home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Internet Info 1997 December
/
Internet_Info_CD-ROM_Walnut_Creek_December_1997.iso
/
drafts
/
draft_ietf_i
/
draft-ietf-ipsec-ah-hmac-md5-96-00.txt
< prev
next >
Wrap
Text File
|
1997-03-21
|
16KB
|
448 lines
Network Working Group M. Oehler (NSA)
R. Glenn (NIST)
Internet Draft March 20, 1997
HMAC-MD5-96 IP Authentication with Replay Prevention
<draft-ietf-ipsec-ah-hmac-md5-96-00.txt>
Status of This Memo
Distribution of this memo is unlimited.
This document is an Internet-Draft. Internet Drafts are working
documents of the Internet Engineering Task Force (IETF), its Areas,
and its Working Groups. Note that other groups may also distribute
working documents as Internet Drafts.
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 not appropriate to use Internet Drafts as
reference material, or to cite them other than as a ``working draft''
or ``work in progress.''
To learn the current status of any Internet-Draft, please check the
``1id-abstracts.txt'' listing contained in the internet-drafts Shadow
Directories on:
ftp.is.co.za (Africa)
nic.nordu.net (Europe)
ds.internic.net (US East Coast)
ftp.isi.edu (US West Coast)
munnari.oz.au (Pacific Rim)
Abstract
This document describes a keyed-MD5 transform to be used in
conjunction with the IP Authentication Header [RFC-1826]. The
particular transform is based on [RFC-2104]. A replay prevention
field is also specified.
Oehler, Glenn [Page 1]
INTERNET DRAFT March 20, 1997 Expires September 1997
Contents
1. Introduction...................................................3
1.1 Terminology.................................................3
1.2 Keys........................................................4
1.3 Data Size...................................................4
2. Packet Format..................................................5
2.1 Replay Prevention...........................................5
2.2 Authentication Data Calculation.............................6
3. Security Considerations........................................7
Acknowledgments....................................................7
References.........................................................8
Authors' Addresses.................................................8
Oehler, Glenn [Page 2]
INTERNET DRAFT March 20, 1997 Expires September 1997
1. Introduction
The Authentication Header (AH) [RFC-1826] provides integrity and
authentication for IP datagrams. The transform specified in this
document uses a keyed-MD5 mechanism [RFC-2104]. The mechanism uses
the (key-less) MD5 hash function [RFC-1321] which produces a message
digest. When combined with an AH Key, Authentication Data is
produced. This value is placed in the Authentication Data field of
the AH [RFC-1826]. This value is also the basis for the data
integrity service offered by the AH protocol.
To provide protection against replay attacks, a Replay Prevention
field is specified as a transform option. This field is used to help
prevent attacks in which a message is stored and re-used later,
replacing or repeating the original. The Security Parameters Index
(SPI) [RFC-1825] is used to determine whether this option is included
in the AH.
Familiarity with the following documents is assumed: "Security
Architecture for the Internet Protocol" [RFC-1825], "IP
Authentication Header" [RFC-1826], and "HMAC: Keyed Hashing for
Message Authentication" [RFC-2104].
All implementations that claim conformance or compliance with the IP
Authentication Header specification [RFC-1826] MUST implement this
HMAC-MD5-96 transform.
1.1 Terminology
In this document, the words that are used to define the
significance of each particular requirement are usually capitalized.
These words are:
- MUST
This word or the adjective "REQUIRED" means that the item is an
absolute requirement of the specification.
- SHOULD
This word or the adjective "RECOMMENDED" means that there might
exist valid reasons in particular circumstances to ignore this item,
but the full implications should be understood and the case carefully
weighed before taking a different course.
- MAY
This word or the adjective "OPTIONAL" means that this item is truly
Oehler, Glenn [Page 3]
INTERNET DRAFT March 20, 1997 Expires September 1997
optional. One vendor might choose to include the item because a
particular marketplace requires it or because it enhances the product,
for example; another vendor may omit the same item.
For the purpose of this specification, the terms conformance and
compliance are synonymous.
1.2 Keys
The "AH Key" is used as a shared secret between two communicating
parties. The Key is not a "cryptographic key" as used in a
traditional sense. Instead, the AH key (shared secret) is hashed with
the transmitted data and thus, assures that an intervening party
cannot duplicate the Authentication Data.
Even though an AH key is not a cryptographic key, the rudimentary
concerns of cryptographic keys still apply. Consider that the
algorithm and most of the data used to produce the output is known.
The strength of the transform lies in the singular mapping of the key
(which needs to be strong) and the IP datagram (which is known) to
the Authentication Data. Thus, implementations should, and as
frequently as possible, change the AH key. Keys need to be chosen at
random, or generated using a cryptographically strong pseudo-random
generator seeded with a random seed. [RFC-2104]
All conforming and compliant implementations MUST support a key
length of 128 bits or less. Implementations SHOULD support longer
key lengths as well. It is advised that the key length be chosen to
be the length of the hash algorithm output, which is 128 bits for
MD5. For other key lengths the following concerns MUST be
considered.
A key length of zero is prohibited and implementations MUST prevent
key lengths of zero from being used with this transform, since no
effective authentication could be provided by a zero-length key.
Keys having a length less than 128 bits are strongly discouraged as
it would decrease the security strength of the function. Keys longer
than 128 bits are acceptable, but the extra length may not
significantly increase the function strength. MD5 operates on 64-
byte blocks. Keys longer than 64-bytes are first hashed using MD5.
The resulting hash is then used to calculate the Authentication Data.
1.3 Data Size
HMAC-MD5 produces a 128-bit value. HMAC-MD5-96 uses the first or
left most 96 bits as the Authentication Data. This procedure is
known as truncation. In the case of this transform, truncation is
used to help maintain 64-bit packet header alignment, eliminate
Oehler, Glenn [Page 4]
INTERNET DRAFT March 20, 1997 Expires September 1997
unnecessary overhead, and potentially provide stronger
authentication. [RFC-2104] provides more information on the
advantages and disadvantages of truncation.
2. Packet Format
+---------------+---------------+---------------+---------------+
| Next Header | Length | RESERVED |
+---------------+---------------+---------------+---------------+
| SPI |
+---------------+---------------+---------------+---------------+
| Replay Prevention |
+---------------+---------------+---------------+---------------+
| |
+ +
| Authentication Data |
+ +
| |
+---------------+---------------+---------------+---------------+
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8
The Next Header, RESERVED, and SPI fields are specified in [RFC-
1826]. The Length field is the length of the Replay Prevention field
and the Authentication Data in 32-bit words. The Length field will
always be set to 4 (128 bits) for HMAC-MD5-96.
2.1 Replay Prevention
The Replay Prevention field is a 32-bit value used to guarantee that
each packet exchanged between two parties is different. Each IPsec
Security Association specifies whether Replay Prevention is used for
that Security Association. The Replay Prevention field is always
included in the calculation of the Authentication Data. If Replay
Prevention is NOT in use, the 32-bit value is set to 0, included in
the calculation of the Authentication Data, and ignored upon receipt
with regard to checking for replay. This field is used to help
prevent attacks in which a message is stored and re-used later,
replacing or repeating the original.
Replay Prevention SHOULD be implemented. If Replay Prevention is not
implemented, the 32-bit field remains are part of the AH and is
treated as if Replay Prevention is NOT in use (i.e. the 32-bit value
is set to 0, included in the calculation of the Authentication Data,
and ignored upon receipt with regard to checking for replay.
The 32-bit field is an up counter starting at a value of 1.
The secret shared key MUST NOT be used for a period of time that
Oehler, Glenn [Page 5]
INTERNET DRAFT March 20, 1997 Expires September 1997
allows the counter to wrap, that is, to transmit more than 2^32
packets using a single key.
Upon receipt, the replay value is assured to be increasing. An
implementation MAY accept out of order packets. If an "out of order
window" is supported, an implementation MUST guarantee that any and
all packets accepted out of order have not arrived before. That is,
an implementation MUST accept any packet, at most, once. The size of
the window is a negotiated value specified by the IPsec Security
Association.
[ESP-DES-MD5] provides more information on negotiated windows sizes,
example code that implements a 32 packet replay window, and a test
routine to show how it could be implemented.
When the destination address is a multicast address and more than one
sender is sharing the same IPsec Security Association to that
multicast destination address, then Replay Prevention SHOULD NOT be
enabled. When Replay Prevention is desired for a multicast session
having multiple senders to the same multicast destination address,
each sender SHOULD have its own IPsec Security Association.
2.2 Authentication Data Calculation
The Authentication Data is the output of the MD5 authentication
algorithm. This value is calculated over the entire IP datagram.
Fields within the datagram that are variant during transit and the
Authentication Data field itself, must contain all zeros prior to the
computation [RFC-1826]. The Replay Prevention field, used or not, is
always included in the calculation.
The definition and reference implementation of MD5 appears in [RFC-
1321]. Let 'text' denote the data to which HMAC-MD5-96 is to be
applied and K be the message authentication secret key shared by the
parties. If K is longer than 64-bytes it MUST first be hashed using
MD5. In this case, K is the resulting hash.
We define two fixed and different strings ipad and opad as follows
(the 'i' and 'o' are mnemonics for inner and outer):
ipad = the byte 0x36 repeated 64 times
opad = the byte 0x5C repeated 64 times.
To compute HMAC-MD5 over the data `text' we perform
MD5(K XOR opad, MD5(K XOR ipad, text))
The result of which is truncated to 96 bits (retaining the left most
Oehler, Glenn [Page 6]
INTERNET DRAFT March 20, 1997 Expires September 1997
bits) to produce HMAC-MD5-96.
The calculation of the Authentication Data consists of the following
steps:
(1) append zeros to the end of K to create a 64 byte string (e.g., if
K is of length 16 bytes it will be appended with 48 zero bytes 0x00)
(2) XOR (bitwise exclusive-OR) the 64 byte string computed in
step (1) with ipad
(3) append the data stream 'text' to the 64 byte string resulting
from step (2)
(4) apply MD5 to the stream generated in step (3)
(5) XOR (bitwise exclusive-OR) the 64 byte string computed in
step (1) with opad
(6) append the MD5 result from step (4) to the 64 byte string
resulting from step (5)
(7) apply MD5 to the stream generated in step (6)
(8) use the left most 96 bits of the result obtained in (7) as the final
result
A similar computation is described in more detail, along with example
code and performance improvements, in [RFC-2104]. Implementers should
consult [RFC-2104] for more information on this technique for keying
a cryptographic hash function.
3. Security Considerations
The security provided by this transform is based on the strength of
MD5, the correctness of the algorithm's implementation, the security
of the key management mechanism and its implementation, the strength
of the associated secret key, and upon the correctness of the
implementations in all of the participating systems. [RFC-2104]
contains a detailed discussion on the strengths and weaknesses of
HMAC algorithms. [HMAC-TESTS] contains test vectors and example code
to assist in verifying the correctness of HMAC-MD5 code.
Acknowledgments
This document is largely based on text written by Hugo Krawczyk. The
format used was derived from work by William Simpson and Perry
Metzger. The text on replay prevention is derived from work by Jim
Hughes.
Oehler, Glenn [Page 7]
INTERNET DRAFT March 20, 1997 Expires September 1997
References
[RFC-1825] R. Atkinson, "Security Architecture for the Internet
Protocol", RFC-1852, Naval Research Laboratory, July 1995.
[RFC-1826] R. Atkinson, "IP Authentication Header",
RFC-1826, August 1995.
[RFC-1828] P. Metzger, W. A. Simpson, "IP Authentication using Keyed MD5",
RFC-1828, August 1995.
[RFC-1321] R. Rivest, "The MD5 Message-Digest Algorithm",
RFC-1321, April 1992.
[RFC-2104] H. Krawczyk, M. Bellare, R. Canetti, "HMAC: Keyed Hashing
for Message Authentication", RFC-2104, February, 1997.
[ESP-DES-MD5] J. Hughes, "Combined DES-CBC, MD5, and Replay Prevention
Security Transform", Internet Draft, September 1996.
[HMAC-TESTS] P. Cheng, R. Glenn, "Test Cases for HMAC-MD5 and HMAC-SHA-1",
Internet Draft, March 1997.
Authors' Addresses
Michael J. Oehler
National Security Agency
Atn: R23, INFOSEC Research and Development
9800 Savage Road
Fort Meade, MD 20755
mjo@tycho.ncsc.mil
Robert Glenn
NIST
Building 820, Room 455
Gaithersburg, MD 20899
rob.glenn@nist.gov
Oehler, Glenn [Page 8]