to the primary or secondary DNS server as Primary Administrator or as superuser.The Primary Administrator role includes the Primary Administrator profile. To create the role and assign the role to a user, see Chapter 2, Working With the Solaris Management Console (Tasks), in System Administration Guide: Basic Administration. Edit the appropriate DNS zone file by adding AAAA records for each IPv6-enabled node:host-name IN AAAA host-address Edit the DNS reverse zone file and add PTR records:host-address IN PTR hostnameFor detailed information on DNS administration, refer to System Administration Guide: Naming and Directory Services (DNS, NIS, and LDAP). This example shows an IPv6 address in the reverse zone file.$ORIGIN ip6.int. 8.2.5.0.2.1.e.f.f.f.9.2.0.0.a.0.6.5.2.9.0.0.0.0.0.0.0.0.2.0.0.0 \ IN PTR vallejo.Eng.apex.COM. You can use the nslookup command to display IPv6 name service information. Under your user account, run the nslookup command.% /usr/sbin/nslookupThe default server name and address appear, followed by the nslookup command's angle bracket prompt. View information about a particular host by typing the following commands at the angle bracket prompt:>set q=any >host-name Type the following command to view only AAAA records:>set q=AAAA hostname Quit the nslookup command by typing exit. This example shows the results of nslookup in an IPv6 network environment.% /usr/sbin/nslookup Default Server: dnsserve.local.com Address: 10.10.50.85 > set q=AAAA > host85 Server: dnsserve.local.com Address: 10.10.50.85 host85.local.com IPv6 address = 2::9256:a00:fe12:528 > exit In this procedure, you use the nslookup command to display PTR records for DNS IPv6. Under your user account, run the nslookup command.% /usr/sbin/nslookupThe default server name and address display, followed by the nslookup command's angle bracket prompt. Type the following at the angle bracket prompt to see the PTR records:>set q=PTR Quit the command by typing exit. The following example shows the PTR record display from the nslookup command.% /usr/sbin/nslookup Default Server: space1999.Eng.apex.COM Address: 192.168.15.78 > set q=PTR > 8.2.5.0.2.1.e.f.f.f.0.2.0.0.a.0.6.5.2.9.0.0.0.0.0.0.0.0.2.0.0.0.ip6.int 8.2.5.0.2.1.e.f.f.f.0.2.0.0.a.0.6.5.2.9.0.0.0.0.0.0.0.0.2.0.0.0.ip6.int name = vallejo.ipv6.Eng.apex.COM ip6.int nameserver = space1999.Eng.apex.COM > exit In this procedure, you use the ypmatch command to display IPv6 information through NIS: Under your user account, type the following to display IPv6 addresses in NIS:% ypmatch hostname hosts .bynameThe information about the specified hostname displays.