| Internet Engineering Task Force | T. Chown |
| Internet-Draft | Jisc |
| Obsoletes: 6434 (if approved) | J. Loughney |
| Intended status: Best Current Practice | Intel |
| Expires: August 31, 2018 | T. Winters |
| UNH-IOL | |
| February 27, 2018 |
IPv6 Node Requirements
draft-ietf-6man-rfc6434-bis-05
This document defines requirements for IPv6 nodes. It is expected that IPv6 will be deployed in a wide range of devices and situations. Specifying the requirements for IPv6 nodes allows IPv6 to function well and interoperate in a large number of situations and deployments.
This document obsoletes RFC 6434, and in turn RFC 4294.
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 August 31, 2018.
Copyright (c) 2018 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 common functionality required by both IPv6 hosts and routers. Many IPv6 nodes will implement optional or additional features, but this document collects and summarizes requirements from other published Standards Track documents in one place.
This document tries to avoid discussion of protocol details and references RFCs for this purpose. This document is intended to be an applicability statement and to provide guidance as to which IPv6 specifications should be implemented in the general case and which specifications may be of interest to specific deployment scenarios. This document does not update any individual protocol document RFCs.
Although this document points to different specifications, it should be noted that in many cases, the granularity of a particular requirement will be smaller than a single specification, as many specifications define multiple, independent pieces, some of which may not be mandatory. In addition, most specifications define both client and server behavior in the same specification, while many implementations will be focused on only one of those roles.
This document defines a minimal level of requirement needed for a device to provide useful internet service and considers a broad range of device types and deployment scenarios. Because of the wide range of deployment scenarios, the minimal requirements specified in this document may not be sufficient for all deployment scenarios. It is perfectly reasonable (and indeed expected) for other profiles to define additional or stricter requirements appropriate for specific usage and deployment environments. For example, this document does not mandate that all clients support DHCP, but some deployment scenarios may deem it appropriate to make such a requirement. For example, government agencies in the USA have defined profiles for specialized requirements for IPv6 in target environments (see [USGv6]).
As it is not always possible for an implementer to know the exact usage of IPv6 in a node, an overriding requirement for IPv6 nodes is that they should adhere to Jon Postel's Robustness Principle: "Be conservative in what you do, be liberal in what you accept from others" [RFC0793].
IPv6 covers many specifications. It is intended that IPv6 will be deployed in many different situations and environments. Therefore, it is important to develop requirements for IPv6 nodes to ensure interoperability.
This document assumes that all IPv6 nodes meet the minimum requirements specified here.
From the Internet Protocol, Version 6 (IPv6) Specification [RFC8200], we have the following definitions:
IPv6 node - a device that implements IPv6.
IPv6 router - a node that forwards IPv6 packets not explicitly
addressed to itself.
IPv6 host - any node that is not a router.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119.
AH Authentication Header DAD Duplicate Address Detection ESP Encapsulating Security Payload ICMP Internet Control Message Protocol IKE Internet Key Exchange MIB Management Information Base MLD Multicast Listener Discovery MTU Maximum Transmission Unit NA Neighbor Advertisement NBMA Non-Broadcast Multiple Access ND Neighbor Discovery NS Neighbor Solicitation NUD Neighbor Unreachability Detection PPP Point-to-Point Protocol
An IPv6 node must include support for one or more IPv6 link-layer specifications. Which link-layer specifications an implementation should include will depend upon what link-layers are supported by the hardware available on the system. It is possible for a conformant IPv6 node to support IPv6 on some of its interfaces and not on others.
As IPv6 is run over new layer 2 technologies, it is expected that new specifications will be issued. In the following, we list some of the layer 2 technologies for which an IPv6 specification has been developed. It is provided for informational purposes only and may not be complete.
In addition to traditional physical link-layers, it is also possible to tunnel IPv6 over other protocols. Examples include:
The Internet Protocol Version 6 is specified in [RFC8200]. This specification MUST be supported.
The node MUST follow the packet transmission rules in RFC 8200.
All conformant IPv6 implementations MUST be capable of sending and receiving IPv6 packets; forwarding functionality MAY be supported. Nodes MUST always be able to send, receive, and process fragment headers. Overlapping fragments MUST be handled as described in [RFC5722].
[RFC6946] discusses IPv6 atomic fragments, and recommends that IPv6 atomic fragments are processed independently of any other fragments, to protect against fragmentation-based attacks. [RFC8021] goes further and recommends the deprecation of atomic fragments. Nodes thus MUST NOT generate atomic fragments.
To mitigate a variety of potential attacks, nodes SHOULD avoid using predictable fragment Identification values in Fragment Headers, as discussed in [RFC7739].
All nodes SHOULD support the setting and use of the IPv6 Flow Label field as defined in the IPv6 Flow Label specification [RFC6437]. Forwarding nodes such as routers and load distributors MUST NOT depend only on Flow Label values being uniformly distributed. It is RECOMMENDED that source hosts support the flow label by setting the Flow Label field for all packets of a given flow to the same value chosen from an approximation to a discrete uniform distribution.
RFC 8200 specifies extension headers and the processing for these headers.
Any unrecognized extension headers or options MUST be processed as described in RFC 8200. Note that where Section 4 of RFC 8200 refers to the action to be taken when a Next Header value in the current header is not recognized by a node, that action applies whether the value is an unrecognized Extension Header or an unrecognized upper layer protocol (ULP).
An IPv6 node MUST be able to process these headers. An exception is Routing Header type 0 (RH0), which was deprecated by [RFC5095] due to security concerns and which MUST be treated as an unrecognized routing type.
Further, [RFC7045] adds specific requirements for processing of Extension Headers, in particular that any forwarding node along an IPv6 packet's path, which forwards the packet for any reason, SHOULD do so regardless of any extension headers that are present.
[RFC7112] discusses issues with oversized IPv6 Extension Header chains, and states that when a node fragments an IPv6 datagram, it MUST include the entire IPv6 Header Chain in the First Fragment.
As stated in RFC8200, extension headers (except for the Hop-by-Hop Options header) are not processed, inserted, or deleted by any node along a packet's delivery path, until the packet reaches the node (or each of the set of nodes, in the case of multicast) identified in the Destination Address field of the IPv6 header.
It should be noted that when future, new Extension Headers are defined, the consistent format described in Section 4 of [RFC6564] MUST be followed.
Per RFC 8200, end hosts are expected to process all extension headers, destination options, and hop-by-hop options in a packet. Given that the only limit on the number and size of extension headers is the MTU, the processing of received packets could be considerable. It is also conceivable that a long chain of extension headers might be used as a form of denial-of-service attack. Accordingly, a host may place limits on the number and sizes of extension headers and options it is willing to process.
A host MAY limit the number of consecutive PAD1 options in destination options or hop-by-hop options to seven. In this case, if the more than seven consecutive PAD1 options are present the packet should be silently discarded. The rationale is that if padding of eight or more bytes is required than the PADN option should be used.
A host MAY limit number of bytes in a PADN option to be less than eight. In such a case, if a PADN option is present that has a length greater than seven then the packet should be silently discarded. The rationale for this guideline is that the purpose of padding is for alignment and eight bytes is the maximum alignment used in IPv6.
A host MAY disallow unknown options in destination options or hop-by-hop options. This should be configurable where the default is to accept unknown options and process them per [RFC8200]. If a packet with unknown options is received and the host is configured to disallow them, then the packet should be silently discarded.
A host MAY impose a limit on the maximum number of non-padding options allowed in a destination options and hop-by-hop extension headers. If this feature is supported the maximum number should be configurable and the default value SHOULD be set to eight. The limits for destination options and hop-by-hop options may be separately configurable. If a packet is received and the number of destination or hop-by-hop optines exceeds the limit, then the packet should be silently discarded.
A host MAY impose a limit on the maximum length of destination options or hop-by-hop options extension header. This value should be configurable and the default is to accept options of any length. If a packet is received and the length of destination or hop-by-hop options extension header exceeds the length limit, then the packet should be silently discarded.
Neighbor Discovery is defined in [RFC4861]; the definition was updated by [RFC5942]. Neighbor Discovery SHOULD be supported. RFC 4861 states:
Some detailed analysis of Neighbor Discovery follows:
Router Discovery is how hosts locate routers that reside on an attached link. Hosts MUST support Router Discovery functionality.
Prefix Discovery is how hosts discover the set of address prefixes that define which destinations are on-link for an attached link. Hosts MUST support Prefix Discovery.
Hosts MUST also implement Neighbor Unreachability Detection (NUD) for all paths between hosts and neighboring nodes. NUD is not required for paths between routers. However, all nodes MUST respond to unicast Neighbor Solicitation (NS) messages.
[RFC7048] discusses NUD, in particular cases where it behaves too impatiently. It states that if a node transmits more than a certain number of packets, then it SHOULD use the exponential backoff of the retransmit timer, up to a certain threshold point.
Hosts MUST support the sending of Router Solicitations and the receiving of Router Advertisements. The ability to understand individual Router Advertisement options is dependent on supporting the functionality making use of the particular option.
[RFC7559] discusses packet loss resliency for Router Solicitations, and requires that nodes MUST use a specific exponential backoff algorithm for RS retransmissions.
All nodes MUST support the sending and receiving of Neighbor Solicitation (NS) and Neighbor Advertisement (NA) messages. NS and NA messages are required for Duplicate Address Detection (DAD).
Hosts SHOULD support the processing of Redirect functionality. Routers MUST support the sending of Redirects, though not necessarily for every individual packet (e.g., due to rate limiting). Redirects are only useful on networks supporting hosts. In core networks dominated by routers, Redirects are typically disabled. The sending of Redirects SHOULD be disabled by default on backbone routers. They MAY be enabled by default on routers intended to support hosts on edge networks.
"IPv6 Host-to-Router Load Sharing" [RFC4311] includes additional recommendations on how to select from a set of available routers. [RFC4311] SHOULD be supported.
SEND [RFC3971] and Cryptographically Generated Addresses (CGAs) [RFC3972] provide a way to secure the message exchanges of Neighbor Discovery. SEND has the potential to address certain classes of spoofing attacks, but it does not provide specific protection for threats from off-link attackers.
There have been relatively few implementations of SEND in common operating systems and platforms, and thus deployment experience has been limited to date.
At this time, SEND is considered optional. Due to the complexity in deploying SEND, and its heavyweight provisioning, its deployment is only likely to be considered where nodes are operating in a particularly strict security environment.
Router Advertisements include an 8-bit field of single-bit Router Advertisement flags. The Router Advertisement Flags Option extends the number of available flag bits by 48 bits. At the time of this writing, 6 of the original 8 single-bit flags have been assigned, while 2 remain available for future assignment. No flags have been defined that make use of the new option, and thus, strictly speaking, there is no requirement to implement the option today. However, implementations that are able to pass unrecognized options to a higher-level entity that may be able to understand them (e.g., a user-level process using a "raw socket" facility) MAY take steps to handle the option in anticipation of a future usage.
"Path MTU Discovery for IP version 6" [RFC8201] SHOULD be supported. From [RFC8200]:
The rules in [RFC8200] and [RFC5722] MUST be followed for packet fragmentation and reassembly.
One operational issue with Path MTU Discovery occurs when, contrary to the guidance in [RFC4890], firewalls block ICMP Packet Too Big messages. Path MTU Discovery relies on such messages to determine what size messages can be successfully sent. "Packetization Layer Path MTU Discovery" [RFC4821] avoids having a dependency on Packet Too Big messages.
While an IPv6 link MTU can be set to 1280 bytes, it is recommended that for IPv6 UDP in particular, which includes DNS operation, the sender use a large MTU if they can, in order to avoid gratuitous fragmentation-caused packet drops.
ICMPv6 [RFC4443] MUST be supported. "Extended ICMP to Support Multi-Part Messages" [RFC4884] MAY be supported.
"Default Router Preferences and More-Specific Routes" [RFC4191] provides support for nodes attached to multiple (different) networks, each providing routers that advertise themselves as default routers via Router Advertisements. In some scenarios, one router may provide connectivity to destinations the other router does not, and choosing the "wrong" default router can result in reachability failures. In order to resolve this scenario IPv6 Nodes MUST implement [RFC4191] and SHOULD implement the Type C host role defined in RFC4191.
In multihomed scenarios, where a host has more than one prefix, each allocated by an upstream network that is assumed to implement BCP 38 ingress filtering, the host may have multiple routers to choose from.
Hosts that may be deployed in such multihomed environments SHOULD follow the guidance given in [RFC8028].
Nodes that need to join multicast groups MUST support MLDv2 [RFC3810]. MLD is needed by any node that is expected to receive and process multicast traffic and in particular MLDv2 is required for support for source-specific multicast (SSM) as per [RFC4607].
Previous versions of this document only required MLDv1 ([RFC2710]) to be implemented on all nodes. Since participation of any MLDv1-only nodes on a link require that all other nodeas on the link then operate in version 1 compatibility mode, the requirement to support MLDv2 on all nodes was upgraded to a MUST. Further, SSM is now the preferred multicast distribution method, rather than ASM.
Note that Neighbor Discovery (as used on most link types -- see Section 5.4) depends on multicast and requires that nodes join Solicited Node multicast addresses.
An ECN-aware router may set a mark in the IP header in order to signal impending congestion, rather than dropping a packet. The receiver of the packet echoes the congestion indication to the sender, which can then reduce its transmission rate as if it detected a dropped packet.
Nodes that may be deployed in environments where they would benefit from such early congestion notification SHOULD implement [RFC3168]. In such cases, the updates presented in [RFC8311] may also be relevant.
The IPv6 Addressing Architecture [RFC4291] MUST be supported.
The current IPv6 Address Architecture is based on a 64-bit boundary for subnet prefixes. The reasoning behind this decision is documented in [RFC7421].
Implementations MUST also support the Multicast flag updates documented in [RFC7371]
Hosts may be configured with addresses through a variety of methods, including SLAAC, DHCPv6, or manual configuration.
[RFC7934] recommends that networks provide general-purpose end hosts with multiple global IPv6 addresses when they attach, and it describes the benefits of and the options for doing so. Routers SHOULD support [RFC7934] for assigning multiple address to a host. Host SHOULD support assigning multiple addresses as described in [RFC7934].
Nodes SHOULD support the capability to be assigned a prefix per host as documented in [RFC8273]. Such an approach can offer improved host isolation and enhanced subscriber management on shared network segments.
Hosts MUST support IPv6 Stateless Address Autoconfiguration. It is recommended, as described in [RFC8064], that unless there is a specific requirement for MAC addresses to be embedded in an IID, nodes follow the procedure in [RFC7217] to generate SLAAC-based addresses, rather than using [RFC4862]. Addresses generated through RFC7217 will be the same whenever a given device (re)appears on the same subnet (with a specific IPv6 prefix), but the IID will vary on each subnet visited.
Nodes that are routers MUST be able to generate link-local addresses as described in [RFC4862].
From RFC 4862:
All nodes MUST implement Duplicate Address Detection. Quoting from Section 5.4 of RFC 4862:
"Optimistic Duplicate Address Detection (DAD) for IPv6" [RFC4429] specifies a mechanism to reduce delays associated with generating addresses via Stateless Address Autoconfiguration [RFC4862]. RFC 4429 was developed in conjunction with Mobile IPv6 in order to reduce the time needed to acquire and configure addresses as devices quickly move from one network to another, and it is desirable to minimize transition delays. For general purpose devices, RFC 4429 remains optional at this time.
[RFC7527] discusses enhanced DAD, and describes an algorithm to automate the detection of looped back IPv6 ND messages used by DAD. Nodes SHOULD implement this behaviour where such detection is beneficial.
A node using Stateless Address Autoconfiguration [RFC4862] to form a globally unique IPv6 address using its MAC address to generate the IID will see that IID remain the same on any visited network, even though the network prefix part changes. Thus it is possible for 3rd party devices such nodes communicate with to track the activities of the node as it moves around the network. Privacy Extensions for Stateless Address Autoconfiguration [RFC4941] address this concern by allowing nodes to configure an additional temporary address where the IID is effectively randomly generated. Privacy addresses are then used as source addresses for new communications initiated by the node.
General issues regarding privacy issues for IPv6 addressing are discussed in [RFC7721].
RFC 4941 SHOULD be supported. In some scenarios, such as dedicated servers in a data center, it provides limited or no benefit, or may complicate network management. Thus devices implementing this specification MUST provide a way for the end user to explicitly enable or disable the use of such temporary addresses.
Note that RFC4941 can be used independently of traditional SLAAC, or of RFC7217-based SLAAC.
Implementers of RFC 4941 should be aware that certain addresses are reserved and should not be chosen for use as temporary addresses. Consult "Reserved IPv6 Interface Identifiers" [RFC5453] for more details.
DHCPv6 [RFC3315] can be used to obtain and configure addresses. In general, a network may provide for the configuration of addresses through SLAAC, DHCPv6, or both. There will be a wide range of IPv6 deployment models and differences in address assignment requirements, some of which may require DHCPv6 for stateful address assignment. Consequently, all hosts SHOULD implement address configuration via DHCPv6.
In the absence of observed Router Advertisement messages, IPv6 nodes MAY initiate DHCP to obtain IPv6 addresses and other configuration information, as described in Section 5.5.2 of [RFC4862].
Where devices are likely to be carried by users and attached to multiple visisted networks, DHCPv6 client anonymity profiles SHOULD be supported as described in [RFC7844] to minimise the disclosure of identifying information. Section 5 of RFC7844 describes operational considerations on the use of such anonymity profiles.
IPv6 nodes will invariably have multiple addresses configured simultaneously, and thus will need to choose which addresses to use for which communications. The rules specified in the Default Address Selection for IPv6 [RFC6724] document MUST be implemented. [RFC8028] updates rule 5.5 from [RFC6724]; implementations SHOULD implement this rule.
DNS is described in [RFC1034], [RFC1035], [RFC3363], and [RFC3596]. Not all nodes will need to resolve names; those that will never need to resolve DNS names do not need to implement resolver functionality. However, the ability to resolve names is a basic infrastructure capability on which applications rely, and most nodes will need to provide support. All nodes SHOULD implement stub-resolver [RFC1034] functionality, as in [RFC1034], Section 5.3.1, with support for:
Those nodes are RECOMMENDED to support DNS security extensions [RFC4033] [RFC4034] [RFC4035].
A6 Resource Records, which were only ever defined with Experimental status in [RFC3363], are now classified as Historic, as per [RFC6563].
DHCP [RFC3315] Specifies a mechanism for IPv6 nodes to obtain address configuration information (see Section 6.5) and to obtain additional (non-address) configuration. If a host implementation supports applications or other protocols that require configuration that is only available via DHCP, hosts SHOULD implement DHCP. For specialized devices on which no such configuration need is present, DHCP may not be necessary.
An IPv6 node can use the subset of DHCP (described in [RFC3736]) to obtain other configuration information.
If an IPv6 node implements DHCP it MUST implement the DNS options [RFC3646] as most deployments will expect these options are available.
There is no defined DHCPv6 Gateway option.
Nodes using the Dynamic Host Configuration Protocol for IPv6 (DHCPv6) are thus expected to determine their default router information and on-link prefix information from received Router Advertisements.
Router Advertisement Options have historically been limited to those that are critical to basic IPv6 functionality. Originally, DNS configuration was not included as an RA option, and DHCP was the recommended way to obtain DNS configuration information. Over time, the thinking surrounding such an option has evolved. It is now generally recognized that few nodes can function adequately without having access to a working DNS resolver, and thus a Standards Track document has been published to provide this capability [RFC8106].
Implementations MUST include support for the DNS RA option [RFC8106].
In IPv6, there are two main protocol mechanisms for propagating configuration information to hosts: Router Advertisements (RAs) and DHCP. RA options have been restricted to those deemed essential for basic network functioning and for which all nodes are configured with exactly the same information. Examples include the Prefix Information Options, the MTU option, etc. On the other hand, DHCP has generally been preferred for configuration of more general parameters and for parameters that may be client-specific. Generally speaking, however, there has been a desire to define only one mechanism for configuring a given option, rather than defining multiple (different) ways of configuring the same information.
One issue with having multiple ways of configuring the same information is that interoperability suffers if a host chooses one mechanism but the network operator chooses a different mechanism. For "closed" environments, where the network operator has significant influence over what devices connect to the network and thus what configuration mechanisms they support, the operator may be able to ensure that a particular mechanism is supported by all connected hosts. In more open environments, however, where arbitrary devices may connect (e.g., a WIFI hotspot), problems can arise. To maximize interoperability in such environments, hosts would need to implement multiple configuration mechanisms to ensure interoperability.
[RFC6762] and [RFC6763] describe multicast DNS (mDNS) and DNS-Based Service Discovery (DNS-SD) respectively. These protocols, collectively commonly referred to as the 'Bonjour' protocols after their naming by Apple, provide the means for devices to discover services within a local link and, in the absence of a unicast DNS service, to exchange naming information.
Where devices are to be deployed in networks where service dicovery would be beneficial, e.g., for users seeking to discover printers or display devices, mDNS and DNS-SD SHOULD be supported.
The IETF dnssd WG is defining solutions for DNS-based service discovery in multi-link networks.
IPv6 nodes MAY support IPv4.
If an IPv6 node implements dual stack and tunneling, then [RFC4213] MUST be supported.
Software that allows users and operators to input IPv6 addresses in text form SHOULD support "A Recommendation for IPv6 Address Text Representation" [RFC5952].
There are a number of IPv6-related APIs. This document does not mandate the use of any, because the choice of API does not directly relate to on-the-wire behavior of protocols. Implementers, however, would be advised to consider providing a common API or reviewing existing APIs for the type of functionality they provide to applications.
"Basic Socket Interface Extensions for IPv6" [RFC3493] provides IPv6 functionality used by typical applications. Implementers should note that RFC3493 has been picked up and further standardized by the Portable Operating System Interface (POSIX) [POSIX].
"Advanced Sockets Application Program Interface (API) for IPv6" [RFC3542] provides access to advanced IPv6 features needed by diagnostic and other more specialized applications.
"IPv6 Socket API for Source Address Selection" [RFC5014] provides facilities that allow an application to override the default Source Address Selection rules of [RFC6724].
"Socket Interface Extensions for Multicast Source Filters" [RFC3678] provides support for expressing source filters on multicast group memberships.
"Extension to Sockets API for Mobile IPv6" [RFC4584] provides application support for accessing and enabling Mobile IPv6 [RFC6275] features.
Mobile IPv6 [RFC6275] and associated specifications [RFC3776] [RFC4877] allow a node to change its point of attachment within the Internet, while maintaining (and using) a permanent address. All communication using the permanent address continues to proceed as expected even as the node moves around. The definition of Mobile IP includes requirements for the following types of nodes:
At the present time, Mobile IP has seen only limited implementation and no significant deployment, partly because it originally assumed an IPv6-only environment rather than a mixed IPv4/IPv6 Internet. Recently, additional work has been done to support mobility in mixed-mode IPv4 and IPv6 networks [RFC5555].
More usage and deployment experience is needed with mobility before any specific approach can be recommended for broad implementation in all hosts and routers. Consequently, [RFC6275], [RFC5555], and associated standards such as [RFC4877] are considered a MAY at this time.
IPv6 for 3GPP [RFC7066] lists a snapshot of required IPv6 Functionalities at the time the document was published that would need to be implemented, going above and beyond the recommendations in this document. Additionally a 3GPP IPv6 Host MAY implement [RFC7278] for delivering IPv6 prefixes on the LAN link.
This section describes the specification for security for IPv6 nodes.
Achieving security in practice is a complex undertaking. Operational procedures, protocols, key distribution mechanisms, certificate management approaches, etc., are all components that impact the level of security actually achieved in practice. More importantly, deficiencies or a poor fit in any one individual component can significantly reduce the overall effectiveness of a particular security approach.
IPsec either can provide end-to-end security between nodes or or can provide channel security (for example, via a site-to-site IPsec VPN), making it possible to provide secure communication for all (or a subset of) communication flows at the IP layer between pairs of internet nodes. IPsec has two standard operating modes, Tunnel-mode and Transport-mode. In Tunnel-mode, IPsec provides network-layer security and protects an entire IP packet by encapsulating the orginal IP packet and then pre-pending a new IP header. In Transport-mode, IPsec provides security for the transport-layer (and above) by encapsulating only the transport-layer (and above) portion of the IP packet (i.e., without adding a 2nd IP header).
Although IPsec can be used with manual keying in some cases, such usage has limited applicability and is not recommended.
A range of security technologies and approaches proliferate today (e.g., IPsec, Transport Layer Security (TLS), Secure SHell (SSH), SSL VPNS, etc.) No one approach has emerged as an ideal technology for all needs and environments. Moreover, IPsec is not viewed as the ideal security technology in all cases and is unlikely to displace the others.
Previously, IPv6 mandated implementation of IPsec and recommended the key management approach of IKE. This document updates that recommendation by making support of the IPsec Architecture [RFC4301] a SHOULD for all IPv6 nodes. Note that the IPsec Architecture requires (e.g., Section 4.5 of RFC 4301) the implementation of both manual and automatic key management. Currently, the default automated key management protocol to implement is IKEv2 [RFC7296].
This document recognizes that there exists a range of device types and environments where approaches to security other than IPsec can be justified. For example, special-purpose devices may support only a very limited number or type of applications, and an application-specific security approach may be sufficient for limited management or configuration capabilities. Alternatively, some devices may run on extremely constrained hardware (e.g., sensors) where the full IPsec Architecture is not justified.
Because most common platforms now support IPv6 and have it enabled by default, IPv6 security is an issue for networks that are ostensibly IPv4-only; see [RFC7123] for guidance on this area.
"Security Architecture for the Internet Protocol" [RFC4301] SHOULD be supported by all IPv6 nodes. Note that the IPsec Architecture requires (e.g., Section 4.5 of [RFC4301]) the implementation of both manual and automatic key management. Currently, the default automated key management protocol to implement is IKEv2. As required in [RFC4301], IPv6 nodes implementing the IPsec Architecture MUST implement ESP [RFC4303] and MAY implement AH [RFC4302].
The current set of mandatory-to-implement algorithms for the IPsec Architecture are defined in "Cryptographic Algorithm Implementation Requirements For ESP and AH" [RFC8221]. IPv6 nodes implementing the IPsec Architecture MUST conform to the requirements in [RFC8221]. Preferred cryptographic algorithms often change more frequently than security protocols. Therefore, implementations MUST allow for migration to new algorithms, as RFC 8221 is replaced or updated in the future.
The current set of mandatory-to-implement algorithms for IKEv2 are defined in "Cryptographic Algorithms for Use in the Internet Key Exchange Version 2 (IKEv2)" [RFC8247]. IPv6 nodes implementing IKEv2 MUST conform to the requirements in [RFC8247] and/or any future updates or replacements to [RFC8247].
This section defines general host considerations for IPv6 nodes that act as routers. Currently, this section does not discuss detailed routing-specific requirements. For the case of typical home routers, [RFC7084] defines basic requirements for customer edge routers.
Further recommendations on router-specific functionality can be found in [I-D.ietf-v6ops-ipv6rtr-reqs].
The IPv6 Router Alert Option [RFC2711] is an optional IPv6 Hop-by-Hop Header that is used in conjunction with some protocols (e.g., RSVP [RFC2205] or Multicast Listener Discovery (MLDv2) [RFC3810]). The Router Alert option will need to be implemented whenever such protocols that mandate its use are implemented. See Section 5.11.
Sending Router Advertisements and processing Router Solicitations MUST be supported.
Section 7 of [RFC6275] includes some mobility-specific extensions to Neighbor Discovery. Routers SHOULD implement Sections 7.3 and 7.5, even if they do not implement Home Agent functionality.
A single DHCP server ([RFC3315] or [RFC4862]) can provide configuration information to devices directly attached to a shared link, as well as to devices located elsewhere within a site. Communication between a client and a DHCP server located on different links requires the use of DHCP relay agents on routers.
In simple deployments, consisting of a single router and either a single LAN or multiple LANs attached to the single router, together with a WAN connection, a DHCP server embedded within the router is one common deployment scenario (e.g., [RFC7084]). There is no need for relay agents in such scenarios.
In more complex deployment scenarios, such as within enterprise or service provider networks, the use of DHCP requires some level of configuration, in order to configure relay agents, DHCP servers, etc. In such environments, the DHCP server might even be run on a traditional server, rather than as part of a router.
Because of the wide range of deployment scenarios, support for DHCP server functionality on routers is optional. However, routers targeted for deployment within more complex scenarios (as described above) SHOULD support relay agent functionality. Note that "Basic Requirements for IPv6 Customer Edge Routers" [RFC7084] requires implementation of a DHCPv6 server function in IPv6 Customer Edge (CE) routers.
Forwarding nodes MUST conform to BCP 198 [RFC7608] and thus IPv6 implementations of nodes that may forward packets MUST conform to the rules specified in Section 5.1 of [RFC4632].
The target for this document is general IPv6 nodes. In the case of constrained nodes, with limited CPU, memory, bandwidth or power, support for certain IPv6 functionality may need to be considered due to those limitations. The requirements of this document are RECOMMENDED for all nodes, including constrained nodes, but compromises may need to be made in certain cases. Where such compromises are made, the interoperability of devices should be strongly considered, paticularly where this may impact other nodes on the same link, e.g., only supporting MLDv1 will affect other nodes.
The IETF 6LowPAN (IPv6 over Low Power LWPAN) WG defined six RFCs, including a general overview and problem statement ([RFC4919], the means by which IPv6 packets are transmitted over IEEE 802.15.4 networks [RFC4944] and ND optimisations for that medium [RFC6775].
If an IPv6 node is concerned about the impact of IPv6 message power consumption, it SHOULD want to implement the recommendations in [RFC7772].
Network management MAY be supported by IPv6 nodes. However, for IPv6 nodes that are embedded devices, network management may be the only possible way of controlling these nodes.
Existing network management protocols include SNMP [RFC3411], NETCONF [RFC6241] and RESTCONF [RFC8040].
[RFC8096] clarifies the obsoleted status of various IPv6-specific MIB modules.
The following two MIB modules SHOULD be supported by nodes that support a Simple Network Management Protocol (SNMP) agent.
The IP Forwarding Table MIB [RFC4292] SHOULD be supported by nodes that support an SNMP agent.
The IP MIB [RFC4293] SHOULD be supported by nodes that support an SNMP agent.
The Interface MIB [RFC2863] SHOULD be supported by nodes the support an SNMP agent.
The following YANG data models SHOULD be supported by nodes that support a NETCONF or RESTCONF agent.
The IP Management YANG Model [I-D.ietf-netmod-rfc7277bis] SHOULD be supported by nodes that support NETCONF or RESTCONF.
The Interface Management YANG Model [I-D.ietf-netmod-rfc7223bis] SHOULD be supported by nodes that support NETCONF or RESTCONF.
This document does not directly affect the security of the Internet, beyond the security considerations associated with the individual protocols.
Security is also discussed in Section 13 above.
This document does not require any IANA actions.
For this version of the IPv6 Node Requirements document, the authors would like to thank Brian Carpenter, Dave Thaler, Tom Herbert, Erik Kline, Mohamed Boucadair, and Michayla Newcombe for their contributions.
Ed Jankiewicz and Thomas Narten were named authors of the previous iteration of this document, RFC6434.
For this version of the document, the authors thanked Hitoshi Asaeda, Brian Carpenter, Tim Chown, Ralph Droms, Sheila Frankel, Sam Hartman, Bob Hinden, Paul Hoffman, Pekka Savola, Yaron Sheffer, and Dave Thaler.
Jari Arkko jari.arkko@ericsson.com Marc Blanchet marc.blanchet@viagenie.qc.ca Samita Chakrabarti samita.chakrabarti@eng.sun.com Alain Durand alain.durand@sun.com Gerard Gastaud gerard.gastaud@alcatel.fr Jun-ichiro Itojun Hagino itojun@iijlab.net Atsushi Inoue inoue@isl.rdc.toshiba.co.jp Masahiro Ishiyama masahiro@isl.rdc.toshiba.co.jp John Loughney john.loughney@nokia.com Rajiv Raghunarayan raraghun@cisco.com Shoichi Sakane shouichi.sakane@jp.yokogawa.com Dave Thaler dthaler@windows.microsoft.com Juha Wiljakka juha.wiljakka@Nokia.com
The original version of this document (RFC 4294) was written by the IPv6 Node Requirements design team:
The authors would like to thank Ran Atkinson, Jim Bound, Brian Carpenter, Ralph Droms, Christian Huitema, Adam Machalek, Thomas Narten, Juha Ollila, and Pekka Savola for their comments. Thanks to Mark Andrews for comments and corrections on DNS text. Thanks to Alfred Hoenes for tracking the updates to various RFCs.
There have been many editorial clarifications as well as significant additions and updates. While this section highlights some of the changes, readers should not rely on this section for a comprehensive list of all changes.
There have been many editorial clarifications as well as significant additions and updates. While this section highlights some of the changes, readers should not rely on this section for a comprehensive list of all changes.