Exploring Usable Path MTU in the Internet

To optimise their transmission, Internet endpoints need to know the largest size of packet they can send across a specific Internet path, the Path Maximum Transmission Unit (PMTU). This paper explores the PMTU size experienced across the Internet core, wired and mobile edge networks. Our results show that MSS Clamping has been widely deployed in edge networks, and some webservers artificially reduce their advertised MSS, both of which we expect help avoid PMTUD failure for TCP. The maximum packet size used by a TCP connection is also constrained by the acMSS. MSS Clamping was observed in over 20% of edge networks tested. We find a significant proportion of webservers that advertise a low MSS can still be reached with a 1500 byte packet. We also find more than half of IPv6 webservers do not attempt PMTUD and clamp the MSS to 1280 bytes. Furthermore, we see evidence of black-hole detection mechanisms implemented by over a quarter of IPv6 webservers and almost 15% of IPv4 webservers. We also consider the implications for UDP - which necessarily can not utilise MSS Clamping. The paper provides useful input to the design of a robust PMTUD method that can be appropriate for the growing volume of UDP-based applications, by determining ICMP quotations can be used as to verify sender authenticity.

[1]  Sally Floyd,et al.  Measuring interactions between transport protocols and middleboxes , 2004, IMC '04.

[2]  Bill Owens,et al.  Inferring and debugging path MTU discovery failures , 2005, IMC '05.

[3]  Carlos Pignataro,et al.  Extended ICMP to Support Multi-Part Messages , 2007, RFC.

[4]  Stephen E. Deering,et al.  Path MTU discovery , 1990, RFC.

[5]  Christian E. Hopps,et al.  Analysis of an Equal-Cost Multi-Path Algorithm , 2000, RFC.

[6]  Jon Postel,et al.  The TCP Maximum Segment Size and Related Topics , 1983, RFC.

[7]  Boris Nechaev,et al.  Netalyzr: illuminating the edge network , 2010, IMC '10.

[8]  Robert T. Braden,et al.  Requirements for Internet Hosts - Communication Layers , 1989, RFC.

[9]  Stephen T. Kent,et al.  IP Encapsulating Security Payload (ESP) , 1995, RFC.

[10]  Gorry Fairhurst,et al.  Exploring DSCP modification pathologies in mobile edge networks , 2017, 2017 Network Traffic Measurement and Analysis Conference (TMA).

[11]  David Malone,et al.  Analysis of ICMP Quotations , 2007, PAM.

[12]  Gorry Fairhurst,et al.  PATHspider: A tool for active measurement of path transparency , 2016, ANRW '16.

[13]  Richard Nelson,et al.  An Analysis of TCP Maximum Segment Sizes , 2010 .

[14]  Matthew J. Luckie,et al.  Scamper: a scalable and extensible packet prober for active measurement of the internet , 2010, IMC '10.

[15]  Vivek Kashyap IP over InfiniBand: Connected Mode , 2006, RFC.

[16]  David A. Borman,et al.  TCP Options and Maximum Segment Size (MSS) , 2012, RFC.

[17]  Ryan Hamilton,et al.  QUIC: A UDP-Based Secure and Reliable Transport for HTTP/2 , 2016 .

[18]  Stephen E. Deering,et al.  Path MTU Discovery for IP version 6 , 1996, RFC.

[19]  Stephen E. Deering,et al.  Internet Protocol, Version 6 (IPv6) Specification , 1995, RFC.

[20]  Kevin Lahey,et al.  TCP Problems with Path MTU Discovery , 2000, RFC.

[21]  Michael Tüxen,et al.  Packetization Layer Path MTU Discovery for Datagram Transports , 2020, RFC.

[22]  Randall Gellens,et al.  The Codecs Parameter for "Bucket" Media Types , 2005, RFC.

[23]  Stephen E. Deering,et al.  Path MTU Discovery for IP version 6 , 2017, RFC.

[24]  Jeffrey C. Mogul,et al.  Fragmentation considered harmful , 1987, CCRV.

[25]  Richard van den Berg,et al.  Over-Zealous Security Administrators Are Breaking the Internet , 2002, LISA.

[26]  Fan Yang,et al.  The QUIC Transport Protocol: Design and Internet-Scale Deployment , 2017, SIGCOMM.

[27]  Carlos Pignataro,et al.  ICMP Extensions for Multiprotocol Label Switching , 2007, RFC.

[28]  Ali Safari Khatouni,et al.  5 MONROE : Measuring Mobile Broadband Networks in Europe , 2017 .

[29]  Matthew J. Luckie,et al.  Measuring path MTU discovery behaviour , 2010, IMC '10.