- 8-1. Comparison of IPv6 and IPv4
- 8-2. IPv6 Addressing
- 8-3. IPv6 Network Settings
- 8-4. Configuring a Router for IPv6
- 8-5. IPv6 Routing
- 8-6. Troubleshooting IPv6 Connection
- Summary
- Questions and Problems
8-6. Troubleshooting IPv6 Connection
One big question that needs to be answered before troubleshooting IPv6 connectivity is: Does the network environment support IPv6? If the answer is yes, the same network troubleshooting techniques and approaches still apply on IPv6 as on IPv4. Remember what has changed is only the network layer on the OSI model. Other layers are still intact and stay the same. You will still need to troubleshoot the physical connections to make sure the physical layer is working properly. The data link layer still needs to be inspected to see if the packets are being forwarded, MAC addresses are still being seen, and hosts are still in the correct VLANs.
The commands such as ipconfig in Windows and ifconfig for Linux or Mac OS X can be used to view the TCP/IP configuration information of a host. This is always a good start in network troubleshooting. First, you have to see what is configured and whether it is configured correctly before you can move on to the next step. The examples of these commands are shown throughout this chapter.
Many basic network tools that are available in IPv4 are available in IPv6 as well. Ping is one of the most commonly used tools to test the connectivity between two hosts. Ping is implemented using ICMP echo and Echo reply for a very simple hello network test. In IPv6, the ICMP version 6 is being used instead; therefore the tool has changed slightly to accommodate the change in the ICMP protocol fields. The command ping6 can be used to explicitly specify the IPv6 address, even though most operating systems have modified the ping command to understand both the IPv4 and IPv6 addresses. An issue of using the ping command in IPv6 is the lengthy address and the time required for entering the destination address. For example, the following is an example. The first part shows the IPv6 address that is assigned to the router’s R1 interface.
R2(config-if)# ipv6 address 2001:C16C:0000:0001:0000:0000:0000:0001/64
The IPv6 address can be simplified using double colon notation, as shown:
R2# ping ipv6 2001:C16C:0:1::11
The IPv6 address is still complicated even with the reduced address length. A solution to this is to assign a hostname to the specified IPv6 address. In this case, the hostname R1-WAN will be assigned to the specified IPv6 address using the command ipv6 host R1-WAN 2001:C16C:0:1::1/64, as shown:
R2(config)# ipv6 host R1-WAN 2001:C16C:0:1::11
Now, the ping command, using the newly assigned hostnames for R1 and R2, can be used. An example is provided:
R2(config)# ipv6 host R1-WAN 2001:C16C:0:1::1 R2# ping R1-WAN Type escape sequence to abort. Sending 5, 100-byte ICMP Echos to 2001:C16C:0:1::1, timeout is 2 seconds: !!!!! Success rate is 100 percent (5/5), round-trip min/avg/max = 0/0/4 ms R2#
Another useful network tool is traceroute or tracert in Windows world. This tool enables the user to see the routing information between the two hosts. The IPv6 version of this tool is traceroute6 or tracert6 in Windows. Like ping, the IPv6 version of traceroute has to understand the ICMP version 6 messages as well. The structure for the traceroute6 command is shown. The Host name and web addresses assume the DNS entries have been made:
traceroute6 <destination address, Host name, or web address>
For example, the following could be entered to run a traceroute:
traceroute6 2001:C16C:0:2::2 traceroute6 www.6bone.net traceroute6 R1-WAN
When will the Internet switch to IPv6? The answer is not clear, but the networking community recognizes that something must be done to address the limited availability of current IP address space. Manufacturers have already incorporated IPv6 capabilities in their routers and operating systems. What about IPv4? The bottom line is that the switch to IPv6 will not come without providing some way for IPv4 networks to still function. Additionally, techniques such as NAT have made it possible for intranets to use the private address space and still be able to connect to the Internet. This has significantly reduced the number of IPv4 addresses required for each network and have delayed the need to immediately switch to IPv6.