draft-ietf-httpbis-encryption-encoding-06.txt   draft-ietf-httpbis-encryption-encoding-07.txt 
HTTP Working Group M. Thomson HTTP Working Group M. Thomson
Internet-Draft Mozilla Internet-Draft Mozilla
Intended status: Standards Track December 22, 2016 Intended status: Standards Track February 13, 2017
Expires: June 25, 2017 Expires: August 17, 2017
Encrypted Content-Encoding for HTTP Encrypted Content-Encoding for HTTP
draft-ietf-httpbis-encryption-encoding-06 draft-ietf-httpbis-encryption-encoding-07
Abstract Abstract
This memo introduces a content coding for HTTP that allows message This memo introduces a content coding for HTTP that allows message
payloads to be encrypted. payloads to be encrypted.
Note to Readers Note to Readers
Discussion of this draft takes place on the HTTP working group Discussion of this draft takes place on the HTTP working group
mailing list (ietf-http-wg@w3.org), which is archived at mailing list (ietf-http-wg@w3.org), which is archived at
skipping to change at page 1, line 41 skipping to change at page 1, line 41
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet- working documents as Internet-Drafts. The list of current Internet-
Drafts is at http://datatracker.ietf.org/drafts/current/. Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
This Internet-Draft will expire on June 25, 2017. This Internet-Draft will expire on August 17, 2017.
Copyright Notice Copyright Notice
Copyright (c) 2016 IETF Trust and the persons identified as the Copyright (c) 2017 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of (http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with respect carefully, as they describe your rights and restrictions with respect
to this document. Code Components extracted from this document must to this document. Code Components extracted from this document must
include Simplified BSD License text as described in Section 4.e of include Simplified BSD License text as described in Section 4.e of
the Trust Legal Provisions and are provided without warranty as the Trust Legal Provisions and are provided without warranty as
skipping to change at page 2, line 24 skipping to change at page 2, line 24
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1. Notational Conventions . . . . . . . . . . . . . . . . . 3 1.1. Notational Conventions . . . . . . . . . . . . . . . . . 3
2. The "aes128gcm" HTTP Content Coding . . . . . . . . . . . . . 3 2. The "aes128gcm" HTTP Content Coding . . . . . . . . . . . . . 3
2.1. Encryption Content Coding Header . . . . . . . . . . . . 5 2.1. Encryption Content Coding Header . . . . . . . . . . . . 5
2.2. Content Encryption Key Derivation . . . . . . . . . . . . 6 2.2. Content Encryption Key Derivation . . . . . . . . . . . . 6
2.3. Nonce Derivation . . . . . . . . . . . . . . . . . . . . 6 2.3. Nonce Derivation . . . . . . . . . . . . . . . . . . . . 6
3. Examples . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3. Examples . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.1. Encryption of a Response . . . . . . . . . . . . . . . . 7 3.1. Encryption of a Response . . . . . . . . . . . . . . . . 7
3.2. Encryption with Multiple Records . . . . . . . . . . . . 8 3.2. Encryption with Multiple Records . . . . . . . . . . . . 8
4. Security Considerations . . . . . . . . . . . . . . . . . . . 8 4. Security Considerations . . . . . . . . . . . . . . . . . . . 8
4.1. Key and Nonce Reuse . . . . . . . . . . . . . . . . . . . 9 4.1. Message Truncation . . . . . . . . . . . . . . . . . . . 9
4.2. Data Encryption Limits . . . . . . . . . . . . . . . . . 9 4.2. Key and Nonce Reuse . . . . . . . . . . . . . . . . . . . 9
4.3. Content Integrity . . . . . . . . . . . . . . . . . . . . 10 4.3. Data Encryption Limits . . . . . . . . . . . . . . . . . 9
4.4. Leaking Information in Headers . . . . . . . . . . . . . 10 4.4. Content Integrity . . . . . . . . . . . . . . . . . . . . 10
4.5. Poisoning Storage . . . . . . . . . . . . . . . . . . . . 11 4.5. Leaking Information in Header Fields . . . . . . . . . . 10
4.6. Sizing and Timing Attacks . . . . . . . . . . . . . . . . 11 4.6. Poisoning Storage . . . . . . . . . . . . . . . . . . . . 11
4.7. Sizing and Timing Attacks . . . . . . . . . . . . . . . . 11
5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 11 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 11
5.1. The "aes128gcm" HTTP Content Coding . . . . . . . . . . . 11 5.1. The "aes128gcm" HTTP Content Coding . . . . . . . . . . . 11
6. References . . . . . . . . . . . . . . . . . . . . . . . . . 12 6. References . . . . . . . . . . . . . . . . . . . . . . . . . 12
6.1. Normative References . . . . . . . . . . . . . . . . . . 12 6.1. Normative References . . . . . . . . . . . . . . . . . . 12
6.2. Informative References . . . . . . . . . . . . . . . . . 12 6.2. Informative References . . . . . . . . . . . . . . . . . 13
Appendix A. JWE Mapping . . . . . . . . . . . . . . . . . . . . 13 Appendix A. JWE Mapping . . . . . . . . . . . . . . . . . . . . 14
Appendix B. Acknowledgements . . . . . . . . . . . . . . . . . . 14 Appendix B. Acknowledgements . . . . . . . . . . . . . . . . . . 14
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 14 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 15
1. Introduction 1. Introduction
It is sometimes desirable to encrypt the contents of a HTTP message It is sometimes desirable to encrypt the contents of a HTTP message
(request or response) so that when the payload is stored (e.g., with (request or response) so that when the payload is stored (e.g., with
a HTTP PUT), only someone with the appropriate key can read it. a HTTP PUT), only someone with the appropriate key can read it.
For example, it might be necessary to store a file on a server For example, it might be necessary to store a file on a server
without exposing its contents to that server. Furthermore, that same without exposing its contents to that server. Furthermore, that same
file could be replicated to other servers (to make it more resistant file could be replicated to other servers (to make it more resistant
skipping to change at page 3, line 15 skipping to change at page 3, line 15
These uses are not met by the use of TLS [RFC5246], since it only These uses are not met by the use of TLS [RFC5246], since it only
encrypts the channel between the client and server. encrypts the channel between the client and server.
This document specifies a content coding (Section 3.1.2 of [RFC7231]) This document specifies a content coding (Section 3.1.2 of [RFC7231])
for HTTP to serve these and other use cases. for HTTP to serve these and other use cases.
This content coding is not a direct adaptation of message-based This content coding is not a direct adaptation of message-based
encryption formats - such as those that are described by [RFC4880], encryption formats - such as those that are described by [RFC4880],
[RFC5652], [RFC7516], and [XMLENC] - which are not suited to stream [RFC5652], [RFC7516], and [XMLENC] - which are not suited to stream
processing, which is necessary for HTTP. The format described here processing, which is necessary for HTTP. The format described here
cleaves more closely to the lower level constructs described in follows more closely to the lower level constructs described in
[RFC5116]. [RFC5116].
To the extent that message-based encryption formats use the same To the extent that message-based encryption formats use the same
primitives, the format can be considered as sequence of encrypted primitives, the format can be considered as sequence of encrypted
messages with a particular profile. For instance, Appendix A messages with a particular profile. For instance, Appendix A
explains how the format is congruent with a sequence of JSON Web explains how the format is congruent with a sequence of JSON Web
Encryption [RFC7516] values with a fixed header. Encryption [RFC7516] values with a fixed header.
This mechanism is likely only a small part of a larger design that This mechanism is likely only a small part of a larger design that
uses content encryption. How clients and servers acquire and uses content encryption. How clients and servers acquire and
skipping to change at page 4, line 6 skipping to change at page 4, line 6
Using this content coding requires knowledge of a key. How this key Using this content coding requires knowledge of a key. How this key
is acquired is not defined in this document. is acquired is not defined in this document.
The "aes128gcm" content coding uses a single fixed set of encryption The "aes128gcm" content coding uses a single fixed set of encryption
primitives. Cipher suite agility is achieved by defining a new primitives. Cipher suite agility is achieved by defining a new
content coding scheme. This ensures that only the HTTP Accept- content coding scheme. This ensures that only the HTTP Accept-
Encoding header field is necessary to negotiate the use of Encoding header field is necessary to negotiate the use of
encryption. encryption.
The "aes128gcm" content coding uses a fixed record size. The final The "aes128gcm" content coding uses a fixed record size. The final
encoding consists of a header (see Section 2.1), zero or more fixed encoding consists of a header (see Section 2.1) and zero or more
size encrypted records, and a partial record. The partial record fixed size encrypted records; the final record can be smaller than
MUST be shorter than the fixed record size. the record size.
The record size determines the length of each portion of plaintext The record size determines the length of each portion of plaintext
that is enciphered, with the exception of the final record, which is that is enciphered. The record size ("rs") is included in the
necessarily smaller. The record size ("rs") is included in the
content coding header (see Section 2.1). content coding header (see Section 2.1).
+-----------+ content of rs octets minus padding +-----------+ content
| data | less padding (2-65537) and tag (16); | data | any length up to rs-17 octets
+-----------+ the last record is smaller +-----------+
| |
v v
+-----+-----------+ add padding to get rs-16 octets; +-----------+-----+ add a delimiter octet (0x01 or 0x02)
| pad | data | the last record contains | data | pad | the 0x00-valued octets to rs-16
+-----+-----------+ up to rs minus 17 octets +-----------+-----+ (or less on the last record)
| |
v v
+--------------------+ encrypt with AEAD_AES_128_GCM; +--------------------+ encrypt with AEAD_AES_128_GCM;
| ciphertext | final size is rs; | ciphertext | final size is rs;
+--------------------+ the last record is smaller +--------------------+ the last record can be smaller
AEAD_AES_128_GCM produces ciphertext 16 octets longer than its input AEAD_AES_128_GCM produces ciphertext 16 octets longer than its input
plaintext. Therefore, the unencrypted content of each record is plaintext. Therefore, the unencrypted content of each record is
shorter than the record size by 16 octets. If the final record ends shorter than the record size by 16 octets. Valid records always
on a record boundary, the encoder MUST append a record that contains contain at least a padding delimiter octet and a 16 octet
contains only padding and is smaller than the full record size. A
receiver MUST fail to decrypt if the final record ciphertext is less
than 18 octets in size or equal to the record size. Valid records
always contain at least a padding length of 2 octets and a 16 octet
authentication tag. authentication tag.
Each record contains a 2 octet padding length and between 0 and 65535 Each record contains a single padding delimiter octet followed by any
octets of padding, inserted into a record before the content. The number of zero octets. The last record uses a padding delimiter
padding length is a two octet unsigned integer in network byte order; octet set to the value 2, all other records have a padding delimiter
padding is that number of zero-valued octets. A receiver MUST fail octet value of 1. A decrypter MUST fail if the unencrypted content
to decrypt if any padding octet is non-zero, or a record has more of a record is all zero-valued. A decrypter MUST fail if the last
padding than the record size can accommodate. record contains a padding delimiter with a value other than 2; a
decrypter MUST fail if any record other than the last contains a
padding delimiter with a value other than 1.
The nonce for each record is a 96-bit value constructed from the The nonce for each record is a 96-bit value constructed from the
record sequence number and the input keying material. Nonce record sequence number and the input keying material. Nonce
derivation is covered in Section 2.3. derivation is covered in Section 2.3.
The additional data passed to each invocation of AEAD_AES_128_GCM is The additional data passed to each invocation of AEAD_AES_128_GCM is
a zero-length octet sequence. a zero-length octet sequence.
A consequence of this record structure is that range requests A consequence of this record structure is that range requests
[RFC7233] and random access to encrypted payload bodies are possible [RFC7233] and random access to encrypted payload bodies are possible
skipping to change at page 5, line 18 skipping to change at page 5, line 15
of a range cannot be decrypted. Thus, it is best if range requests of a range cannot be decrypted. Thus, it is best if range requests
start and end on record boundaries. Note however that random access start and end on record boundaries. Note however that random access
to specific parts of encrypted data could be confounded by the to specific parts of encrypted data could be confounded by the
presence of padding. presence of padding.
Selecting the record size most appropriate for a given situation Selecting the record size most appropriate for a given situation
requires a trade-off. A smaller record size allows decrypted octets requires a trade-off. A smaller record size allows decrypted octets
to be released more rapidly, which can be appropriate for to be released more rapidly, which can be appropriate for
applications that depend on responsiveness. Smaller records also applications that depend on responsiveness. Smaller records also
reduce the additional data required if random access into the reduce the additional data required if random access into the
ciphertext is needed. Applications that depend on being able to pad ciphertext is needed.
by arbitrary amounts cannot increase the record size beyond 65537
octets.
Applications that don't depending on streaming, random access, or Applications that don't depending on streaming, random access, or
arbitrary padding can use larger records, or even a single record. A arbitrary padding can use larger records, or even a single record. A
larger record size reduces the processing and data overheads. larger record size reduces processing and data overheads.
2.1. Encryption Content Coding Header 2.1. Encryption Content Coding Header
The content coding uses a header block that includes all parameters The content coding uses a header block that includes all parameters
needed to decrypt the content (other than the key). The header block needed to decrypt the content (other than the key). The header block
is placed in the body of a message ahead of the sequence of records. is placed in the body of a message ahead of the sequence of records.
+-----------+--------+-----------+---------------+ +-----------+--------+-----------+---------------+
| salt (16) | rs (4) | idlen (1) | keyid (idlen) | | salt (16) | rs (4) | idlen (1) | keyid (idlen) |
+-----------+--------+-----------+---------------+ +-----------+--------+-----------+---------------+
skipping to change at page 5, line 47 skipping to change at page 5, line 42
"aes128gcm" content coding header. The same "salt" parameter "aes128gcm" content coding header. The same "salt" parameter
value MUST NOT be reused for two different payload bodies that value MUST NOT be reused for two different payload bodies that
have the same input keying material; generating a random salt for have the same input keying material; generating a random salt for
every application of the content coding ensures that content every application of the content coding ensures that content
encryption key reuse is highly unlikely. encryption key reuse is highly unlikely.
rs: The "rs" or record size parameter contains an unsigned 32-bit rs: The "rs" or record size parameter contains an unsigned 32-bit
integer in network byte order that describes the record size in integer in network byte order that describes the record size in
octets. Note that it is therefore impossible to exceed the octets. Note that it is therefore impossible to exceed the
2^36-31 limit on plaintext input to AEAD_AES_128_GCM. Values 2^36-31 limit on plaintext input to AEAD_AES_128_GCM. Values
smaller than 19 are invalid. smaller than 18 are invalid.
keyid: The "keyid" parameter can be used to identify the keying keyid: The "keyid" parameter can be used to identify the keying
material that is used. Recipients that receive a message are material that is used. Recipients that receive a message are
expected to know how to retrieve keys; the "keyid" parameter might expected to know how to retrieve keys; the "keyid" parameter might
be input to that process. A "keyid" parameter SHOULD be a UTF-8 be input to that process. A "keyid" parameter SHOULD be a UTF-8
[RFC3629] encoded string, particularly where the identifier might [RFC3629] encoded string, particularly where the identifier might
need to appear in a textual form. need to appear in a textual form.
2.2. Content Encryption Key Derivation 2.2. Content Encryption Key Derivation
skipping to change at page 6, line 43 skipping to change at page 6, line 40
(CEK), so the length (L) parameter to HKDF is 16. The second step of (CEK), so the length (L) parameter to HKDF is 16. The second step of
HKDF can therefore be simplified to the first 16 octets of a single HKDF can therefore be simplified to the first 16 octets of a single
HMAC: HMAC:
CEK = HMAC-SHA-256(PRK, cek_info || 0x01) CEK = HMAC-SHA-256(PRK, cek_info || 0x01)
2.3. Nonce Derivation 2.3. Nonce Derivation
The nonce input to AEAD_AES_128_GCM is constructed for each record. The nonce input to AEAD_AES_128_GCM is constructed for each record.
The nonce for each record is a 12 octet (96 bit) value that is The nonce for each record is a 12 octet (96 bit) value that is
produced from the record sequence number and a value derived from the derived from the record sequence number, input keying material, and
input keying material. salt.
The input keying material and salt values are input to HKDF with The input keying material and salt values are input to HKDF with
different info and length parameters. different info and length parameters.
The length (L) parameter is 12 octets. The info parameter for the The length (L) parameter is 12 octets. The info parameter for the
nonce is the ASCII-encoded string "Content-Encoding: nonce", nonce is the ASCII-encoded string "Content-Encoding: nonce",
terminated by a a single zero octet: terminated by a a single zero octet:
nonce_info = "Content-Encoding: nonce" || 0x00 nonce_info = "Content-Encoding: nonce" || 0x00
skipping to change at page 7, line 35 skipping to change at page 7, line 31
This section shows a few examples of the encrypted content coding. This section shows a few examples of the encrypted content coding.
Note: All binary values in the examples in this section use base64url Note: All binary values in the examples in this section use base64url
encoding [RFC7515]. This includes the bodies of requests. encoding [RFC7515]. This includes the bodies of requests.
Whitespace and line wrapping is added to fit formatting constraints. Whitespace and line wrapping is added to fit formatting constraints.
3.1. Encryption of a Response 3.1. Encryption of a Response
Here, a successful HTTP GET response has been encrypted. This uses a Here, a successful HTTP GET response has been encrypted. This uses a
record size of 4096 and no padding (just the 2 octet padding length), record size of 4096 and no padding (just the single octet padding
so only a partial record is present. The input keying material is delimiter), so only a partial record is present. The input keying
identified by an empty string (that is, the "keyid" field in the material is identified by an empty string (that is, the "keyid" field
header is zero octets in length). in the header is zero octets in length).
The encrypted data in this example is the UTF-8 encoded string "I am The encrypted data in this example is the UTF-8 encoded string "I am
the walrus". The input keying material is the value the walrus". The input keying material is the value "yqdlZ-
"B33e_VeFrOyIHwFTIfmesA" (in base64url). The content body contains a tYemfogSmv7Ws5PQ" (in base64url). The 54 octet content body contains
single record and is shown here using base64url encoding for a single record and is shown here using 71 base64url characters for
presentation reasons. presentation reasons.
HTTP/1.1 200 OK HTTP/1.1 200 OK
Content-Type: application/octet-stream Content-Type: application/octet-stream
Content-Length: 54 Content-Length: 54
Content-Encoding: aes128gcm Content-Encoding: aes128gcm
sJvlboCWzB5jr8hI_q9cOQAAEAAANSmxkSVa0-MiNNuF77YHSs-iwaNe_OK0qfmO I1BsxtFttlv3u_Oo94xnmwAAEAAA-NAVub2qFgBEuQKRapoZu-IxkIva3MEB1PD-
c7NT5WSW ly8Thjg
Note that the media type has been changed to "application/octet- Note that the media type has been changed to "application/octet-
stream" to avoid exposing information about the content. stream" to avoid exposing information about the content.
Alternatively (and equivalently), the Content-Type header field can Alternatively (and equivalently), the Content-Type header field can
be omitted. be omitted.
Intermediate values for this example (all shown in base64): Intermediate values for this example (all shown using base64url):
salt (from header) = sJvlboCWzB5jr8hI_q9cOQ salt (from header) = I1BsxtFttlv3u_Oo94xnmw
PRK = MLAQxt_DHjM15cdlyU1oUnjq7TFlzToGTkdRmvvxVBw PRK = zyeH5phsIsgUyd4oiSEIy35x-gIi4aM7y0hCF8mwn9g
CEK = v31u7VGV3soO3wNaMaIdhg CEK = _wniytB-ofscZDh4tbSjHw
NONCE = XOaygzko98zjUFTJ NONCE = Bcs8gkIRKLI8GeI8
plaintext = AABJIGFtIHRoZSB3YWxydXM plaintext = SSBhbSB0aGUgd2FscnVzAg
3.2. Encryption with Multiple Records 3.2. Encryption with Multiple Records
This example shows the same message with input keying material of This example shows the same message with input keying material of
"BO3ZVPxUlnLORbVGMpbT1Q". In this example, the plaintext is split "BO3ZVPxUlnLORbVGMpbT1Q". In this example, the plaintext is split
into records of 26 octets each (that is, the "rs" field in the header into records of 25 octets each (that is, the "rs" field in the header
is 26). The first record includes a single octet of padding. This is 25). The first record includes one 0x00 padding octet. This
means that there are 7 octets of message in the first record, and 8 means that there are 7 octets of message in the first record, and 8
in the second. This causes the end of the content to align with a in the second. A key identifier of the UTF-8 encoded string "a1" is
record boundary, forcing the creation of a third record that contains also included in the header.
only two octets of the padding length.
HTTP/1.1 200 OK HTTP/1.1 200 OK
Content-Length: 93 Content-Length: 73
Content-Encoding: aes128gcm Content-Encoding: aes128gcm
uNCkWiNYzKTnBN9ji3-qWAAAABoCYTGHOqYFz-0in3dpb-VE2GfBngkaPy6bZus_ uNCkWiNYzKTnBN9ji3-qWAAAABkCYTHOG8chz_gnvgOqdGYovxyjuqRyJFjEDyoF
qLF79s6zQyTSsA0iLOKyd3JqVIwprNzVatRCWZGUx_qsFbJBCQu62RqQuR2d 1Fvkj6hQPdPHI51OEUKEpgz3SsLWIqS_uA
4. Security Considerations 4. Security Considerations
This mechanism assumes the presence of a key management framework This mechanism assumes the presence of a key management framework
that is used to manage the distribution of keys between valid senders that is used to manage the distribution of keys between valid senders
and receivers. Defining key management is part of composing this and receivers. Defining key management is part of composing this
mechanism into a larger application, protocol, or framework. mechanism into a larger application, protocol, or framework.
Implementation of cryptography - and key management in particular - Implementation of cryptography - and key management in particular -
can be difficult. For instance, implementations need to account for can be difficult. For instance, implementations need to account for
the potential for exposing keying material on side channels, such as the potential for exposing keying material on side channels, such as
might be exposed by the time it takes to perform a given operation. might be exposed by the time it takes to perform a given operation.
The requirements for a good implementation of cryptographic The requirements for a good implementation of cryptographic
algorithms can change over time. algorithms can change over time.
4.1. Key and Nonce Reuse As a content coding, presence of the "aes128gcm" coding might be
transparent to a consumer of a message. Recipients that depend on
content origin authentication using this mechanism MUST reject
messages that don't include the "aes128gcm" content coding.
4.1. Message Truncation
This content encoding is designed to permit the incremental
processing of large messages. It also permits random access to
plaintext in a limited fashion. The content encoding permits a
receiver to detect when a message is truncated.
A partially delivered message MUST NOT be processed as though the
entire message was successfully delivered. For instance, a partially
delivered message cannot be cached as though it were complete.
An attacker might exploit willingness to process partial messages to
cause a receiver to remain in a specific intermediate state.
Implementations performing processing on partial messages need to
ensure that any intermediate processing states don't advantage an
attacker.
4.2. Key and Nonce Reuse
Encrypting different plaintext with the same content encryption key Encrypting different plaintext with the same content encryption key
and nonce in AES-GCM is not safe [RFC5116]. The scheme defined here and nonce in AES-GCM is not safe [RFC5116]. The scheme defined here
uses a fixed progression of nonce values. Thus, a new content uses a fixed progression of nonce values. Thus, a new content
encryption key is needed for every application of the content coding. encryption key is needed for every application of the content coding.
Since input keying material can be reused, a unique "salt" parameter Since input keying material can be reused, a unique "salt" parameter
is needed to ensure a content encryption key is not reused. is needed to ensure a content encryption key is not reused.
If a content encryption key is reused - that is, if input keying If a content encryption key is reused - that is, if input keying
material and salt are reused - this could expose the plaintext and material and salt are reused - this could expose the plaintext and
the authentication key, nullifying the protection offered by the authentication key, nullifying the protection offered by
encryption. Thus, if the same input keying material is reused, then encryption. Thus, if the same input keying material is reused, then
the salt parameter MUST be unique each time. This ensures that the the salt parameter MUST be unique each time. This ensures that the
content encryption key is not reused. An implementation SHOULD content encryption key is not reused. An implementation SHOULD
generate a random salt parameter for every message; a counter could generate a random salt parameter for every message; a counter could
achieve the same result. achieve the same result.
4.2. Data Encryption Limits 4.3. Data Encryption Limits
There are limits to the data that AEAD_AES_128_GCM can encipher. The There are limits to the data that AEAD_AES_128_GCM can encipher. The
maximum value for the record size is limited by the size of the "rs" maximum value for the record size is limited by the size of the "rs"
field in the header (see Section 2.1), which ensures that the 2^36-31 field in the header (see Section 2.1), which ensures that the 2^36-31
limit for a single application of AEAD_AES_128_GCM is not reached limit for a single application of AEAD_AES_128_GCM is not reached
[RFC5116]. In order to preserve a 2^-40 probability of [RFC5116]. In order to preserve a 2^-40 probability of
indistinguishability under chosen plaintext attack (IND-CPA), the indistinguishability under chosen plaintext attack (IND-CPA), the
total amount of plaintext that can be enciphered MUST be less than total amount of plaintext that can be enciphered with the key derived
2^44.5 blocks of 16 octets [AEBounds]. from the same input keying material and salt MUST be less than 2^44.5
blocks of 16 octets [AEBounds].
If the record size is a multiple of 16 octets, this means 398 If the record size is a multiple of 16 octets, this means 398
terabytes can be encrypted safely, including padding and overhead. terabytes can be encrypted safely, including padding and overhead.
However, if the record size is not a multiple of 16 octets, the total However, if the record size is not a multiple of 16 octets, the total
amount of data that can be safely encrypted is reduced because amount of data that can be safely encrypted is reduced because
partial AES blocks are encrypted. The worst case is a record size of partial AES blocks are encrypted. The worst case is a record size of
19 octets, for which at most 74 terabytes of plaintext can be 18 octets, for which at most 74 terabytes of plaintext can be
encrypted, of which at least two-thirds is padding. encrypted, of which at least half is padding.
4.3. Content Integrity 4.4. Content Integrity
This mechanism only provides content origin authentication. The This mechanism only provides content origin authentication. The
authentication tag only ensures that an entity with access to the authentication tag only ensures that an entity with access to the
content encryption key produced the encrypted data. content encryption key produced the encrypted data.
Any entity with the content encryption key can therefore produce Any entity with the content encryption key can therefore produce
content that will be accepted as valid. This includes all recipients content that will be accepted as valid. This includes all recipients
of the same HTTP message. of the same HTTP message.
Furthermore, any entity that is able to modify both the Encryption Furthermore, any entity that is able to modify both the Content-
header field and the HTTP message body can replace the contents. Encoding header field and the HTTP message body can replace the
Without the content encryption key or the input keying material, contents. Without the content encryption key or the input keying
modifications to or replacement of parts of a payload body are not material, modifications to or replacement of parts of a payload body
possible. are not possible.
4.4. Leaking Information in Headers 4.5. Leaking Information in Header Fields
Because only the payload body is encrypted, information exposed in Because only the payload body is encrypted, information exposed in
header fields is visible to anyone who can read the HTTP message. header fields is visible to anyone who can read the HTTP message.
This could expose side-channel information. This could expose side-channel information.
For example, the Content-Type header field can leak information about For example, the Content-Type header field can leak information about
the payload body. the payload body.
There are a number of strategies available to mitigate this threat, There are a number of strategies available to mitigate this threat,
depending upon the application's threat model and the users' depending upon the application's threat model and the users'
skipping to change at page 11, line 5 skipping to change at page 11, line 13
in other representations, etc.), omit the relevant headers, and/ in other representations, etc.), omit the relevant headers, and/
or normalize them. In the case of Content-Type, this could be or normalize them. In the case of Content-Type, this could be
accomplished by always sending Content-Type: application/octet- accomplished by always sending Content-Type: application/octet-
stream (the most generic media type), or no Content-Type at all. stream (the most generic media type), or no Content-Type at all.
3. If it is considered sensitive information and it is not possible 3. If it is considered sensitive information and it is not possible
to convey it elsewhere, encapsulate the HTTP message using the to convey it elsewhere, encapsulate the HTTP message using the
application/http media type (Section 8.3.2 of [RFC7230]), application/http media type (Section 8.3.2 of [RFC7230]),
encrypting that as the payload of the "outer" message. encrypting that as the payload of the "outer" message.
4.5. Poisoning Storage 4.6. Poisoning Storage
This mechanism only offers encryption of content; it does not perform This mechanism only offers encryption of content; it does not perform
authentication or authorization, which still needs to be performed authentication or authorization, which still needs to be performed
(e.g., by HTTP authentication [RFC7235]). (e.g., by HTTP authentication [RFC7235]).
This is especially relevant when a HTTP PUT request is accepted by a This is especially relevant when a HTTP PUT request is accepted by a
server; if the request is unauthenticated, it becomes possible for a server; if the request is unauthenticated, it becomes possible for a
third party to deny service and/or poison the store. third party to deny service and/or poison the store.
4.6. Sizing and Timing Attacks 4.7. Sizing and Timing Attacks
Applications using this mechanism need to be aware that the size of Applications using this mechanism need to be aware that the size of
encrypted messages, as well as their timing, HTTP methods, URIs and encrypted messages, as well as their timing, HTTP methods, URIs and
so on, may leak sensitive information. so on, may leak sensitive information.
This risk can be mitigated through the use of the padding that this This risk can be mitigated through the use of the padding that this
mechanism provides. Alternatively, splitting up content into mechanism provides. Alternatively, splitting up content into
segments and storing the separately might reduce exposure. HTTP/2 segments and storing them separately might reduce exposure. HTTP/2
[RFC7540] combined with TLS [RFC5246] might be used to hide the size [RFC7540] combined with TLS [RFC5246] might be used to hide the size
of individual messages. of individual messages.
Developing a padding strategy is difficult. A good padding strategy Developing a padding strategy is difficult. A good padding strategy
can depend on context. Common strategies include padding to a small can depend on context. Common strategies include padding to a small
set of fixed lengths, padding to multiples of a values, or padding to set of fixed lengths, padding to multiples of a value, or padding to
powers of 2. Even a good strategy can still cause size information powers of 2. Even a good strategy can still cause size information
to leak if processing activity of a recipient can be observed. This to leak if processing activity of a recipient can be observed. This
is especially true if the trailing records of a message contain only is especially true if the trailing records of a message contain only
padding. Distributing non-padding data is recommended to avoid padding. Distributing non-padding data is recommended to avoid
leaking size information. leaking size information.
5. IANA Considerations 5. IANA Considerations
5.1. The "aes128gcm" HTTP Content Coding 5.1. The "aes128gcm" HTTP Content Coding
skipping to change at page 13, line 48 skipping to change at page 14, line 9
[XMLENC] Eastlake, D., Reagle, J., Hirsch, F., Roessler, T., [XMLENC] Eastlake, D., Reagle, J., Hirsch, F., Roessler, T.,
Imamura, T., Dillaway, B., Simon, E., Yiu, K., and M. Imamura, T., Dillaway, B., Simon, E., Yiu, K., and M.
Nystroem, "XML Encryption Syntax and Processing", W3C Nystroem, "XML Encryption Syntax and Processing", W3C
Recommendation REC-xmlenc-core1-20130411 , January 2013, Recommendation REC-xmlenc-core1-20130411 , January 2013,
<https://www.w3.org/TR/2013/REC-xmlenc-core1-20130411>. <https://www.w3.org/TR/2013/REC-xmlenc-core1-20130411>.
Appendix A. JWE Mapping Appendix A. JWE Mapping
The "aes128gcm" content coding can be considered as a sequence of The "aes128gcm" content coding can be considered as a sequence of
JSON Web Encryption (JWE) objects [RFC7516], each corresponding to a JSON Web Encryption (JWE) objects [RFC7516], each corresponding to a
single fixed size record that includes leading padding. The single fixed size record that includes trailing padding. The
following transformations are applied to a JWE object that might be following transformations are applied to a JWE object that might be
expressed using the JWE Compact Serialization: expressed using the JWE Compact Serialization:
o The JWE Protected Header is fixed to the value { "alg": "dir", o The JWE Protected Header is fixed to the value { "alg": "dir",
"enc": "A128GCM" }, describing direct encryption using AES-GCM "enc": "A128GCM" }, describing direct encryption using AES-GCM
with a 128-bit content encryption key. This header is not with a 128-bit content encryption key. This header is not
transmitted, it is instead implied by the value of the Content- transmitted, it is instead implied by the value of the Content-
Encoding header field. Encoding header field.
o The JWE Encrypted Key is empty, as stipulated by the direct o The JWE Encrypted Key is empty, as stipulated by the direct
skipping to change at page 14, line 27 skipping to change at page 14, line 35
Section 2.3). This value is also not transmitted. Section 2.3). This value is also not transmitted.
o The final value is the concatenated header, JWE Ciphertext, and o The final value is the concatenated header, JWE Ciphertext, and
JWE Authentication Tag, all expressed without base64url encoding. JWE Authentication Tag, all expressed without base64url encoding.
The "." separator is omitted, since the length of these fields is The "." separator is omitted, since the length of these fields is
known. known.
Thus, the example in Section 3.1 can be rendered using the JWE Thus, the example in Section 3.1 can be rendered using the JWE
Compact Serialization as: Compact Serialization as:
eyAiYWxnIjogImRpciIsICJlbmMiOiAiQTEyOEdDTSIgfQ..31iQYc1v4a36EgyJ. eyAiYWxnIjogImRpciIsICJlbmMiOiAiQTEyOEdDTSIgfQ..Bcs8gkIRKLI8GeI8.
NSmxkSVa0-MiNNuF77YHSs8.osGjXvzitKn5jnOzU-Vklg -NAVub2qFgBEuQKRapoZuw.4jGQi9rcwQHU8P6XLxOGOA
Where the first line represents the fixed JWE Protected Header, an Where the first line represents the fixed JWE Protected Header, an
empty JWE Encrypted Key, and the algorithmically-determined JWE empty JWE Encrypted Key, and the algorithmically-determined JWE
Initialization Vector. The second line contains the encoded body, Initialization Vector. The second line contains the encoded body,
split into JWE Ciphertext and JWE Authentication Tag. split into JWE Ciphertext and JWE Authentication Tag.
Appendix B. Acknowledgements Appendix B. Acknowledgements
Mark Nottingham was an original author of this document. Mark Nottingham was an original author of this document.
The following people provided valuable input: Richard Barnes, David The following people provided valuable input: Richard Barnes, David
Benjamin, Peter Beverloo, JR Conlin, Mike Jones, Stephen Farrell, Benjamin, Peter Beverloo, JR Conlin, Mike Jones, Stephen Farrell,
Adam Langley, John Mattsson, Julian Reschke, Eric Rescorla, Jim Adam Langley, James Manger, John Mattsson, Julian Reschke, Eric
Schaad, and Magnus Westerlund. Rescorla, Jim Schaad, and Magnus Westerlund.
Author's Address Author's Address
Martin Thomson Martin Thomson
Mozilla Mozilla
Email: martin.thomson@gmail.com Email: martin.thomson@gmail.com
 End of changes. 41 change blocks. 
94 lines changed or deleted 111 lines changed or added

This html diff was produced by rfcdiff 1.45. The latest version is available from http://tools.ietf.org/tools/rfcdiff/