NAT64 Operational Experiences
draft-ietf-v6ops-nat64-experience-04
The information below is for an old version of the document.
Document | Type |
This is an older version of an Internet-Draft that was ultimately published as RFC 7269.
|
|
---|---|---|---|
Authors | Gang Chen , Zhen Cao , Chongfeng Xie , David Binet | ||
Last updated | 2013-10-13 | ||
Replaces | draft-chen-v6ops-nat64-experience | ||
RFC stream | Internet Engineering Task Force (IETF) | ||
Formats | |||
Reviews | |||
Additional resources | Mailing list discussion | ||
Stream | WG state | WG Document | |
Document shepherd | (None) | ||
IESG | IESG state | Became RFC 7269 (Informational) | |
Consensus boilerplate | Unknown | ||
Telechat date | (None) | ||
Responsible AD | (None) | ||
Send notices to | (None) |
draft-ietf-v6ops-nat64-experience-04
Internet Engineering Task Force G. Chen Internet-Draft Z. Cao Intended status: Informational China Mobile Expires: April 17, 2014 C. Xie China Telecom D. Binet France Telecom-Orange October 14, 2013 NAT64 Operational Experiences draft-ietf-v6ops-nat64-experience-04 Abstract This document summarizes NAT64 function deployment scenarios and operational experience. Both NAT64 Carrier Grade NAT (NAT64-CGN) and NAT64 server Front End (NAT64-FE) are considered in this document. Status of This Memo 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 http://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 April 17, 2014. Copyright Notice Copyright (c) 2013 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 (http://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 Chen, et al. Expires April 17, 2014 [Page 1] Internet-Draft NAT64 Experiences October 2013 the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 3 3. NAT64 Networking Experiences . . . . . . . . . . . . . . . . 4 3.1. NAT64-CGN Consideration . . . . . . . . . . . . . . . . . 4 3.1.1. NAT64-CGN Usages . . . . . . . . . . . . . . . . . . 4 3.1.2. DNS64 Deployment . . . . . . . . . . . . . . . . . . 4 3.1.3. NAT64 Placement . . . . . . . . . . . . . . . . . . . 4 3.1.4. Co-existence of NAT64 and NAT44 . . . . . . . . . . . 5 3.2. NAT64-FE Consideration . . . . . . . . . . . . . . . . . 6 4. High Availability . . . . . . . . . . . . . . . . . . . . . . 7 4.1. Redundancy Design . . . . . . . . . . . . . . . . . . . . 7 4.2. Load Balancing . . . . . . . . . . . . . . . . . . . . . 8 5. Source Address Transparency . . . . . . . . . . . . . . . . . 9 5.1. Traceability . . . . . . . . . . . . . . . . . . . . . . 9 5.2. Geo-location . . . . . . . . . . . . . . . . . . . . . . 10 6. Quality of Experience . . . . . . . . . . . . . . . . . . . . 10 6.1. Service Reachability . . . . . . . . . . . . . . . . . . 11 6.2. Resource Reservation . . . . . . . . . . . . . . . . . . 11 7. MTU Considerations . . . . . . . . . . . . . . . . . . . . . 12 8. ULA Usages . . . . . . . . . . . . . . . . . . . . . . . . . 12 9. Security Considerations . . . . . . . . . . . . . . . . . . . 13 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 14 11. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 14 12. Additional Author List . . . . . . . . . . . . . . . . . . . 14 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 14 13.1. Normative References . . . . . . . . . . . . . . . . . . 15 13.2. Informative References . . . . . . . . . . . . . . . . . 16 Appendix A. Testing Results of Application Behavior . . . . . . 18 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 19 1. Introduction IPv6 is the only sustainable solution for numbering nodes on Internet due to the IPv4 depletion. Network operators have to deploy IPv6-only networks in order to meet the needs of the expanding internet without available IPv4 addresses. Single stack IPv6 network deployment can simplify network's provisioning. Some justifications have been described in 464xlat [RFC6877]. As an example, IPv6-only connectivity confers some benefits to mobile operators. In such mobile context, it enables the use of a single IPv6 Packet Data Protocol(PDP) context or Evolved Packet System (EPS) bearer if Long Term Evolution (LTE) network is Chen, et al. Expires April 17, 2014 [Page 2] Internet-Draft NAT64 Experiences October 2013 considered, which eliminates significant network costs caused by doubling the number of PDP contexts in some cases and the need of IPv4 addresses to be assigned to customers. In broadband networks overall, it can allow for the scaling of edge-network growth decoupled from IPv4 numbering limitations. In a transition scenario, some existing networks are likely to be IPv4-only configured for quite a long time. IPv6 networks and hosts will need to coexist with IPv4 numbered resources. Widespread dual- stack deployments have not materialized at the anticipated rate over the last 10 years, one possible conclusion being that legacy networks will not make the jump quickly. The Internet will include nodes that are dual-stack, nodes that remain IPv4-only, and nodes that can be deployed as IPv6-only nodes. A translation mechanism based on a NAT64[RFC6146] [RFC6145]function is likely to be a key element of the Internet for IPv6-IPv4 interoperability. [RFC6036] reports at least 30% of operators plan to run some kind of translator (presumably NAT64/DNS64). Advice on NAT64 deployment and operations are therefore of some importance. [RFC6586] documents the implications for IPv6 only networks. This document intends to be specific to NAT64 network planning. 2. Terminology In regards to IPv4/IPv6 translation, [RFC6144] has described a framework of enabling networks to make interworking possible between IPv4 and IPv6 networks. This document has further categorized different NAT64 function locations and use cases. The principle distinction of location is if the NAT64 is located in a Carrier Grade NAT or server Front End. The terms of NAT-CGN/FE are understood to be a topological distinction indicating different features employed in a NAT64 deployment. NAT64-CGN: A NAT64-CGN is placed in an ISP network. IPv6 subscribers leverage the NAT64-CGN to access existing IPv4 internet services. The ISP as an administrative entity takes full control on the IPv6 side, but has limited or no control on the IPv4 side. NAT64-CGN may have to consider the IPv4 Internet environment and services to make appropriate configurations. NAT64-FE: A NAT64-FE is generally a device with NAT64 functionality in a content provider or data center network. It could be for example a traffic load balancer or a firewall. The operator of the NAT64-FE has full control over the IPv4 network within the data center, but only limited influence or control over the external IPv6 network. Chen, et al. Expires April 17, 2014 [Page 3] Internet-Draft NAT64 Experiences October 2013 3. NAT64 Networking Experiences 3.1. NAT64-CGN Consideration 3.1.1. NAT64-CGN Usages Fixed network operators and mobile operators may locate NAT64 in access networks or in mobile core networks. It can be built into various devices, including routers, gateways or firewalls in order to connect IPv6 users to the IPv4 Internet. With regard to the numbers of users and the shortage of public IPv4 addresses, stateful NAT64[RFC6146] is more adapted to perform some maximal sharing of public IPv4 addresses. The usage of stateless NAT64 can be seen with better transparency features [I-D.ietf-softwire-stateless-4v6-motivation], while it has to be coordinated with A+P[RFC6346] processes as specified in [I-D.ietf-softwire-map-t]and [I-D.ietf-softwire-4rd] in order to cope with IPv4 shortage. 3.1.2. DNS64 Deployment DNS64[RFC6147] is recommended for use in combination with stateful NAT64, and will likely be an essential part of an IPv6 single-stack network that couples to the IPv4 Internet. 464xlat[RFC6877] is proposed to enable access of IPv4 only applications or applications that call IPv4 literal addresses. Using DNS64 will help 464xlat to automatically discover NAT64 prefix through [I-D.ietf-behave-nat64-discovery-heuristic]. Berkeley Internet Name Daemon (BIND) software supports the function. It&Perkins, et al. Expires September 25, 2015 [Page 27] Internet-Draft AODVv2 March 2015 length shorter than the number of bits of the address family for OrigAddr. OrigSeqNum OrigSeqNum is REQUIRED and carries the destination sequence number associated with OrigNode. TargSeqNum TargSeqNum is optional and carries a destination sequence number associated with TargNode. MetricList The MetricList data element is REQUIRED, and carries the route metric information associated with OrigAddr. MetricType The MetricType element defines the type of Metric associated with the entries in the MetricList. ValidityTimeList The ValidityTimeList is optional and carries the length of time that the sender is willing to offer a route towards OrigAddr. RREQ messages carry information about OrigAddr and TargAddr, as identified in the context of the RREQ_Gen. The OrigSeqNum MUST appear. Both MAY appear in the same RREQ when SeqNum is available for both OrigAddr and TargAddr. The OrigSeqNum data element in a RteMsg MUST apply only to OrigAddr. The other address in the AddressList is TargAddr. If the TargSeqNum data element appears, then it MUST apply only to TargAddr. The other address in the AddressList is OrigAddr. 9.1.1. RREQ Generation Upon receiving an IP packet from one of its Router Clients, it often happens that an AODVv2 router has no valid route to the destination. In this case the AODVv2 router is responsible for generating a RREQ and associated data elements on behalf of its client OrigNode. The router is referred to as RREQ_Gen. Before creating a RREQ, RREQ_Gen should check if an RREQ has recently been sent for this destination and a response is awaited, or if the limit of AODVv2 RREQ retries has been reached. In constructing the RREQ, RREQ_Gen uses AddressList, OrigSeqNum, MetricList, and optionally PrefixLengthList, TargSeqNum, MetricType, and ValidityTime. Perkins, et al. Expires September 25, 2015 [Page 28] Internet-Draft AODVv2 March 2015 RREQ_Gen follows the steps in this section. OrigAddr MUST be a unicast address. The order of data elements is illustrated schematically in Figure 1. RREQ_Gen SHOULD include TargSeqNum, if a previous value of the TargAddr's SeqNum is known (e.g. from an invalid route table entry using longest-prefix matching). If TargSeqNum is not included, AODVv2 routers handling the RREQ assume that RREQ_Gen does not have that information. 1. Set msg_hop_limit to MAX_HOPCOUNT. 2. Set msg_hop_count to zero, if including it. 3. Set AddressList := {OrigAddr, TargAddr}. 4. For the PrefixLengthList: * If OrigAddr resides on a subnet of Router Clients, set PrefixLengthList := { OrigAddr subnet's prefix, null }. * Otherwise, the PrefixLengthList is omitted. 5. For the Sequence Number List: * Increment the SeqNum as specified in Section 6.4. * Set OrigSeqNum to the new value of SeqNum. * If an Invalid route exists matching TargAddr using longest prefix matching, include TargSeqNum and set it to the sequence number on the Invalid route. Otherwise omit TargSeqNum. 6. Set MetricList := { Route[OrigAddr].Metric, null }. 7. Include the MetricType data element if requesting a route for a non-default metric type. 8. If the RREQ_Gen wishes to limit the time that the route to OrigAddr may be used, include the ValidityTime data element. 9.1.2. RREQ Reception Upon receiving an RREQ, an AODVv2 router performs the following steps. 1. A router MUST handle RREQs only from neighbors. RREQs from nodes that are not neighbors MUST be disregarded. Perkins, et al. Expires September 25, 2015 [Page 29] Internet-Draft AODVv2 March 2015 2. Check whether the sender is on the blacklist of AODVv2 routers (see Section 6.2). If not, continue processing. Otherwise, check the Blacklist Remove Time. * If Current_Time < Remove Time, ignore this RREQ for further processing. * If Current_Time >= Remove Time, remove the Blacklist entry and continue processing. 3. Verify that the message contains the required data elements: msg_hop_limit, OrigAddr, TargAddr, OrigSeqNum, OrigAddrMetric, and verify that OrigAddr and TargAddr are valid addresses (routable and unicast). If not, ignore this message for further processing. 4. If the MetricType data element is present, check that the MetricType is known. * If not, ignore this RREQ for further processing. * Otherwise continue processing . 5. Verify that OrigAddrMetric <= {MAX_METRIC[MetricType] - Cost(Link)}. * If not, ignore this RREQ for further processing. * Otherwise continue processing . 6. Process the route to OrigAddr as specified in Section 8.1. 7. Check if the message is a duplicate or redundant by comparing to entries in the RteMsg table as described in Section 8.6. * If duplicate or redundant, ignore this RREQ for further processing. * Otherwise save the information in the RteMsg table to identify future duplicates and continue processing. 8. Check if the TargAddr belongs to one of the Router Clients. * If so, generate a RREP as specified in Section 9.2.1. * Otherwise, continue to RREQ regeneration. Perkins, et al. Expires September 25, 2015 [Page 30] Internet-Draft AODVv2 March 2015 9.1.3. RREQ Regeneration Unless the router is prepared to advertise the new route, it halts processing. By sending a RREQ, a router advertises that it will forward packets to the OrigAddr contained in the RREQ according to the information enclosed. The router MAY choose not to regenerate the RREQ, though this could decrease connectivity in the network or result in non-optimal paths. The circumstances under which a router MAY choose not to regenerate a RREQ are not specified in this document. Some examples may include the router being heavily loaded and not advertising routing for more traffic, or being low on energy and having to reduce energy expended for sending AODVv2 messages or packet forwarding. The procedure for RREQ regeneration is as follows: 1. Check the msg_hop_limit. * If it is zero, do not regenerate. * Otherwise, decrement the value by one. 2. Check if msg_hop_count is present and greater than or equal to MAX_HOPCOUNT * If so, do not regenerate. * Otherwise, increment msg_hop_count by one. 3. Change OrigAddrMetric to match the route table entry for OrigAddr, which should match the advertised value in the received RREQ plus the cost of the link to the router which forwarded the RREQ. 4. If the incoming RREQ contains a ValidityTimeList, it MUST be copied into the regenerated RREQ. If not present, and the regenerating router wishes to limit the time that its route to OrigAddr may be used, set ValidityTimeList := {ValidityTime for OrigAddr, null}. If the received RREQ was unicast, the regenerated RREQ can be unicast to the next hop address of the route towards TargAddr, if known. Otherwise, the RREQ SHOULD be multicast to the LL-MANET-Routers IP and MAC address [RFC5498], [RFC4291]. Perkins, et al. Expires September 25, 2015 [Page 31] Internet-Draft AODVv2 March 2015 9.2. RREP Messages RREP messages are used to offer a route to a target address, and are sent in response to a RREQ message. RREP messages have the following general structure: +-----------------------------------------------------------------+ | msg_hop_limit, msg_hop_count | +-----------------------------------------------------------------+ | AckReq (optional) | +-----------------------------------------------------------------+ | AddressList := {OrigAddr,TargAddr} | +-----------------------------------------------------------------+ | PrefixLengthList := {null, PrefixLength for TargAddr(optional)} | +-----------------------------------------------------------------+ | TargSeqNum | +-----------------------------------------------------------------+ | MetricList := {null, metric for TargAddr} | +-----------------------------------------------------------------+ | MetricType (optional) | +-----------------------------------------------------------------+ | ValidityTimeList := {null, ValidityTime for TargAddr}(optional) | +-----------------------------------------------------------------+ Figure 2: RREP message structure RREP Data Elements msg_hop_limit The remaining number of hops allowed for dissemination of the RREP message. msg_hop_count The number of hops already traversed during dissemination of the RREP message. AckReq Acknowledgement Requested by sender (optional). AddressList AddressList contains OrigAddr and TargAddr. PrefixLengthList PrefixLengthList contains the length of the prefix for TargAddr, if TargAddr resides on a Client Network with a prefix length shorter than the number of bits of the address family for TargAddr. Perkins, et al. Expires September 25, 2015 [Page 32] Internet-Draft AODVv2 March 2015 TargSeqNum TargSeqNum is REQUIRED and carries the destination sequence number associated with TargNode. MetricList The MetricList data element is REQUIRED, and carries the route metric information associated with TargAddr. MetricType The MetricType element defines the type of Metric associated with the entries in the MetricList. ValidityTimeList The ValidityTimeList is optional and carries the length of time that the sender is willing to offer a route towards TargAddr. RREP messages carry information about OrigAddr and TargAddr, as known in the context of the RREP_Gen. The TargSeqNum MUST appear. It MUST apply only to TargAddr. The other address in the AddressList is OrigAddr. 9.2.1. RREP Generation This section specifies the generation of an RREP by an AODVv2 router (RREP_Gen) that provides connectivity for TargAddr, thus enabling the establishment of a route between OrigAddr and TargAddr. In constructing the RREP, AODVv2 uses AddressList, TargSeqNumber List, MetricList, and optionally AckReq, PrefixLengthList and/or ValidityTimeList. These elements are then used to create a RFC5444 message; see Section 10 for details. The AckReq data element indicates that an acknowledgement to the RREP has been requested. If no corresponding RREP_Ack is received within the RREP_Ack_SENT_TIMEOUT, the next hop is added to the blacklist as discussed in Section 6.2. The procedure for RREP generation is as follows: 1. Set msg_hop_limit to the msg_hop_count from the received RREQ message. 2. Set msg_hop_count, if including it, to zero. 3. Include the AckReq data element if RREP_Ack is requested from the next hop (as described in Section 6.2). 4. Include the MetricType data element and set the type accordingly. Perkins, et al. Expires September 25, 2015 [Page 33] Internet-Draft AODVv2 March 2015 5. Set the Address List := {OrigAddr, TargAddr}. 6. For the PrefixLengthList: * If TargAddr resides on a subnet of Router Clients, set PrefixLengthList := {null, TargAddr subnet's prefix}. * Otherwise, no PrefixLengthList is needed. 7. For the TargSeqNum: * RREP_Gen increments its SeqNum as specified in Section 6.4. * Set TargSeqNum := the new value of SeqNum. 8. Set MetricList := { null, Route[TargAddr].Metric }. 9. If the RREP_Gen wishes to limit the time that the route to TargAddr may be used, set ValidityTimeList := {null, TargAddr ValidityTime}. By default, the RREP is sent by unicast to the IP address of the next hop of the RREP_Gen's route to OrigAddr. 9.2.2. RREP Reception Upon receiving an RREP, an AODVv2 router performs the following steps. 1. Verify that the RREP message contains the required data elements: msg_hop_limit, OrigAddr, TargAddr, TargAddrMetric, TargSeqNum, and verify that OrigAddr and TargAddr are valid addresses (routable and unicast). If not, ignore this RREP message for further processing. 2. Check that the MetricType is known. * If not, ignore this RREP for further processing. * Otherwise continue processing . 3. Verify that TargAddrMetric <= {MAX_METRIC[MetricType] - Cost(Link)}. * If not, ignore this RREP for further processing. * Otherwise continue processing . Perkins, et al. Expires September 25, 2015 [Page 34] Internet-Draft AODVv2 March 2015 4. Process the route to TargAddr as specified in Section 8.1. 5. If the AckReq data element is present, send a RREP_Ack as specified in Section 9.4. 6. Check if the message is a duplicate or redundant by comparing to entries in the RREP table as described in Section 8.6. * If duplicate or redundant, ignore this RREP for further processing. * Otherwise save the information in the RREP table to identify future duplicates and continue processing. 7. Check if the OrigAddr belongs to one of the Router Clients. * If so, the RREP satisfies a previously sent RREQ. Processing is complete and data can now be forwarded along the route. Any packets from OrigAddr that were buffered for later delivery SHOULD be transmitted. * Otherwise, continue to RREP regeneration. 9.2.3. RREP Regeneration Similar to rules for RREQ regeneration, unless the router is prepared to advertise the route to TargAddr, it halts processing. By forwarding a RREP, a router advertises that it will forward packets to the TargAddr contained in the RREP according to the information enclosed. The router MAY choose not to regenerate the RREP, for the same reasons as mentioned under RREQ regeneration Section 9.1.3, though this could decrease connectivity in the network or result in non-optimal paths. If no valid route exists to OrigAddr, a RERR SHOULD be transmitted to TargAddr as specified in Section 9.3.1 and the RREP should not be regenerated. The procedure for RREP regeneration is as follows: 1. Check the msg_hop_limit. * If it is zero, do not regenerate. * Otherwise, decrement the value by one. 2. If msg_hop_count is present, then: Perkins, et al. Expires September 25, 2015 [Page 35] Internet-Draft AODVv2 March 2015 * If msg_hop_count >= MAX_HOPCOUNT, do not regenerate. * Otherwise, increment msg_hop_count by one. 3. The RREP SHOULD be unicast to the next hop on the route to OrigAddr. If no valid route exists to OrigAddr, a RERR SHOULD be transmitted to TargAddr as specified in Section 9.3.1. 4. Change TargAddrMetric to match the route table entry for TargAddr, which should match the advertised value in the received RREP plus the cost of the link to the router which forwarded the RREP. 5. Include the AckReq data element if this device requires acknowledgement of the RREP message. 6. If the incoming RREP contains a ValidityTimeList, it MUST be copied into the regenerated RREP. If not present, and the regenerating router wishes to limit the time that its route to TargAddr may be used, set ValidityTimeList := {null, ValidityTime for TargAddr}. The RREP SHOULD be unicast to the next hop on the route to OrigAddr. 9.3. RERR Messages An RERR message is generated by a AODVv2 router (i.e., RERR_Gen) in order to notify upstream routers that packets cannot be delivered to one or more destinations. An RERR message has the following general structure: +-----------------------------------------------------------------+ | msg_hop_limit | +-----------------------------------------------------------------+ | PktSource (optional) | +-----------------------------------------------------------------+ | RERR AddressList | +-----------------------------------------------------------------+ | PrefixLengthList for UnreachableAddresses (optional) | +-----------------------------------------------------------------+ | SeqNumList (one entry per address) | +-----------------------------------------------------------------+ | MetricType (optional) | +-----------------------------------------------------------------+ Figure 3: RERR message structure RERR Data Elements Perkins, et al. Expires September 25, 2015 [Page 36] Internet-Draft AODVv2 March 2015 msg_hop_limit The remaining number of hops allowed for dissemination of the RERR message. PktSource The IP address of the unreachable destination triggering RERR generation. If this RERR message was triggered by a broken link, the PktSource data element is not required. RERR AddressList A list of IP addresses not reachable by the AODVv2 router transmitting the RERR. PrefixLengthList PrefixLengthList contains the prefix lengths associated with the addresses in the RERR AddressList, if any of them reside on a Client Network with a prefix length shorter than the number of bits of their address family. MetricType If MetricType != DEFAULT_METRIC_TYPE, the MetricType associated with routes affected by a broken link. SeqNumList The list of sequence numbers associated with the UnreachableAddresses in the RERR AddressList. 9.3.1. RERR Generation There are two types of events which trigger generation of a RERR message. The first is the arrival of a packet for which there is no route to the destination address. This can be a packet forwarded by the routing process, or a RREP when there is no route to OrigAddr. In this case, exactly one UnreachableAddress will be included in RERR's AddressList (either the Destination Address of the IP header from a data packet, or the OrigAddr found in the AddressList of an RREP message). RERR_Gen MUST discard the packet or message that triggered generation of the RERR. The second type of event happens when a link breaks. All routes (whether valid or not) that use the broken link MUST be marked as Invalid. If the broken link was not used by any Active route, no RERR message is generated. Every Invalid route reported in the RERR MUST have the same MetricType. If the broken link affects routes to destinations that have different MetricTypes, multiple RERR messages must be generated. Perkins, et al. Expires September 25, 2015 [Page 37] Internet-Draft AODVv2 March 2015 If an AODVv2 router receives an ICMP packet to or from the address of one of its client nodes, it simply forwards the ICMP packet, and does not generate any RERR message. In constructing the RERR, AODVv2 uses MetricType, AddressList, SeqNumList, and in some cases PktSource and PrefixLengthList. These elements are then used to create a RFC5444 message; see Section 10 for details. The procedure for RERR generation is as follows: 1. Set msg_hop_limit to MAX_HOPCOUNT. 2. If the RERR was triggered by an Undeliverable Packet, the PktSource data element MUST be included, containing the source IP address of the Undeliverable Packet. 3. Include the MetricType data element if reporting a Invalid route for a non-default metric type. 4. For the RERR AddressList: * If the RERR was triggered by an undeliverable packet, insert the destination IP address of the undeliverable packet, or if the packet was a RREP, insert the OrigAddr. * If the RERR was triggered by a broken link, include the addresses of all previously Active routes which are now Invalid, up to the limit imposed by the MTU (interface "Maximum Transfer Unit") of the physical medium. If there are too many such previously Active routes, additional RERR messages should be constructed and transmitted to contain the remaining addresses. If the configuration option ENABLE_IDLE_IN_RERR is enabled, include any previously Idle routes which are now Invalid, as long as the packet size of the RERR does not exceed the MTU. 5. If there are destinations reported in the RERR AddressList that have associated subnet prefixes in the route table, insert those prefixes in the PrefixLengthList; otherwise, omit the PrefixLengthList. 6. If known, the sequence numbers associated with the routes to the addresses in the RERR AddressList SHOULD be included in the SeqNumList; otherwise, omit the SeqNumList. If the RERR is sent in response to an Undeliverable Packet: Perkins, et al. Expires September 25, 2015 [Page 38] Internet-Draft AODVv2 March 2015 o It SHOULD be sent unicast to the next hop towards the source IP address of the packet which triggered the RERR. o Otherwise the RERR MUST be sent to the multicast IP and MAC address for LL-MANET-Routers. If the RERR is sent in response to a broken link: o If precursor lists are maintained for the addresses in the RERR AddressList (see Section 12.2), the RERR SHOULD be unicast to the precursors. o Otherwise the RERR MUST be sent to the multicast IP and MAC address for LL-MANET-Routers. 9.3.2. RERR Reception Upon receiving an RERR, the following steps are performed. 1. If the message does not contain the msg_hop_limit and at least one UnreachableAddress, do not process the RERR. 2. If the MetricType data element is present, check that the MetricType is known. * If not, ignore this RERR for further processing. * Otherwise continue processing . 3. For each UnreachableAddress, * Check that the address is valid (routable and unicast). * Check that there is a valid route with the same MetricType matching the address using longest prefix matching. * Check that the route's next hop is the sender of the RERR. * Check that the route's next hop interface is the interface on which the RERR was received. * Check that the Unreachable Address SeqNum is either unknown, or is greater than the route's SeqNum. * If any of the above are false, the UnreachableAddress does not need to be advertised in a regenerated RERR. * If all of the above are true: Perkins, et al. Expires September 25, 2015 [Page 39] Internet-Draft AODVv2 March 2015 + If the route's prefix length is the same as the UnreachableAddress's prefix length, set the route state to Invalid. + If the prefix length is shorter than the original route, the route MUST be expunged from the routing table, since it is a sub-route of the larger route which is reported to be Invalid. + If the prefix length is different, create a new route with the UnreachableAddress and its prefix, and set the state to Invalid. If there are no UnreachableAddresses which need to be advertised in a regenerated RERR, take no further action. Otherwise regenerate the RERR as specified in Section 9.3.3. 9.3.3. RERR Regeneration The procedure for RERR regeneration is as follows: 1. Check the msg_hop_limit. * If it is zero, do not regenerate. * Otherwise, decrement the value by one. 2. If the PktSource data element was included in the original RERR, copy it into the regenerated RERR. 3. For the RERR AddressList, include all UnreachableAddresses which have been determined to need regeneration. 4. For the PrefixLengthList, insert the prefix lengths associated with the addresses in the RERR AddressList. 5. For the SeqNumList, include the sequence numbers corresponding to the addresses in the RERR AddressList. If the original RERR contained the PktSource data element, and a route exists to the source address, the regenerated RERR MUST be sent unicast to the next hop of the route towards PktSource. Otherwise, if precursor lists are maintained, the regenerated RERR SHOULD be sent to the active precursors of the Invalid routes as specified in Section 12.2. Perkins, et al. Expires September 25, 2015 [Page 40] Internet-Draft AODVv2 March 2015 Otherwise the regenerated RERR MUST be sent to the multicast IP and MAC address for LL-MANET-Routers. 9.4. RREP_Ack Messages RREP_Ack is modeled on the RREP_Ack message type from AODV [RFC3561]. RREP_Ack messages have the following general structure: +-----------------------------------------------------------------+ | msg_hop_limit := 1 | +-----------------------------------------------------------------+ Figure 4: RREP_Ack message structure RREP_Ack Data Elements msg_hop_limit The remaining number of hops allowed for dissemination of the RREP_Ack message. 9.4.1. RREP_Ack Generation This section specifies the generation of an RREP_Ack by an AODVv2 router. The procedure is as follows: 1. Set msg_hop_limit := 1. The RREP_Ack is sent by unicast to the IP address of router that inserted a AckReq data element into a RREP message. 9.4.2. RREP_Ack Reception Upon receiving an RREP_Ack, an AODVv2 router performs the following steps. 1. The router checks whether the sender's IP address is in the blacklist. If so, the IP address is deleted from the blacklist. 2. The router checks whether an RREP_Ack message was expected from the sending IP address, in response to an AckReq data element that the router included in a preceding RREP message as specified in Section 9.2.1. If so, the router records that the required RREP_Ack has been received and cancels the associated timeout. Perkins, et al. Expires September 25, 2015 [Page 41] Internet-Draft AODVv2 March 2015 10. Representing AODVv2 data elements using RFC 5444 AODVv2 specifies that all control plane messages between Routers SHOULD use the Generalised Mobile Ad-hoc Network Packet and Message Format [RFC5444], which provides a multiplexed transport for multiple protocols. AODVv2 therefore specifies Route Messages comprising data elements that map to message elements in RFC5444 but, in line with the concept of use, does not specify which order the messages should be arranged in an RFC5444 packet. An implementation of an RFC5444 multiplexer may choose to optimise the content of certain message elements to reduce control plane overhead. For handling of messages that contain unknown TLV types, the multiplexer SHOULD ignore the information for processing, but preserve it unmodified for forwarding. Here is a brief summary of the RFC 5444 format. 1. A packet formatted according to RFC 5444 contains zero or more messages. 2. A message contains a message header, message TLV block, and zero or more address blocks. 3. Each address block MAY also have one TLV blocks; each TLV block MAY encode any number of TLVs (including zero). Each TLV value in an Address TLV block is associated with exactly one of the addresses in the address block. The following table shows how AODVv2 data elements are represented in RFC 5444 messages. Perkins, et al. Expires September 25, 2015 [Page 42] Internet-Draft AODVv2 March 2015 +------------------------+------------------------------------------+ | Data Element | RFC 5444 Message Representation | +------------------------+------------------------------------------+ | msg_hop_limit | RFC 5444 Message Header <msg-hop-limit> | | msg_hop_count | RFC 5444 Message Header <msg-hop-count> | | AckReq | Acknowledgement Request Message TLV | | PktSource | The Packet Source Message TLV | | RteMsg AddressList | RFC 5444 Address Block | | - OrigAddr | | | - TargAddr | | | - PrefixLengthList | | | RERR AddressList | RFC 5444 Address Block | | - UnreachableAddress | | | - PrefixLengthList | | | SeqNumList | Sequence Number Address Block TLV | | - SeqNum | | | OrigSeqNum | Originating Node Sequence Number Address | | | Block TLV | | TargSeqNum | Target Node Sequence Number Address | | | Block TLV | | MetricType | Extension byte of Metric Address Block | | | TLV | | MetricList | Metric Address Block TLV | | - OrigAddrMetric | - corresponds to OrigAddr | | - TargAddrMetric | - corresponds to TargAddr | | ValidityTimeList | VALIDITY_TIME Address Block TLV | | - ValidityTime | | +------------------------+------------------------------------------+ Table 3 AODVv2 neither requires any inclusion nor uses any information from the packet header. The length of an address (32 bits for IPv4 and 128 bits for IPv6) inside an AODVv2 message is indicated by the msg- addr-length (MAL) in the msg-header. Although the addresses in an Address Block may appear in any order, each TLV value in a TLV Block is associated with exactly one Address in the Address Block. So, for instance, the ordering of the OrigAddrMetric and TargAddrMetric values in the MetricList is determined by the order of OrigAddr and TargAddr in the preceding RteMsg Address List. See Section 14.2 for more information about AODVv2 Message TLVs. See Section 14.3 for more information about AODVv2 Address Block TLVs. 11. Simple Internet Attachment Simple Internet attachment means attachment of a stub (i.e., non- transit) network of AODVv2 routers to the Internet via a single Internet AODVv2 router (called IAR). Perkins, et al. Expires September 25, 2015 [Page 43] Internet-Draft AODVv2 March 2015 As in any Internet-attached network, AODVv2 routers, and their clients, wishing to be reachable from hosts on the Internet MUST have IP addresses within the IAR's important to note that DNS64 generates the synthetic AAAA reply when services only register A records. Operators should not expect to access IPv4 parts of a dual-stack server using NAT64/DNS64. The traffic is forwarded on IPv6 paths if dual-stack servers are targeted. IPv6 traffic may be routed not going through NAT64. Only the traffic going to IPv4-only service would traverse NAT64. In some sense, it encourages IPv6 transmission and restrains NAT uses compared to NAT44(if used), on which all traffic flows have to be traversed and translated. In some cases, NAT64-CGN may serve double roles, i.e. a translator and IPv6 forwarder. Some failure cases may be occurred once NAT64 serves a IPv6 gateway while is configured only with IPv4 on WAN links. We tested on Top100 websites (referring to [Alexa] statistics) in such condition. 43% of websites are failed to be connected since those websites have both AAAA and A records. Therefore, it's recommended to enable NAT64 WAN links with dual-stack connections in such case. 3.1.3. NAT64 Placement Chen, et al. Expires April 17, 2014 [Page 4] Internet-Draft NAT64 Experiences October 2013 All connections to IPv4 services from IPv6-only clients must traverse the NAT64-CGN. It can be advantageous from the vantage-point of troubleshooting and traffic engineering to carry the IPv6 traffic natively for as long as possible within an access network and translate packets only at or near the network egress. NAT64 can be considered as a feature of the Autonomous System (AS) border in fixed networks. And, it is likely to be deployed in an IP node beyond the Gateway GPRS Support Node (GGSN) or Public Data Network- Gateway (PDN-GW) in mobile networks or directly in the gateway itself in some situations. This allows consistent attribution and traceability within the service provider network. It has been observed that the process of correlating log information is problematic from multiple- vendor's equipment due to inconsistent formats of log records. Placing NAT64 in a centralized location may reduce diversity of log format and simplify the network provisioning. Moreover, since NAT64 is only targeted at serving traffic flows from IPv6 to IPv4-only services, the user traffic volume should not be as high as in a NAT44 scenario, and therefore, the gateway's capacity in such location may not be as much of a concern or a hurdle to deployment. On the other hand, the placement in a centralized way would require more strict high availability (HA) design. It would also make geo-location based on IPv4 addresses rather inaccurate as it is currently the case for NAT44 CGN already deployed in ISP networks. More considerations or workarounds on HA and traceability could be found at Section 4 and Section 5. 3.1.4. Co-existence of NAT64 and NAT44 NAT64 could likely co-exist with NAT44 in a dual-stack network mostly because IPv4 private addresses are allocated to customers. The coexistence has already appeared in mobile networks, in which dual stack mobile phones normally initiate some dual-stack PDN/PDP Type[RFC6459] to query both IPv4/IPv6 address and IPv4 allocated addresses are very often private ones. [RFC6724] always prioritizes IPv6 connections regardless of whether the end-to-end path is native IPv6 or IPv6 translated to IPv4 via NAT64/DNS64. Conversely, Happy Eyeballs[RFC6555] will direct some IP flows across IPv4 paths. The selection of IPv4/IPv6 paths may depend on particular implementation choices or settings on a host-by-host basis, and may differ from an operator's deterministic scheme. Our tests verified that hosts may find themselves switching between IPv4 and IPv6 paths as they access identical service, but at different times [I-D.kaliwoda-sunset4-dual-ipv6-coexist]. Since the topology on each path is different, it may cause unstable user experiences and some degradation of Quality of Experience (QoE) when fallback to the other protocol is not powerful enough for example. It's also difficult for operators to debug the issue and make optimal resource usages on both NAT44 and NAT64. It's desirable to find the solutions that will Chen, et al. Expires April 17, 2014 [Page 5] Internet-Draft NAT64 Experiences October 2013 allow introducing IPv6/IPv4 translation service to IPv6-only hosts while keeping dual-stack hosts unaffected and IPv4 service unchanged. 3.2. NAT64-FE Consideration Some Internet Content Providers (ICPs) may locate NAT64 in front of an Internet Data Center (IDC), for example co-located with load balancing function. Load balancers are employed to connect different IP family domains, meanwhile distribute workloads across multiple domains or internal servers actually. In some cases, IPv4 addresses exhaustion may not be a problem in some IDC's networks. IPv6 support for some applications may require some investments and workloads so IPv6 support may not be a priority. The use of NAT64 may be served to support widespread IPv6 adoption on the Internet while maintaining IPv4-only applications access. Different strategy has been described in [RFC6883]referred to as "inside out" and "outside in". An IDC operator may implement the following practices in the NAT64-FE networking. o Some ICPs who already have satisfactory operational experiences would adopt single stack IPv6 operations to build up their data center network, servers and applications since it allows new services delivery without having to integrate consideration of IPv4 NAT and address limitations of IPv4 networks. Stateless NAT64[RFC6145]is used to provide services for IPv4-only subscribers. [I-D.anderson-siit-dc]has provided further descriptions and guidelines. o ICPs who attempt to offer customers IPv6 support in their application farms at an early stage may likely run some proxies, which are configured to handle incoming IPv6 flows and proxy them to IPv4 back-end systems. Many load balancers have already integrated some proxy functionality. IPv4 addresses configured in the proxy can be multiplexed like a stateful NAT64 performs. A similar challenge exists once increasingly numerous users in IPv6 Internet access an IPv4 network. High loads on load-balancers may be apt to cause additional latency, IPv4 pool exhaustion, etc. Therefore, this approach is only reasonable at an early stage. ICPs may learn from the experiences and move on to dual-stack or IPv6 single stack in a further stage, since the native IPv6 is always more desirable than transition solutions. [RFC6144] recommends that AAAA records of load-balancers or application servers can be directly registered in the authoritative DNS servers requiring to populate these servers with corresponding AAAA records. In this case, there is no need to deploy DNS64 servers. Those AAAA records can be some native IPv6 addresses or Chen, et al. Expires April 17, 2014 [Page 6] Internet-Draft NAT64 Experiences October 2013 some IPv4-converted IPv6 addresses[RFC6052]. The type of IPv6 address does not give the possibility to nodes to get any information about NAT64 presence on communication path and the possibility to prefer IPv4 path or the IPv6 path in dual-stack networks. Using an independent sub domain e.g. ipv6exp.xxx.xxx may help to identify experimental ipv6 services to users. How to design the FQDN for the IPv6 service is out-of-scope of this document. 4. High Availability 4.1. Redundancy Design High Availability (HA) is a major requirement for every service and network services. The deployment of redundancy mechanism is an essential approach to avoid single-point failure and significantly increase the network reliability. It's not only useful to stateful NAT64 cases, but also to stateless NAT64 gateways. Three redundancy modes are mainly used hereafter: cold standby, warm standby and hot standby. o Cold standby can't replicate the NAT64 states from the primary equipment to the backup. Administrators switch on the backup NAT64 only if the primary NAT64 fails. As the results, all the existing established sessions will be disconnected. The internal hosts are required to re-establish sessions to the external hosts. Since the backup NAT64 is manually configured to switch over to active NAT64, it may have unpredictable impacts to the ongoing services. Normally, the handover would take several minutes so as to wait for the whole process of NAT64 bootstrap loader. o Warm standby is a flavor of the cold standby mode. Backup NAT64 would keep running once the primary NAT64 is working. This makes warm standby less time consuming during the traffic failover. Virtual Router Redundancy Protocol (VRRP)[RFC5798] can be a solution to enable automatic handover in the warm standby. It was tested that the handover takes as maximum as 1 minute if the backup NAT64 needs to take over routing and re-construct the Binding Information Bases (BIBs) for 30 million sessions. In deployment phase, operators could balance loads on distinct NAT64s devices. Those NAT64s make a warm backup of each other. o Hot standby must synchronize the BIBs between the primary NAT64 and backup. When the primary NAT64 fails, backup NAT64 would take over and maintain the state of all existing sessions. The internal hosts don't have to re-connect the external hosts. The handover time has been extremely reduced. Thanks to Bidirectional Forwarding Detection (BFD) [RFC5880] combining with VRRP, a delay Chen, et al. Expires April 17, 2014 [Page 7] Internet-Draft NAT64 Experiences October 2013 of only 35ms for 30 million sessions handover was observed during testing. In some sense, it could guarantee the session continuity for every service. In order to timely transmit states information, operators may have to deploy extra transport links between primary NAT64 and distant backup. In general, cold-standby and warm-standby is simpler and less resource intensive, but it requires clients to re-establish sessions when a fail-over occurs. Hot standby doubles resource's consumption to synchronize the states, but it achieve seamless handover. The consideration of redundancy mode for stateless NAT64 is simple, because it doesn't have to consider time consuming for states maintenance. The warm standby is sufficient for stateless NAT64. In regards to stateful NAT64, it maybe useful to investigate performance tolerance of applications and the traffic characteristics in a particular network. Some testing results are shown in the Appendix A. Our statistics in a mobile network shown that almost 91.21% of amount of traffic is accounted by browsing services. Those services don't require session continuity. The handover time of warm standby is qualified to the delay tolerance. Hot-standby does not offer much benefit for those sessions on this point. In a fixed network, HTTP streaming, p2p and online games would be the major traffic[Cisco-VNI]. Consideration should be given to the importance of maintaining bindings for those sessions across failover. Operators may also consider the Average Revenue Per User (ARPU) factors to deploy suitable redundancy mode. Warm standby may still be adopted to cover most services while hot standby could be used to upgrade Quality of Experience (QoE) using DNS64 with different synthetic responses for limited traffic. Further considerations are discussed at Section 6. 4.2. Load Balancing Load balancing is used to accompany redundancy design so that better scalability and resiliency could be achieved. Stateless NAT64s allow asymmetric routing while anycast-based solutions are recommended in [I-D.ietf-softwire-map-deployment]. The deployment of load balancing may make more sense to stateful NAT64s for the sake of single-point failure avoidance. Since the NAT64-CGN and NAT64-FE have distinct facilities, the following lists the considerations for each case. o NAT64-CGN equipment doesn't implement load balancer functions on a board card. Therefore, the gateways have to resort to DNS64 or internal host's behavior. Once DNS64 is deployed, the load balancing can be performed by synthesizing AAAA response with different IPv6 prefixes. For the applications not requiring DNS Chen, et al. Expires April 17, 2014 [Page 8] Internet-Draft NAT64 Experiences October 2013 resolver, internal hosts could learn multiple IPv6 prefixes through the approaches defined in[I-D.ietf-behave-nat64-discovery-heuristic] and then select one based on a given prefix selection policy. o A dedicated Load Balancer could be deployed at front of a NAT64-FE farm. Load Balancer uses proxy mode to redirect the flows to the appropriate NAT64 instance. Stateful NAT64s require a deterministic pattern to arrange the traffic in order to ensure outbound/inbound flows traverse the identical NAT64. Therefore, static scheduling algorithms, for example source-address based policy, is preferred. A dynamic algorithm, for example Round- Robin, may have impacts on applications seeking session continuity, which described in the Table 1. 5. Source Address Transparency 5.1. Traceability Traceability is required in many cases such as identifying malicious attacks sources and accounting requirements. Operators are asked to record the NAT64 log information for specific periods of time. In our lab testing, the log information from 200,000 subscribers have been collected from a stateful NAT64 gateway for 60 days. Syslog[RFC5424] has been adopted to transmit log message from NAT64 to a log station. Each log message contains transport protocol, source IPv6 address:port, translated IPv4 address: port and timestamp. It takes almost 125 bytes long in ASCII format. It has been verified that the volume of recorded information reach up to 42.5 terabytes in the raw format and 29.07 terabytes in a compact format. Operators have to build up dedicated transport links, storage system and servers for the purpose. There are also several implementations to mitigate the issue. For example, stateful NAT64 could configure with bulk port allocation method. Once a subscriber creates the first session, a number of ports are pre-allocated. A bulk allocation message is logged indicating this allocation. Subsequent session creations will use one of the pre-allocated port and hence does not require logging. The log volume in this case may be only one thousandth of dynamic port allocation. Some implementations may adopt static port-range allocations [I-D.donley-behave-deterministic-cgn] which eliminates the need for per-subscriber logging. As a side effect, the IPv4 multiplexing efficiency is decreased regarding to those methods. For example, the utilization ratio of public IPv4 address is dropped approximately to 75% when NAT64 gateway is configured with bulk port allocation (The lab testing allocates each subscriber with 400 ports). In addition, port-range based allocation should also consider port randomization described in [RFC6056] . A trade-off among address multiplexing Chen, et al. Expires April 17, 2014 [Page 9] Internet-Draft NAT64 Experiences October 2013 efficiency, logging storage compression and port allocation complexity should be considered. More discussions could be found in [I-D.chen-sunset4-cgn-port-allocation].Basically, the decision depends on usable IPv4 resource and investments of log systems. 5.2. Geo-location IP addresses are usually used as inputs to geo-location services. The use of address sharing will prevent these systems from resolving the location of a host based on IP address alone. Applications that assume such geographic information may not work as intended. The possible solutions listed in [RFC6967] are intended to bridge the gap. However, those solutions can only provide a sub-optimal substitution to solve the problem of host identification, in particular it may not today solve problems with source identification through translation. The following lists current practices to mitigate the issue. o Operators who adopt NAT64-FE may leverage the application layer proxies, e.g. X-Forwarded-For (XFF) [I-D.ietf-appsawg-http-forwarded], to convey the IPv6 source address in HTTP headers. Those messages would be passed on to web-servers. The queried server can lookup Radius servers for the target subscribers based on IPv6 addresses included in XFF HTTP headers. XFF is the de facto standard which has been integrated in most Load Balancers. Therefore, it may be superior to use in a NAT-FE environment. In the downsides, XFF is specific to HTTP. It restricts the usages so that the solution can't be applied to requests made over HTTPs. This makes geo-location problematic for HTTPs based services. o The NAT64-CGN equipment may not implement XFF. Geo-location based on shared IPv4 address is rather inaccurate in that case. Operators could subdivide the outside IPv4 address pool so an IPv6 address can be translated depending on their geographical locations. As consequence, location information can be identified from a certain IPv4 address range. [RFC6967] also enumerates several options to reveal the host identifier. Each solution likely has their-own specific usage. For the geo-location systems relying on a Radius database[RFC5580], we have investigated to deliver NAT64 BIBs and Session Table Entrys (STEs) to a Radius server[I-D.chen-behave-nat64-radius-extension]. This method could provide geo-location system with an internal IPv6 address to identify each user. It can get along with [RFC5580] to convey original source address through same message bus. 6. Quality of Experience Chen, et al. Expires April 17, 2014 [Page 10] Internet-Draft NAT64 Experiences October 2013 6.1. Service Reachability NAT64 is providing a translation capability between IPv6 and IPv4 end-nodes. In order to provide the reachability between two IP address families, NAT64-CGN has to implement appropriate application aware functions, i.e. Application Layer Gateway (ALG), where address translation is not itself sufficient and security mechanisms do not render it infeasible. Most NAT64-CGNs mainly provide FTP- ALG[RFC6384]. NAT64-FEs may have functional richness on Load Balancer, for example HTTP-ALG, HTTPs-ALG, RTSP-ALG and SMTP-ALG have been supported. It should be noted that ALGs may impact the performance on a NAT64 box to some extent. ISPs as well as content providers might choose to avoid situations where the imposition of an ALG might be required. At the same time, it is also important to remind customers and application developers that IPv6 end-to-end usage does not require ALG imposition and therefore results in a better overall user experience. 6.2. Resource Reservation Session status normally is managed by a static timer. For example, the value of the "established connection idle-timeout" for TCP sessions must not be less than 2 hours 4 minutes[RFC5382] and 5 minutes for UDP sessions[RFC4787]. In some cases, NAT resource maybe significantly consumed by largely inactive users. The NAT translator and other customers would suffer from service degradation due to port consummation by other subscribers using the same NAT64 device. A flexible NAT session control is desirable to resolve the issues. PCP[RFC6887] could be a candidate to provide such capability. A NAT64-CGN should integrate with a PCP server, to allocate available IPv4 address/port resources. Resources could be assigned to PCP clients through PCP MAP/PEER mode. Such ability can be considered to upgrade user experiences, for example assigning different sizes of port ranges for different subscribers. Those mechanisms are also helpful to minimize terminal battery consumption and reduce the number of keep-alive messages to be sent by mobile terminal devices. Subscribers can also benefit from network reliability. It has been discussed that hot-standby offers satisfactory experience once outage of primary NAT64 is occurred. Operators may rightly be concerned about the considerable investment required for NAT64 equipment relative to low ARPU income. For example, transport links may cost much, because primary NAT64 and backup are normally located at different locations, separated by a relatively large distance. Additional maintenance has to be spent to ensure the connectivity quality. However, that may be necessary to some applications, which are delay-sensitive and seek session continuity, for example on-line games and live-streaming. Operators may be able to get added-values Chen, et al. Expires April 17, 2014 [Page 11] Internet-Draft NAT64 Experiences October 2013 from those services by offering first-class services. It can be pre- configured on the gateway to hot-standby modes depending on subscriber's profile. The rest of other sessions can be covered by cold/warm standby. 7. MTU Considerations IPv6 requires that every link in the internet have an Maximum Transmission Unit (MTU) of 1280 octets or greater[RFC2460]. However, in case of NAT64 translation deployment, some IPv4 MTU constrained link will be used in some communication path and originating IPv6 nodes may therefore receive an ICMP Packet Too Big (PTB) message, reporting a Next-Hop MTU less than 1280 bytes. The result would be that IPv6 allows packets to contain a fragmentation header, without the packet being fragmented into multiple pieces. A NAT64 would receive IPv6 packets with fragmentation header in which "M" flag equal to 0 and "Fragment Offset" equal to 0. Those packets likely impact other fragments already queued with the same set of {IPv6 Source Address, IPv6 Destination Address, Fragment Identification}. If the NAT64 box is compliant with [RFC5722], there is risk that all the fragments have to be dropped. [RFC6946] discusses how this situation could be exploited by an attacker to perform fragmentation-based attacks, and also proposes an improved handling of such packets. It required enhancements on NAT64 gateway implementations to isolate packet's processing. NAT64 should follow the recommendation and take steps to prevent the risks of fragmentation. Another approach that potentially avoids this issue is to configure IPv4 MTU more than 1260 bytes. It would forbid the occurrence of PTB smaller than 1280 bytes. Such an operational consideration is hard to universally apply to the legacy "IPv4 Internet" NAT64-CGN bridged. However, it's a feasible approach in NAT64-FE cases, since a IPv4 network NAT64-FE connected is rather well-organized and operated by a IDC operator or content provider. Therefore, the MTU of IPv4 network in NAT64-FE case are strongly recommended to set to more than 1260 bytes. 8. ULA Usages Chen, et al. Expires April 17, 2014 [Page 12] Internet-Draft NAT64 Experiences October 2013 Unique Local Addresses (ULAs) are defined in [RFC4193] to be renumbered within a network site for local communications. Operators may use ULAs as NAT64 prefixes to provide site-local IPv6 connectivity. Those ULA prefixes are stripped when the packets going to the IPv4 Internet, therefore ULAs are only valid in the IPv6 site. The use of ULAs could help in identifying the translation traffic.[I-D.ietf-v6ops-ula-usage-recommendations] has provided further guidance for the ULAs usages. We configure ULAs as NAT64 prefixes on a NAT64-CGN. If a host is only assigned with an IPv6 address and connected to NAT64-CGN, when connect to an IPv4 service, it would receive AAAA record generated by the DNS64 with the ULA prefix. A Global Unicast Address (GUA) will be selected as the source address to the ULA destination address. When the host has both IPv4 and IPv6 address, it would initiate both A and AAAA record lookup, then both original A record and DNS64-generated AAAA record would be received. A host, which is compliant with [RFC6724], will never prefer ULA over IPv4. An IPv4 path will be always selected. It may be undesirable because the NAT64-CGN will never be used. Operators may consider to add additional site-specific rows to the default table to steer traffic flows going through NAT64-CGN. However, it involves significant costs to change terminal's behavior. Therefore, operators are not suggested to configure ULAs on a NAT64-CGN. ULAs can't work when hosts transit the Internet to connect with NAT64. Therefore, ULAs are inapplicable to the case of NAT64-FE. 9. Security Considerations his document presents the deployment experiences of NAT64 in CGN and FE scenarios. In general, RFC 6146[RFC6146] provides TCP-tracking, address-dependent filtering mechanisms to protect NAT64 from Distributed Denial of Service (DDoS). In NAT64-CGN cases, operators also could adopt unicast Reverse Path Forwarding (uRPF)[RFC3704] and black/white-list to enhance the security by specifying access policies. For example, NAT64-CGN should forbid establish NAT64 BIB for incoming IPv6 packets if uRPF in Strict or Loose mode check does not pass or whose source IPv6 address is associated to black-lists. The stateful NAT64-FE creates state and maps that connection to an internally-facing IPv4 address and port. An attacker can consume the resources of the NAT64-FE device by sending an excessive number of connection attempts. Without a DDoS limitation mechanism, the NAT64-FE is exposed to attacks. Load Balancer is recommended to enable the capabilities of line rate DDOS defense, such as the employment of SYN PROXY-COOKIE. Security domain division is necessary as well in this case. Therefore, Load Balancers could not Chen, et al. Expires April 17, 2014 [Page 13] Internet-Draft NAT64 Experiences October 2013 only serve for optimization of traffic distribution, but also prevent service from quality deterioration due to security attacks. 10. IANA Considerations This memo includes no request to IANA. 11. Acknowledgements The authors would like to thank Jari Arkko, Dan Wing, Remi Despres, Fred Baker, Hui Deng, Lee Howard, Iljitsch van Beijnum, Philip Matthews, Randy Bush, Mikael Abrahamsson, Lorenzo Colitti and Sheng Jiang for their helpful comments. Many thanks to Wesley George and Satoru Matsushima for their detailed reviews. The authors especially thank Joel Jaeggli and Ray Hunter for his efforts and contributions on editing which substantially improves the legibility of the document. Thanks to Cameron Byrne who was an active co-author of some earlier versions of this draft. 12. Additional Author List The following are extended authors who contributed to the effort: Qiong Sun China Telecom Room 708, No.118, Xizhimennei Street Beijing 100035 P.R.China Phone: +86-10-58552936 Email: sunqiong@ctbri.com.cn QiBo Niu ZTE 50,RuanJian Road. YuHua District, Nan Jing 210012 P.R.China Email: niu.qibo@zte.com.cn 13. References Chen, et al. Expires April 17, 2014 [Page 14] Internet-Draft NAT64 Experiences October 2013 13.1. Normative References [I-D.ietf-appsawg-http-forwarded] Petersson, A. and M. Nilsson, "Forwarded HTTP Extension", draft-ietf-appsawg-http-forwarded-10 (work in progress), October 2012. [I-D.ietf-behave-nat64-discovery-heuristic] Savolainen, T., Korhonen, J., and D. Wing, "Discovery of the IPv6 Prefix Used for IPv6 Address Synthesis", draft- ietf-behave-nat64-discovery-heuristic-17 (work in progress), April 2013. [RFC2460] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6) Specification", RFC 2460, December 1998. [RFC3704] Baker, F. and P. Savola, "Ingress Filtering for Multihomed Networks", BCP 84, RFC 3704, March 2004. [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast Addresses", RFC 4193, October 2005. [RFC4787] Audet, F. and C. Jennings, "Network Address Translation (NAT) Behavioral Requirements for Unicast UDP", BCP 127, RFC 4787, January 2007. [RFC5382] Guha, S., Biswas, K., Ford, B., Sivakumar, S., and P. Srisuresh, "NAT Behavioral Requirements for TCP", BCP 142, RFC 5382, October 2008. [RFC5424] Gerhards, R., "The Syslog Protocol", RFC 5424, March 2009. [RFC5580] Tschofenig, H., Adrangi, F., Jones, M., Lior, A., and B. Aboba, "Carrying Location Objects in RADIUS and Diameter", RFC 5580, August 2009. [RFC5722] Krishnan, S., "Handling of Overlapping IPv6 Fragments", RFC 5722, December 2009. [RFC5798] Nadas, S., "Virtual Router Redundancy Protocol (VRRP) Version 3 for IPv4 and IPv6", RFC 5798, March 2010. [RFC5880] Katz, D. and D. Ward, "Bidirectional Forwarding Detection (BFD)", RFC 5880, June 2010. [RFC6052] Bao, C., Huitema, C., Bagnulo, M., Boucadair, M., and X. Li, "IPv6 Addressing of IPv4/IPv6 Translators", RFC 6052, October 2010. Chen, et al. Expires April 17, 2014 [Page 15] Internet-Draft NAT64 Experiences October 2013 [RFC6145] Li, X., Bao, C., and F. Baker, "IP/ICMP Translation Algorithm", RFC 6145, April 2011. [RFC6146] Bagnulo, M., Matthews, P., and I. van Beijnum, "Stateful NAT64: Network Address and Protocol Translation from IPv6 Clients to IPv4 Servers", RFC 6146, April 2011. [RFC6147] Bagnulo, M., Sullivan, A., Matthews, P., and I. van Beijnum, "DNS64: DNS Extensions for Network Address Translation from IPv6 Clients to IPv4 Servers", RFC 6147, April 2011. [RFC6384] van Beijnum, I., "An FTP Application Layer Gateway (ALG) for IPv6-to-IPv4 Translation", RFC 6384, October 2011. [RFC6555] Wing, D. and A. Yourtchenko, "Happy Eyeballs: Success with Dual-Stack Hosts", RFC 6555, April 2012. [RFC6724] Thaler, D., Draves, R., Matsumoto, A., and T. Chown, "Default Address Selection for Internet Protocol Version 6 (IPv6)", RFC 6724, September 2012. [RFC6887] Wing, D., Cheshire, S., Boucadair, M., Penno, R., and P. Selkirk, "Port Control Protocol (PCP)", RFC 6887, April 2013. [RFC6946] Gont, F., "Processing of IPv6 "Atomic" Fragments", RFC 6946, May 2013. 13.2. Informative References [Alexa] Alexa, "http://www.alexa.com/topsites", April 2013. [Cisco-VNI] Cisco, "Cisco Visual Networking Index: Forecast and Methodology, 2012-2017, http://ciscovni.com/forecast-widget/index.html", May 2013. [I-D.anderson-siit-dc] Anderson, T., "Stateless IP/ICMP Translation in IPv6 Data Centre Environments", draft-anderson-siit-dc-00 (work in progress), November 2012. [I-D.chen-behave-nat64-radius-extension] Chen, G. and D. Binet, "Radius Attributes for Stateful NAT64", draft-chen-behave-nat64-radius-extension-00 (work in progress), July 2013. Chen, et al. Expires April 17, 2014 [Page 16] Internet-Draft NAT64 Experiences October 2013 [I-D.chen-sunset4-cgn-port-allocation] Chen, G., "Analysis of NAT64 Port Allocation Method", draft-chen-sunset4-cgn-port-allocation-02 (work in progress), July 2013. [I-D.donley-behave-deterministic-cgn] Donley, C., Grundemann, C., Sarawat, V., Sundaresan, K., and O. Vautrin, "Deterministic Address Mapping to Reduce Logging in Carrier Grade NAT Deployments", draft-donley- behave-deterministic-cgn-06 (work in progress), July 2013. [I-D.ietf-softwire-4rd] Despres, R., Jiang, S., Penno, R., Lee, Y., Chen, G., and M. Chen, "IPv4 Residual Deployment via IPv6 - a Stateless Solution (4rd)", draft-ietf-softwire-4rd-07 (work in progress), October 2013. [I-D.ietf-softwire-map-deployment] Sun, Q., Chen, M., Chen, G., Tsou, T., and S. Perreault, "Mapping of Address and Port (MAP) - Deployment Considerations", draft-ietf-softwire-map-deployment-02 (work in progress), July 2013. [I-D.ietf-softwire-map-t] Li, X., Bao, C., Dec, W., Troan, O., Matsushima, S., and T. Murakami, "Mapping of Address and Port using Translation (MAP-T)", draft-ietf-softwire-map-t-04 (work in progress), September 2013. [I-D.ietf-softwire-stateless-4v6-motivation] Boucadair, M., Matsushima, S., Lee, Y., Bonness, O., Borges, I., and G. Chen, "Motivations for Carrier-side Stateless IPv4 over IPv6 Migration Solutions", draft-ietf- softwire-stateless-4v6-motivation-05 (work in progress), November 2012. [I-D.ietf-v6ops-ula-usage-recommendations] Liu, B., Jiang, S., and C. Byrne, "Recommendations of Using Unique Local Addresses", draft-ietf-v6ops-ula-usage- recommendations-00 (work in progress), May 2013. [I-D.kaliwoda-sunset4-dual-ipv6-coexist] Kaliwoda, A. and D. Binet, "Co-existence of both dual- stack and IPv6-only hosts", draft-kaliwoda-sunset4-dual- ipv6-coexist-01 (work in progress), October 2012. [RFC3484] Draves, R., "Default Address Selection for Internet Protocol version 6 (IPv6)", RFC 3484, February 2003. Chen, et al. Expires April 17, 2014 [Page 17] Internet-Draft NAT64 Experiences October 2013 [RFC6036] Carpenter, B. and S. Jiang, "Emerging Service Provider Scenarios for IPv6 Deployment", RFC 6036, October 2010. [RFC6056] Larsen, M. and F. Gont, "Recommendations for Transport- Protocol Port Randomization's routable and topologically correct prefix (e.g. 191.0.2.0/24). /-------------------------\ / +----------------+ \ / | AODVv2 Router | \ | | 191.0.2.2/32 | | | +----------------+ | Routable | +-----+--------+ Prefix | | Internet | /191.0.2/24 | | AODVv2 Router| / | | 191.0.2.1 |/ /---------------\ | | serving net +------+ Internet \ | | 191.0.2/24 | \ / | +-----+--------+ \---------------/ | +----------------+ | | | AODVv2 Router | | | | 191.0.2.3/32 | | \ +----------------+ / \ / \-------------------------/ Figure 5: Simple Internet Attachment Example When an AODVv2 router within the AODVv2 MANET wants to discover a route toward a node on the Internet, it uses the normal AODVv2 route discovery for that IP Destination Address. The IAR MUST respond to RREQ on behalf of all Internet destinations. When a packet from a node on the Internet destined for a node in the AODVv2 MANET reaches the IAR, if the IAR does not have a route toward that destination it will perform normal AODVv2 route discovery for that destination. 12. Optional Features Some optional features of AODVv2, associated with AODV, are not required by minimal implementations. These features are expected to apply in networks with greater mobility, or larger node populations, or requiring reduced latency for application launches. The optional features are as follows: o Expanding Rings Multicast o Precursor lists. Perkins, et al. Expires September 25, 2015 [Page 44] Internet-Draft AODVv2 March 2015 o Multicast RREP Response to RREQ o Intermediate RREPs (iRREPs): Without iRREP, only the destination can respond to a RREQ. o Message Aggregation Delay. 12.1. Expanding Rings Multicast For multicast RREQ, msg_hop_limit MAY be set in accordance with an expanding ring search as described in [RFC3561] to limit the RREQ propagation to a subset of the local network and possibly reduce route discovery overhead. 12.2. Precursor Lists and Notifications This section specifies an interoperable enhancement to AODVv2 (and possibly other reactive routing protocols) enabling more economical RERR notifications to traffic sources upon determination that a route needed to forward such traffic to its destination has become Invalid. 12.2.1. Overview In many circumstances, there can be several sources of traffic for a certain destination. Each such source of traffic is known as a "precursor" for the destination, as well as all upstream routers between the forwarding AODVv2 router and the traffic source. There is no need to keep track of upstream routers any farther away than the next hop. For each destination, an AODVv2 router MAY choose to keep track of the upstream neighbors that have provided traffic for that destination. Moreover, any particular link to an adjacent AODVv2 router may be a path component of multiple routes towards various destinations. The precursors for all destinations using the next hop across any link are collectively known as the precursors for that next hop. When an AODVv2 router marks a route as Invalid, the precursors of the Invalid route should be notified (using RERR) about the change in status of their route to the destination of that Invalid route. 12.2.2. Precursor Notification Details During normal operation, each AODVv2 router wishing to maintain precursor lists as described above, maintains a precursor table and updates the table whenever the node forwards traffic to one of the destinations in its route table. For each precursor in the precursor list, a record must be maintained to indicate whether the precursor Perkins, et al. Expires September 25, 2015 [Page 45] Internet-Draft AODVv2 March 2015 has been used for recent traffic (in other words, whether the precursor is an Active precursor). So, when traffic arrives from a precursor, the Current_Time is used to mark the time of last use for the precursor list element associated with that precursor. When an AODVv2 router detects that a link is broken, then for each Active precursor using that next hop, the node MAY notify the precursor using either unicast or multicast RERR: unicast RERR to each Active precursor This option is applicable when there are few Active precursors compared to the number of neighboring AODVv2 routers. multicast RERR to RERR_PRECURSORS RERR_PRECURSORS is, by default, LL-MANET-Routers [RFC5498]. This option is typically preferable when there are many precursors, since fewer packet transmissions are required. Each neighbor receiving the RERR MAY then execute the same procedure until all upstream routers have received the RERR notification. 12.3. Multicast RREP Response to RREQ The RREQ Target Router (RREP_Gen) MAY, as an alternative to unicasting a RREP, be configured to use multicast to distribute routing information about the route toward TargAddr. RREP_Gen does this as described in Section 9.2.1, but multicasting the RREP to LL- MANET-Routers [RFC5498]. Routers receiving the multicast RREP must perform RteMsg suppression (see Section 8.6). Broadcast RREP response to incoming RREQ was originally specified to handle unidirectional links, but it is expensive. Due to the significant overhead, AODVv2 routers MUST NOT use multicast RREP unless configured to do so by setting the administrative parameter USE_MULTICAST_RREP. This technique can be used to find the best return path rather than follow the same path as the RREQ took. 12.4. Intermediate RREP This specification has been published as a separate Internet Draft [I-D.perkins-irrep]. 12.5. Message Aggregation Delay The aggregation of multiple messages into a packet is specified in RFC 5444 [RFC5444]. Perkins, et al. Expires September 25, 2015 [Page 46] Internet-Draft AODVv2 March 2015 Implementations MAY choose to briefly delay transmission of messages for the purpose of aggregation (into a single packet) or to improve performance by using jitter [RFC5148]. 13. Administratively Configurable Parameters and Timer Values AODVv2 uses various configurable parameters of various types: o Timers o Protocol constants o Administrative (functional) controls o Other administrative parameters and lists The tables in the following sections show the parameters along their definitions and default values (if any). Note: several fields have limited size (bits or bytes). These sizes and their encoding may place specific limitations on the values that can be set. For example, <msg-hop-count> is a 8-bit field and therefore MAX_HOPCOUNT cannot be larger than 255. 13.1. Timers AODVv2 requires certain timing information to be associated with route table entries. The default values are as follows: +------------------------+---------------+ | Name | Default Value | +------------------------+---------------+ | ACTIVE_INTERVAL | 5 second | | MAX_IDLETIME | 200 seconds | | MAX_BLACKLIST_TIME | 200 seconds | | MAX_SEQNUM_LIFETIME | 300 seconds | | RteMsg_ENTRY_TIME | 12 seconds | | RREQ_WAIT_TIME | 2 seconds | | RREP_Ack_SENT_TIMEOUT | 1 second | | RREQ_HOLDDOWN_TIME | 10 seconds | +------------------------+---------------+ Table 4: Timing Parameter Values The above timing parameter values have worked well for small and medium well-connected networks with moderate topology changes. The timing parameters SHOULD be administratively configurable for the network where AODVv2 is used. Ideally, for networks with frequent Perkins, et al. Expires September 25, 2015 [Page 47] Internet-Draft AODVv2 March 2015 topology changes the AODVv2 parameters should be adjusted using either experimentally determined values or dynamic adaptation. For example, in networks with infrequent topology changes MAX_IDLETIME may be set to a much larger value. 13.2. Protocol Constants AODVv2 protocol constants typically do not require changes. The following table lists these constants, along with their values and a reference to the specification describing their use. +------------------------+-----------------+------------------------+ | Name | Default Value | Description | +------------------------+-----------------+------------------------+ | DISCOVERY_ATTEMPTS_MAX | 3 | Section 8.5 | | MAX_HOPCOUNT | 20 hops | Section 7 | | MAX_METRIC[i] | Specified only | Section 7 | | | for HopCount | | | MAXTIME | [TBD] | Maximum expressible | | | | clock time Section 8.4 | +------------------------+-----------------+------------------------+ Table 5: Parameter Values These values MUST have the same values for all AODVv2 routers in the ad hoc network. If the configured values are different, the following consequences may be observed: o DISCOVERY_ATTEMPTS_MAX: some nodes are likely to be more successful at finding routes, but at the cost of additional control traffic for unsuccessful attempts. o MAX_HOPCOUNT: If some nodes use a value that is too small, they would not be able to discover routes to distant addresses. o MAX_METRIC[DEFAULT_METRIC_TYPE]: MUST always be the maximum expressible metric of type DEFAULT_METRIC_TYPE. No interoperability problems due to variations on different nodes, but if a lesser value is used, route comparisons may exhibit overly restrictive behavior. o MAXTIME: Variations on different nodes would not cause problems for interoperability. If a lesser value is used, route state management may exhibit overly restrictive behavior. Perkins, et al. Expires September 25, 2015 [Page 48] Internet-Draft AODVv2 March 2015 13.3. Administrative (functional) controls The following administrative controls may be used to change the operation of the network, by enabling optional behaviors. These options are not required for correct routing behavior, although they may potentially reduce AODVv2 protocol messaging in certain situations. The default behavior is typically to NOT enable the options. Inconsistent settings at different nodes in the network will not result in protocol errors. In the case of inconsistent settings for DEFAULT_METRIC_TYPE, inconsistent setting might result in messages specifying metric types unknown to some nodes and consequent poor performance. +------------------------+------------------------------------+ | Name | Description | +------------------------+------------------------------------+ | DEFAULT_METRIC_TYPE | 3 (i.e, Hop Count (see [RFC6551])) | | ENABLE_IDLE_IN_RERR | Section 9.3.1 | | ENABLE_IRREP | Section 9.1.1 | | USE_MULTICAST_RREP | Section 12.3 | +------------------------+------------------------------------+ Table 6: Administratively Configured Controls 13.4. Other administrative parameters and lists The following table lists contains AODVv2 parameters which should be administratively configured for each node. +-----------------------+-----------------------+-----------------+ | Name | Default Value | Cross Reference | +-----------------------+-----------------------+-----------------+ | AODVv2_INTERFACES | | Section 4 | | BUFFER_SIZE_PACKETS | 2 | Section 8.5 | | BUFFER_SIZE_BYTES | MAX_PACKET_SIZE [TBD] | Section 8.5 | | CLIENT_ADDRESSES | AODVv2_INTERFACES | Section 6.3 | | CONTROL_TRAFFIC_LIMIT | TBD [50 packets/sec?] | Section 9 | +-----------------------+-----------------------+-----------------+ Table 7: Other Administrative Parameters 14. IANA Considerations This section specifies several RFC 5444 message types, message tlv- types, and address tlv-types. Also, a new registry of 16-bit alternate metric types is specified. Perkins, et al. Expires September 25, 2015 [Page 49] Internet-Draft AODVv2 March 2015 14.1. AODVv2 Message Types Specification +----------------------------------------+----------+ | Name of AODVv2 Message | Type | +----------------------------------------+----------+ | Route Request (RREQ) | 10 (TBD) | | Route Reply (RREP) | 11 (TBD) | | Route Error (RERR) | 12 (TBD) | | Route Reply Acknowledgement (RREP_Ack) | 13 (TBD) | +----------------------------------------+----------+ Table 8: AODVv2 Message Types 14.2. Message TLV Type Specification +-----------------------------+----------+----------+---------------+ | Name of Message TLV | Type | Length | Cross | | | | (octets) | Reference | +-----------------------------+----------+----------+---------------+ | AckReq (Acknowledgment | 10 (TBD) | 0 | Section 6.2 | | Request) | | | | | PktSource (Packet Source) | 11 (TBD) | 4 or 16 | Section 9.3.1 | +-----------------------------+----------+----------+---------------+ Table 9: Message TLV Types 14.3. Address Block TLV Specification +----------------------------+-----------+------------+-------------+ | Name of Address Block TLV | Type | Length | Value | +----------------------------+-----------+------------+-------------+ | Metric | 10 (TBD) | depends on | Section 9.1 | | | | MetricType | | | Sequence Number (SeqNum) | 11 (TBD) | 2 octets | Section 9.1 | | Originating Node Sequence | 12 (TBD) | 2 octets | Section 9.1 | | Number (OrigSeqNum) | | | | | Target Node Sequence | 13 (TBD) | 2 octets | Section 9.1 | | Number (TargSeqNum) | | | | | VALIDITY_TIME | 1 | 1 octet | [RFC5497] | +----------------------------+-----------+------------+-------------+ Table 10: Address Block TLV (AddrTLV) Types 14.4. MetricType Number Allocation Metric types are identified according to the assignments as specified in [RFC6551]. The metric type of the Hop Count metric is assigned to Perkins, et al. Expires September 25, 2015 [Page 50] Internet-Draft AODVv2 March 2015 be 3, in order to maintain compatibility with that existing table of values from RFC 6551. +-----------------------+----------+-------------+ | Name of MetricType | Type | Metric Size | +-----------------------+----------+-------------+ | Unallocated | 0 -- 2 | TBD | | Hop Count | 3 - TBD | 1 octet | | Unallocated | 4 -- 254 | TBD | | Reserved | 255 | Undefined | +-----------------------+----------+-------------+ Table 11: Metric Types 15. Security Considerations The objective of the AODVv2 protocol is for each router to communicate reachability information about addresses for which it is responsible. Positive routing information (i.e. a route exists) is distributed via RREQ and RREP messages. Negative routing information (i.e. a route does not exist) is distributed via RERRs. AODVv2 routers store the information contained in these messages in order to properly forward data packets, and they generally provide this information to other AODVv2 routers. This section describes various security considerations and potential avenues to secure AODVv2 routing. Security for authentication of AODVv2 routers, and/or encryption of traffic is dealt with by the underlying transport mechanism (e.g., by using the techniques for Authentication, Integrity, and Confidentiality documented in [RFC5444]). The most important security mechanism for AODVv2 routing is integrity/authentication. In situations where routing information are suspect, integrity and authentication techniques SHOULD be applied to AODVv2 messages. In these situations, routing information that is distributed over multiple hops SHOULD also verify the integrity of information based on originator of the routing information. In situations where confidentiality of AODVv2 messages is important, cryptographic techniques can be applied. In certain situations, for example sending a RREP or RERR, an AODVv2 router could include proof that it has previously received valid routing information to reach the destination, at one point of time in the past. In situations where routers are suspected of transmitting maliciously erroneous information, the original routing information along with its security credentials SHOULD be included. Perkins, et al. Expires September 25, 2015 [Page 51] Internet-Draft AODVv2 March 2015 Note that if multicast is used, any confidentiality and integrity algorithms used MUST permit multiple receivers to handle the message [RFC7182]. Routing protocols, however, are prime targets for impersonation attacks. In networks where the node membership is not known, it is difficult to determine the occurrence of impersonation attacks, and security prevention techniques are difficult at best. However, when the network membership is known and there is a danger of such attacks, AODVv2 messages must be protected by the use of authentication techniques, such as those involving generation of unforgeable and cryptographically strong message digests or digital signatures. Most AODVv2 messages are transmitted to the multicast address LL- MANET-Routers [RFC5498]. It is therefore required for security that AODVv2 neighbors exchange security information that can be used to insert an ICV [RFC7182] into the AODVv2 message block [RFC5444]. This enables hop-by-hop security. For destination-only RREP discovery procedures, AODVv2 routers that share a security association SHOULD use the appropriate mechanisms as specified in [RFC7182]. The establishment of these security associations is out of scope for this document. 16. Acknowledgments AODVv2 is a descendant of the design of previous MANET on-demand protocols, especially AODV [RFC3561] and DSR [RFC4728]. Changes to previous MANET on-demand protocols stem from research and implementation experiences. Thanks to Elizabeth Belding and Ian Chakeres for their long time authorship of AODV. Additional thanks to Derek Atkins, Emmanuel Baccelli, Abdussalam Baryun, Ramon Caceres, Thomas Clausen, Christopher Dearlove, Ulrich Herberg, Henner Jakob, Luke Klein-Berndt, Lars Kristensen, Tronje Krop, Koojana Kuladinithi, Kedar Namjoshi, Alexandru Petrescu, Henning Rogge, Fransisco Ros, Pedro Ruiz, Christoph Sommer, Lotte Steenbrink, Romain Thouvenin, Richard Trefler, Jiazi Yi, Seung Yi, and Cong Yuan, for their reviews AODVv2 and DYMO, as well as numerous specification suggestions. 17. References 17.1. Normative References [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, March 1997. [RFC4291] Hinden, R. and S. Deering, "IP Version 6 Addressing Architecture", RFC 4291, February 2006. Perkins, et al. Expires September 25, 2015 [Page 52] Internet-Draft AODVv2 March 2015 [RFC5082] Gill, V., Heasley, J., Meyer, D., Savola, P., and C. Pignataro, "The Generalized TTL Security Mechanism (GTSM)", RFC 5082, October 2007. [RFC5444] Clausen, T., Dearlove, C., Dean, J., and C. Adjih, "Generalized Mobile Ad Hoc Network (MANET) Packet/Message Format", RFC 5444, February 2009. [RFC5497] Clausen, T. and C. Dearlove, "Representing Multi-Value Time in Mobile Ad Hoc Networks (MANETs)", RFC 5497, March 2009. [RFC5498] Chakeres, I., "IANA Allocations for Mobile Ad Hoc Network (MANET) Protocols", RFC 5498, March 2009. [RFC6551] Vasseur, JP., Kim, M., Pister, K., Dejean, N., and D. Barthel, "Routing Metrics Used for Path Calculation in Low-Power and Lossy Networks", RFC 6551, March 2012. 17.2. Informative References [I-D.perkins-irrep] Perkins, C. and I. Chakeres, "Intermediate RREP for dynamic MANET On-demand (AODVv2) Routing", draft-perkins- irrep-02 (work in progress), November 2012. [Perkins94] Perkins, C. and P. Bhagwat, "Highly Dynamic Destination- Sequenced Distance-Vector Routing (DSDV) for Mobile Computers", Proceedings of the ACM SIGCOMM '94 Conference on Communications Architectures, Protocols and Applications, London, UK, pp. 234-244, August 1994. [Perkins99] Perkins, C. and E. Royer, "Ad hoc On-Demand Distance Vector (AODV) Routing", Proceedings of the 2nd IEEE Workshop on Mobile Computing Systems and Applications, New Orleans, LA, pp. 90-100, February 1999. [RFC2501] Corson, M. and J. Macker, "Mobile Ad hoc Networking (MANET): Routing Protocol Performance Issues and Evaluation Considerations", RFC 2501, January 1999. [RFC3561] Perkins, C., Belding-Royer, E., and S. Das, "Ad hoc On- Demand Distance Vector (AODV) Routing", RFC 3561, July 2003. Perkins, et al. Expires September 25, 2015 [Page 53] Internet-Draft AODVv2 March 2015 [RFC4193] Hinden, R. and B. Haberman, "Unique Local IPv6 Unicast Addresses", RFC 4193, October 2005. [RFC4728] Johnson, D., Hu, Y., and D. Maltz, "The Dynamic Source Routing Protocol (DSR) for Mobile Ad Hoc Networks for IPv4", RFC 4728, February 2007. [RFC4861] Narten, T., Nordmark, E., Simpson, W., and H. Soliman, "Neighbor Discovery for IP version 6 (IPv6)", RFC 4861, September 2007. [RFC5148] Clausen, T., Dearlove, C., and B. Adamson, "Jitter Considerations in Mobile Ad Hoc Networks (MANETs)", RFC 5148, February 2008. [RFC6130] Clausen, T., Dearlove, C., and J. Dean, "Mobile Ad Hoc Network (MANET) Neighborhood Discovery Protocol (NHDP)", RFC 6130, April 2011. [RFC6621] Macker, J., "Simplified Multicast Forwarding", RFC 6621, May 2012. [RFC7182] Herberg, U., Clausen, T., and C. Dearlove, "Integrity Check Value and Timestamp TLV Definitions for Mobile Ad Hoc Networks (MANETs)", RFC 7182, April 2014. Appendix A. Example Algorithms for AODVv2 Protocol Operations The following subsections show example algorithms for protocol operations required by AODVv2, including RREQ, RREP, RERR, and RREP_Ack. Processing for RREQ, RREP, and RERR messages follows the following general outline: 1. Receive incoming message. 2. Update route table as appropriate. 3. Respond as needed, often regenerating the incoming message with updated information. Once the route table has been updated, the information contained there is known to be the most recent available information for any fields in the outgoing message. For this reason, the algorithms are written as if outgoing message field values are assigned from the route table information, even though it is often equally appropriate to use fields from the incoming message. Perkins, et al. Expires September 25, 2015 [Page 54] Internet-Draft AODVv2 March 2015 AODVv2_algorithms: o Process_Routing_Info o Fetch_Route_Table_Entry o Update_Route_Table_Entry o Create_Route_Table_Entry o LoopFree o o Update_Rte_Msg_Table o o Generate_RREQ o Receive_RREQ o Regenerate_RREQ o o Generate_RREP o Receive_RREP o Regenerate_RREP o o Generate_RERR o Receive_RERR o Regenerate_RERR o o Generate_RREP_Ack o Receive_RREP_Ack o Timeout RREP_Ack Perkins, et al. Expires September 25, 2015 [Page 55] Internet-Draft AODVv2 March 2015 The following lists indicate the meaning of the field names used in subsequent sections to describe message processing for the above algorithms. RteMsg parameters, where rteMsg can be inRREQ, outRREQ, inRREP or outRREP: rteMsg.hopLimit rteMsg.hopCount rteMsg.ackReq (RREP only, optional) rteMsg.metricType (optional) rteMsg.origAddr rteMsg.targAddr rteMsg.origPrefixLen (optional) rteMsg.targPrefixLen (optional) rteMsg.origSeqNum (RREQ only) rteMsg.targSeqNum (optional in RREQ) rteMsg.origAddrMetric (RREQ only) rteMsg.targAddrMetric (RREP only) rteMsg.validityTime rteMsg.nbrIP AdvRte has the following properties as described in Section 8.1: AdvRte.Address = OrigAddr (in a RREQ) or TargAddr (in a RREP) AdvRte.PrefixLength = PrefixLength for OrigAddr (in a RREQ) or TargAddr (in a RREP), or if not present, the maximum address length for the address family of AdvRte.Address AdvRte.SeqNum = SeqNum for OrigAddr (in a RREQ) or for TargAddr (in a RREP) AdvRte.MetricType = RteMsg.MetricType Perkins, et al. Expires September 25, 2015 [Page 56] Internet-Draft AODVv2 March 2015 AdvRte.Metric = RteMsg.Metric AdvRte.Cost = AdvRte.Metric + Cost(L) according to the indicated MetricType, where L is the link from the advertising router AdvRte.ValidityTime = ValidityTime in the RteMsg, if present AdvRte.NextHopIP = IP source of the RteMsg AdvRte.NextHopIntf = interface the RteMsg was received on AdvRte.HopCount = value from RteMsg header AdvRte.HopLimit = value from RteMsg header AdvRte.AckReq = true/false whether present in RteMsg (optional in RREP) A route table entry has properties as described in Section 6.1: Route.Address Route.PrefixLength Route.SeqNum Route.NextHop Route.NextHopInterface Route.LastUsed Route.LastSeqNum Route.ExpirationTime Route.MetricType Route.Metric Route.State Route.Timed Route.Precursors (optional) Perkins, et al. Expires September 25, 2015 [Page 57] Internet-Draft AODVv2 March 2015 A.1. Subroutines for AODVv2 Operations A.1.1. Process_Routing_Info /* Compare incoming route information to stored route, maybe use linkMetric: either Cost(inRREQ.netif) or (inRREP.netif) */ Process_Routing_Info (advRte) { rte := Fetch_Route_Table_Entry (advRte); if (!rte exists) { rte := Create_Route_Table_Entry(advRte); return rte; } /* rule from 8.1 */ if ( (AdvRte.SeqNum > Route.SeqNum) /* stored route is stale */ OR ((AdvRte.SeqNum == Route.SeqNum) /* same SeqNum */ AND [( (Route.State == Invalid) AND (LoopFree(advRte, rte))) /* advRte can repair stored */ OR (AdvRte.Cost < Route.Metric)])) /* advRte is better */ { Update_Route_Table_Entry (rte, advRte); } return rte; } Perkins, et al. Expires September 25, 2015 [Page 58] Internet-Draft AODVv2 March 2015 A.1.2. Fetch_Route_Table_Entry /* lookup a route table entry matching an advertised route */ Fetch_Route_Table_Entry (advRte) { foreach (rteTableEntry in rteTable) { if (rteTableEntry.Address == advRte.Address AND rteTableEntry.MetricType == advRte.MetricType) return rteTableEntry; } return null; } /* lookup a route table entry matching address and metric type */ Fetch_Route_Table_Entry (destination, metricType) { foreach (rteTableEntry in rteTable) { if (rteTableEntry.Address == destination AND rteTableEntry.MetricType == metricType) return rteTableEntry; } return null; } Perkins, et al. Expires September 25, 2015 [Page 59] Internet-Draft AODVv2 March 2015 A.1.3. Update_Route_Table_Entry /* update a route table entry using AdvRte in received RteMsg */ Update_Route_Table_Entry (rte, advRte); { rte.SeqNum := advRte.SeqNum; rte.NextHop := advRte.NextHopIp; rte.NextHopInterface := advRte.NextHopIntf; rte.LastUsed := Current_Time; rte.LastSeqNum := Current_Time; if (validityTime) { rte.ExpirationTime := Current_Time + advRte.validityTime; rte.Timed := true; } else { rte.Timed := false; rte.ExpirationTime := MAXTIME; } rte.Metric := advRte.Cost; if (rte.State == Invalid) rte.State := Idle; } A.1.4. Create_Route_Table_Entry /* Create a route table entry from address and prefix length */ Create_Route_Table_Entry (address, prefixLength, seqNum, metricType) { rte := allocate_memory(); rte.Address := address; rte.PrefixLength := prefixLength; rte.SeqNum := seqNum; rte.MetricType := metricType; } Perkins, et al. Expires September 25, 2015 [Page 60] Internet-Draft AODVv2 March 2015 /* Create a route table entry from the advertised route */ Create_Route_Table_Entry(advRte) { rte := allocate_memory(); rte.Address := advRte.Address; if (advRte.PrefixLength) rte.PrefixLength := advRte.PrefixLength; else rte.PrefixLength := maxPrefixLenForAddressFamily; rte.SeqNum := advRte.SeqNum; rte.NextHop := advRte.NextHopIp; rte.NextHopInterface := advRte.NextHopIntf; rte.LastUsed := Current_Time rte.LastSeqnum := Current_Time if (validityTime) { rte.ExpirationTime := Current_Time + advRte.ValidityTime; rte.Timed := true; } else { rte.Timed := false; rte.ExpirationTime := MAXTIME; } rte.MetricType := advRte.MetricType; rte.Metric := advRte.Metric; rte.State := Idle; } A.1.5. LoopFree /* return TRUE if the route advRte is LoopFree compared to rte */ LoopFree(advRte, rte) { if (advRte.Cost <= rte.Cost) return true; else return false; } Perkins, et al. Expires September 25, 2015 [Page 61] Internet-Draft AODVv2 March 2015 A.1.6. Fetch_Rte_Msg_Table_Entry /* Find an entry in the RteMsg table matching the given message's msg-type, OrigAddr, TargAddr, MetricType */ Fetch_Rte_Msg_Table_Entry (rteMsg) { foreach (entry in RteMsgTable) { if (entry.msg-type == rteMsg.msg-type AND entry.OrigAddr == rteMsg.OrigAddr AND entry.TargAddr == rteMsg.TargAddr AND entry.MetricType == rteMsg.MetricType) { return entry; } } return NULL; } A.1.7. Update_Rte_Msg_Table /* update the multicast route message suppression table based on the received RteMsg, return true if it was created or the SeqNum was updated (i.e. it needs to be regenerated) */ Update_Rte_Msg_Table(rteMsg) { /* search for a comparable entry */ entry := Fetch_Rte_Msg_Table_Entry(rteMsg) /* if there is none, create one (see 6.5 and 8.6) */ if (entry does not exist) { entry.MessageType := rteMsg.msg_type entry.OrigAddr := rteMsg.OrigAddr entry.TargAddr := rteMsg.TargAddr entry.OrigSeqNum := rteMsg.origSeqNum (if present) entry.TargSeqNum := rteMsg.targSeqNum (if present) entry.MetricType := rteMsg.MetricType entry.Metric := rteMsg.origAddrMetric(for RREQ) or rteMsg.targAddrMetric(for RREP) entry.Timestamp := Current_Time return true; } Perkins, et al. Expires September 25, 2015 [Page 62] Internet-Draft AODVv2 March 2015 /* if current entry is stale */ if ( (rteMsg.msg-type == RREQ AND entry.OrigSeqNum < rteMsg.OrigSeqNum) OR (rteMsg.msg-type == RREP AND entry.TargSeqNum < rteMsg.TargSeqNum)) { entry.OrigSeqNum := rteMsg.OrigSeqNum (if present) entry.TargSeqNum := rteMsg.TargSeqNum (if present) entry.Timestamp := Current_Time return true; } /* if received rteMsg is stale */ if ( (rteMsg.msg-type == RREQ AND entry.OrigSeqNum > rteMsg.OrigSeqNum) OR (rteMsg.msg-type == RREP AND entry.TargSeqNum > rteMsg.TargSeqNum)) { entry.Timestamp := Current_Time return false; } /* if same SeqNum but rteMsg has lower metric */ if (entry.Metric > rteMsg.Metric) entry.Metric := rteMsg.Metric entry.Timestamp := Current_Time return false; } Perkins, et al. Expires September 25, 2015 [Page 63] Internet-Draft AODVv2 March 2015 A.1.8. Build_RFC_5444_message_header /* This pseudocode shows possible RFC 5444 actions, and would not be performed by the AODVv2 implementation. It is shown only to provide more understanding about the AODVv2 message that will be constructed by RFC 5444 */ Build_RFC_5444_message_header (msgType, Flags, AddrFamily, Size, hopLimit, hopCount, tlvLength) { /* Build RFC 5444 message header fields */ msg-type := msgType MF (Message Flags) := Flags MAL (Message Address Length) := 3 for IPv4, 15 for IPv6 msg-size := Size (octets - counting MsgHdr, AddrBlk, AddrTLVs) msg-hop-limit := hopLimit if (hopCount != 0) /* hopCount == 0 means do not include */ msg-hop-count := hopCount msg.tlvs-length := tlvLength } A.2. Example Algorithms for AODVv2 RREQ Operations A.2.1. Generate_RREQ Generate_RREQ { /* Increment sequence number */ mySeqNum := (1 + mySeqNum) /* from nonvolatile storage */ /* Marshall parameters */ outRREQ.hopLimit := MAX_HOPCOUNT /* RFC 5444 */ outRREQ.hopCount := (if included) 0 outRREQ.metricType := if not DEFAULT_METRIC_TYPE, metric type needed by application outRREQ.origAddr := IP address of Router Client which generated the packet to be forwarded outRREQ.targAddr := destination IP address in the packet to be forwarded outRREQ.origPrefixLen := if included, the prefix length associated with the Router Client outRREQ.origSeqNum := mySeqNum outRREQ.targSeqNum := if known from route table, target sequence number outRREQ.origAddrMetric := 0 (default) or MIN_METRIC(outRREQ.metricType) outRREQ.validityTime := if included, the validity time for route to OrigAddr Perkins, et al. Expires September 25, 2015 [Page 64] Internet-Draft AODVv2 March 2015 /* Build Address Blk */ AddrBlk := outRREQ.origAddr and outRREQ.targAddr addresses /* using prefix length information from outRREQ.origPrefixLen if necessary */ /* Include each available Sequence Number in appropriate Address Block TLV */ /* OrigSeqNum Address Block TLV */ origSeqNumAddrBlkTlv.value := outRREQ.origSeqNum /* TargSeqNum Address Block TLV */ if (outRREQ.targSeqNum is known) { targSeqNumAddrBlkTlv.value := outRREQ.targSeqNum } /* Build Metric Address Block TLV */ metricAddrBlkTlv.value := outRREQ.origAddrMetric if (outRREQ.metricType != DEFAULT_METRIC_TYPE) { /* include Metric AddrBlkTlv Extension byte */ metricAddrBlkTlv.typeExtension := outRREQ.MetricType } if (outRREQ.validityTime is required) { /* Build VALIDITY_TIME Address Block TLV */ VALIDITY_TIMEAddrBlkTlv.value := outRREQ.validityTime } /* multicast RFC 5444 message to LL-MANET-Routers */ } A.2.2. Receive_RREQ Receive_RREQ (inRREQ) { if (inRREQ.nbrIP present in blacklist) { if (blacklist_expiration_time < current_time) return; /* don't process or regenerate RREQ... */ else remove nbrIP from blacklist; } if (inRREQ does not contain msg_hop_limit, OrigAddr, TargAddr, OrigSeqNum, OrigAddrMetric) return; if (inRREQ.origAddr and inRREQ.targAddr are not valid routable and unicast addresses) Perkins, et al. Expires September 25, 2015 [Page 65] Internet-Draft AODVv2 March 2015 quot;, BCP 156, RFC 6056, January 2011. [RFC6144] Baker, F., Li, X., Bao, C., and K. Yin, "Framework for IPv4/IPv6 Translation", RFC 6144, April 2011. [RFC6269] Ford, M., Boucadair, M., Durand, A., Levis, P., and P. Roberts, "Issues with IP Address Sharing", RFC 6269, June 2011. [RFC6346] Bush, R., "The Address plus Port (A+P) Approach to the IPv4 Address Shortage", RFC 6346, August 2011. [RFC6459] Korhonen, J., Soininen, J., Patil, B., Savolainen, T., Bajko, G., and K. Iisakkila, "IPv6 in 3rd Generation Partnership Project (3GPP) Evolved Packet System (EPS)", RFC 6459, January 2012. [RFC6586] Arkko, J. and A. Keranen, "Experiences from an IPv6-Only Network", RFC 6586, April 2012. [RFC6877] Mawatari, M., Kawashima, M., and C. Byrne, "464XLAT: Combination of Stateful and Stateless Translation", RFC 6877, April 2013. [RFC6883] Carpenter, B. and S. Jiang, "IPv6 Guidance for Internet Content Providers and Application Service Providers", RFC 6883, March 2013. [RFC6967] Boucadair, M., Touch, J., Levis, P., and R. Penno, "Analysis of Potential Solutions for Revealing a Host Identifier (HOST_ID) in Shared Address Deployments", RFC 6967, June 2013. Appendix A. Testing Results of Application Behavior We test several application behaviors in a lab environment to evaluate the impact when a primary NAT64 is out of service. In this testing, participants are asked to connect a IPv6-only WiFi network using laptops, tablets or mobile phones. NAT64 is deployed as the gateway to connect Internet service. The tested applications are shown in the below table. Cold standby, warm standby and hot standby are taken truns to be tested. The partipants may experience service interruption due to the NAT64 handover. Different interruption Chen, et al. Expires April 17, 2014 [Page 18] Internet-Draft NAT64 Experiences October 2013 intervals are tested to gauge application behaviors. The results are illuminated as below. Table 1: The acceptable delay of applications +----------------+------------------------+-------------------------+ | APPs | Acceptable Interrupt | Session Continuity | | | Recovery | | +----------------+------------------------+-------------------------+ | Web Browse |As maximum as 6s | No | +----------------+------------------------+-------------------------+ |Http streaming |As maximum as 10s(cache)| Yes | +----------------+------------------------+-------------------------+ | Gaming | 200ms~400ms | Yes | +----------------+------------------------+-------------------------+ | P2P | 10~16s | Yes | +----------------+------------------------+-------------------------+ |Instant Message |1 minute | Yes | +----------------+------------------------+-------------------------+ |Mail |30 seconds | No | +----------------+------------------------+-------------------------+ |Downloading |1 minutes | No | +----------------+------------------------+-------------------------+ Authors' Addresses Gang Chen China Mobile 53A,Xibianmennei Ave., Xuanwu District, Beijing 100053 China Email: phdgang@gmail.com Zhen Cao China Mobile 53A,Xibianmennei Ave., Xuanwu District, Beijing 100053 China Email: caozhen@chinamobile.com Chen, et al. Expires April 17, 2014 [Page 19] Internet-Draft NAT64 Experiences October 2013 Chongfeng Xie China Telecom Room 708 No.118, Xizhimenneidajie Beijing 100035 P.R.China Email: xiechf@ctbri.com.cn David Binet France Telecom-Orange Rennes 35000 France Email: david.binet@orange.com Chen, et al. Expires April 17, 2014 [Page 20]