| RFC 9164 | CBOR Tags for IP | December 2021 |
| Richardson & Bormann | Standards Track | [Page] |
This specification defines two Concise Binary Object Representation (CBOR) tags for use with IPv6 and IPv4 addresses and prefixes.¶
This is an Internet Standards Track document.¶
This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Further information on Internet Standards is available in Section 2 of RFC 7841.¶
Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at https://www.rfc-editor.org/info/rfc9164.¶
Copyright (c) 2021 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 (https://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 Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
[RFC8949] defines a number of CBOR tags for common items. Tags 260 and 261 were later defined in drafts listed with IANA [IANA.cbor-tags]. These tags were intended to cover addresses (260) and prefixes (261). Tag 260 distinguishes between IPv6, IPv4, and MAC [RFC7042] addresses only through the length of the byte string, making it impossible, for example, to drop trailing zeros in the encoding of IP addresses. Tag 261 was not documented well enough for use.¶
This specification defines tags 54 and 52 to explicitly indicate use of IPv6 or IPv4 by the tag number. These new tags are intended to be used in preference to tags 260 and 261. They provide formats for IPv6 and IPv4 addresses, prefixes, and addresses with prefixes, while explicitly indicating use of IPv6 or IPv4. The prefix format omits trailing zeroes in the address part. (Due to the complexity of testing, the value of omitting trailing zeros for the pure address format was considered nonessential, and support for that is not provided in this specification.) This specification does not deal with MAC addresses (Section 2 of [RFC7042]).¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
These tags can be applied to byte strings to represent a single address.¶
This form is called the "Address Format".¶
When applied to an array that starts with an unsigned integer, the tags represent a CIDR-style prefix of that length.¶
When the Address Format (i.e., without prefix) appears in a context where a prefix is expected, then it is to be assumed that all bits are relevant. That is, for IPv4, a /32 is implied, and for IPv6, a /128 is implied.¶
This form is called the "Prefix Format".¶
When applied to an array that starts with a byte string, which stands
for an IP address, followed by an unsigned integer giving the bit
length of a prefix built out of the first length bits of the
address, the tags represent information that is commonly used to specify
both the network prefix and the IP address of an interface.¶
The length of the byte string is always 16 bytes (for IPv6) and 4 bytes (for IPv4).¶
This form is called the "Interface Format".¶
Interface Format definitions support an optional third element to the array, which is to be used as the IPv6 link-local zone identifier from Section 6 of [RFC4007]; for symmetry, this is also provided for IPv4 as in [RFC4001] and [RFC6991]. The zone identifier may be an integer, in which case it is to be interpreted as the interface index. It may be a text string, in which case it is to be interpreted as an interface name.¶
As explained in [RFC4007], the zone identifiers are strictly local to the node. They are useful for communications within a node about connected addresses (for instance, where a link-local peer is discovered by one daemon and another daemon needs to be informed). They may also have utility in some management protocols.¶
In the cases where the Interface Format is being used to represent only an address with a zone identifier and no interface prefix information, the prefix length may be replaced with the CBOR "null" (0xF6).¶
IANA has allocated tag 54 for IPv6 uses. (This is the ASCII code for '6'.)¶
An IPv6 address is to be encoded as a sixteen-byte byte string (Section 3.1 of [RFC8949], major type 2), enclosed in tag number 54.¶
For example:¶
54(h'20010db81234deedbeefcafefacefeed')¶
An IPv6 prefix, such as 2001:db8:1234::/48, is to be encoded as a two-element array, with the length of the prefix first. See Section 4 for the detailed construction of the second element.¶
For example:¶
54([48, h'20010db81234'])¶
An IPv6 address combined with a prefix length, such as one used for configuring an interface, is to be encoded as a two-element array, with the (full-length) IPv6 address first and the length of the associated network the prefix next; a third element can be added for the zone identifier.¶
For example:¶
54([h'20010db81234deedbeefcafefacefeed', 56])¶
The address-with-prefix form can be reliably distinguished from the prefix form only in the sequence of the array elements.¶
An example of a link-local IPv6 address with a 64-bit prefix:¶
54([h'fe8000000000020202fffffffe030303', 64, 'eth0'])¶
with a numeric zone identifier:¶
54([h'fe8000000000020202fffffffe030303', 64, 42])¶
An IPv6 link-local address without a prefix length:¶
54([h'fe8000000000020202fffffffe030303', null, 42])¶
Zone identifiers may be used with any kind of IP address, not just link-local addresses. In particular, they are valid for multicast addresses, and there may still be some significance for Globally Unique Addresses (GUAs).¶
IANA has allocated tag 52 for IPv4 uses. (This is the ASCII code for '4'.)¶
An IPv4 address is to be encoded as a four-byte byte string (Section 3.1 of [RFC8949], major type 2), enclosed in tag number 52.¶
For example:¶
52(h'c0000201')¶
An IPv4 prefix, such as 192.0.2.0/24, is to be encoded as a two-element array, with the length of the prefix first. See Section 4 for the detailed construction of the second element.¶
For example:¶
52([24, h'c00002'])¶
An IPv4 address combined with a prefix length, such as being used for configuring an interface, is to be encoded as a two-element array, with the (full-length) IPv4 address first and the length of the associated network the prefix next; a third element can be added for the zone identifier.¶
For example, 192.0.2.1/24 is to be encoded as a two-element array, with the length of the prefix (implied 192.0.2.0/24) last.¶
52([h'c0000201', 24])¶
The address-with-prefix form can be reliably distinguished from the prefix form only in the sequence of the array elements.¶
This section discusses when tag 54 or tag 52 is valid (Section 5.3.2 of [RFC8949]). As with all CBOR tags, validity checking can be handled in a generic CBOR library or in the application. A generic CBOR library needs to document whether and how it handles validity checking.¶
The rule ip-address-or-prefix in Figure 1 shows how to check the
overall structure of these tags and their content, the ranges of
integer values, and the lengths of byte strings.
An instance of tag 52 or 54 is valid if it matches that rule and, for
ipv6-prefix and ipv4-prefix, the considerations of Sections
4.2 and 4.3.¶
The tag validity rules, combined with the rules in Section 4.2.1 of [RFC8949], lead to deterministic encoding for tags 54 and 52 and require no further additional deterministic encoding considerations as per Section 4.2.2 of [RFC8949].¶
For the byte strings used as the second element in the array representing a prefix:¶
(1) An encoder MUST set any unused bytes and any unused bits in the
final byte, if any, to zero.
Unused bytes (or bits) are bytes (or bits) that are not covered by the prefix length given.
So, for example, 2001:db8:1230::/44 MUST be encoded as:¶
54([44, h'20010db81230'])¶
even though variations like:¶
54([44, h'20010db81233']) 54([44, h'20010db8123f']) 54([44, h'20010db8123012'])¶
start with the same 44 bits but are not valid.¶
(Analogous examples can be constructed for IPv4 prefixes.)¶
(2) An encoder MUST then omit any right-aligned (trailing) sequence of bytes in which the bytes are all zeros.¶
There is no relationship between the number of bytes omitted and the prefix length. For instance, the prefix 2001:db8::/64 is encoded as:¶
54([64, h'20010db8'])¶
A decoder MUST check that all unused bits encoded in the byte string ipv6-prefix-bytes/ipv4-prefix-bytes, i.e., the bits to the right of the prefix length, are zero.¶
A decoder MUST also check that the byte string does not end in a zero byte.¶
Since encoders are required to remove zero-valued trailing bytes, a decoder MUST handle cases where a prefix length specifies that more bits are relevant than are actually present in the byte string.¶
As an example, ::/128 is encoded as¶
54([128, h''])¶
A recommendation for prefix decoder implementations is to first create an array of 16 (or 4) zero bytes.¶
Then, taking whichever is smaller between (a) the length of the included byte string and (b) the number of bytes covered by the prefix length rounded up to the next multiple of 8, fail if that number is greater than 16 (or 4) and then copy that many bytes from the byte string into the byte array.¶
Finally, when looking at the number of unused bits in the last byte (if any) of the range covered by the prefix length, check that any unused bits in the byte string are zero:¶
unused_bits = (8 - (prefix_length_in_bits % 8)) % 8;
if (length_in_bytes > 0 &&
(address_bytes[length_in_bytes - 1] & ~(0xFF << unused_bits))
!= 0)
fail();
¶
For use with Concise Data Definition Language (CDDL) [RFC8610], the type names defined in Figure 1 are recommended:¶
ip-address-or-prefix = ipv6-address-or-prefix /
ipv4-address-or-prefix
ipv6-address-or-prefix = #6.54(ipv6-address /
ipv6-address-with-prefix /
ipv6-prefix)
ipv4-address-or-prefix = #6.52(ipv4-address /
ipv4-address-with-prefix /
ipv4-prefix)
ipv6-address = bytes .size 16
ipv4-address = bytes .size 4
ipv6-address-with-prefix = [ipv6-address,
ipv6-prefix-length / null,
?ip-zone-identifier]
ipv4-address-with-prefix = [ipv4-address,
ipv4-prefix-length / null,
?ip-zone-identifier]
ipv6-prefix-length = 0..128
ipv4-prefix-length = 0..32
ipv6-prefix = [ipv6-prefix-length, ipv6-prefix-bytes]
ipv4-prefix = [ipv4-prefix-length, ipv4-prefix-bytes]
ipv6-prefix-bytes = bytes .size (uint .le 16)
ipv4-prefix-bytes = bytes .size (uint .le 4)
ip-zone-identifier = uint / text
This document provides a CBOR encoding for IPv4 and IPv6 address information. Any applications using these encodings will need to consider the security implications of this data in their specific context. For example, identifying which byte sequences in a protocol are addresses may allow an attacker or eavesdropper to better understand what parts of a packet to attack.¶
Applications need to check the validity (Section 4) of a tag before acting on any of its contents. If the validity checking is not done in the generic CBOR decoder, it needs to be done in the application; in any case, it needs to be done before the tag is transformed into a platform-specific representation that could conceal validity errors.¶
The right-hand bits of the prefix, after the prefix length, are set to zero by this protocol. (Otherwise, a malicious party could use them to transmit covert data in a way that would not affect the primary use of this encoding. Such abuse is detected by tag validity checking and can also be detected by examination of the raw protocol bytes.)¶
IANA has allocated two tags from the Specification Required [RFC8126] area of the "Concise Binary Object Representation (CBOR) Tags" registry [IANA.cbor-tags]:¶
Roman Danyliw, Donald Eastlake, Ben Kaduk, Barry Leiba, and Éric Vyncke reviewed the document and provided suggested text. Jürgen Schönwälder helped find the history of IPv4 zone identifiers.¶