Website Monitoring

Internet Engineering Task Force (IETF) introduced IPv6 (Internet Protocol version six) in 1994 to replace version four of the communication protocol. IPv6 uses a 128-bit address space that fixes the most fundamental problem with IPv4 of the quickly exhausted addresses available in the 32-bit based system.

What is an IP address?

An Internet Protocol address is a unique identifier for devices connected to a network. The unique identifier makes it possible for devices to find and communicate with each other. Initially, the main types of devices that required an IP address included network devices, such as computers, servers, routers, and printers. However, with the Internet of Things, the list includes cell phones, televisions, refrigerators, automobiles, light bulbs, or anything else capable of receiving information over a network.

What is an IPv4 address?

Introduced in 1983, IPv4 was the first public version of the Internet Protocol, and it remains the most widely used version of the protocol. An IPv4 address space consists of four octets of numbers from zero to 255 separated by a decimal.

For example:

Google’s IPv4 address: 172.217.168.196

Using the above numeric addressing system allows for 2³² unique addresses or approximately 4.3 billion addresses. The adoption of several different conventions, such as classes and Classless Inter-Domain Routing, expanded the number of available addresses and pushed the need for IPv6 back, and IPv4 remains the most widely used protocol. However, the rapid increase in needed addresses due to the Internet of Things has accelerated the adoption of IPv6.

How is IPv6 different from IPv4?

IPv6 corrects many problems inherent in IPv4. To start, IPv6 is a 128-bit alphanumeric character set rather than numeric.

Over 340 undecillions available addresses

IPv6 has a nearly inexhaustible number of available addresses at over 340 undecillions or 340,282,366,920,938,463,463,374,607,431,768,211,456. The quantity of available addresses, although finite, can scale nearly indefinitely.

Instead of points or decimals separating the groups of hexadecimal digits, IPv6 uses colons. A blank section represents a section that consists of all zeros, as shown in the address below.

For example:

Google’s IPv6 address: 2a00:1450:400e:80c::2004

IPv6 has many advantages over IPv4

Beyond the number of addresses, IPv6 also tackles other problems with IPv4.

• More efficient routing IPv6 uses a hierarchical approach that reduces the size of routing tables by allowing Internet service providers (ISP) to bundle customer prefixes into one prefix. Learn more about IPv6 routing.
• Simplified packet header IPv6 improves the efficiency of the packet processing by eliminating the IP-level checksum. Error checking at the IP level is redundant since other link-layers and transport layers already have a checksum for error detection. Learn more about IPv6 packet headers.
• Network configuration simplified Address auto-configuration is part of IPv6. A host can create an IP address by appending its MAC address in EUI (extended universal identifier) 64-bit format to the 64-bit local link prefix. Learn more about IPv6 network configuration/autoconfiguration.
• Removal of Network Address Translation (NAT) NAT is a fix to the IPv4 address depletion problem where two networks may have incompatible addressing. NAT allows them to connect. Because IPv6 doesn’t need NAT, peer-to-peer networks are easier to build, and other services become more robust. However, some network engineers still do use NAT with IPv6.
• Multicast instead of broadcast IPv6 uses multicast to direct data to multiple recipients at once. IPv4 uses broadcast where a host sends data to every other host on the network whether they want it or not. Broadcast uses excessive bandwidth and requires hosts to process packets not intended for them.
• IPsec built-in IPv6 has Internet protocol security (IPsec) built-in to avoid some of the security issues with IPv4, such as the passing of malware with the packets. A common use of IPsec is for VPN connections. Learn more about IPsec.

Can IPv6 and IPv4 coexist?

IPv6 and IPv4 use completely different infrastructures. An IPv4 network cannot operate over an IPv6 connection and vice versa. However, servers and routers can support both protocols (called dual-stack IP).

What is dual-stack IP?

A network device (server, router, switches) uses Dual-stack IP to process both protocols. The DNS server returns either or both IP addresses depending on the destination. Configure a preferred protocol on the host or the DNS. The IETF published an algorithm called Happy Eyeballs to make dual-stack applications more responsive. Learn more about dual-stack IP.

What is Happy Eyeballs?

IETF’s Happy Eyeballs algorithm helps dual-stack choose the best IP version for the user. Because IPv6 connectivity isn’t available for every user, Happy Eyeballs sends requests using both IP versions. The application uses the first to receive a response. Thus, making human users happy to see a faster response. Learn more about Happy Eyeballs.

IPv6 acceptance

According to Google, IPv6 acceptance increases approximately five percent each year with a 33 percent availability rate as of July 2020. However, IPv6 is still not available in many areas of Africa and the Middle-East. India currently has the highest adoption rate of 58 percent.

Monitoring IPv6 websites and services

Many brands offer their websites or services over both protocols, while a limited few only support IPv6. When supporting both protocols, it is important to monitor both versions even if they share the same host.

DNS issues can affect one or both protocols

A single DNS record can respond with either an IPv4 (A record) or an IPv6 (AAAA record) address or both. DNS issues could arise on either protocol. A site could continue to respond via IPv4 while failing on requests using IPv6. Set up two DNS monitors to resolve the domain name: one to the IPv4 address and one for the IPv6 address. Make sure when setting up the monitor to set the proper IP version. Learn more about DNS monitoring.

Availability monitoring with both IPv6 and IPv4

Beyond the domain name resolution, other server and network issues can arise that may affect a user’s ability to reach a host. Setting up HTTP/HTTPS availability monitoring verifies responses and allows for content checks. Content checks make sure the server responded with the correct content. Learn more about Website Uptime Monitoring and Web Service Uptime Monitoring.

Because availability monitors do not use a browser, you can designate the IP version used for the check making it easy to check both. Plus, with tests every minute, availability monitors capture even small or intermittent outages.

Browser-based Performance monitoring and IPv6

IPv6, in theory, is faster than IPv4 due to the improved routing. However, the infrastructure for IPv6 is not complete. So, as described earlier, a dual-stack system will use the Happy Eyeballs algorithm to choose the connection protocol. Your performance monitoring results will include both IPv4 and IPv6 results due to varying checkpoint locations and the Internet infrastructure in which they exist. Learn more about Web Performance Monitoring.

Key points

• IPv6 replaces IPv4
• IPv6 fixes the address depletion problem with IPv4
• IPv6 uses a 128-bit hexadecimal addressing system.
• IPv6 and IPv4 can exist on the same device with dual-stack enabled.
• Happy Eyeballs is an algorithm that allows the device or browser to choose the faster protocol from a destination.
• IPv6 adoption grows by 5% each year.
• A website or service while available on one protocol may error on the other. Monitor both for availability.
• Browser-based Performance monitors will use Happy Eyeballs to choose the protocol, so your data on a dual-stack system will contain performance results from both.