Datagram Convergence Layers for the DTN Bundle and LTP Protocols
draft-irtf-dtnrg-dgram-clayer-00

Abstract

This document specifies the preferred method for transporting DTN protocol data over the Internet using datagrams. The specification covers convergence layers for the Bundle Protocol as well as the transportation of LTP segments. UDP and DCCP are the candidate datagram protocols discussed. UDP can only be used on a local network, or in cases where the DTN node implements explicit congestion control. DCCP does address the congestion control problem; however, the availability of implementations is limited.

Status of This Memo

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at http:/⁠/⁠datatracker.ietf.org/⁠drafts/⁠current/⁠.

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."

Copyright Notice

Copyright (c) 2012 IETF Trust and the persons identified as the document authors. All rights reserved.

This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http:/⁠/⁠trustee.ietf.org/⁠license-⁠info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.

Table of Contents

• 1. Introduction
• 1.1. Requirements Language
• 2. General Recommendation
• 3. Recommendations for Implementers
• 3.1. How and Where to Deal with Fragmentation
• 3.1.1. DCCP
• 3.1.2. UDP
• 3.2. Bundle Protocol over a Datagram Convergence Layer
• 3.2.1. DCCP
• 3.2.2. UDP
• 3.3. LTP over a Datagram Convergence Layer
• 3.3.1. DCCP
• 3.3.2. UDP
• 3.4. Keep Alive Option
• 3.5. Checksums
• 3.5.1. DCCP
• 3.5.2. UDP
• 3.6. DCCP Availability
• 3.7. DCCP Congestion Control Modules
• 4. Acknowledgements
• 5. IANA Considerations
• 6. Security Considerations
• 7. References
• 7.1. Normative References
• 7.2. Informative References
• Authors' Addresses

1. Introduction

Delay/Disruption Tolerant Network (DTN) communication protocols include the Bundle Protocol described in RFC 5050 [RFC5050], which provides reliable transmission of application data blocks (bundles) through optional intermediate custody transfer, and the Licklider Transmission Protocol (LTP), RFCs 5325 [RFC5325], 5326 [RFC5326], and 5327 [RFC5327] which can be used to transmit bundles reliably and efficiently over a point to point link. It is often desirable to test these protocols over Internet Protocol links. draft-irtf-dtnrg-tcp-clayer [I-D.irtf-dtnrg-tcp-clayer] defines a method for transporting bundles over TCP. This draft specifies the preferred method for transmitting either bundles or LTP blocks across the Internet using datagrams in place of TCP.

1.1. Requirements Language

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119].

2. General Recommendation

In order to utilize DTN protocols across the Internet, whether for testing purposes or as part of a larger network path, it is necessary to encapsulate them into a standard Internet protocol so that they travel easily across the Internet. This is particularly true for LTP, which provides no endpoint addressing. This encapsulation choice needs to be made carefully in order to avoid redundancy, since DTN protocols may provide their own reliability mechanisms.

TCP, a logical choice, guarantees reliability and provides congestion control. Congestion control is vital to the continued functioning of the Internet, particularly for situations where data will be sent at arbitrarily fast data rates. Because the Bundle Protocol offers neither congestion control nor reliability, TCP is the RECOMMENDED choice for its encapsulation. draft-irtf-dtnrg-tcp-clayer [I-D.irtf-dtnrg-tcp-clayer] defines the method for transporting bundles over TCP.

LTP, on the other hand, offers it's own form of reliability. Particularly for testing purposes, it makes no sense to run LTP over a protocol, like TCP, that offers reliability already. In addition, running LTP over TCP would reduce the flexibility available to users, since LTP offers more control over what data is delivered reliably and what data is delivered best effort, a feature that TCP lacks. As such, it would be better to run LTP over an unreliable protocol.

One solution would be to use UDP. UDP provides no reliability, allowing LTP to manage that itself. However, UDP does not provide congestion control. Because LTP is designed to run over fixed rate radio links it does provides rate control, but not congestion control. Lack of congestion control in network connections is a major problem that can cause artificially high loss rates and/or serious fairness issues. Previous standards documents are unanimous in recommending congestion control for protocols to be used on the Internet, see RFCs 2914 [RFC2914], 5405 [RFC5405], and 2309 [RFC2309], among others. RFC 5405 [RFC5405], in particular, calls congestion control "vital" for "applications that can operate at higher, potentially unbounded data rates". Therefore, any application using UDP to transport LTP segements or Bundles MUST implement congestion control consistent with RFC 5405.

Alternatively, the Datagram Congestion Control Protocol (DCCP) [RFC4340] was designed specifically to provide congestion control without reliability for those applications that traverse the Internet but do not desire to retransmit lost data. As such, it is RECOMMENDED that, if possible, DCCP be used to transport LTP segments across the Internet.

3. Recommendations for Implementers

3.1. How and Where to Deal with Fragmentation

The Bundle Protocol allows bundles with sizes limited only by node resource constraints. In IPv4, the maximum size of a UDP datagram is nearly 64KB. In IPv6, when using jumbograms [RFC2675], UDP datagrams can be up to 4GB in size [RFC2147]. It is well understood that sending large IP datagrams that must be fragmented by the network has enormous efficiency penalties [Kent88]. The primary efficiency penalty is increased loss probability. When a large datagram is broken into a number of fragments, the original datagram can only be recreated if all the fragments arrive at the ultimate destination for reassembly. When transmitted over a network with a packet loss probability of 2%, for example, a single, unfragmented datagram will arrive with probability 98%; a large datagram fragmented into 10 fragments will have all of its fragments arrive with probability 98%**10, giving a datagram arrival probability of only 81.7%. The higher-level protocol using UDP for delivery can retransmit lost UDP datagrams, but cannot retransmit lost IP datagram fragments. Therefore, retransmitting large, lost datagrams because of a small number of missing fragments can require many more packets than retransmitting a number of smaller, unfragmented datagrams because only the missing pieces need to be retransmitted. The other efficiency penalty paid by fragmentation that would be significant for DTN is the resources (time, complexity, and memory) required for IP reassembly. If the Bundle Protocol is being encapsulated in DCCP or UDP, the bundle protocol specification provides a bundle fragmentation concept [RFC5050] that allows a large bundle to be divided into bundle fragments, each of which SHOULD be created of sufficiently small size that it can then be encapsulated into a datagram that will not need to be fragmented.

3.1.1. DCCP

Because DCCP implementations are not required to support IP fragmentation and are not allowed to enable it by default, a DCCP CL MUST NOT accept data segments that cannot be sent as one MTU sized datagram.

3.1.2. UDP

When an LTP CL is using UDP for datagram delivery, it SHOULD NOT create segments that will result in UDP datagrams that will need to be fragmented, as discussed above.

Without information from elsewhere in the networking stack about path MTU, the protocol can assume a minimum path MTU that would allow 512 bytes of UDP data [RFC0791] over IPv4 or (1280-(UDP and IP header sizes)) bytes [RFC1883] over IPv6.

3.2. Bundle Protocol over a Datagram Convergence Layer

In general, the use of the bundle protocol over a datagram CL is discouraged. Bundles can be of (almost) arbitrary length, and the bundle protocol does not include an effective retransmission mechanism. Whenever possible the bundle protocol SHOULD be operated over the TCP Convergence Layer or over LTP.

If a datagram CL is used for transmission of bundles, every packet MUST contain exactly one bundle or four zero octets as a keep-alive. The CL SHOULD use available operating system services to obtain the largest supported packet size, and MAY use the default packet size limit if path-specific information is not available. For bundles that are too large for the supported packet size, the bundle protocol fragmentation process SHOULD be used to transmit the large bundle.

3.2.1. DCCP

The DCCP CL for bundle protocol use SHOULD use the IANA assigned port 4556/DCCP and service code 1685351985; the use of other port numbers and service codes is implementation specific.

3.2.2. UDP

The UDP CL for bundle protocol use SHOULD use the IANA assigned port 4556/UDP; the use of other port numbers is implementation specific.

3.3. LTP over a Datagram Convergence Layer

LTP is designed as a point to point protocol within DTN, and it provides intrinsic acknowledgement and retransmission facilities. Transmission of LTP over a datagram CL is therefore the most appropriate choice. When a datagram CL is used to transmit LTP data, every packet MUST contain exactly one LTP segment or four zero octets as a keep-alive. The CL SHOULD use available operating system services to obtain the largest supported packet size, and MAY use the default packet size limit if path-specific information is not available. LTP MUST perform segmentation in such a way as to insure that every LTP segments fits into a single packet.

3.3.1. DCCP

The DCCP CL for LTP SHOULD use the IANA assigned port 1113/DCCP and service code 7107696; the use of other port numbers and service codes is implementation specific.

3.3.2. UDP

The UDP CL for LTP SHOULD use the IANA assigned port 1113/UDP; the use of other port numbers is implementation specific.

3.4. Keep Alive Option

It may be desirable for a UDP or DCCP CL to send "keep-alive" packets during extended idle periods. This may be needed to refresh a contact table entry at the destination, or to maintain an address mapping in a NAT or a dynamic access rule in a firewall. Therefore, the CL MAY send a packet containing exactly 4 octets of zero bits. The CL receiving such a packet MUST discard this packet; the receiving CL may then perform local maintenance of its state tables, these maintenance functions are not covered in this draft. Note that "real" CL packets will always contain more than 4 octets of information (either the bundle or the LTP header); keep-alive packets will therefore never be mistaken for actual data packets.

3.5. Checksums

Both the core bundle protocol specification and core LTP specification assume that they are transmitting over an erasure channel, i.e. a channel that either delivers packets correctly or not at all.

3.5.1. DCCP

A DCCP CL transmitter MUST, therefore, ensure that the entire packet is checksummed by setting the Checksum Coverage to 0. Likewise, the DCCP CL receiver MUST ignore all packets with partial checksum coverage.

3.5.2. UDP

A UDP CL transmitter therefore MUST NOT disable UDP checksums, and the UDP CL receiver MUST NOT disable checking of received UDP checksums.

Even when UDP checksums are enabled a small probability of UDP packet corruption remains. In some environments it may be acceptable for LTP or the bundle protocol to occasionally receive corrupted input. In general, however, a UDP CL implementation SHOULD use optional security extensions available in the bundle protocol or LTP to protect against message corruption.

3.6. DCCP Availability

As of this writing, the most mature DCCP implementation seems to be the one in the Linux Kernel. DCCP has, unfortunately, been slow in making it's way into most of the major platforms. As a result, if no DCCP implementation is available for a target platform, tunneling LTP over UDP is acceptable. In such a case, the UDP CL either MUST NOT be used outside an isolated network for the transmission of any non-trivial amounts of data, or it MUST implement congestion control procedures as outlined in RFC 5405 [RFC5405].

3.7. DCCP Congestion Control Modules

DCCP supports pluggable congestion control modules in order to optimize it's behavior to particular environments. The two most common congestion control modules (CCIDs) are TCP-like Congestion Control (CCID2) [RFC4341] and TCP-Friendly Rate Control (CCID3) [RFC4342]. TCP-like Congestion Control is designed to emulate TCP's congestion control as much as possible. It is recommended for applications that want to send data as quickly as possible, while TCP-Friendly Rate Control is aimed at applications that want to avoid sudden changes in sending rate. DTN use cases seem to fit more into the first case so DCCP CL's SHOULD use TCP-like Congestion Control (CCID2) by default.

4. Acknowledgements

5. IANA Considerations

Port number assignments 1113/UDP and 4556/UDP have been registered with IANA. Port numbers 1113/DCCP for the transport of LTP, and 4556/DCCP for the transport of bundles have been requested. DCCP Service Codes 7107696 for tunneling LTP and 1685351985 for tunneling Bundle Protocol have been requested.

6. Security Considerations

This memo describes the use of datagrams to transport DTN application data. Hosts may be in the position of having to accept and process packets from unknown sources; the DTN Endpoint ID can be discovered only after the bundle has been retrieved from the DCCP or UDP packet. Hosts SHOULD use authentication methods available in the DTN specifications to prevent malicious hosts from inserting unknown data into the application.

Hosts need to listen for and process DCCP or UDP data on the known LTP or bundle protocol ports. A denial of service scenario exists where a malicious host sends datagrams at a high rate, forcing the receiving hosts to use its resources to process and attempt to authenticate this data. Whenever possible, hosts SHOULD use IP address filtering to limit the origin of packets to known hosts.

7. References

7.1. Normative References

7.2. Informative References

Authors' Addresses