Browse Prior Art Database

Method and System for Enabling Routers to Truncate TCP Segments Size to Adjust an Outgoing Maximum Transmit Unit (MTU) Disclosure Number: IPCOM000243053D
Publication Date: 2015-Sep-10
Document File: 2 page(s) / 44K

Publishing Venue

The Prior Art Database


A method and system is disclosed for enabling routers to truncate TCP segments size to adjust an outgoing Maximum Transmit Unit (MTU).

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 2

Method and System for Enabling Routers to Truncate TCP Segments Size to Adjust an Outgoing Maximum Transmit Unit (




Issues arising from IP Fragmentation, Maximum Transmit Unit (MTU), Path Maximum Transmit Unit Discovery (PMTUD), and Dynamic path changes can result in TCP connection failures. The growing adoption of Network Function Virtualization (NFV) and deployment of tunnels are bringing those problems to the forefront .

More specifically, links enforce a MTU whose value depend on the link types (e.g., 1500 bytes for Ethernet), and IP fragmentation allows routers to fragment IP datagrams to fit the MTU of the links from a sender to a receiver . However, IP fragmentation is often disabled (e.g., for performance reasons) and dropped (e.g., for security reasons). As a result, the IETF standardized the PMTUD to avoid fragmentation along the paths between endpoints. The PMTUD dynamically determines the lowest MTU, and works as follows:
1. Senders typically send an IP datagram with the Don't Fragment (DF) bit set, since fragmentation is bad for performance .

2. When that IP datagram arrives at a router whose outgoing link has a smaller MTU than the size of the datagram, because the DF bit is set, the router cannot fragment the datagram. Instead, the router drops the datagram and sends an ICMP message (type
3, code 4) to the sender.

3. Upon receiving the ICMP message, the sender should lower the size of the IP datagram, and retransmit the packet.

The PMTUD was as such designed to solve the MTU issues. However, today, ICMP messages are commonly dropped by routers and firewalls. Consequently, the IP datagrams cannot reach the receivers, and the sender is not informed of the reasons . In other words, the TCP connection fails.

The above problem is becoming more important with NFV . NFV frequently requires tunneling, and tunnels add encapsulation overheads, ultimately reducing the MTU, and increasing the risks for the above issues. Also, tunneling technologies (GRE, IP in IP,

VXLAN) have different overhead making the pr...