Internet Engineering Task Force (IETF) M. Boucadair, Ed. Request for Comments: 8512 Orange Category: Standards Track S. Sivakumar ISSN: 2070-1721 Cisco Systems C. Jacquenet Orange S. Vinapamula Juniper Networks Q. Wu Huawei January 2019
A YANG Module for Network Address Translation (NAT) and Network Prefix Translation (NPT)
Abstract
This document defines a YANG module for the Network Address Translation (NAT) function.
Network Address Translation from IPv4 to IPv4 (NAT44), Network Address and Protocol Translation from IPv6 Clients to IPv4 Servers (NAT64), customer-side translator (CLAT), Stateless IP/ICMP Translation (SIIT), Explicit Address Mappings (EAM) for SIIT, IPv6-to-IPv6 Network Prefix Translation (NPTv6), and Destination NAT are covered in this document.
Status of This Memo
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/rfc8512.
Boucadair, et al. Standards Track [Page 1]
RFC 8512 YANG Module for NAT and NPT January 2019
Copyright Notice
Copyright (c) 2019 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 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.
This document defines a data model for Network Address Translation (NAT) and Network Prefix Translation (NPT) capabilities using the YANG data modeling language [RFC7950].
Traditional NAT is defined in [RFC2663], while Carrier Grade NAT (CGN) is defined in [RFC6888]. Unlike traditional NAT, the CGN is used to optimize the usage of global IP address space at the scale of a domain: a CGN is not managed by end users but by service providers instead. This document covers both traditional NATs and CGNs.
This document also covers NAT64 [RFC6146], customer-side translator (CLAT) [RFC6877], Stateless IP/ICMP Translation (SIIT) [RFC7915], Explicit Address Mappings (EAM) for SIIT [RFC7757], IPv6 Network Prefix Translation (NPTv6) [RFC6296], and Destination NAT. The full set of translation schemes that are in scope is included in Section 2.2.
Some examples are provided in Appendix A. These examples are not intended to be exhaustive.
o Basic Network Address Translation from IPv4 to IPv4 (NAT44): translation is limited to IP addresses alone (Section 2.1 of [RFC3022]).
o Network Address Port Translator (NAPT): translation in NAPT is extended to include IP addresses and transport identifiers (such as a TCP/UDP port or ICMP query ID); refer to Section 2.2 of [RFC3022]. A NAPT may use an extra identifier, in addition to the five transport tuples, to disambiguate bindings [RFC6619].
o Destination NAT: is a translation that acts on the destination IP address and/or destination port number. This flavor is usually deployed in load balancers or at devices in front of public servers.
o Port-restricted IPv4 address: an IPv4 address with a restricted port set. Multiple hosts may share the same IPv4 address; however, their port sets must not overlap [RFC7596].
Boucadair, et al. Standards Track [Page 4]
RFC 8512 YANG Module for NAT and NPT January 2019
o Restricted port set: a non-overlapping range of allowed external ports to use for NAT operation. Source ports of IPv4 packets translated by a NAT must belong to the assigned port set. The port set is used for all port-aware IP protocols [RFC7596].
o Internal host: a host that may need to use a translation capability to send to and receive traffic from the Internet.
o Internal address/prefix: the IP address/prefix of an internal host.
o External address: the IP address/prefix assigned by a translator to an internal host; this is the address that will be seen by a remote host on the Internet.
o Mapping: denotes a state at the translator that is necessary for network address and/or port translation.
o Dynamic implicit mapping: is created implicitly as a side effect of processing a packet (e.g., an initial TCP SYN packet) that requires a new mapping. A validity lifetime is associated with this mapping.
o Dynamic explicit mapping: is created as a result of an explicit request, e.g., a Port Control Protocol (PCP) message [RFC6887]. A validity lifetime is associated with this mapping.
o Static explicit mapping: is created using, e.g., a command-line interface (CLI). This mapping is likely to be maintained by the NAT function till an explicit action is executed to remove it.
The usage of the term NAT in this document refers to any translation flavor (NAT44, NAT64, etc.) indifferently.
This document uses the term "session" as defined in [RFC2663] and [RFC6146] for NAT64.
This document follows the guidelines of [RFC8407], uses the common YANG types defined in [RFC6991], and adopts the Network Management Datastore Architecture (NMDA). The meaning of the symbols in tree diagrams is defined in [RFC8340].
The NAT YANG module is designed to cover dynamic implicit mappings and static explicit mappings. The required functionality to instruct dynamic explicit mappings is defined in separate documents such as [YANG-PCP]. Considerations about instructing by explicit dynamic means (e.g., [RFC6887], [RFC6736], or [RFC8045]) are out of scope. As a reminder, REQ-9 of [RFC6888] requires that a CGN must implement a protocol giving subscribers explicit control over NAT mappings; that protocol should be the Port Control Protocol [RFC6887].
A single NAT device can have multiple NAT instances; each of these instances can be provided with its own policies (e.g., be responsible for serving a group of hosts). This document does not make any assumption about how internal hosts or flows are associated with a given NAT instance.
The NAT YANG module assumes that each NAT instance can be enabled/ disabled, be provisioned with a specific set of configuration data, and maintain its own mapping tables.
The NAT YANG module allows for a NAT instance to be provided with multiple NAT policies (/nat/instances/instance/policy). The document does not make any assumption about how flows are associated with a given NAT policy of a given NAT instance. Classification filters are out of scope.
Defining multiple NAT instances or configuring multiple NAT policies within one single NAT instance is implementation and deployment specific.
This YANG module does not provide any method to instruct a NAT function to enable the logging feature or to specify the information to be logged for administrative or regulatory reasons (Section 2.3 of [RFC6908] and REQ-12 of [RFC6888]). Those considerations are out of the scope of this document.
o Basic NAT44 o NAPT o Destination NAT o Port-restricted NAT o Stateful NAT64 (including with destination-based Pref64::/n [RFC7050]) o SIIT o CLAT o EAM o NPTv6 o Combination of Basic NAT/NAPT and Destination NAT o Combination of port-restricted and Destination NAT o Combination of NAT64 and EAM o Stateful and Stateless NAT64
[RFC8513] specifies an extension to the NAT YANG module to support Dual-Stack Lite (DS-Lite).
The YANG "feature" statement is used to indicate which of the different translation modes is relevant for a specific data node. Table 1 lists defined features:
The following translation modes do not require that dedicated features be defined:
o Port-restricted NAT: This mode corresponds to supplying port- restriction policies to a NAPT or NAT64 (port-set-restrict). o Combination of Basic NAT/NAPT and Destination NAT: This mode corresponds to setting 'dst-nat-enable' for Basic NAT44 or NAPT.
Boucadair, et al. Standards Track [Page 7]
RFC 8512 YANG Module for NAT and NPT January 2019
o Combination of port-restricted and Destination NAT: This mode can be achieved by configuring a NAPT with port restriction policies (port-set-restrict) together with a destination IP address pool (dst-ip-address-pool). o Combination of NAT64 and EAM: This mode corresponds to configuring static mappings for NAT64. o Stateful and stateless NAT64: A NAT64 implementation can be instructed to behave in the stateless mode for a given prefix by setting the parameter (nat64-prefixes/stateless-enable). A NAT64 implementation may behave in both stateful and stateless modes if, in addition to appropriately setting the parameter (nat64-prefixes/stateless-enable), an external IPv4 address pool is configured.
The NAT YANG module provides a method to retrieve the capabilities of a NAT instance (including a list of supported translation modes, a list of supported protocols, the supported NAT mapping types, the supported NAT filtering types, the behavior for handling fragments (all, out-of-order, in-order), and the support statuses for the following: port restriction, port range allocation, port parity preservation, and port preservation).
The module is structured to support protocols other than UDP, TCP, and ICMP. Concretely, the module allows the operator to enable translation for other transport protocols when required (/nat/instances/instance/policy/transport-protocols). Moreover, the mapping table is designed so that it can indicate any transport protocol. For example, this module may be used to manage a NAT capable of the Datagram Congestion Control Protocol (DCCP) that adheres to [RFC5597].
Future extensions may be needed to cover NAT-related considerations that are specific to other transport protocols such as the Stream Control Transmission Protocol (SCTP) [NAT-SUPP]. Typically, the mapping entry can be extended to record two optional SCTP-specific parameters: the Internal Verification Tag (Int-VTag) and External Verification Tag (Ext-VTag).
Boucadair, et al. Standards Track [Page 8]
RFC 8512 YANG Module for NAT and NPT January 2019
This document only specifies transport-protocol-specific timers for UDP, TCP, and ICMP. While some timers could potentially be generalized for other connection-oriented protocols, this document does not follow such an approach because there is no standard document specifying such generic behavior. Future documents may be edited to clarify how to reuse TCP-specific timers when needed.
The NAT YANG module assumes that blocks of IP external addresses (external-ip-address-pool) can be provisioned to the NAT function. These blocks may be contiguous or not.
This behavior is aligned with [RFC6888], which specifies that a NAT function should not have any limitations on the size or the contiguity of the external address pool. In particular, the NAT function must be configurable with contiguous or non-contiguous external IPv4 address ranges. To accommodate traditional NAT, the module allows for a single IP address to be configured for external- ip-address-pool.
Likewise, one or multiple IP address pools may be configured for Destination NAT (dst-ip-address-pool).
Port numbers can be assigned by a NAT individually (that is, a single port is assigned on a per-session basis), but this port allocation scheme may not be optimal for logging purposes (Section 12 of [RFC6269]). A NAT function should be able to assign port sets (e.g., [RFC7753]) to optimize the volume of the logging data (REQ-14 of [RFC6888]). Both allocation schemes are supported in the NAT YANG module.
When port-set assignment is activated (i.e., port-allocation- type==port-range-allocation), the NAT can be provided with the size of the port set to be assigned (port-set-size).
Some NATs restrict the source port numbers (e.g., Lightweight 4over6 [RFC7596] and Mapping of Address and Port with Encapsulation (MAP-E) [RFC7597]). Two schemes of port-set assignments (port-set-restrict) are supported in this document:
o Simple port range: is defined by two port values, the start and the end of the port range [RFC8045].
Boucadair, et al. Standards Track [Page 9]
RFC 8512 YANG Module for NAT and NPT January 2019
o Algorithmic: an algorithm is defined in [RFC7597] to characterize the set of ports that can be used.
As a reminder, all the ICMP Query messages contain an 'Identifier' field, which is referred to in this document as the 'ICMP Identifier'.
To cover TCP, UDP, and ICMP, the NAT YANG module assumes the following structure of a mapping entry:
type: Indicates how the mapping was instantiated. For example, it may indicate whether a mapping is dynamically instantiated by a packet or statically configured.
transport-protocol: Indicates the transport protocol (e.g., UDP, TCP, and ICMP) of a given mapping.
internal-src-address: Indicates the source IP address/prefix as used by an internal host.
internal-src-port: Indicates the source port number (or ICMP identifier) as used by an internal host.
external-src-address: Indicates the source IP address/prefix as assigned by the NAT.
external-src-port: Indicates the source port number (or ICMP identifier) as assigned by the NAT.
internal-dst-address: Indicates the destination IP address/prefix as used by an internal host when sending a packet to a remote host.
Boucadair, et al. Standards Track [Page 10]
RFC 8512 YANG Module for NAT and NPT January 2019
internal-dst-port: Indicates the destination port number as used by an internal host when sending a packet to a remote host.
external-dst-address: Indicates the destination IP address/prefix used by a NAT when processing a packet issued by an internal host towards a remote host.
external-dst-port: Indicates the destination port number used by a NAT when processing a packet issued by an internal host towards a remote host.
In order to cover both NAT64 and NAT44 flavors, the NAT mapping structure allows for the inclusion of an IPv4 or an IPv6 address as an internal IP address. Remaining fields are common to both NAT schemes.
For example, the mapping that will be created by a NAT64 upon receipt of a TCP SYN from source address 2001:db8:aaaa::1 and source port number 25636 to destination IP address 2001:db8:1234::198.51.100.1 and destination port number 8080 is shown in Table 2. This example assumes Endpoint-Dependent Mapping (EDM).
+-----------------------+-------------------------------------------+ | Mapping Entry | Value | | Attribute | | +-----------------------+-------------------------------------------+ | type | dynamic implicit mapping | | transport-protocol | 6 (TCP) | | internal-src-address | 2001:db8:aaaa::1 | | internal-src-port | 25636 | | external-src-address | T (an IPv4 address configured on the | | | NAT64) | | external-src-port | t (a port number that is chosen by the | | | NAT64) | | internal-dst-address | 2001:db8:1234::198.51.100.1 | | internal-dst-port | 8080 | | external-dst-address | 198.51.100.1 | | external-dst-port | 8080 | +-----------------------+-------------------------------------------+
Table 2: Example of an EDM NAT64 Mapping
Boucadair, et al. Standards Track [Page 11]
RFC 8512 YANG Module for NAT and NPT January 2019
The mappings that will be created by a NAT44 upon receipt of an ICMP request from source address 198.51.100.1 and ICMP identifier (ID1) to destination IP address 198.51.100.11 is depicted in Table 3. This example assumes Endpoint-Independent Mapping (EIM).
+----------------------+--------------------------------------------+ | Mapping-Entry | Value | | Attribute | | +----------------------+--------------------------------------------+ | type | dynamic implicit mapping | | transport-protocol | 1 (ICMP) | | internal-src-address | 198.51.100.1 | | internal-src-port | ID1 | | external-src-address | T (an IPv4 address configured on the | | | NAT44) | | external-src-port | ID2 (an ICMP identifier that is chosen by | | | the NAT44) | +----------------------+--------------------------------------------+
Table 3: Example of an EIM NAT44 Mapping Entry
The mapping that will be created by a NAT64 (EIM mode) upon receipt of an ICMP request from source address 2001:db8:aaaa::1 and ICMP identifier (ID1) to destination IP address 2001:db8:1234::198.51.100.1 is shown in Table 4.
+----------------------+--------------------------------------------+ | Mapping-Entry | Value | | Attribute | | +----------------------+--------------------------------------------+ | type | dynamic implicit mapping | | transport-protocol | 58 (ICMPv6) | | internal-src-address | 2001:db8:aaaa::1 | | internal-src-port | ID1 | | external-src-address | T (an IPv4 address configured on the | | | NAT64) | | external-src-port | ID2 (an ICMP identifier that is chosen by | | | the NAT64) | +----------------------+--------------------------------------------+
Table 4: Example of an EIM NAT64 Mapping Entry
Boucadair, et al. Standards Track [Page 12]
RFC 8512 YANG Module for NAT and NPT January 2019
Note that a mapping table is maintained only for stateful NAT functions. Particularly:
o No mapping table is maintained for NPTv6 given that it is stateless and transport-agnostic.
o The double translations are stateless in CLAT if a dedicated IPv6 prefix is provided for CLAT. If not, a stateful NAT44 will be required.
o No per-flow mapping is maintained for EAM [RFC7757].
o No mapping table is maintained for Stateless IPv4/IPv6 translation. As a reminder, in such deployments, internal IPv6 nodes are addressed using IPv4-translatable IPv6 addresses, which enable them to be accessed by IPv4 nodes [RFC6052].
In order to comply with CGN deployments in particular, the NAT YANG module allows limiting the number of external ports per subscriber (port-quota) and the amount of state memory allocated per mapping and per subscriber (mapping-limits and connection-limits). According to [RFC6888], the module is designed to allow for the following:
o Per-subscriber limits are configurable by the NAT administrator.
o Per-subscriber limits are configurable independently per the transport protocol.
o Administrator-adjustable thresholds to prevent a single subscriber from consuming excessive CPU resources from the NAT (e.g., rate- limit the subscriber's creation of new mappings) can be configured.
Boucadair, et al. Standards Track [Page 13]
RFC 8512 YANG Module for NAT and NPT January 2019
Table 5 lists the various limits that can be set using the NAT YANG module. Once a limit is reached, packets that would normally trigger new port mappings or be translated because they match existing mappings, are dropped by the translator.
+-------------------+-----------------------------------------------+ | Limit | Description | +-------------------+-----------------------------------------------+ | port-quota | Specifies a port quota to be assigned per | | | subscriber. It corresponds to the maximum | | | number of ports to be used by a subscriber. | | | The port quota can be configured to apply to | | | all protocols or to a specific protocol. | | | Distinct port quota may be configured per | | | protocol. | +-------------------+-----------------------------------------------+ | fragments-limit | In order to prevent denial-of-service (DoS) | | | attacks that can be caused by fragments, this | | | parameter is used to limit the number of out- | | | of-order fragments that can be handled by a | | | translator. | +-------------------+-----------------------------------------------+ | mapping-limits | This parameter can be used to control the | | | maximum number of subscribers that can be | | | serviced by a NAT instance (limit-subscriber) | | | and the maximum number of address and/or port | | | mappings that can be maintained by a NAT | | | instance (limit-address-mappings and limit- | | | port-mappings). Also, limits specific to | | | protocols (e.g., TCP, UDP, ICMP) can also be | | | specified (limit-per-protocol). | +-------------------+-----------------------------------------------+ | connection-limits | In order to prevent exhausting the resources | | | of a NAT implementation and to ensure | | | fairness usage among subscribers, various | | | rate limits can be specified. Rate-limiting | | | can be enforced per subscriber (limit- | | | subscriber), per NAT instance (limit-per- | | | instance), and/or be specified for each | | | supported protocol (limit-per-protocol). | +-------------------+-----------------------------------------------+
Table 5: NAT Limits
Boucadair, et al. Standards Track [Page 14]
RFC 8512 YANG Module for NAT and NPT January 2019
Table 6 describes limits that, once exceeded, will trigger notifications to be generated:
+--------------------------+----------------------------------------+ | Notification Threshold | Description | +--------------------------+----------------------------------------+ | high-threshold | Used to notify high address | | | utilization of a given pool. When | | | exceeded, a nat-pool-event | | | notification will be generated. | +--------------------------+----------------------------------------+ | low-threshold | Used to notify low address utilization | | | of a given pool. An administrator is | | | supposed to configure low-threshold so | | | that it can reflect an abnormal usage | | | of NAT resources. When exceeded, a | | | nat-pool-event notification will be | | | generated. | +--------------------------+----------------------------------------+ | notify-addresses-usage | Used to notify high address | | | utilization of all pools configured to | | | a NAT instance. When exceeded, a nat- | | | instance-event will be generated. | +--------------------------+----------------------------------------+ | notify-ports-usage | Used to notify high port allocation | | | taking into account all pools | | | configured to a NAT instance. When | | | exceeded, a nat-instance-event | | | notification will be generated. | +--------------------------+----------------------------------------+ | notify-subscribers-limit | Used to notify a high number of active | | | subscribers that are serviced by a NAT | | | instance. When exceeded, a nat- | | | instance-event notification will be | | | generated. | +--------------------------+----------------------------------------+
Table 6: Notification Thresholds
Boucadair, et al. Standards Track [Page 15]
RFC 8512 YANG Module for NAT and NPT January 2019
In order to prevent a NAT implementation from generating frequent notifications, the NAT YANG module supports the following limits (Table 7) used to control how frequent notifications can be generated. That is, notifications are subject to rate-limiting imposed by these intervals.
+-------------------------------------+-----------------------------+ | Interval | Description | +-------------------------------------+-----------------------------+ | notify-pool-usage/notify-interval | Indicates the minimum | | | number of seconds between | | | successive notifications | | | for a given address pool. | +-------------------------------------+-----------------------------+ | notification-limits/notify-interval | Indicates the minimum | | | number of seconds between | | | successive notifications | | | for a NAT instance. | +-------------------------------------+-----------------------------+
Table 7: Notification Intervals
2.10. Binding the NAT Function to an External Interface
The module is designed to specify an external realm on which the NAT function must be applied (external-realm). The module supports indicating an interface as an external realm [RFC8343], but the module is extensible so that other choices can be indicated in the future (e.g., Virtual Routing and Forwarding (VRF) instance).
Distinct external realms can be provided as a function of the NAT policy (see, for example, Section 4 of [RFC7289]).
If no external realm is provided, this assumes that the system is able to determine the external interface (VRF instance, etc.) on which the NAT will be applied. Typically, the WAN and LAN interfaces of Customer Premises Equipment (CPE) are determined by the CPE.
Section of 5.1 of [RFC7659] indicates that the NATV2-MIB assumes that the following information is configured on the NAT by some means, which is not specified in [RFC7659]:
o The set of address realms to which the device connects.
Boucadair, et al. Standards Track [Page 16]
RFC 8512 YANG Module for NAT and NPT January 2019
o For the CGN case, per-subscriber information including the subscriber index, address realm, assigned prefix or address, and (possibly) policies regarding address pool selection in the various possible address realms to which the subscriber may connect.
o The set of NAT instances running on the device, identified by NAT instance index and name.
o The port mapping, filtering, pooling, and fragment behaviors for each NAT instance.
o The set of protocols supported by each NAT instance.
o Address pools for each NAT instance, including for each pool the pool index, address realm, and minimum and maximum port numbers.
o Static address and port mapping entries.
All the above parameters can be configured by means of the NAT YANG module.
Unlike the NATV2-MIB, the NAT YANG module allows the configuration of multiple policies per NAT instance.
import ietf-inet-types { prefix inet; reference "Section 4 of RFC 6991"; } import ietf-yang-types { prefix yang; reference "Section 3 of RFC 6991"; } import ietf-interfaces { prefix if; reference "RFC 8343: A YANG Data Model for Interface Management"; }
organization "IETF OPSAWG (Operations and Management Area Working Group)"; contact "WG Web: <https://datatracker.ietf.org/wg/opsawg/> WG List: <mailto:opsawg@ietf.org>
description "This module is a YANG module for NAT implementations.
Boucadair, et al. Standards Track [Page 24]
RFC 8512 YANG Module for NAT and NPT January 2019
NAT44, Network Address and Protocol Translation from IPv6 Clients to IPv4 Servers (NAT64), customer-side translator (CLAT), Stateless IP/ICMP Translation (SIIT), Explicit Address Mappings (EAM) for SIIT, IPv6 Network Prefix Translation (NPTv6), and Destination NAT are covered.
Copyright (c) 2018 IETF Trust and the persons identified as authors of the code. All rights reserved.
Redistribution and use in source and binary forms, with or without modification, is permitted pursuant to, and subject to the license terms contained in, the Simplified BSD License set forth in Section 4.c of the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info).
This version of this YANG module is part of RFC 8512; see the RFC itself for full legal notices.";
revision 2019-01-10 { description "Initial revision."; reference "RFC 8512: A YANG Module for Network Address Translation (NAT) and Network Prefix Translation (NPT)"; }
/* * Definitions */
typedef percent { type uint8 { range "0 .. 100"; } description "Percentage"; }
/* * Features */
feature basic-nat44 { description "Basic NAT44 translation is limited to IP addresses alone."; reference "RFC 3022: Traditional IP Network Address Translator
Boucadair, et al. Standards Track [Page 25]
RFC 8512 YANG Module for NAT and NPT January 2019
(Traditional NAT)"; }
feature napt44 { description "Network Address Port Translator (NAPT): translation is extended to include IP addresses and transport identifiers (such as a TCP/UDP port or ICMP query ID).
If the internal IP address is not sufficient to uniquely disambiguate NAPT44 mappings, an additional attribute is required. For example, that additional attribute may be an IPv6 address (a.k.a., DS-Lite) or a Layer 2 identifier (a.k.a., Per-Interface NAT)"; reference "RFC 3022: Traditional IP Network Address Translator (Traditional NAT)"; }
feature dst-nat { description "Destination NAT is a translation that acts on the destination IP address and/or destination port number. This flavor is usually deployed in load balancers or at devices in front of public servers."; }
feature nat64 { description "NAT64 translation allows IPv6-only clients to contact IPv4 servers using, e.g., UDP, TCP, or ICMP. One or more public IPv4 addresses assigned to a NAT64 translator are shared among several IPv6-only clients."; reference "RFC 6146: Stateful NAT64: Network Address and Protocol Translation from IPv6 Clients to IPv4 Servers"; }
feature siit { description "The Stateless IP/ICMP Translation Algorithm (SIIT), which translates between IPv4 and IPv6 packet headers (including ICMP headers).
In the stateless mode, an IP/ICMP translator converts IPv4 addresses to IPv6, and vice versa, solely based on the configuration of the stateless IP/ICMP translator and information contained within the packet being translated.
Boucadair, et al. Standards Track [Page 26]
RFC 8512 YANG Module for NAT and NPT January 2019
The translator must support the stateless address mapping algorithm defined in RFC 6052, which is the default behavior."; reference "RFC 7915: IP/ICMP Translation Algorithm"; }
feature clat { description "CLAT is customer-side translator that algorithmically translates 1:1 private IPv4 addresses to global IPv6 addresses, and vice versa.
When a dedicated /64 prefix is not available for translation from DHCPv6-PD, the CLAT may perform NAT44 for all IPv4 LAN packets so that all the LAN-originated IPv4 packets appear from a single IPv4 address and are then statelessly translated to one interface IPv6 address that is claimed by the CLAT via the Neighbor Discovery Protocol (NDP) and defended with Duplicate Address Detection."; reference "RFC 6877: 464XLAT: Combination of Stateful and Stateless Translation"; }
feature eam { description "Explicit Address Mapping (EAM) is a bidirectional coupling between an IPv4 prefix and an IPv6 prefix."; reference "RFC 7757: Explicit Address Mappings for Stateless IP/ICMP Translation"; }
feature nptv6 { description "NPTv6 is a stateless transport-agnostic IPv6-to-IPv6 prefix translation."; reference "RFC 6296: IPv6-to-IPv6 Network Prefix Translation"; }
/* * Identities */
identity nat-type { description "Base identity for nat type.";
Boucadair, et al. Standards Track [Page 27]
RFC 8512 YANG Module for NAT and NPT January 2019
}
identity basic-nat44 { base nat:nat-type; description "Identity for Basic NAT support."; reference "RFC 3022: Traditional IP Network Address Translator (Traditional NAT)"; }
identity napt44 { base nat:nat-type; description "Identity for NAPT support."; reference "RFC 3022: Traditional IP Network Address Translator (Traditional NAT)"; }
identity dst-nat { base nat:nat-type; description "Identity for Destination NAT support."; }
identity nat64 { base nat:nat-type; description "Identity for NAT64 support."; reference "RFC 6146: Stateful NAT64: Network Address and Protocol Translation from IPv6 Clients to IPv4 Servers"; }
identity siit { base nat:nat-type; description "Identity for SIIT support."; reference "RFC 7915: IP/ICMP Translation Algorithm"; }
identity clat { base nat:nat-type; description "Identity for CLAT support."; reference
Boucadair, et al. Standards Track [Page 28]
RFC 8512 YANG Module for NAT and NPT January 2019
"RFC 6877: 464XLAT: Combination of Stateful and Stateless Translation"; }
identity eam { base nat:nat-type; description "Identity for EAM support."; reference "RFC 7757: Explicit Address Mappings for Stateless IP/ICMP Translation"; }
identity nptv6 { base nat:nat-type; description "Identity for NPTv6 support."; reference "RFC 6296: IPv6-to-IPv6 Network Prefix Translation"; }
/* * Grouping */
grouping port-number { description "An individual port number or a range of ports. When only start-port-number is present, it represents a single port number."; leaf start-port-number { type inet:port-number; description "Beginning of the port range."; reference "Section 3.2.9 of RFC 8045"; } leaf end-port-number { type inet:port-number; must '. >= ../start-port-number' { error-message "The end-port-number must be greater than or equal to start-port-number."; } description "End of the port range."; reference "Section 3.2.10 of RFC 8045";
Boucadair, et al. Standards Track [Page 29]
RFC 8512 YANG Module for NAT and NPT January 2019
} }
grouping port-set { description "Indicates a set of port numbers.
It may be a simple port range, or use the Port Set Identifier (PSID) algorithm to represent a range of transport-layer port numbers that will be used by a NAPT."; choice port-type { default "port-range"; description "Port type: port-range or port-set-algo."; case port-range { uses port-number; } case port-set-algo { leaf psid-offset { type uint8 { range "0..15"; } description "The number of offset bits (a.k.a., 'a' bits). Specifies the numeric value for the excluded port range/offset bits.
Allowed values are between 0 and 15."; reference "Section 5.1 of RFC 7597"; } leaf psid-len { type uint8 { range "0..15"; } mandatory true; description "The length of PSID, representing the sharing ratio for an IPv4 address.
(also known as 'k').
The address-sharing ratio would be 2^k."; reference "Section 5.1 of RFC 7597"; } leaf psid {
Boucadair, et al. Standards Track [Page 30]
RFC 8512 YANG Module for NAT and NPT January 2019
type uint16; mandatory true; description "PSID value, which identifies a set of ports algorithmically."; reference "Section 5.1 of RFC 7597"; } } reference "RFC 7597: Mapping of Address and Port with Encapsulation (MAP-E)"; } }
If an attribute is not stored in the mapping/session table, it means the corresponding field of a packet that matches this entry is not rewritten by the NAT or this information is not required for NAT filtering purposes."; leaf index { type uint32; description "A unique identifier of a mapping entry. This identifier can be automatically assigned by the NAT instance or be explicitly configured."; } leaf type { type enumeration { enum static { description "The mapping entry is explicitly configured (e.g., via a command-line interface)."; } enum dynamic-implicit { description "This mapping is created implicitly as a side effect of processing a packet that requires a new mapping."; } enum dynamic-explicit { description "This mapping is created as a result of an explicit request, e.g., a PCP message."; } }
Boucadair, et al. Standards Track [Page 31]
RFC 8512 YANG Module for NAT and NPT January 2019
description "Indicates the type of a mapping entry. For example, a mapping can be: static, implicit dynamic, or explicit dynamic."; } leaf transport-protocol { type uint8; description "The upper-layer protocol associated with this mapping. Values are taken from the IANA Protocol Numbers registry: <https://www.iana.org/assignments/protocol-numbers/>.
For example, this field contains 6 for TCP, 17 for UDP, 33 for DCCP, or 132 for SCTP.
If this leaf is not instantiated, then the mapping applies to any protocol."; } leaf internal-src-address { type inet:ip-prefix; description "Corresponds to the source IPv4/IPv6 address/prefix of the packet received on an internal interface."; } container internal-src-port { description "Corresponds to the source port of the packet received on an internal interface.
It is also used to indicate the internal source ICMP identifier.
As a reminder, all the ICMP Query messages contain an 'Identifier' field, which is referred to in this document as the 'ICMP Identifier'."; uses port-number; } leaf external-src-address { type inet:ip-prefix; description "Source IP address/prefix of the packet sent on an external interface of the NAT."; } container external-src-port { description "Source port of the packet sent on an external interface of the NAT.
Boucadair, et al. Standards Track [Page 32]
RFC 8512 YANG Module for NAT and NPT January 2019
It is also used to indicate the external source ICMP identifier."; uses port-number; } leaf internal-dst-address { type inet:ip-prefix; description "Corresponds to the destination IP address/prefix of the packet received on an internal interface of the NAT.
For example, some NAT implementations support the translation of both source and destination addresses and port numbers, sometimes referred to as 'Twice NAT'."; } container internal-dst-port { description "Corresponds to the destination port of the IP packet received on the internal interface.
It is also used to include the internal destination ICMP identifier."; uses port-number; } leaf external-dst-address { type inet:ip-prefix; description "Corresponds to the destination IP address/prefix of the packet sent on an external interface of the NAT."; } container external-dst-port { description "Corresponds to the destination port number of the packet sent on the external interface of the NAT.
It is also used to include the external destination ICMP identifier."; uses port-number; } leaf lifetime { type uint32; units "seconds"; description "When specified, it is used to track the connection that is fully formed (e.g., once the three-way handshake
Boucadair, et al. Standards Track [Page 33]
RFC 8512 YANG Module for NAT and NPT January 2019
TCP is completed) or the duration for maintaining an explicit mapping alive. The mapping entry will be removed by the NAT instance once this lifetime is expired.
When reported in a get operation, the lifetime indicates the remaining validity lifetime.
Static mappings may not be associated with a lifetime. If no lifetime is associated with a static mapping, an explicit action is required to remove that mapping."; } }
/* * NAT Module */
container nat { description "NAT module"; container instances { description "NAT instances"; list instance { key "id"; description "A NAT instance. This identifier can be automatically assigned or explicitly configured."; leaf id { type uint32; must '. >= 1'; description "NAT instance identifier.
The identifier must be greater than zero."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)"; } leaf name { type string; description "A name associated with the NAT instance."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)"; }
Boucadair, et al. Standards Track [Page 34]
RFC 8512 YANG Module for NAT and NPT January 2019
leaf enable { type boolean; description "Status of the NAT instance."; } container capabilities { config false; description "NAT capabilities."; leaf-list nat-flavor { type identityref { base nat-type; } description "Supported translation type(s)."; } leaf-list per-interface-binding { type enumeration { enum unsupported { description "No capability to associate a NAT binding with an extra identifier."; } enum layer-2 { description "The NAT instance is able to associate a mapping with a Layer 2 identifier."; } enum dslite { description "The NAT instance is able to associate a mapping with an IPv6 address (a.k.a., DS-Lite)."; } } description "Indicates the capability of a NAT to associate a particular NAT session not only with the five tuples used for the transport connection on both sides of the NAT but also with the internal interface on which the user device is connected to the NAT."; reference "Section 4 of RFC 6619"; } list transport-protocols { key "protocol-id"; description "List of supported protocols.";
Boucadair, et al. Standards Track [Page 35]
RFC 8512 YANG Module for NAT and NPT January 2019
leaf protocol-id { type uint8; mandatory true; description "The upper-layer protocol associated with a mapping.
Values are taken from the IANA Protocol Numbers registry.
For example, this field contains 6 for TCP, 17 for UDP, 33 for DCCP, or 132 for SCTP."; } leaf protocol-name { type string; description "The name of the upper-layer protocol associated with this mapping.
For example, TCP, UDP, DCCP, and SCTP."; } } leaf restricted-port-support { type boolean; description "Indicates source port NAT restriction support."; reference "RFC 7596: Lightweight 4over6: An Extension to the Dual-Stack Lite Architecture"; } leaf static-mapping-support { type boolean; description "Indicates whether static mappings are supported."; } leaf port-randomization-support { type boolean; description "Indicates whether port randomization is supported."; reference "Section 4.2.1 of RFC 4787"; } leaf port-range-allocation-support { type boolean; description "Indicates whether port range allocation is supported."; reference "Section 1.1 of RFC 7753"; }
Boucadair, et al. Standards Track [Page 36]
RFC 8512 YANG Module for NAT and NPT January 2019
leaf port-preservation-suport { type boolean; description "Indicates whether port preservation is supported."; reference "Section 4.2.1 of RFC 4787"; } leaf port-parity-preservation-support { type boolean; description "Indicates whether port parity preservation is supported."; reference "Section 8 of RFC 7857"; } leaf address-roundrobin-support { type boolean; description "Indicates whether address allocation round robin is supported."; } leaf paired-address-pooling-support { type boolean; description "Indicates whether paired-address-pooling is supported"; reference "REQ-2 of RFC 4787"; } leaf endpoint-independent-mapping-support { type boolean; description "Indicates whether endpoint-independent- mapping is supported."; reference "Section 4 of RFC 4787"; } leaf address-dependent-mapping-support { type boolean; description "Indicates whether address-dependent-mapping is supported."; reference "Section 4 of RFC 4787"; } leaf address-and-port-dependent-mapping-support { type boolean; description
Boucadair, et al. Standards Track [Page 37]
RFC 8512 YANG Module for NAT and NPT January 2019
"Indicates whether address-and-port-dependent-mapping is supported."; reference "Section 4 of RFC 4787"; } leaf endpoint-independent-filtering-support { type boolean; description "Indicates whether endpoint-independent-filtering is supported."; reference "Section 5 of RFC 4787"; } leaf address-dependent-filtering { type boolean; description "Indicates whether address-dependent-filtering is supported."; reference "Section 5 of RFC 4787"; } leaf address-and-port-dependent-filtering { type boolean; description "Indicates whether address-and-port-dependent is supported."; reference "Section 5 of RFC 4787"; } leaf fragment-behavior { type enumeration { enum unsupported { description "No capability to translate incoming fragments. All received fragments are dropped."; } enum in-order { description "The NAT instance is able to translate fragments only if they are received in order. That is, in particular the header is in the first packet. Fragments received out of order are dropped. "; } enum out-of-order { description "The NAT instance is able to translate a fragment even if it is received out of order.
Boucadair, et al. Standards Track [Page 38]
RFC 8512 YANG Module for NAT and NPT January 2019
This behavior is recommended."; reference "REQ-14 of RFC 4787"; } } description "The fragment behavior is the NAT instance's capability to translate fragments received on the external interface of the NAT."; } } leaf type { type identityref { base nat-type; } description "Specify the translation type. Particularly useful when multiple translation flavors are supported.
If one type is supported by a NAT, this parameter is by default set to that type."; } leaf per-interface-binding { type enumeration { enum disabled { description "Disable the capability to associate an extra identifier with NAT mappings."; } enum layer-2 { description "The NAT instance is able to associate a mapping with a Layer 2 identifier."; } enum dslite { description "The NAT instance is able to associate a mapping with an IPv6 address (a.k.a., DS-Lite)."; } } description "A NAT that associates a particular NAT session not only with the five tuples used for the transport connection on both sides of the NAT but also with the internal interface on which the user device is connected to the NAT.
If supported, this mode of operation should be
Boucadair, et al. Standards Track [Page 39]
RFC 8512 YANG Module for NAT and NPT January 2019
configurable, and it should be disabled by default in general-purpose NAT devices. If one single per-interface binding behavior is supported by a NAT, this parameter is by default set to that behavior."; reference "Section 4 of RFC 6619"; } list nat-pass-through { if-feature "basic-nat44 or napt44 or dst-nat"; key "id"; description "IP prefix NAT pass-through."; leaf id { type uint32; description "An identifier of the IP prefix pass-through."; } leaf prefix { type inet:ip-prefix; mandatory true; description "The IP addresses that match should not be translated.
It must be possible to administratively turn off translation for specific destination addresses and/or ports."; reference "REQ-6 of RFC 6888"; } leaf port { type inet:port-number; description "It must be possible to administratively turn off translation for specific destination addresses and/or ports.
If no prefix is defined, the NAT pass-through bound to a given port applies for any destination address."; reference "REQ-6 of RFC 6888"; } } list policy { key "id"; description "NAT parameters for a given instance"; leaf id {
Boucadair, et al. Standards Track [Page 40]
RFC 8512 YANG Module for NAT and NPT January 2019
type uint32; description "An identifier of the NAT policy. It must be unique within the NAT instance."; } container clat-parameters { if-feature "clat"; description "CLAT parameters."; list clat-ipv6-prefixes { key "ipv6-prefix"; description "464XLAT double-translation treatment is stateless when a dedicated /64 is available for translation on the CLAT. Otherwise, the CLAT will have both stateful and stateless translation since it requires NAT44 from the LAN to a single IPv4 address and then stateless translation to a single IPv6 address."; reference "RFC 6877: 464XLAT: Combination of Stateful and Stateless Translation"; leaf ipv6-prefix { type inet:ipv6-prefix; description "An IPv6 prefix used for CLAT."; } } list ipv4-prefixes { key "ipv4-prefix"; description "Pool of IPv4 addresses used for CLAT. 192.0.0.0/29 is the IPv4 service continuity prefix."; reference "RFC 7335: IPv4 Service Continuity Prefix"; leaf ipv4-prefix { type inet:ipv4-prefix; description "464XLAT double-translation treatment is stateless when a dedicated /64 is available for translation on the CLAT. Otherwise, the CLAT will have both stateful and stateless translation since it requires NAT44 from the LAN to a single IPv4 address and then stateless translation to a single IPv6 address. The CLAT performs NAT44 for all IPv4 LAN packets so that all the LAN-originated IPv4 packets appear from a single IPv4 address and are then statelessly translated to one
Boucadair, et al. Standards Track [Page 41]
RFC 8512 YANG Module for NAT and NPT January 2019
interface IPv6 address that is claimed by the CLAT.
An IPv4 address from this pool is also provided to an application that makes use of literals."; reference "RFC 6877: 464XLAT: Combination of Stateful and Stateless Translation"; } } } list nptv6-prefixes { if-feature "nptv6"; key "internal-ipv6-prefix"; description "Provides one or a list of (internal IPv6 prefix, external IPv6 prefix) required for NPTv6.
In its simplest form, NPTv6 interconnects two network links: one is an 'internal' network link attached to a leaf network within a single administrative domain, and the other is an 'external' network with connectivity to the global Internet."; reference "RFC 6296: IPv6-to-IPv6 Network Prefix Translation"; leaf internal-ipv6-prefix { type inet:ipv6-prefix; mandatory true; description "An IPv6 prefix used by an internal interface of NPTv6."; reference "RFC 6296: IPv6-to-IPv6 Network Prefix Translation"; } leaf external-ipv6-prefix { type inet:ipv6-prefix; mandatory true; description "An IPv6 prefix used by the external interface of NPTv6."; reference "RFC 6296: IPv6-to-IPv6 Network Prefix Translation"; } } list eam { if-feature "eam";
Boucadair, et al. Standards Track [Page 42]
RFC 8512 YANG Module for NAT and NPT January 2019
key "ipv4-prefix"; description "The Explicit Address Mapping Table is a conceptual table in which each row represents an EAM.
Each EAM describes a mapping between IPv4 and IPv6 prefixes/addresses."; reference "Section 3.1 of RFC 7757"; leaf ipv4-prefix { type inet:ipv4-prefix; mandatory true; description "The IPv4 prefix of an EAM."; reference "Section 3.2 of RFC 7757"; } leaf ipv6-prefix { type inet:ipv6-prefix; mandatory true; description "The IPv6 prefix of an EAM."; reference "Section 3.2 of RFC 7757"; } } list nat64-prefixes { if-feature "siit or nat64 or clat"; key "nat64-prefix"; description "Provides one or a list of NAT64 prefixes with or without a list of destination IPv4 prefixes. It allows mapping IPv4 address ranges to IPv6 prefixes. For example: 192.0.2.0/24 is mapped to 2001:db8:122:300::/56. 198.51.100.0/24 is mapped to 2001:db8:122::/48."; reference "Section 5.1 of RFC 7050"; leaf nat64-prefix { type inet:ipv6-prefix; mandatory true; description "A NAT64 prefix. Can be a Network-Specific Prefix (NSP) or a Well-Known Prefix (WKP).
Organizations deploying stateless IPv4/IPv6 translation should assign an NSP to their IPv4/IPv6 translation service.
Boucadair, et al. Standards Track [Page 43]
RFC 8512 YANG Module for NAT and NPT January 2019
For stateless NAT64, IPv4-translatable IPv6 addresses must use the selected NSP.
Both IPv4-translatable IPv6 addresses and IPv4-converted IPv6 addresses should use the same prefix."; reference "Sections 3.3 and 3.4 of RFC 6052"; } list destination-ipv4-prefix { key "ipv4-prefix"; description "An IPv4 prefix/address."; leaf ipv4-prefix { type inet:ipv4-prefix; description "An IPv4 address/prefix."; } } leaf stateless-enable { type boolean; default "false"; description "Enable explicitly stateless NAT64."; } } list external-ip-address-pool { if-feature "basic-nat44 or napt44 or nat64"; key "pool-id"; description "Pool of external IP addresses used to service internal hosts.
A pool is a set of IP prefixes."; leaf pool-id { type uint32; must '. >= 1'; description "An identifier that uniquely identifies the address pool within a NAT instance.
The identifier must be greater than zero."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)"; } leaf external-ip-pool { type inet:ipv4-prefix;
Boucadair, et al. Standards Track [Page 44]
RFC 8512 YANG Module for NAT and NPT January 2019
mandatory true; description "An IPv4 prefix used for NAT purposes."; } } container port-set-restrict { if-feature "napt44 or nat64"; description "Configures contiguous and non-contiguous port ranges.
The port set is used to restrict the external source port numbers used by the translator."; uses port-set; } leaf dst-nat-enable { if-feature "basic-nat44 or napt44"; type boolean; default "false"; description "Enable/disable Destination NAT.
A NAT44 may be configured to enable Destination NAT, too."; } list dst-ip-address-pool { if-feature "dst-nat"; key "pool-id"; description "Pool of IP addresses used for Destination NAT."; leaf pool-id { type uint32; description "An identifier of the address pool."; } leaf dst-in-ip-pool { type inet:ip-prefix; description "Is used to identify an internal destination IP prefix/address to be translated."; } leaf dst-out-ip-pool { type inet:ip-prefix; mandatory true; description "IP address/prefix used for Destination NAT."; } } list transport-protocols {
Boucadair, et al. Standards Track [Page 45]
RFC 8512 YANG Module for NAT and NPT January 2019
if-feature "napt44 or nat64 or dst-nat"; key "protocol-id"; description "Configure the transport protocols to be handled by the translator.
TCP and UDP are supported by default."; leaf protocol-id { type uint8; mandatory true; description "The upper-layer protocol associated with this mapping.
Values are taken from the IANA Protocol Numbers registry.
For example, this field contains 6 for TCP, 17 for UDP, 33 for DCCP, or 132 for SCTP."; } leaf protocol-name { type string; description "The name of the upper-layer protocol associated with this mapping.
For example, TCP, UDP, DCCP, and SCTP."; } } leaf subscriber-mask-v6 { type uint8 { range "0 .. 128"; } description "The subscriber mask is an integer that indicates the length of significant bits to be applied on the source IPv6 address (internal side) to unambiguously identify a user device (e.g., CPE).
Subscriber mask is a system-wide configuration parameter that is used to enforce generic per-subscriber policies (e.g., port-quota).
The enforcement of these generic policies does not require the configuration of every subscriber's prefix.
Example: suppose the 2001:db8:100:100::/56 prefix
Boucadair, et al. Standards Track [Page 46]
RFC 8512 YANG Module for NAT and NPT January 2019
is assigned to a NAT64-serviced CPE. Suppose also that 2001:db8:100:100::1 is the IPv6 address used by the client that resides in that CPE. When the NAT64 receives a packet from this client, it applies the subscriber-mask-v6 (e.g., 56) on the source IPv6 address to compute the associated prefix for this client (2001:db8:100:100::/56). Then, the NAT64 enforces policies based on that prefix (2001:db8:100:100::/56), not on the exact source IPv6 address."; } list subscriber-match { if-feature "basic-nat44 or napt44 or dst-nat"; key "match-id"; description "IP prefix match. A subscriber is identified by a subnet."; leaf match-id { type uint32; description "An identifier of the subscriber match."; } leaf subnet { type inet:ip-prefix; mandatory true; description "The IP address subnets that match should be translated. For example, all addresses that belong to the 192.0.2.0/24 prefix must be processed by the NAT."; } } leaf address-allocation-type { type enumeration { enum arbitrary { if-feature "basic-nat44 or napt44 or nat64"; description "Arbitrary pooling behavior means that the NAT instance may create the new port mapping using any address in the pool that has a free port for the protocol concerned."; } enum roundrobin { if-feature "basic-nat44 or napt44 or nat64"; description "Round-robin allocation."; } enum paired {
Boucadair, et al. Standards Track [Page 47]
RFC 8512 YANG Module for NAT and NPT January 2019
if-feature "napt44 or nat64"; description "Paired address pooling informs the NAT that all the flows from an internal IP address must be assigned the same external address. This is the recommended behavior for NAPT/NAT64."; reference "RFC 4787: Network Address Translation (NAT) Behavioral Requirements for Unicast UDP"; } } description "Specifies how external IP addresses are allocated."; } leaf port-allocation-type { if-feature "napt44 or nat64"; type enumeration { enum random { description "Port randomization is enabled. A NAT port allocation scheme should make it hard for attackers to guess port numbers"; reference "REQ-15 of RFC 6888"; } enum port-preservation { description "Indicates whether the NAT should preserve the internal port number."; } enum port-parity-preservation { description "Indicates whether the NAT should preserve the port parity of the internal port number."; } enum port-range-allocation { description "Indicates whether the NAT assigns a range of ports for an internal host. This scheme allows the minimizing of the log volume."; reference "REQ-14 of RFC 6888"; } } description "Indicates the type of port allocation."; }
Boucadair, et al. Standards Track [Page 48]
RFC 8512 YANG Module for NAT and NPT January 2019
leaf mapping-type { if-feature "napt44 or nat64"; type enumeration { enum eim { description "endpoint-independent-mapping."; reference "Section 4 of RFC 4787"; } enum adm { description "address-dependent-mapping."; reference "Section 4 of RFC 4787"; } enum edm { description "address-and-port-dependent-mapping."; reference "Section 4 of RFC 4787"; } } description "Indicates the type of NAT mapping."; } leaf filtering-type { if-feature "napt44 or nat64"; type enumeration { enum eif { description "endpoint-independent-filtering."; reference "Section 5 of RFC 4787"; } enum adf { description "address-dependent-filtering."; reference "Section 5 of RFC 4787"; } enum edf { description "address-and-port-dependent-filtering"; reference "Section 5 of RFC 4787"; } } description
Boucadair, et al. Standards Track [Page 49]
RFC 8512 YANG Module for NAT and NPT January 2019
"Indicates the type of NAT filtering."; } leaf fragment-behavior { if-feature "napt44 or nat64"; type enumeration { enum drop-all { description "All received fragments are dropped."; } enum in-order { description "Translate fragments only if they are received in order."; } enum out-of-order { description "Translate a fragment even if it is received out of order.
This behavior is recommended."; reference "REQ-14 of RFC 4787"; } } description "The fragment behavior instructs the NAT about the behavior to follow to translate fragments received on the external interface of the NAT."; } list port-quota { if-feature "napt44 or nat64"; key "quota-type"; description "Configures a port quota to be assigned per subscriber. It corresponds to the maximum number of ports to be used by a subscriber."; leaf port-limit { type uint16; description "Configures a port quota to be assigned per subscriber. It corresponds to the maximum number of ports to be used by a subscriber."; reference "REQ-4 of RFC 6888"; } leaf quota-type { type uint8; description
Boucadair, et al. Standards Track [Page 50]
RFC 8512 YANG Module for NAT and NPT January 2019
"Indicates whether the port quota applies to all protocols (0) or to a specific protocol."; } } container port-set { when "../port-allocation-type = 'port-range-allocation'"; if-feature "napt44 or nat64"; description "Manages port-set assignments."; leaf port-set-size { type uint16; mandatory true; description "Indicates the size of assigned port sets."; } leaf port-set-timeout { type uint32; units "seconds"; description "inactivity timeout for port sets."; } } container timers { if-feature "napt44 or nat64"; description "Configure values of various timeouts."; leaf udp-timeout { type uint32; units "seconds"; default "300"; description "UDP inactivity timeout. That is the time a mapping will stay active without packets traversing the NAT."; reference "RFC 4787: Network Address Translation (NAT) Behavioral Requirements for Unicast UDP"; } leaf tcp-idle-timeout { type uint32; units "seconds"; default "7440"; description "TCP idle timeout should be 2 hours and 4 minutes."; reference "RFC 5382: NAT Behavioral Requirements for TCP"; } leaf tcp-trans-open-timeout { type uint32;
Boucadair, et al. Standards Track [Page 51]
RFC 8512 YANG Module for NAT and NPT January 2019
units "seconds"; default "240"; description "The value of the transitory open connection idle-timeout.
A NAT should provide different configurable parameters for configuring the open and closing idle timeouts.
To accommodate deployments that consider a partially open timeout of 4 minutes as being excessive from a security standpoint, a NAT may allow the configured timeout to be less than 4 minutes.
However, a minimum default transitory connection idle-timeout of 4 minutes is recommended."; reference "Section 2.1 of RFC 7857"; } leaf tcp-trans-close-timeout { type uint32; units "seconds"; default "240"; description "The value of the transitory close connection idle-timeout.
A NAT should provide different configurable parameters for configuring the open and closing idle timeouts."; reference "Section 2.1 of RFC 7857"; } leaf tcp-in-syn-timeout { type uint32; units "seconds"; default "6"; description "A NAT must not respond to an unsolicited inbound SYN packet for at least 6 seconds after the packet is received. If during this interval the NAT receives and translates an outbound SYN for the connection the NAT must silently drop the original unsolicited inbound SYN packet."; reference
Boucadair, et al. Standards Track [Page 52]
RFC 8512 YANG Module for NAT and NPT January 2019
"RFC 5382 NAT Behavioral Requirements for TCP"; } leaf fragment-min-timeout { when "../../fragment-behavior='out-of-order'"; type uint32; units "seconds"; default "2"; description "As long as the NAT has available resources, the NAT allows the fragments to arrive over the fragment-min-timeout interval. The default value is inspired from RFC 6146."; } leaf icmp-timeout { type uint32; units "seconds"; default "60"; description "An ICMP Query session timer must not expire in less than 60 seconds. It is recommended that the ICMP Query session timer be made configurable"; reference "RFC 5508: NAT Behavioral Requirements for ICMP"; } list per-port-timeout { key "port-number"; description "Some NATs are configurable with short timeouts for some ports, e.g., as 10 seconds on port 53 (DNS) and 123 (NTP), and longer timeouts on other ports."; leaf port-number { type inet:port-number; description "A port number."; } leaf protocol { type uint8; description "The upper-layer protocol associated with this port.
Values are taken from the IANA Protocol Numbers registry.
If no protocol is indicated, it means 'any protocol'."; }
Boucadair, et al. Standards Track [Page 53]
RFC 8512 YANG Module for NAT and NPT January 2019
leaf timeout { type uint32; units "seconds"; mandatory true; description "Timeout for this port number"; } } leaf hold-down-timeout { type uint32; units "seconds"; default "120"; description "Hold-down timer.
Ports in the hold-down pool are not reassigned until hold-down-timeout expires.
The length of time and the maximum number of ports in this state must be configurable by the administrator.
This is necessary in order to prevent collisions between old and new mappings and sessions. It ensures that all established sessions are broken instead of redirected to a different peer."; reference "REQ-8 of RFC 6888"; } leaf hold-down-max { type uint32; description "Maximum ports in the hold-down port pool."; reference "REQ-8 of RFC 6888"; } } leaf fragments-limit { when "../fragment-behavior='out-of-order'"; type uint32; description "Limits the number of out-of-order fragments that can be handled."; reference "Section 11 of RFC 4787"; } list algs { key "name"; description
Boucadair, et al. Standards Track [Page 54]
RFC 8512 YANG Module for NAT and NPT January 2019
"Features related to the Application Layer Gateway (ALG)."; leaf name { type string; description "The name of the ALG."; } leaf transport-protocol { type uint32; description "The transport protocol used by the ALG (e.g., TCP and UDP)."; } container dst-transport-port { uses port-number; description "The destination port number(s) used by the ALG. For example, - 21 for the FTP ALG - 53 for the DNS ALG."; } container src-transport-port { uses port-number; description "The source port number(s) used by the ALG."; } leaf status { type boolean; description "Enable/disable the ALG."; } } leaf all-algs-enable { type boolean; description "Disable/enable all ALGs.
When specified, this parameter overrides the one that may be indicated, eventually, by the 'status' of an individual ALG."; } container notify-pool-usage { if-feature "basic-nat44 or napt44 or nat64"; description "Notification of pool usage when certain criteria are met."; leaf pool-id { type uint32;
Boucadair, et al. Standards Track [Page 55]
RFC 8512 YANG Module for NAT and NPT January 2019
description "Pool-ID for which the notification criteria is defined"; } leaf low-threshold { type percent; description "Notification must be generated when the defined low threshold is reached.
For example, if a notification is required when the pool utilization reaches below 10%, this configuration parameter must be set to 10.
0% indicates that low-threshold notification is disabled."; } leaf high-threshold { type percent; must '. >= ../low-threshold' { error-message "The high threshold must be greater than or equal to the low threshold."; } description "Notification must be generated when the defined high threshold is reached.
For example, if a notification is required when the pool utilization reaches 90%, this configuration parameter must be set to 90.
Setting the same value as low-threshold is equivalent to disabling high-threshold notification."; } leaf notify-interval { type uint32 { range "1 .. 3600"; } units "seconds"; default "20"; description "Minimum number of seconds between successive notifications for this pool."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)"; }
Boucadair, et al. Standards Track [Page 56]
RFC 8512 YANG Module for NAT and NPT January 2019
} container external-realm { description "Identifies the external realm of the NAT instance."; choice realm-type { description "Can be an interface, VRF instance, etc."; case interface { description "External interface."; leaf external-interface { type if:interface-ref; description "Name of the external interface."; } } } } } container mapping-limits { if-feature "napt44 or nat64"; description "Information about the configuration parameters that limits the mappings based upon various criteria."; leaf limit-subscribers { type uint32; description "Maximum number of subscribers that can be serviced by a NAT instance.
A subscriber is identified by a given prefix."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)"; } leaf limit-address-mappings { type uint32; description "Maximum number of address mappings that can be handled by a NAT instance.
When this limit is reached, packets that would normally trigger translation will be dropped."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)"; } leaf limit-port-mappings {
Boucadair, et al. Standards Track [Page 57]
RFC 8512 YANG Module for NAT and NPT January 2019
type uint32; description "Maximum number of port mappings that can be handled by a NAT instance.
When this limit is reached, packets that would normally trigger translation will be dropped."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)"; } list limit-per-protocol { if-feature "napt44 or nat64 or dst-nat"; key "protocol-id"; description "Configure limits per transport protocol"; leaf protocol-id { type uint8; mandatory true; description "The upper-layer protocol.
Values are taken from the IANA Protocol Numbers registry.
For example, this field contains 6 for TCP, 17 for UDP, 33 for DCCP, or 132 for SCTP."; } leaf limit { type uint32; description "Maximum number of protocol-specific NAT mappings per instance."; } } } container connection-limits { if-feature "basic-nat44 or napt44 or nat64"; description "Information about the configuration parameters that rate-limit the translation based upon various criteria."; leaf limit-per-subscriber { type uint32; units "bits/second"; description "Rate-limit the number of new mappings and sessions per subscriber."; }
Boucadair, et al. Standards Track [Page 58]
RFC 8512 YANG Module for NAT and NPT January 2019
leaf limit-per-instance { type uint32; units "bits/second"; description "Rate-limit the number of new mappings and sessions per instance."; } list limit-per-protocol { if-feature "napt44 or nat64"; key "protocol-id"; description "Configure limits per transport protocol"; leaf protocol-id { type uint8; mandatory true; description "The upper-layer protocol.
Values are taken from the IANA Protocol Numbers registry.
For example, this field contains 6 for TCP, 17 for UDP, 33 for DCCP, or 132 for SCTP."; } leaf limit { type uint32; description "Limit the number of protocol-specific mappings and sessions per instance."; } } } container notification-limits { description "Sets notification limits."; leaf notify-interval { if-feature "basic-nat44 or napt44 or nat64"; type uint32 { range "1 .. 3600"; } units "seconds"; default "10"; description "Minimum number of seconds between successive notifications for this NAT instance."; reference "RFC 7659: Definitions of Managed Objects for Network Address Translators (NATs)";
Boucadair, et al. Standards Track [Page 59]
RFC 8512 YANG Module for NAT and NPT January 2019
} leaf notify-addresses-usage { if-feature "basic-nat44 or napt44 or nat64"; type percent; description "Notification of address mappings usage over the whole NAT instance.
Notification must be generated when the defined threshold is reached.
For example, if a notification is required when the address mappings utilization reaches 90%, this configuration parameter must be set to 90."; } leaf notify-ports-usage { if-feature "napt44 or nat64"; type percent; description "Notification of port mappings usage over the whole NAT instance.
Notification must be generated when the defined threshold is reached.
For example, if a notification is required when the port mappings utilization reaches 90%, this configuration parameter must be set to 90."; } leaf notify-subscribers-limit { if-feature "basic-nat44 or napt44 or nat64"; type uint32; description "Notification of active subscribers per NAT instance.
Notification must be generated when the defined threshold is reached."; } } container mapping-table { if-feature "basic-nat44 or napt44 or nat64 " + "or clat or dst-nat"; description "NAT mapping table. Applicable for functions that maintain static and/or dynamic mappings, such as NAT44, Destination NAT, NAT64, or CLAT.";
Boucadair, et al. Standards Track [Page 60]
RFC 8512 YANG Module for NAT and NPT January 2019
list mapping-entry { key "index"; description "NAT mapping entry."; uses mapping-entry; } } container statistics { config false; description "Statistics related to the NAT instance."; leaf discontinuity-time { type yang:date-and-time; mandatory true; description "The time on the most recent occasion at which the NAT instance suffered a discontinuity. This must be initialized when the NAT instance is configured or rebooted."; } container traffic-statistics { description "Generic traffic statistics."; leaf sent-packets { type yang:zero-based-counter64; description "Number of packets sent."; } leaf sent-bytes { type yang:zero-based-counter64; units "bytes"; description "Counter for sent traffic in bytes."; } leaf rcvd-packets { type yang:zero-based-counter64; description "Number of received packets."; } leaf rcvd-bytes { type yang:zero-based-counter64; units "bytes"; description "Counter for received traffic in bytes."; } leaf dropped-packets { type yang:zero-based-counter64; description
Boucadair, et al. Standards Track [Page 61]
RFC 8512 YANG Module for NAT and NPT January 2019
"Number of dropped packets."; } leaf dropped-bytes { type yang:zero-based-counter64; units "bytes"; description "Counter for dropped traffic in bytes."; } leaf dropped-fragments { if-feature "napt44 or nat64"; type yang:zero-based-counter64; description "Number of dropped fragments on the external realm."; } leaf dropped-address-limit-packets { if-feature "basic-nat44 or napt44 or nat64"; type yang:zero-based-counter64; description "Number of dropped packets because an address limit is reached."; } leaf dropped-address-limit-bytes { if-feature "basic-nat44 or napt44 or nat64"; type yang:zero-based-counter64; units "bytes"; description "Counter of dropped packets because an address limit is reached, in bytes."; } leaf dropped-address-packets { if-feature "basic-nat44 or napt44 or nat64"; type yang:zero-based-counter64; description "Number of dropped packets because no address is available for allocation."; } leaf dropped-address-bytes { if-feature "basic-nat44 or napt44 or nat64"; type yang:zero-based-counter64; units "bytes"; description "Counter of dropped packets because no address is available for allocation, in bytes."; } leaf dropped-port-limit-packets { if-feature "napt44 or nat64"; type yang:zero-based-counter64; description
Boucadair, et al. Standards Track [Page 62]
RFC 8512 YANG Module for NAT and NPT January 2019
"Number of dropped packets because a port limit is reached."; } leaf dropped-port-limit-bytes { if-feature "napt44 or nat64"; type yang:zero-based-counter64; units "bytes"; description "Counter of dropped packets because a port limit is reached, in bytes."; } leaf dropped-port-packets { if-feature "napt44 or nat64"; type yang:zero-based-counter64; description "Number of dropped packets because no port is available for allocation."; } leaf dropped-port-bytes { if-feature "napt44 or nat64"; type yang:zero-based-counter64; units "bytes"; description "Counter of dropped packets because no port is available for allocation, in bytes."; } leaf dropped-subscriber-limit-packets { if-feature "basic-nat44 or napt44 or nat64"; type yang:zero-based-counter64; description "Number of dropped packets because the subscriber limit per instance is reached."; } leaf dropped-subscriber-limit-bytes { if-feature "basic-nat44 or napt44 or nat64"; type yang:zero-based-counter64; units "bytes"; description "Counter of dropped packets because the subscriber limit per instance is reached, in bytes."; } } container mappings-statistics { description "Mappings statistics."; leaf total-active-subscribers { if-feature "basic-nat44 or napt44 or nat64"; type yang:gauge32;
Boucadair, et al. Standards Track [Page 63]
RFC 8512 YANG Module for NAT and NPT January 2019
description "Total number of active subscribers (that is, subscribers for which the NAT maintains active mappings).
A subscriber is identified by a subnet, subscriber-mask, etc."; } leaf total-address-mappings { if-feature "basic-nat44 or napt44 or nat64 " + "or clat or dst-nat"; type yang:gauge32; description "Total number of address mappings present at a given time. It includes both static and dynamic mappings."; reference "Section 3.3.8 of RFC 7659"; } leaf total-port-mappings { if-feature "napt44 or nat64"; type yang:gauge32; description "Total number of NAT port mappings present at a given time. It includes both static and dynamic mappings."; reference "Section 3.3.9 of RFC 7659"; } list total-per-protocol { if-feature "napt44 or nat64"; key "protocol-id"; description "Total mappings for each enabled/supported protocol."; leaf protocol-id { type uint8; mandatory true; description "The upper-layer protocol. For example, this field contains 6 for TCP, 17 for UDP, 33 for DCCP, or 132 for SCTP."; } leaf total { type yang:gauge32; description "Total number of a protocol-specific mappings present at a given time. The protocol is identified by protocol-id."; }
Boucadair, et al. Standards Track [Page 64]
RFC 8512 YANG Module for NAT and NPT January 2019
} } container pools-stats { if-feature "basic-nat44 or napt44 or nat64"; description "Statistics related to address/prefix pools usage"; leaf addresses-allocated { type yang:gauge32; description "Number of all allocated addresses."; } leaf addresses-free { type yang:gauge32; description "Number of unallocated addresses of all pools at a given time. The sum of unallocated and allocated addresses is the total number of addresses of the pools."; } container ports-stats { if-feature "napt44 or nat64"; description "Statistics related to port numbers usage."; leaf ports-allocated { type yang:gauge32; description "Number of allocated ports from all pools."; } leaf ports-free { type yang:gauge32; description "Number of unallocated addresses from all pools."; } } list per-pool-stats { if-feature "basic-nat44 or napt44 or nat64"; key "pool-id"; description "Statistics related to address/prefix pool usage"; leaf pool-id { type uint32; description "Unique identifier that represents a pool of addresses/prefixes."; } leaf discontinuity-time { type yang:date-and-time;
Boucadair, et al. Standards Track [Page 65]
RFC 8512 YANG Module for NAT and NPT January 2019
mandatory true; description "The time on the most recent occasion at which this pool counter suffered a discontinuity. This must be initialized when the address pool is configured."; } container pool-stats { description "Statistics related to address/prefix pool usage"; leaf addresses-allocated { type yang:gauge32; description "Number of allocated addresses from this pool."; } leaf addresses-free { type yang:gauge32; description "Number of unallocated addresses in this pool."; } } container port-stats { if-feature "napt44 or nat64"; description "Statistics related to port numbers usage."; leaf ports-allocated { type yang:gauge32; description "Number of allocated ports from this pool."; } leaf ports-free { type yang:gauge32; description "Number of unallocated addresses from this pool."; } } } } } } } }
/* * Notifications */
notification nat-pool-event {
Boucadair, et al. Standards Track [Page 66]
RFC 8512 YANG Module for NAT and NPT January 2019
if-feature "basic-nat44 or napt44 or nat64"; description "Notifications must be generated when the defined high/low threshold is reached. Related configuration parameters must be provided to trigger the notifications."; leaf id { type leafref { path "/nat/instances/instance/id"; } mandatory true; description "NAT instance identifier."; } leaf policy-id { type leafref { path "/nat/instances/instance/policy/id"; } description "Policy identifier."; } leaf pool-id { type leafref { path "/nat/instances/instance/policy" + "/external-ip-address-pool/pool-id"; } mandatory true; description "Pool Identifier."; } leaf notify-pool-threshold { type percent; mandatory true; description "A threshold (high threshold or low threshold) has been fired."; } }
notification nat-instance-event { if-feature "basic-nat44 or napt44 or nat64"; description "Notifications must be generated when notify-addresses-usage and/or notify-ports-usage thresholds are reached."; leaf id { type leafref { path "/nat/instances/instance/id"; } mandatory true;
Boucadair, et al. Standards Track [Page 67]
RFC 8512 YANG Module for NAT and NPT January 2019
description "NAT instance identifier."; } leaf notify-subscribers-threshold { type uint32; description "The notify-subscribers-limit threshold has been fired."; } leaf notify-addresses-threshold { type percent; description "The notify-addresses-usage threshold has been fired."; } leaf notify-ports-threshold { type percent; description "The notify-ports-usage threshold has been fired."; } } }
Security considerations related to address and prefix translation are discussed in [RFC6888], [RFC6146], [RFC6877], [RFC6296], and [RFC7757].
The YANG module specified in this document defines a schema for data that is designed to be accessed via network management protocols such as NETCONF [RFC6241] or RESTCONF [RFC8040]. The lowest NETCONF layer is the secure transport layer, and the mandatory-to-implement secure transport is Secure Shell (SSH) [RFC6242]. The lowest RESTCONF layer is HTTPS, and the mandatory-to-implement secure transport is TLS [RFC8446].
The Network Configuration Access Control Model (NACM) [RFC8341] provides the means to restrict access for particular NETCONF or RESTCONF users to a preconfigured subset of all available NETCONF or RESTCONF protocol operations and content.
All data nodes defined in the YANG module that can be created, modified, and deleted (i.e., config true, which is the default) are considered sensitive. Write operations (e.g., edit-config) applied to these data nodes without proper protection can negatively affect network operations. The NAT YANG module provides a method to set parameters to prevent a user from aggressively using NAT resources
Boucadair, et al. Standards Track [Page 68]
RFC 8512 YANG Module for NAT and NPT January 2019
(port-quota), rate-limit connections as a guard against DoS, or to enable notifications so that appropriate measures are enforced to anticipate traffic drops. Nevertheless, an attacker who is able to access the NAT can undertake various attacks, such as:
o Set a high or low resource limit to cause a DoS attack:
* /nat/instances/instance/policy/port-quota
* /nat/instances/instance/policy/fragments-limit
* /nat/instances/instance/mapping-limits
* /nat/instances/instance/connection-limits
o Set a low notification threshold to cause useless notifications to be generated:
[RFC7597] Troan, O., Ed., Dec, W., Li, X., Bao, C., Matsushima, S., Murakami, T., and T. Taylor, Ed., "Mapping of Address and Port with Encapsulation (MAP-E)", RFC 7597, DOI 10.17487/RFC7597, July 2015, <https://www.rfc-editor.org/info/rfc7597>.
[RFC7857] Penno, R., Perreault, S., Boucadair, M., Ed., Sivakumar, S., and K. Naito, "Updates to Network Address Translation (NAT) Behavioral Requirements", BCP 127, RFC 7857, DOI 10.17487/RFC7857, April 2016, <https://www.rfc-editor.org/info/rfc7857>.
[NAT-SUPP] Stewart, R., Tuexen, M., and I. Ruengeler, "Stream Control Transmission Protocol (SCTP) Network Address Translation Support", Work in Progress, draft-ietf-tsvwg-natsupp-12, July 2018.
[YANG-PCP] Boucadair, M., Jacquenet, C., Sivakumar, S., and S. Vinapamula, "YANG Modules for the Port Control Protocol (PCP)", Work in Progress, draft-boucadair-pcp-yang-05, October 2017.
Traditional NAT44 is a Basic NAT44 or NAPT that is used to share the same IPv4 address among hosts that are owned by the same subscriber. This is typically the NAT that is embedded in CPE devices.
This NAT is usually provided with one single external IPv4 address; disambiguating connections is achieved by rewriting the source port number. The XML snippet to configure the external IPv4 address in such case together with a mapping entry is depicted below:
The following shows the XML excerpt depicting a dynamic UDP mapping entry maintained by a traditional NAPT44. In reference to this example, the UDP packet received with a source IPv4 address (192.0.2.1) and source port number (1568) is translated into a UDP packet having a source IPv4 address (198.51.100.1) and source port (15000). The remaining lifetime of this mapping is 300 seconds.
The following XML snippet shows the example of a CGN that is provisioned with one contiguous pool of external IPv4 addresses (198.51.100.0/24). Further, the CGN is instructed to limit the number of allocated ports per subscriber to 1024. Ports can be allocated by the CGN by assigning ranges of 256 ports (that is, a subscriber can be allocated up to four port ranges of 256 ports each).
Figure 1 illustrates an example of the CGN pass-through feature.
X1:x1 X1':x1' X2:x2 +---+from X1:x1 +---+from X1:x1 +---+ | C | to X2:x2 | | to X2:x2 | S | | l |>>>>>>>>>>>>| C |>>>>>>>>>>>>>>| e | | i | | G | | r | | e |<<<<<<<<<<<<| N |<<<<<<<<<<<<<<| v | | n |from X2:x2 | |from X2:x2 | e | | t | to X1:x1 | | to X1:x1 | r | +---+ +---+ +---+
Figure 1: CGN Pass-Through
For example, in order to disable NAT for communications issued by the client (192.0.2.1), the following configuration parameter must be set:
Let's consider the example of a NAT64 that should use 2001:db8:122:300::/56 to perform IPv6 address synthesis [RFC6052]. The XML snippet to configure the NAT64 prefix in such case is depicted below:
Let's now consider the example of a NAT64 that should use 2001:db8:122::/48 to perform IPv6 address synthesis [RFC6052] only if the destination address matches 198.51.100.0/24. The XML snippet to configure the NAT64 prefix in such case is shown below:
Let's consider the example of a stateless translator that is configured with 2001:db8:100::/40 to perform IPv6 address synthesis [RFC6052]. Similar to the NAT64 case, the XML snippet to configure the NAT64 prefix in such case is depicted below:
When the translator receives an IPv6 packet, for example, with a source address (2001:db8:1c0:2:21::) and destination address (2001:db8:1c6:3364:2::), it extracts embedded IPv4 addresses following rules per RFC 6052 with 2001:db8:100::/40 as the NSP:
o 192.0.2.33 is extracted from 2001:db8:1c0:2:21::
o 198.51.100.2 is extracted from 2001:db8:1c6:3364:2::
The translator transforms the IPv6 header into an IPv4 header using the IP/ICMP Translation Algorithm [RFC7915]. The IPv4 packets will include 192.0.2.33 as the source address and 198.51.100.2 as the destination address.
Also, a NAT64 can be instructed to behave in the stateless mode by providing the following configuration. The same NAT64 prefix is used for constructing both IPv4-translatable IPv6 addresses and IPv4-converted IPv6 addresses (see Section 3.3 of [RFC6052]).
The following example shows a static mapping that instructs a NAT to translate packets issued from 192.0.2.1 and with source ports in the 100-500 range to 198.51.100.1:1100-1500.
The following XML snippet shows an example of a Destination NAT that is instructed to translate all packets having 192.0.2.1 as a destination IP address to 198.51.100.1.
In order to instruct a NAT to translate TCP packets destined to '192.0.2.1:80' to '198.51.100.1:8080', the following XML snippet shows the static mapping configured on the NAT:
In order to instruct a NAT to translate TCP packets destined to '192.0.2.1:80' (HTTP traffic) to 198.51.100.1 and '192.0.2.1:22' (SSH traffic) to 198.51.100.2, the following XML snippet shows the static mappings configured on the NAT:
The NAT may also be instructed to proceed with both source and Destination NAT. To do so, in addition to the above example to configure Destination NAT, the NAT may be provided, for example with a pool of external IP addresses (198.51.100.0/24) to use for source address translation. An example of the corresponding XML snippet is provided hereafter:
Instead of providing an external IP address to share, the NAT may be configured with static mapping entries that modify the internal IP address and/or port number.
The following XML snippet shows the example of a CLAT that is configured with 2001:db8:1234::/96 as a PLAT-side IPv6 prefix and 2001:db8:aaaa::/96 as a CLAT-side IPv6 prefix. The CLAT is also provided with 192.0.0.1/32 (which is selected from the IPv4 service continuity prefix defined in [RFC7335]).
Let's consider the example of an NPTv6 translator that should rewrite packets with the source prefix (fd03:c03a:ecab::/48) with the external prefix (2001:db8:1::/48). The internal interface is "eth0" while the external interface is "eth1" (Figure 2).
Figure 3 shows an example of an NPTv6 translator that interconnects two internal networks (fd03:c03a:ecab::/48 and fda8:d5cb:14f3::/48); each is translated using a dedicated prefix (2001:db8:1::/48 and 2001:db8:6666::/48, respectively).
Internal Prefix = fda8:d5cb:14f3::/48 -------------------------------------- V | External Prefix V |eth1 2001:db8:1::/48 V +---------+ ^ V | NPTv6 | ^ V | | ^ V +---------+ ^ External Prefix |eth0 ^ 2001:db8:6666::/48 | ^ -------------------------------------- Internal Prefix = fd03:c03a:ecab::/48
Figure 3: Connecting Two Peer Networks
Boucadair, et al. Standards Track [Page 91]
RFC 8512 YANG Module for NAT and NPT January 2019
To that aim, the following configuration is provided to the NPTv6 translator:
Many thanks to Dan Wing, Tianran Zhou, Tom Petch, Warren Kumari, and Benjamin Kaduk for their review.
Thanks to Juergen Schoenwaelder for the comments on the YANG structure and the suggestion to use NMDA. Mahesh Jethanandani provided useful comments.
Thanks to Lee Howard and Jordi Palet for the CLAT comments, Fred Baker for the NPTv6 comments, Tore Anderson for the EAM SIIT review, and Kristian Poscic for the CGN review.