A Brief History of MPLS Usage in IPv6

Recent researches have stated the fast deployment of IPv6. It has been demonstrated that IPv6 grows much faster, being so more and more adopted by both Internet service providers but also by servers and end-hosts. In parallel, researches have been conducted to discover and assess the usage of MPLS tunnels. Indeed, recent developments in the ICMP protocol make certain categories of MPLS tunnels transparent to traceroute probing. However, these studies focus only on IPv4, where MPLS is strongly deployed. In this paper, we provide a first look at how MPLS is used under IPv6 networks using traceroute data collected by CAIDA. At first glance, we observe that the MPLS deployment and usage seem to greatly differ between IPv4 and IPv6, in particular in the way MPLS label stacks are used. While label stacks with at least two labels are marginal in IPv4 (and mostly correspond to a VPN usage), they are prevalent in IPv6. After a deeper analysis of the label stack typical content in IPv6, we show that such tunnels result from the use of 6PE. This is not really surprising since this mechanism was specifically designed to forward IPv6 traffic using MPLS tunnels through networks that are not fully IPv6 compliant. However, we show that it does not result from non dual-stack routers but rather from the absence of native IPv6 MPLS signaling protocols. Finally, we investigate a large Tier-1 network, Cogent, that stands out with an original set-up.