Internet-Draft | icmpext-xlat-source | July 2024 |
Lamparter & Linkova | Expires 8 January 2025 | [Page] |
This document proposes allocating a dedicated IPv4 address for translating arbitrary (non-IPv4-translatable) IPv6 addresses and suggests the creation of an ICMP Multi-part Extension to carry the original IPv6 source address of ICMPv6 messages translated to ICMP by stateless (RFC7915) or stateful (RFC 6146) protocol translators.¶
This note is to be removed before publishing as an RFC.¶
Source, version control history, and issue tracker for this draft can be found at https://github.com/eqvinox/icmpext-clat-source.¶
(Note the draft was renamed (clat → xlat) prior to submission but changing the repository name on github breaks too many things to be worth the effort.)¶
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 https://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."¶
This Internet-Draft will expire on 8 January 2025.¶
Copyright (c) 2024 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.¶
To allow communication between IPv6-only and IPv4-only devices, IPv4/IPv6 translators translate IPv6 and IPv4 packet headers according to the IP/ICMP Translation Algorithm defined in [RFC7915]. For example, 464XLAT ([RFC6877]) defines an architecture for providing IPv4 connectivity across an IPv6-only network. The solution contains two key elements: provider-side translator (PLAT) and customer-side translator (CLAT). CLAT implementations translate private IPv4 addresses to global IPv6 addresses, and vice versa, as defined in [RFC7915].¶
To map IPv4 addresses to IPv6 ones the translators use the translation prefix (either a well-known or a network-specific one, see [RFC6052]). The resulting IPv6 addresses can be statelessly translated back to IPv4. However it's not the case for an arbitrary global IPv6 addresses. Those addresses can only be translated to IPv4 by a stateful translators.¶
One of scenarios when it might be required but not currently possible is translating ICMPv6 error messages send by intermediate nodes to the CLAT address in the 464XLAT envinronment. The most typical example is a diagnistic tool like traceroute sending packets to an IPv4 destination from an IPv6-only host. Received ICMPv6 Time Exceeded are translated to ICMP Time Exceeded. If those packets were originated from an IPv4 address and translated to ICMPv6 by the PLAT (NAT64) device, then the source address of such packet can be mapped back to IPv4 by removing the translation prefix. However ICMPv6 error messages sent by devices located between the IPv6-only host and the NAT64 device have "native" IPv6 source addresses, which can not be mapped back to IPv4. Those packets are usually dropped and tools like traceroute can not represent IPv6 intermediate hops in any meaningful way. Such behaviour complicates troubleshooting. It's also confusing for users and increases operational costs, as users report packet loss in the network based on traceroute output.¶
Some CLAT implementations are known to workaround this issue by representing IPv6 addresses in IPv4 traceroute by using a reserved IPv4 address space and using the hop limit as the last octet, so an IPv6 device 5 hops away is shown as 225.0.0.5 etc.¶
It should be noted that the similar issue occurs in IPv6 Data Center Environments when an ICMPv6 error message needs to be sent to an IPv4-only client. As per Section 4.8 of [RFC7755], ICMPv6 error packets are usually dropped by the translator.¶
This document introduces a dedicated IPv4 address to translate arbitrary global IPv6 addresses in ICMPv6 error messages, and proposes an ICMP extension so original IPv6 address of an ICMPv6 error message can be included into IMCP message and therefore passed to an application.¶
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.¶
Whenever a translator translates an ICMPv4 ([RFC0792]) packet from an ICMPv6 ([RFC4443]) one, and the IPv6 source address in the outermost IPv6 header does not match the NAT64 prefix (and is therefore not mappable to an IPv4 address), the translator SHOULD:¶
Append an extension object ([RFC4884]) described in this document to the ICMPv6 packet before translation.¶
Perform stateful translation of the IPv6 source address to 192.0.0.11.¶
The translator SHOULD NOT use 192.0.0.11/32 to translate the source IPv6 address and SHOULD NOT add the extension if the packet IPv6 source address is an IPv4 address mapped to an IPv6 address using the translation prefix known to the translator, of if the translator has a stateful entry for the given IPv6 address mapped to another IPv4 address.¶
This document proposed allocating a dedicated IPv4 192.0.0.11/32 to use for statefully translating global IPv6 addresses in ICMPv6 error messages in scenarious, when the transator doesn't have any existing entry to use for translating such an address.¶
The suggested encoding to be appended[RFC4884] to ICMP messages is as follows:¶
0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Length = 20 | Class TBD1 | C-Type = 0 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | + + | original IPv6 source address | + 16 octets + | | + + | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
The Length, Class and C-Type fields are as defined in [RFC4884] and filled in accordingly. This document describes only one encoding and uses C-Type 0 for it. C-Type MUST be set to 0 when appending this extension. On receipt, this extension MUST be ignored if C-Type is not 0 or if Length is not 20.¶
The original IPv6 source address field is always 16 octets in length and filled as described in this document. It may contain any source address possibly seen in an ICMPv6 packet. This notably includes link-local addresses, the IPv6 loopback address, and mapped IPv4 addresses. Receivers MUST NOT reject addresses solely due to the address not being a globally scoped IPv6 addresses.¶
IPv6 Original Source Extension Object SHOULD be added when translating"¶
ICMPv6 Destination Unreachable to ICMPv4 Destination Unreachable¶
ICMPv6 Time Exceeded to ICMPv4 Time Exceeded.¶
and the IPv6 source address in the outermost IPv6 header does not match the NAT64 prefix.¶
IPv6 Original Source Extension Object MUST NOT be added in any other cases.¶
If the original ICMPv6 message does not contain an ICMP Extension Structure (as defined in Section 7 of [RFC4884]), the translator SHOULD append a new ICMP Extension Structure to the resulting ICMPv4 message. When adding the new Extension Structure, the translator MUST:¶
Create a new ICMP Extension Structure, containing one Extension Header and one IPv6 Original Source Extension object.¶
Append that Extension Structure to the original ICMPv6 message.¶
If the resulting packet size exceeds the minimum IPv6 MTU: truncate the embedded invoking packet by removing the trailing 24 octets (to accomodate for 4 octets of the extension header and 20 octets of the extension object).¶
Set the length field of the ICMPv6 message to the length of the padded "original datagram" field, measured in 32-bit words.¶
Translate the resulting ICMPv6 message to IPv6 as per [RFC7915].¶
If the original ICMPv6 message already contains an ICMP Extension Structure, the translator SHOULD append an IPv6 Original Source Extension object to that structure. When appending the object, the translator MUST:¶
Append an IPv6 Original Source Extension object to the Extension Structure.¶
Update the checksum field of the Extension Header accordingly.¶
If the resulting packet size exceeds the minimum IPv6 MTU: truncate the embedded invoking packet by removing the trailing 20 octets (to accomodate for 20 octets of the extension object) and update the length field of the ICMPv6 message¶
Translate the resulting ICMPv6 message to IPv6 as per [RFC7915].¶
The translator SHOULD use the dedicated IPv4 192.0.0.11/32 for translating IPv6 source addresses if all of the following conditions are met:¶
The packet being translated is one of the following:¶
the IPv6 source address in the outermost IPv6 header does not match the NAT64 prefix (and is therefore not mappable to an IPv4 address).¶
The translator does not have an explicit address mapping ([RFC7757]) configured for the given global IPv6 address to an IPv4 address .¶
it should be noted that the IPv6 Original Source Extension Object can not be added to ICMPv6 Packet Too Big messages (see Section 4.6 of [RFC4884]). At the same time it's highly desirable to ensure that translators translate those messages even if the IPv6 source address in the outermost IPv6 header does not match the NAT64 prefix (and is therefore not mappable to an IPv4 address), as those messages are crucial for Path MTU Discovery.¶
This document makes the following changes to Section 5.1 of [RFC7915]:¶
The text in RFC7915 is as follows:¶
Source Address: Mapped to an IPv4 address based on the algorithms presented in Section 6.¶
This document updates the text as follows:¶
Source Address: Mapped to an IPv4 address based on the algorithms presented in Section 6. When translating ICMPv4 error messages to ICMPv6 error messages and the valid IPv6 source address in the outermost IPv6 header does not match the prefix used in algorithmic mapping, the translator SHOULD follow the recommendations in draft-equinox-intarea-icmpext-xlat-source.¶
This document also adds the following paragraph before the very last paragraph of Section 5.2 of [RFC7915] (before "Error payload:"):¶
If valid IPv6 source address in the outermost IPv6 header of the ICMPv6 messages does not match the prefix used in algorithmic mapping, the translator SHOULD follow the recommendations in draft-equinox-intarea-icmpext-xlat-source.¶
TBD. Should probably be local-only.¶
This document requests that IANA allocates a "Class Value" from the "ICMP Extension Object Classes and Class Sub-types" registry created by [RFC4884] for use as described above. The following entry should be appended:¶
Class Value | Class Name | Reference |
---|---|---|
TBD1 | Original IPv6 Source Address | [THIS DOCUMENT] |
This document also requests that IANA allocates an IPv4 address 192.0.0.11/32) and updates the IPv4 Special-Purpose Address Registry available at (http://www.iana.org/assignments/iana-ipv4-special-registry/) with the following:¶
Address Block: 192.0.0.11/32¶
Name: [TBA IN THE NEXT VERSION OF THE DRAFT]¶
RFC: [THIS DOCUMENT]¶
Allocation Date: TBD3¶
Termination Date: N/A¶
Source: True¶
Destination: False¶
Forwardable: False¶
Global: False¶
Reserved-by-Protocol: False¶
This document is the result of discussions with Thomas Jensen. The authors would like to thank Darren Dukes, Bill Fenner for their feedback, comments and guidance.¶