--- 1/draft-ietf-dmm-distributed-mobility-anchoring-02.txt 2016-12-15 10:13:14.151605430 -0800 +++ 2/draft-ietf-dmm-distributed-mobility-anchoring-03.txt 2016-12-15 10:13:14.255608009 -0800 @@ -1,132 +1,156 @@ DMM H. Chan, Ed. Internet-Draft X. Wei Intended status: Informational Huawei Technologies -Expires: March 27, 2017 J. Lee +Expires: June 17, 2017 J. Lee Sangmyung University S. Jeon Sungkyunkwan University A. Petrescu CEA, LIST F. Templin Boeing Research and Technology - September 23, 2016 + December 14, 2016 Distributed Mobility Anchoring - draft-ietf-dmm-distributed-mobility-anchoring-02 + draft-ietf-dmm-distributed-mobility-anchoring-03 Abstract - This document defines distributed mobility anchoring to meet diverse - mobility needs in 5G Wireless and beyond. Multiple anchors and nodes - with mobility functions work together to provide IP mobility support. - A network or network slice may be configured with distributed - mobility anchoring depending on the needs of mobility support. In - the distributed mobility anchoring environment, multiple anchors are - available for mid-session switching of an IP prefix anchor. Without - an ongoing session, i.e., no IP session continuity required, a flow - of a mobile node can be re-started using a new IP prefix which is - allocated from a new network of the mobile node and is therefore - anchored to the new network. With an ongoing session, the anchoring - of the prior IP prefix may be relocated to the new network to enable - IP session continuity. + This document defines distributed mobility anchoring in terms of the + different configurations, operations and parameters of mobility + functions to provide different IP mobility support for the diverse + mobility needs in 5G Wireless and beyond. A network or network slice + may be configured with distributed mobility anchoring functions + according to the needs of mobility support. In the distributed + mobility anchoring environment, multiple anchors are available for + mid-session switching of an IP prefix anchor. To start a new flow or + to handle a flow not requiring IP session continuity as a mobile node + moves to a new network, the flow can be started or re-started using a + new IP prefix which is allocated from and is therefore anchored to + the new network. For a flow requiring IP session continuity, the + anchoring of the prior IP prefix may be moved to the new network. + The mobility functions and their operations and parameters are + general for different configurations. The mobility signaling may be + between anchors and nodes in the network in a network-based mobility + solution. It may also be between the anchors and the mobile node in + a host-based solution. The mobile node may be a host, but may also + be a router carrying a network requiring network mobility support. 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 March 27, 2017. + + This Internet-Draft will expire on June 17, 2017. Copyright Notice Copyright (c) 2016 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 the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 - 2. Conventions and Terminology . . . . . . . . . . . . . . . . . 4 - 3. Distributed Mobility Anchoring . . . . . . . . . . . . . . . 6 - 3.1. Configurations for Different Networks or Network Slices . 6 + 2. Conventions and Terminology . . . . . . . . . . . . . . . . . 5 + 3. Distributed Mobility Anchoring . . . . . . . . . . . . . . . 7 + 3.1. Configurations for Different Networks or Network Slices . 7 3.1.1. Network-based Mobility Support for a Flat Network . . 7 3.1.2. Network-based Mobility Support for a Hierarchical - Network . . . . . . . . . . . . . . . . . . . . . . . 8 + Network . . . . . . . . . . . . . . . . . . . . . . . 9 3.1.3. Host-based Mobility Support . . . . . . . . . . . . . 11 3.1.4. NEtwork MObility (NEMO) Basic Support . . . . . . . . 13 3.2. Operations and Parameters . . . . . . . . . . . . . . . . 15 3.2.1. Location Management . . . . . . . . . . . . . . . . . 16 3.2.2. Forwarding Management . . . . . . . . . . . . . . . . 18 4. IP Mobility Handling in Distributed Anchoring Environments - - Mobility Support Only When Needed . . . . . . . . . . . . . . 24 - 4.1. No Need of IP Mobility: Changing to New IP Prefix/Address 25 + Mobility Support Only When Needed . . . . . . . . . . . . . . 26 + 4.1. No Need of IP Mobility: Changing to New IP Prefix/Address 27 4.1.1. Guidelines for IPv6 Nodes: Changing to New IP - Prefix/Address . . . . . . . . . . . . . . . . . . . 27 - 4.2. Need of IP Mobility . . . . . . . . . . . . . . . . . . . 28 - 4.2.1. Guidelines for IPv6 Nodes: Need of IP Mobility . . . 30 + Prefix/Address . . . . . . . . . . . . . . . . . . . 29 + 4.2. Need of IP Mobility . . . . . . . . . . . . . . . . . . . 30 + 4.2.1. Guidelines for IPv6 Nodes: Need of IP Mobility . . . 31 5. IP Mobility Handling in Distributed Mobility Anchoring - Environments - Anchor Switching to the New Network . . . . . 31 - 5.1. IP Prefix/Address Anchor Switching for Flat Network . . . 31 + Environments - Anchor Switching to the New Network . . . . . 33 + 5.1. IP Prefix/Address Anchor Switching for Flat Network . . . 33 5.1.1. Guidelines for IPv6 Nodes: Switching Anchor for Flat - Network . . . . . . . . . . . . . . . . . . . . . . . 32 + Network . . . . . . . . . . . . . . . . . . . . . . . 34 + 5.2. IP Prefix/Address Anchor Switching for Flat Network with - Centralized Control Plane . . . . . . . . . . . . . . . . 33 + Centralized Control Plane . . . . . . . . . . . . . . . . 36 5.2.1. Additional Guidelines for IPv6 Nodes: Switching - Anchor with Centralized CP . . . . . . . . . . . . . 36 - 5.3. IP Prefix/Address Anchor Switching for a Hierarchical - Network . . . . . . . . . . . . . . . . . . . . . . . . . 37 - 5.3.1. Additional Guidelines for IPv6 Nodes: No Anchoring - Change with a Hierarchical Network . . . . . . . . . 39 + Anchor with Centralized CP . . . . . . . . . . . . . 38 + 5.3. Hierarchical Network . . . . . . . . . . . . . . . . . . 39 + 5.3.1. Additional Guidelines for IPv6 Nodes: Hierarchical + Network with No Anchor Relocation . . . . . . . . . . 41 5.4. IP Prefix/Address Anchor Switching for a Hierarchical - Network . . . . . . . . . . . . . . . . . . . . . . . . . 39 + Network . . . . . . . . . . . . . . . . . . . . . . . . . 42 5.4.1. Additional Guidelines for IPv6 Nodes: Switching - Anchor with Hierarchical Network . . . . . . . . . . 41 - 6. Security Considerations . . . . . . . . . . . . . . . . . . . 41 - 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 41 - 8. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 41 - 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 42 - 9.1. Normative References . . . . . . . . . . . . . . . . . . 42 - 9.2. Informative References . . . . . . . . . . . . . . . . . 44 - Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 44 + Anchor with Hierarchical Network . . . . . . . . . . 44 + 5.5. Network Mobility . . . . . . . . . . . . . . . . . . . . 44 + 5.5.1. Additional Guidelines for IPv6 Nodes: Network + mobility . . . . . . . . . . . . . . . . . . . . . . 46 + 6. Security Considerations . . . . . . . . . . . . . . . . . . . 47 + 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 47 + 8. Contributors . . . . . . . . . . . . . . . . . . . . . . . . 47 + 9. References . . . . . . . . . . . . . . . . . . . . . . . . . 48 + 9.1. Normative References . . . . . . . . . . . . . . . . . . 48 + 9.2. Informative References . . . . . . . . . . . . . . . . . 50 + Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 50 1. Introduction A key requirement in distributed mobility management [RFC7333] is to enable traffic to avoid traversing a single mobility anchor far from an optimal route. Distributed mobility management solutions do not - make use of centrally deployed mobility anchor for a data plane + rely on a centrally deployed mobility anchor in the data plane [Paper-Distributed.Mobility]. As such, the traffic of a flow SHOULD be able to change from traversing one mobility anchor to traversing another mobility anchor as a mobile node (MN) moves, or when changing operation and management requirements call for mobility anchor - switching, thus avoiding non-optimal routes. This draft proposes - distributed mobility anchoring to enable making such route changes. + switching, thus avoiding non-optimal routes. + + Companion distributed mobility management documents are already + addressing the architecture and deployment + [I-D.ietf-dmm-deployment-models], source address selection + [I-D.ietf-dmm-ondemand-mobility], and control-plane data-plane + signaling [I-D.ietf-dmm-fpc-cpdp]. Yet in 5G Wireless and beyond, + the mobility requirements are diverse, and IP mobility support is no + longer by default with a one-size-fit-all solution. In different + networks or network slices, different kinds of mobility support are + possible depending on the needs. It may not always be obvious on how + to best configure and use only the needed mobility functions to + provide the specific mobility support. This draft defines different + configurations, functional operations and parameters for distributed + mobility anchoring and explains how to use them to make the route + changes to avoid unnecessarily long routes. Distributed mobility anchoring employs multiple anchors in the data plane. In general, control plane functions may be separate from data plane functions and be centralized but may also be co-located with the data plane functions at the distributed anchors. Different configurations of distributed mobility anchoring are described in Section 3.1. For instance, the configurations for network-based mobility support in a flat network, for network-based mobility support in a hierarchical network, for host-based mobility support, and for NEtwork MObility (NEMO) basic support are described @@ -154,21 +178,21 @@ prefix/address change, the flow can be reinitiated with a new IP address anchored in the new network as shown in Section 4.1. On the other hand, if the ongoing IP flow cannot cope with such change, mobility support is needed as shown in Section 4.2. A network or network slice supporting a mix of flows requiring and not requiring IP mobility support will need to distinguish these flows. The guidelines for such network or network slice are described in Section 4.1.1. The general guidelines for such network or network slice to provide IP mobility support are described in Section 4.2.1. - Specifically, IP mobility support can be provided by changing the + Specifically, IP mobility support can be provided by relocating the anchoring of the IP prefix/address of the flow from the home network of the flow to the new network of attachment. The basic case may be with network-based mobility for a flat network configuration described in Section 5.1 with the guidelines described in Section 5.1.1. This case is discussed further with a centralized control plane in Section 5.2 with additional guidelines described in Section 5.2.1. A level of hierarchy of nodes may then be added to the network configuration. Mobility involving change in the Data Plane Node (DPN) without changing the Data Plane Anchor (DPA) is described in Section 5.3 with additional guidelines described in @@ -261,31 +285,29 @@ 3.1. Configurations for Different Networks or Network Slices The mobility functions may be implemented in different configurations of distributed mobility anchoring in architectures separating the control and data planes. The separation described in [I-D.ietf-dmm-deployment-models] has defined the home control plane anchor (Home-CPA), home data plane anchor (Home-DPA), access control plane node (Access-CPN), and access data plane node (Access-DPN), which are respectively abbreviated as CPA, DPA, CPN, and DPN here. - Some configurations are described in - [I-D.sijeon-dmm-deployment-models]. Different networks or different network slices may have different configurations in distributed mobility anchoring. The configurations also differ depending on the desired mobility supports: network-based mobility support for a flat network in Section 3.1.1, network-based mobility support for a hierarchical - network in Section 3.1.2, host-based mobility support - (Section 3.1.3), and NEtwork MObility (NEMO) based support in + network in Section 3.1.2, host-based mobility support in + Section 3.1.3, and NEtwork MObility (NEMO) based support in Section 3.1.4. 3.1.1. Network-based Mobility Support for a Flat Network Figure 1 shows two different configurations of network-based mobility management for a flat network. (a) (b) +-----+ |LMs | @@ -536,29 +558,27 @@ +------------+ +------------+ +------------+ +------------+ +------------+ +------------+ |DPA(IPa1): | |DPA(IPa2): | |DPA(IPa1): | |DPA(IPa2): | |anchors IP1 | |anchors IP2 | |anchors IP1 | |anchors IP2 | |DHCPv6-PD | |DHCPv6-PD | ... |DHCPv6-PD | |DHCPv6-PD | ... | IPn1| | IPn2| | IPn1| | IPn2| |FM-DP | |FM-DP | |FM-DP | |FM-DP | +------------+ +------------+ +------------+ +------------+ +------------+ +------------+ - |FM-CP LMc | |FM-CP LMc | - |- - - - - - | |- - - - - - | |MR(IP1) | |MR(IP1) | |anchors IPn1| |anchors IPn1| - |FM-DP | |FM-DP | + |FM, LMc | |FM, LMc | +------------+ +------------+ +------------+ +------------+ - |MNN(IPn1) | |MR(IP1n1) | + |MNN(IPn1) | |MNN(IPn1) | |flow(IPn1,.)| |flow(IPn1,.)| +------------+ +------------+ Figure 4. Configurations of NEMO basic support for a MR. (a) FM-CP and LMs at CPA, FM-DP at DPA, FM and LMc at MR; (b) Separate LMs, FM- CP and LMp at CPA, FM-DP at DPA, FM and LMc at MR. Figure 4 shows 2 configurations of host-based mobility management for a MR with multiple instances of DPA for a distributed mobility anchoring environment. @@ -643,102 +663,116 @@ LM-cfg:1 LMs may co-locate with LMc at CPA in a flat network with network-based mobility as shown in Figure 1(a) in Section 3.1.1. LM-cfg:2 LMs may be a separate server whereas LMc is implemented in CPA in a flat network with network-based mobility as shown in Figure 1(b) in Section 3.1.1. LM-cfg:3 LMs may be implemented at CPA, whereas LMc is implemented at CPN in a hierarchical network with network-based - mobility as shown in Figure 2(a) in Section 3.1.2 or at MN + mobility as shown in Figure 2(a) in Section 3.1.2, at MN for host-based mobility as shown in Figure 3(a) in - Section 3.1.3. + Section 3.1.3, or at MR for network mobility as shown in + Figure 4(a) in Section 3.1.4. LM-cfg:4 LMs may be a separate server with LMp implemented at CPA whereas LMc is implemented at CPN in a hierarchical network with network-based mobility as shown in Figure 2(b) in - Section 3.1.2 or at MN for host-based mobility as shown in - Figure 3(b) in Section 3.1.3. + Section 3.1.2, at MN for host-based mobility as shown in + Figure 3(b) in Section 3.1.3, or at MR for network mobility + as shown in Figure 4(b) in Section 3.1.4. LM-db: LM may manage the location information in a client-server database system. Example LM database functions are as follows: LM-db:1 LMc may query LMs about location information for a prefix of MN (pull). Parameters: + - IP prefix of MN: integrity support required and privacy support may be required. LM-db:2 LMs may reply to LMc query about location information for a prefix of MN (pull). Parameters: + - IP prefix of MN: integrity support required and privacy - support may be required + support may be required, - IP address of FM-DP/DPA/DPN to forward the packets of the flow: integrity support required. LM-db:3 LMs may inform LMc about location information for a prefix of MN (push). Parameters: + - IP prefix of MN: integrity support required and privacy - support may be required + support may be required, - IP address of FM-DP/DPA/DPN to forward the packets of the - flow. + flow: integrity support required. This function in the PMIPv6 protocol is the Update Notification (UPN) together with the Update Notification Acknowledgment (UPA) as defined in [RFC7077]. LM-db:4 LMc may inform LMs about update location information for a prefix of MN. Parameters: + - IP prefix of MN: integrity support required and privacy - support may be required + support may be required, - IP address of FM-DP/DPA/DPN to forward the packets of the - flow: integrity support required + flow: integrity support required. This function in the MIPv6 / PMIPv6 protocol is the Binding Update (BU) / Proxy Binding Update (PBU) together with the Binding Acknowledgment (BA) / Proxy Binding Acknowledgment (PBA) as defined in [RFC6275] / [RFC5213] respectively. LM-db:5 The MN may be a host or a router. When the MN is an MR, the - prefix information may include the MNP delegated to the MR. + prefix information may include the IP prefix delegated to + the MR. + Additional parameters: - MNP: integrity support required and privacy support may be - required + + - IP prefix delegated to MR: integrity support required and + privacy support may be required, + - IP prefix/address of the MR to forward the packets of the + prefix delegated to the MR: integrity support required. LM-svr: The LM may be a distributed database with multiple LMs servers. For example: LM-svr:1 A LMs may join a pool of LMs servers. Parameters: - - IP address of the LMs: integrity support required + + - IP address of the LMs: integrity support required, - IP prefixes for which the LMs will host the primary location information: integrity support required. LM-svr:2 LMs may query a peer LMs about location information for a prefix of MN. Parameters: + - IP prefix: integrity support required and privacy support may be required. LM-svr:3 LMs may reply to a peer LMs about location information for a prefix of MN. Parameters: + - IP prefix of MN: integrity support required and privacy - support may be required + support may be required, - IP address of FM-DP/DPA/DPN to forward the packets of the flow: integrity support required. The parameters indicated above are only the minimal. In a specific mobility protocol, additional parameters should be added as needed. Examples of these additional parameters are those passed in the mobility options of the mobility header for MIPv6 [RFC6275] and for PMIPv6 [RFC5213]. 3.2.2. Forwarding Management @@ -762,74 +796,62 @@ FM-cfg:2 FM-CP may be implemented at both CPA and CPN and FM-DP is implemented at both DPA and DPN in a hierarchical network with network-based mobility as shown in Figure 2(a) and Figure 2(b) in Section 3.1.2. FM-cfg:3 FM-CP and FM-DP may be implemented at CPA and DPA respectively and also both implemented at MN for host-based mobility as shown in Figure 3(a) and Figure 3(b) in Section 3.1.3. + FM-cfg:4 FM-CP and FM-DP may be implemented at CPA and DPA + respectively and also both implemented at MR for network + mobility as shown in Figure 4(a) and Figure 4(b) in + Section 3.1.4. + Forwarding management operations and parameters: FM-find:1 An anchor may discover and be discovered such as through an anchor registration system as follows: FM-find:2 FM registers and authenticates itself with a centralized mobility controller. Parameters: + - IP address of DPA and its CPA: integrity support - required + required, - IP prefix anchored to the DPA: integrity support - required + required. - registration reply: acknowledge of registration and echo + Registration reply: acknowledge of registration and echo the input parameters. FM-find:3 FM discovers the FM of another IP prefix by querying the mobility controller based on the IP prefix. Parameters: + - IP prefix of MN: integrity support required and privacy - support may be required + support may be required. - FM-find:4 when making anchor discovery FM expects the answer + FM-find:4 When making anchor discovery FM expects the answer parameters: + - IP address of DPA to which IP prefix of MN is anchored: - integrity support required + integrity support required, - IP prefix of the corresponding CPA: integrity support - required + required. FM-flow:1 The FM may be carried out on the packets to/from an MN up to the granularity of a flow. FM-flow:2 Example matching parameters are in the 5-tuple of a flow. - FM-cpdp: With separation of control plane function and data plane - function, FM-CP and FM-DP communicate with each other. Such - communication may be realized by the appropriate messages in - [I-D.ietf-dmm-fpc-cpdp]. - - For example: - - FM-cpdp:1 CPA/FM-CP sends forwarding table updates to DPA/FM-DP. - Parameters: - - New forwarding table entries to add: integrity support - required - - Expired forwarding table entries to delete: integrity - support required - - FM-cpdp:2 DPA/FM-DP sends to CPA/FM-CP about its status and load. - Parameters: - - State of forwarding function being active or not: - integrity support required - - Loading percentage: integrity support required - FM-path:1 FM may change the forwarding path of a flow upon a change of point of attachment of a MN. Prior to the changes, packets coming from the CN to the MN would traverse from the CN to the home network anchor of the flow for the MN before reaching the MN. Changes are from this original forwarding path or paths to a new forwarding path or paths from the CN to the current AR of the MN and then the MN itself. FM-path:2 As an incoming packet is forwarded from the CN to the MN, @@ -850,188 +872,273 @@ anchoring is to explain the use of multiple anchors to avoid unnecessarily long route which may be encountered in centralized mobility anchoring. It is therefore not the emphasis of this document on which particular mechanism to choose from. FM-path-tbl:4 With forwarding table updates, changes to the forwarding table are needed at each of the affected forwarding switches in order to change the forwarding path of the packets for the flow from that originally - between the CN and the home network anchor to that - between the CN and the new AR. - - Forwarding table updates may be achieved through BGP - update as described in [I-D.templin-aerolink], - [I-D.mccann-dmm-flatarch] and also for 3GPP Evolved - Packet Core (EPC) network in - [I-D.matsushima-stateless-uplane-vepc] when the scope - and response time can be managed. Alternatively, a - centralized control plane may be used. + between the CN and the home network anchor or previous + AR to that between the CN and the new AR. - When the control plane is centralized, forwarding - table updates may be achieved through messaging - between the centralized control plane and the - distributed forwarding switches as described above - (FM-cpdp) in this section. + Specifically, such forwarding table updates may + include: (1) addition of forwarding table entries + needed to forward the packets destined to the MN to + the new AR; (2) deletion of forwarding table entries + to forward the packets destined to the MN to the home + network anchor or to the previous AR. - Forwarding table updates may be triggered using + FM-path-tbl:5 Forwarding table updates may be triggered using DHCPv6-PD prefix delegation to change the role of IP - anchoring from the home network anchor (with FM-DP) to - the new anchor (with FM-DP) to which the MN is - currently attached. The new anchor will then - advertise routes for the delegated prefix. + anchoring from the home network anchor or previous AR + (with FM-DP) to the new anchor (with FM-DP) to which + the MN is currently attached. The new anchor will + then advertise routes for the delegated prefix. With a distributed routing protocol, the updates spread out from neighbors to neighbors and will affect all the forwarding switches such that the packets sent from "any" node to MN will go to the new AR. - Yet the scope of such updates for a given flow may be - confined to only those forwarding switches such that - the packets sent only from the "CN" to MN will go to - the new AR. Such confinement may be made when using a - centralized central plane possessing a global view of - all the forwarding switches. + FM-path-tbl:6 Forwarding table updates may also be achieved through + BGP update as described in [I-D.templin-aerolink], + [I-D.mccann-dmm-flatarch] and also for 3GPP Evolved + Packet Core (EPC) network in + [I-D.matsushima-stateless-uplane-vepc] when the scope + and response time can be managed. - FM-path-tbl:5 FM reverts the changes previously made to the + FM-path-tbl:7 Alternatively, with a centralized control plane, + forwarding table updates may be achieved through + messaging between the centralized control plane and + the distributed forwarding switches as described above + (FM-cpdp) in this section. + + FM-path-tbl:8 To reduce excessive signaling, the scope of such + updates for a given flow may be confined to only those + forwarding switches such that only the packets sent + from the "CN" to the MN will go to the new AR. Such + confinement may be made when using a centralized + central plane possessing a global view of all the + forwarding switches. + + FM-path-tbl:9 FM reverts the changes previously made to the forwarding path of a flow when such changes are no longer needed, e.g., when all the ongoing flows using an IP prefix/address requiring IP session continuity have closed. When using DHCPv6-PD, the forwarding paths will be reverted upon prefix lease expiration. - FM-path-ind:6 Indirection forwards the incoming packets of the flow + FM-path-ind:10 Indirection forwards the incoming packets of the flow from the DPA at the far end to a DPA/DPN at the near - end of indirection. Both ends of the indirection - needs to know the LM information of the MN for the - flow and also needs to possess FM capability to - perform indirection. + end of indirection. Both ends of the indirection need + to know the LM information of the MN for the flow and + also need to possess FM capability to perform + indirection. - FM-path-ind:7 The mechanism of changing the forwarding path in - [RFC6275] and [RFC5213] is tunneling. In the control - plane, the FM-CP sets up the tunnel by instructing the - FM-DP at both ends of the tunnel. In the data plane, - the FM-DP at the start of the tunnel performs packet - encapsulation, whereas the FM-DP at the end of the - tunnel decapsulates the packet. + FM-path-ind:11 The mechanism of changing the forwarding path in MIPv6 + [RFC6275] and PMIPv6 [RFC5213] is tunneling. In the + control plane, the FM-CP sets up the tunnel by + instructing the FM-DP at both ends of the tunnel. In + the data plane, the FM-DP at the start of the tunnel + performs packet encapsulation, whereas the FM-DP at + the end of the tunnel decapsulates the packet. Note that in principle the ends of the indirection path can be any pair of network elements with the FM- DP function. - FM-path-ind:8 FM reverts the changes previously made to the + FM-path-ind:12 FM reverts the changes previously made to the forwarding path of a flow when such changes are no longer needed, e.g., when all the ongoing flows using an IP prefix/address requiring IP session continuity have closed. When tunneling is used, the tunnels will be torn down when they are no longer needed. - FM-DPA:1 Recall from above that for the incoming packets from the - CN, forwarding path change by FM is from the DPA at the far - end which may be at any forwarding switch (or even CN - itself) in the original forwarding path to the near end - DPA/DPN. + FM-cpdp: With separation of control plane function and data plane + function, FM-CP and FM-DP communicate with each other. Such + communication may be realized by the appropriate messages in + [I-D.ietf-dmm-fpc-cpdp]. - It is necessary that any incoming packet from the CN of the + For example: + + FM-cpdp:1 CPA/FM-CP sends forwarding table updates to DPA/FM-DP. + Parameters: + + - New forwarding table entries to add: integrity support + required, + + - Expired forwarding table entries to delete: integrity + support required. + + FM-cpdp:2 DPA/FM-DP sends to CPA/FM-CP about its status and load. + Parameters: + + - State of forwarding function being active or not: + integrity support required, + - Loading percentage: integrity support required. + + FM-CPA: The CPA possesses FM-CP function to make the changes to the + forwarding path as described in FM-path, and the changes may + be implemented through forwarding table changes or through + indirection as described respectively in FM-path-tbl and FM- + path-ind above. + + The FM-CP communicates with the FM-DP using the appropirate + messages in [I-D.ietf-dmm-fpc-cpdp] as described in FM-cpdp + above so that it may instruct the FM-DP to perform the + changed forwarding tasks. + + FM-DPA: The DPA possesses FM-DP function to forward packets according + to the changed forwarding path as described in FM-path, and + also FM-path-tbl or FM-path-ind depending on whether + forwarding table changes or indirection is used. + + The FM-DP communicates with the FM-CP using the appropirate + messages in [I-D.ietf-dmm-fpc-cpdp] as described in FM-cpdp + above so that it may perform the changed forwarding tasks. + + The operations and their parameters for the DPA to perform + distributed mobility management are described below: + + FM-DPA:1 The DPAs perform the needed functions such that for the + incoming packets from the CN, forwarding path change by FM + is from the DPA at the far end which may be at any + forwarding switch (or even CN itself) in the original + forwarding path to the near end DPA/DPN. + + FM-DPA:2 It is necessary that any incoming packet from the CN of the flow must traverse the DPA (or at least one of the DPAs, e.g., in the case of anycast) at the far end in order for - the packet to detour to a new forwarding path. - - Therefore a convenient design is to locate the far end DPA - at a unique location which is always in the forwarding - path. This is the case in a centralized mobility design - where the DPA at the far end is the home network anchor of - the flow. + the packet to detour to a new forwarding path. Therefore a + convenient design is to locate the far end DPA at a unique + location which is always in the forwarding path. This is + the case in a centralized mobility design where the DPA at + the far end is the home network anchor of the flow. Distributed mobility however may place the far end DPA at other locations in order to avoid unnecessarily long route. - FM-DPA:2 With multiple nodes possessing DPA capabilities, the role + FM-DPA:3 With multiple nodes possessing DPA capabilities, the role of FM to begin path change for the incoming packets of a flow at the home network DPA at the far end may be passed to or added to that of another DPA. In particular, this DPA role may be moved upstream from the home network DPA in the original forwarding path from CN to MN. - FM-DPA:3 Optimization of the new forwarding path may be achieved + FM-DPA:4 Optimization of the new forwarding path may be achieved when the path change for the incoming packets begins at a DPA where the original path and the direct IPv6 path overlaps. Then the new forwarding path will resemble the direct IPv6 path from the CN to the MN. - FM-DPA-tbl:4 Forwarding table updates, such as that triggered using - DHCPv6-PD to change the role of IP anchoring from the - home network anchor (DPA with FM-DP) to the new anchor - (DPA with FM-DP), may put the near end of the path - change at the new DPA. Subsequent forwarding table - updates may propagates upstream up to a far end where - the original path and the direct IPv6 path overlaps. + FM-DPA-tbl:5 One method to support IP mobility is through forwarding + table changes triggered using DHCPv6-PD to change the + role of IP anchoring from the home network anchor (DPA + with FM-DP) to the new anchor (DPA with FM-DP). It + therefore puts the near end of the path change at the + new DPA. Forwarding table updates will subsequently + propagate upstream from this DPA up to a far end DPA + where the original path and the direct IPv6 path + overlap. When that far end is too far upstream the signaling of forwarding table updates may become excessive. An alternative is to use indirection (see FM-DPA-ind) from that far end to the new DPA at the near end. Still another alternative is to combine forwarding table update with indirection. - FM-DPA-tbl:5 Changes made by FM to the following tables, which are + FM-DPA-tbl:6 Changes made by FM to the forwarding tables, which are IPv6 nodes, at the ends of the path change for a flow will be reverted when the mobility support for the flow is no longer needed, e.g., when the flows have terminated. - FM-DPA-ind:6 With indirection, locating or moving the FM function to + FM-DPA-ind:7 An alternative mobility support is indirection from the + far end DPA to the near end DPA. Both DPAs need to be + capable to performing indirection. For incoming + packets from the CN to the MN, the far end DPA needs to + start the indirection towards the near end DPA, which + will be the receiving end of indirection. In addition, + the near end DPA needs to continue the forwarding of + the packet towards the MN, such as through L2 + forwarding or through another indirection towards the + MN. + + FM-DPA-ind:8 With indirection, locating or moving the FM function to begin indirection upstream along the forwarding path from CN to MN again may help to reduce unnecessarily long path. - FM-DPA-ind:7 Changes made by FM to establish indirection at the DPA + FM-DPA-ind:9 Changes made by FM to establish indirection at the DPA and DPN, which are IPv6 nodes, at the ends of the path change for a flow will be reverted when the mobility support for the flow is no longer needed, e.g., when the flows have terminated. FM-state:1 In addition to the above, a flow/session may contain states with the required information for QoS, charging, etc. as needed. These states need to be transferred from the old anchor to the new anchor. - FM-buffer:1 An anchor can buffer packets of a flow in a mobility + FM-buffer: An anchor can buffer packets of a flow in a mobility event: - FM-buffer:2 CPA/FM-CP informs DPA/FM-DP to buffer packets of a flow. + + FM-buffer:1 CPA/FM-CP informs DPA/FM-DP to buffer packets of a flow. Trigger: + - MN leaves DPA in a mobility event. + Parameters: + - IP prefix of the flow for which packets need to be buffered: integrity support required - FM-buffer:3 CPA/FM-CP on behalf of a new DPA/FM-DP informs the CPA/ + FM-buffer:2 CPA/FM-CP on behalf of a new DPA/FM-DP informs the CPA/ FM-CP of the prior DPA/FM-DP that it is ready to receive any buffered packets of a flow. Parameters: + - Destination IP prefix of the flow's packets: integrity - support required - - IP address of the new DPA: integrity support required + support required, + - IP address of the new DPA: integrity support required. - FM-mr:1 When the MN is a mobile router the access router anchoring - the IP prefix of MR will also anchor the IP prefix or - prefixes delegated to the MR. + FM-mr:1 When the MN is a mobile router (MR) the access router + anchoring the IP prefix of the MR will also own the IP + prefix or prefixes to be delegated to the MR. The MNNs in + the network carried by the MR obtains IP prefixes from the + MR. + + FM-mr:2 When the MR moves from a previous AR to a new AR, the MNNs + moves with the MR. Network mobility support for these MNNs + may be provided by forwarding table updates such that + packets destined to the MNNs will be forwarded towards the + new AR instead of towards the old AR. + + Changes to forwarding table entries may occur at the new AR, + the aggregate router, and other affected switch/routers such + that packets destined to the MNNs will be forwarded to the + new AR. + + Meanwhile, changes to forwarding table entries may also + occur at the old AR, the aggregate router, and other + affected switch/routers such that packets destined to the + MNNs will not be forwarded to the old AR. 4. IP Mobility Handling in Distributed Anchoring Environments - Mobility Support Only When Needed - IP Mobility Support Only When Needed: - IP mobility support may be provided only when needed instead of being provided by default. The LM and FM functions in the different configurations shown in Section 3.1 are then utilized only when needed. A straightforward choice of mobility anchoring is for a flow to use the IP prefix of the network to which the MN is attached when the flow is initiated [I-D.seite-dmm-dma]. The IP prefix/address at the MN's side of a flow may be anchored at @@ -1056,30 +1163,30 @@ 4.1. No Need of IP Mobility: Changing to New IP Prefix/Address When IP mobility support is not needed for a flow, the LM and FM functions are not utilized so that the configurations in Section 3.1 are simplified as shown in Figure 5. Net1 Net2 +---------------+ +---------------+ -|AR1 | |AR2 | -+---------------+ +---------------+ +|AR1 | AR is changed |AR2 | ++---------------+ -------> +---------------+ |CPA: | |CPA: | |---------------| |---------------| |DPA(IPa1): | |DPA(IPa2): | |anchors IP1 | |anchors IP2 | +---------------+ +---------------+ +...............+ +---------------+ -.MN(IP1) . move |MN(IP2) | +.MN(IP1) . MN moves |MN(IP2) | .flow(IP1,...) . =======> |flow(IP2,...) | +...............+ +---------------+ Figure 5. Changing to the new IP prefix/address. MN running a flow using IP1 in a network Net1 changes to running a flow using IP2 in Net2. When there is no need to provide IP mobility to a flow, the flow may use a new IP address acquired from a new network as the MN moves to the new network. @@ -1090,69 +1197,74 @@ network. When IP session continuity is needed, even if a flow is ongoing as the MN moves, it may still be desirable for the flow to change to using the new IP prefix configured in the new network. The flow may then close and then restart using a new IP address configured in the new network. Such a change in the IP address of the flow may be enabled using a higher layer mobility support which is not in the scope of this document. - In Figure 5, a flow initiated while the MN was in a network Net1 has + In Figure 5, a flow initiated while the MN was using the IP prefix + IP1 anchored to a previous access router AR1 in network Net1 has terminated before the MN moves to a new network Net2. After moving - to Net2, the MN uses the new IP prefix anchored in Net2 to start a - new flow. The packets may then be forwarded without requiring IP - layer mobility support. + to Net2, the MN uses the new IP prefix IP2 anchored to a new access + router AR2 in network Net2 to start a new flow. The packets may then + be forwarded without requiring IP layer mobility support. An example call flow is outlined in Figure 6. - MN p-AR n-AR CN - |MN attaches to p-AR: | | | + MN AR1 AR2 CN + |MN attaches to AR1: | | | |acquire MN-ID and profile | | |--RS---------------->| | | | | | | - |<----------RA(HNP1)--| | | + |<----------RA(IP1)---| | | | | | | -Allocated prefix HNP1 -IP1 address configuration +Allocated prefix IP1 | | | +IP1 address configuration | | | | | | |<-Flow(IP1,IPcn,...)-+--------------------------------------------->| | | | | - |MN detaches from p-AR| | | - |MN attaches to n-AR | | | + |MN detaches from AR1 | | | + |MN attaches to AR2 | | | | | | | |--RS------------------------------>| | | | | | - |<--------------RA(HNP2)------------| | + |<--------------RA(IP2)-------------| | | | | | -Allocated prefix HNP2 -IP2 address configuration +Allocated prefix IP2 | | | +IP2 address configuration | | | | | | |<-new Flow(IP2,IPcn,...)-----------+------------------------------->| | | | | Figure 6. Re-starting a flow to use the IP allocated from the network at which the MN is attached. 4.1.1. Guidelines for IPv6 Nodes: Changing to New IP Prefix/Address A network or network slice may not need IP mobility support. For example, a network slice for stationary sensors only will never encounter mobility. The standard functions in IPv6 already include dropping the old IPv6 prefix/address and acquiring new IPv6 prefix/address when the node changes its point of attachment to a new network. Therefore, a network or network slice not providing IP mobility support at all will not need any of the functions with the mobility operations and messages described in Section 3.2. + On the other hand, a network or network slice supporting a mix of + flows requiring and not requiring IP mobility support will still need + the mobility functions, which it will invoke or not invoke as needed. + The guidelines for the IPv6 nodes in a network or network slice supporting a mix of flows requiring and not requiring IP mobility support include the following: GL-cfg:1 A network or network slice supporting a mix of flows requiring and not requiring mobility support may take any of the configurations described in Section 3.1 and need to implement in the appropriate IPv6 nodes the mobility functions LM and FM as described respectively in LM-cfg and FM-cfg in Section 3.2 according to the configuration @@ -1207,46 +1319,46 @@ prior network's IP address to avoid having to invoke IP mobility support. This may be the case where a dynamic IP prefix/address rather than a permanent one is used. The flow may then use the new IP prefix in the network where the flow is being initiated. Routing is again kept simpler without employing IP mobility and will remain so as long as the MN which is now in the new network has not moved again and left to another new network. An example call flow in this case is outlined in Figure 7. - MN p-AR n-AR CN - |MN attaches to p-AR: | | | + MN AR1 AR2 CN + |MN attaches to AR1: | | | |acquire MN-ID and profile | | |--RS---------------->| | | | | | | - |<----------RA(HNP1)--| | | + |<----------RA(IP1)---| | | | | | | -Allocated prefix HNP1 -IP1 address configuration +Allocated prefix IP1 | | | +IP1 address configuration | | | | | | |<-Flow(IP1,IPcn,...)-+--------------------------------------------->| | | | | - |MN detach from p-AR | | | - |MN attach to n-AR | | | + |MN detach from AR1 | | | + |MN attach to AR2 | | | | | | | |--RS------------------------------>| | IP mobility support such as that described in next sub-section - |<--------------RA(HNP2,HNP1)-------| | + |<--------------RA(IP2,IP1)---------| | | | | | |<-Flow(IP1,IPcn,...)---------------+------------------------------->| | | | | -Allocated prefix HNP2 -IP2 address configuration +Allocated prefix IP2 | | | +IP2 address configuration | | | | | | -Flow(IP1,IPcn) terminates +Flow(IP1,IPcn) terminates | | | | | | |<-new Flow(IP2,IPcn,...)-----------+------------------------------->| | | | | Figure 7. A flow continues to use the IP from its home network after MN has moved to a new network. 4.2.1. Guidelines for IPv6 Nodes: Need of IP Mobility The configuration guidelines of distributed mobility for the IPv6 @@ -1274,36 +1386,58 @@ GL-mix:5 The distributed anchors may need to message with each other. When such messaging is needed, the anchors may need to discover each other as described in the FM operations and mobility message parameters (FM-find) in Section 3.2.2. GL-mix:6 The anchors may need to provide mobility support on a per- flow basis as described in the FM operations and mobility message parameters (FM-flow) in Section 3.2.2. GL-mix:7 Then the anchors need to properly forward the packets of - the flows as described in the FM operations and mobility - message parameters (FM-path, FM-path-tbl, FM-DPA, FM-DPA- - tbl) in Section 3.2.2. + the flows in the appropriate FM operations and mobility + message parameters depending on the specific mobility + mechanism as described in Section 3.2.2. - GL-mix:8 If there are in-flight packets toward the old anchor while + GL-mix:8 When using a mechanism of changing forwarding table + entries, the FM operations and mobility message parameters + are described in FM-path, FM-path-tbl, FM-DPA, and FM-DPA- + tbl in Section 3.2.2. + + The forwarding table updates will take place at AR1, AR2, + the far end DPA, and other affected switches/routers such + that the packet from the CN to the MN will traverse from + the far end DPA towards AR2 instead of towards AR1. + + Therefore new entries to the forwarding table will be added + between AR2, the far end DPA and other affected routers so + that these packets will traverse towards AR2. Meanwhile, + changes to the forwarding table entries will also occur + between AR1, the far end DPA and other affected routers so + that if these packets ever reaches any of them, the they + will not traverse towards AR1 but will traverse towards + AR2. Section 3.2.2. + + GL-mix:9 Alternatively when using a mechanism of indirection, the FM + operations and mobility message parameters are described in + FM-path, FM-path-ind, FM-DPA, and FM-DPA-ind in + Section 3.2.2. + + GL-mix:10 If there are in-flight packets toward the old anchor while the MN is moving to the new anchor, it may be necessary to buffer these packets and then forward to the new anchor after the old anchor knows that the new anchor is ready. Such are described in the FM operations and mobility message parameters (FM-buffer) in Section 3.2.2. 5. IP Mobility Handling in Distributed Mobility Anchoring Environments - Anchor Switching to the New Network - IP Prefix/Address Anchor Switching to the New Network: - IP mobility is invoked to enable IP session continuity for an ongoing flow as the MN moves to a new network. Here the anchoring of the IP address of the flow is in the home network of the flow, which is not in the current network of attachment. A centralized mobility management mechanism may employ indirection from the anchor in the home network to the current network of attachment. Yet it may be difficult to avoid unnecessarily long route when the route between the MN and the CN via the anchor in the home network is significantly longer than the direct route between them. An alternative is to switch the IP prefix/address anchoring to the new network. @@ -1315,27 +1449,27 @@ 1(a) and 1(b) in Section 3.1 are implemented as shown in Figure 8. Net1 Net2 +---------------+ +---------------+ |AR1 | |AR2 | +---------------+ +---------------+ |CPA: | |CPA: | |LM:IP1<-->IPa2 | |LM:IP1<-->IPa2 | |---------------| |---------------| -|DPA(IPa1): | |DPA(IPa2): | -|anchors IP1 | move |anchors IP2,IP1| -|FM:DHCPv6-PD | =======> |FM:DHCPv6-PD | +|DPA(IPa1): | anchoring of IP1 is moved |DPA(IPa2): | +|anchored IP1 | =======> |anchors IP2,IP1| +|FM:DHCPv6-PD | |FM:DHCPv6-PD | +---------------+ +---------------+ +...............+ +---------------+ -.MN(IP1) . move |MN(IP2,IP1) | +.MN(IP1) . MN moves |MN(IP2,IP1) | .flow(IP1,...) . =======> |flow(IP1,...) | +...............+ +---------------+ Figure 8. IP prefix/address anchor switching to the new network. MN with flow using IP1 in Net1 continues to run the flow using IP1 as it moves to Net2. As an MN with an ongoing session moves to a new network, the flow may preserve IP session continuity by moving the anchoring of the original IP prefix/address of the flow to the new network. BGP @@ -1370,33 +1504,33 @@ IP prefix from an AR in the original network is being used by a flow in which AR in a new network. Such information needs to be deleted or updated when such flows have closed so that the IP prefix is no longer used in a different network. The LM operations are described in Section 3.2.1. GL-switch:2 The FM functions are implemented through the DHCPv6-PD protocol. Here the anchor operations to properly forward the packets for a flow as described in the FM - operations and mobility message parameters in - Section 3.2.2 FM-path, FM-path-tbl, FM-DPA, FM-DPA-tbl - are realized by changing the anchor with DHCPv6-PD and - also by reverting such changes later after the - application has already closed and when the DHCPv6-PD - timer expires. If there are in-flight packets toward - the old anchor while the MN is moving to the new anchor, - it may be necessary to buffer these packets and then - forward to the new anchor after the old anchor knows - that the new anchor is ready as are described in - Section 3.2.2 (FM-buffer). The anchors may also need to - discover each other as described also in the FM - operations and mobility message parameters (FM-find). + operations and mobility message parameters in FM-path, + FM-path-tbl, FM-DPA, and FM-DPA-tbl in Section 3.2.2 are + realized by changing the anchor with DHCPv6-PD and also + by reverting such changes later after the application + has already closed and when the DHCPv6-PD timer expires. + If there are in-flight packets toward the old anchor + while the MN is moving to the new anchor, it may be + necessary to buffer these packets and then forward to + the new anchor after the old anchor knows that the new + anchor is ready as are described in FM-buffer in + Section 3.2.2. The anchors may also need to discover + each other as described also in the FM operations and + mobility message parameters (FM-find). GL-switch:3 The security management function in the anchor node at a new network must allow to assign the original IP prefix/ address used by the mobile node at the previous (original) network. As the assigned original IP prefix/ address is to be used in the new network, the security management function in the anchor node must allow to advertise the prefix of the original IP address and also allow the mobile node to send and receive data packets with the original IP address. @@ -1417,122 +1551,123 @@ control and data planes ([I-D.liu-dmm-deployment-scenario] and [I-D.matsushima-stateless-uplane-vepc]), where the controller may send to the switches/routers the updated information of the forwarding tables with the IP address anchoring of the original IP prefix/address at AR1 moved to AR2 in the new network. That is, the IP address anchoring in the original network which was advertising the prefix will need to move to the new network. As the anchoring in the new network advertises the prefix of the original IP address in the new network, the forwarding tables will be updated so that packets of the flow will be forwarded according to the updated - forwarding tables. The configurations in Figures 1(a) and 1(b) in - Section 3.1 for which FM-CP and LM are centralized and FM-DP's are - distributed apply here. Figure 9 shows its implementation where LM - is a binding between the original IP prefix/address of the flow and - the IP address of the new DPA, whereas FM uses the DHCPv6-PD - protocol. + forwarding tables. + + The configurations in Figures 1(a) and 1(b) in Section 3.1 for which + the FM-CP and the LM are centralized and the FM-DP's are distributed + apply here. Figure 9 shows its implementation where the LM is a + binding between the original IP prefix/address of the flow and the IP + address of the new DPA, whereas the FM uses the DHCPv6-PD protocol. Net1 Net2 +----------------------------------------------------------------------+ | CPA: | | LM:IP1<-->IPa2 | | FM-CP | +----------------------------------------------------------------------+ +---------------+ +---------------+ |AR1 | |AR2 | +---------------+ +---------------+ -|DPA(IPa1): | |DPA(IPa2): | -|anchors IP1 | move |anchors IP2,IP1| -|FM:DHCPv6-PD | =======> |FM:DHCPv6-PD | +|DPA(IPa1): | anchoring of IP1 is moved |DPA(IPa2): | +|anchored IP1 | =======> |anchors IP2,IP1| +|FM:DHCPv6-PD | |FM:DHCPv6-PD | +---------------+ +---------------+ +...............+ +---------------+ -.MN(IP1) . move |MN(IP2,IP1) | +.MN(IP1) . MN moves |MN(IP2,IP1) | .flow(IP1,...) . =======> |flow(IP1,...) | +...............+ +---------------+ Figure 9. IP prefix/address anchor switching to the new network with - with LM and FM-CP in a centralized control plane whereas the FM-DP's - are distributed. + with the LM and the FM-CP in a centralized control plane whereas the + FM-DP's are distributed. - The example call flow in Figure 10 shows that MN is allocated HNP1 - when it attaches to the p-AR. A flow running in MN and needing IP + The example call flow in Figure 10 shows that MN is allocated IP1 + when it attaches to the AR1 A flow running in MN and needing IP mobility may continue to use the previous IP prefix by moving the anchoring of the IP prefix to the new network. Yet a new flow to be initiated in the new network may simply use a new IP prefix allocated from the new network. - MN p-AR n-AR DHCPv6 Servers CN - |MN attaches to p-AR: | | | | + MN AR1 AR2 DHCPv6 Servers CN + |MN attaches to AR1: | | | | |acquire MN-ID and profile | | | |--RS---------------->| | | | - |<----------RA(HNP1)--| | | | - | | | Allocate MN-HNP1 | + |<----------RA(IP1)---| | | | + | | | Allocate MN:IP1 | IP addr config | | | | | | | | | |<-Flow(IP1,IPcn,...)-+--------------------------------------------->| | | | | | - |MN detach from p-AR | | | | - |MN attach to n-AR | | | | + |MN detach from AR1 | | | | + |MN attach to AR2 | | | | | | | | | |--RS------------------------------>| | | | | | | | | |------DHCPv6 release-------------->| | | | | | | | | |--DHCPv6 PD request->| | | | |<-DHCPv6 PD reply--->| | | | | | | | forwarding table updates | | | | | | | - |<--------------RA(HNP2,HNP1)-------| | | - | | | Allocate MN-HNP2 | + |<--------------RA(IP2,IP1)---------| | | + | | | Allocate MN:IP2 | IP addr config | | | | | | | | | |<-Flow(IP1,IPcn,...)---------------+------------------------------->| | | | | | | Flow(IP1,IPcn,...) terminates | | | | | | | | | | DHCPv6-PD timeout | | | | | | | | forwarding table updates | | | | | | | | | | | | |<-new Flow(IP2,IPcn,...)-----------+------------------------------->| | | | | | Figure 10. DMM solution. MN with flow using IP1 in Net1 continues to run the flow using IP1 as it moves to Net2. - As the MN moves from p-AR to n-AR, the p-AR as a DHCPv6 client may - send a DHCPv6 release message to release the HNP1. It is now - necessary for n-AR to learn the IP prefix of the MN from the previous - network so that it will be possible for Net2 to allocate both the - previous network prefix and the new network prefix. The network may - learn the previous prefix in different methods. For example, the MN - may provide its previous network prefix information by including it - to the RS message [I-D.jhlee-dmm-dnpp]. + As the MN moves from AR1 to AR2, the AR1 as a DHCPv6 client may send + a DHCPv6 release message to release the IP1. It is now necessary for + AR2 to learn the IP prefix of the MN from the previous network so + that it will be possible for Net2 to allocate both the previous + network prefix and the new network prefix. The network may learn the + previous prefix in different methods. For example, the MN may + provide its previous network prefix information by including it to + the RS message [I-D.jhlee-dmm-dnpp]. - Knowing that MN is using HNP1, the n-AR sends to a DHCPv6 server a - DHCPv6-PD request to move the HNP1 to n-AR. The server sends to n-AR - a DHCPv6-PD reply to move the HNP1. Then forwarding tables updates - will take place here. + Knowing that MN is using IP1, the AR2 sends to a DHCPv6 server a + DHCPv6-PD request to move the IP1 to AR2. The server sends to AR2 a + DHCPv6-PD reply to move the IP1. Then forwarding tables updates will + take place here. - In addition, the MN also needs a new HNP in the new network. The - n-AR may now send RA to n-AR, with prefix information that includes - HNP1 and HNP2. The MN may then continue to use IP1. In addition, - the MN is allocated the prefix HNP2 with which it may configure its - IP addresses. Now for flows using IP1, packets destined to IP1 will - be forwarded to the MN via n-AR. + In addition, the MN also needs a new IP in the new network. The AR2 + may now send RA to AR2, with prefix information that includes IP1 and + IP2. The MN may then continue to use IP1. In addition, the MN is + allocated the prefix IP2 with which it may configure its IP + addresses. Now for flows using IP1, packets destined to IP1 will be + forwarded to the MN via AR2. - As such flows have terminated and DHCPv6-PD has timed out, HNP1 goes - back to Net1. MN will then be left with HNP2 only, which it will use + As such flows have terminated and DHCPv6-PD has timed out, IP1 goes + back to Net1. MN will then be left with IP2 only, which it will use when it now starts a new flow. 5.2.1. Additional Guidelines for IPv6 Nodes: Switching Anchor with Centralized CP The configuration guideline for a flat network or network slice with centralized control plane and supporting a mix of flows requiring and not requiring IP mobility support is: GL-cfg:4 Multiple instances of DPAs (at access routers) which are @@ -1541,63 +1676,65 @@ Figure 1(b)in Section 3.1 with centralized control plane for a flat network. The appropriate IPv6 nodes (CPA, DPA) are to be implemented the mobility functions LM and FM as described respectively in LM-cfg:1 or LM-cfg:2 and FM-cfg:1 in Section 3.2. The guidelines (GL-mix) in Section 4.1.1 and in Section 4.2.1 for the IPv6 nodes for a network or network slice supporting a mix of flows requiring and not requiring IP mobility support apply here. The - guidelines (GL-mix) in Section 5.1.1 also apply here. In addition, - the following are required. + guidelines (GL-mix) in Section 5.1.1 for moving anchoring for a flat + network also apply here. In addition, the following are required. - GL-switch:5 The anchor operations to properly forward the packets - for a flow as described in the FM operations and - mobility message parameters in Section 3.2.2 FM-path, - FM-path-tbl, FM-DPA, FM-DPA-tbl is realized by changing - the anchoring with DHCPv6-PD and undoing such changes - later when its timer expires and the application has - already closed. With the anchors being separated in - control and data planes with LMs and FM-CP centralized - in the same control plane, messaging between anchors and - the discovery of anchors become internal to the control - plane as described in Section 3.2.2 FM-cpdp. However, - the centralized FM-CP needs to communicate with the - distributed FM-DP as described as described in the FM - operations and mobility message parameters (FM-find). - Such may be realized by the appropriate messages in - [I-D.ietf-dmm-fpc-cpdp]. + GL-switch:5 It was already mentioned that the anchor operations to + properly forward the packets for a flow as described in + the FM operations and mobility message parameters in FM- + path, FM-path-tbl, FM-DPA, and FM-DPA-tbl in + Section 3.2.2 is realized by changing the anchoring with + DHCPv6-PD and undoing such changes later when its timer + expires and the application has already closed. Here + however, with separation of control and data planes for + the anchors and where the LMs and the FM-CP are + centralized in the same control plane, messaging between + anchors and the discovery of anchors become internal to + the control plane. - GL-switch:6 It was already mentioned before that, if there are in- - flight packets toward the old anchor while the MN is - moving to the new anchor, it may be necessary to buffer - these packets and then forward to the new anchor after - the old anchor knows that the new anchor is ready Here, - however, the corresponding FM operations and mobility - message parameters as described in Section 3.2.2 (FM- - buffer) can be realized by the internal operations in - the control plane together with signaling between the - control plane and distributed data plane. These - signaling may be realized by the appropriate messages in - [I-D.ietf-dmm-fpc-cpdp]. + GL-switch:6 The centralized FM-CP needs to communicate with the + distributed FM-DP using the FM operations and mobility + message parameters as described in FM-cpdp in + Section 3.2.2. Such may be realized by the appropriate + messages in [I-D.ietf-dmm-fpc-cpdp]. -5.3. IP Prefix/Address Anchor Switching for a Hierarchical Network + GL-switch:7 It was also already mentioned before that, if there are + in-flight packets toward the previous anchor while the + MN is moving to the new anchor, it may be necessary to + buffer these packets and then forward to the new anchor + after the old anchor knows that the new anchor is ready + Here however, the corresponding FM operations and + mobility message parameters as described in + Section 3.2.2 (FM-buffer) can be realized by the + internal operations in the control plane together with + signaling between the control plane and distributed data + plane. These signaling may be realized by the + appropriate messages in [I-D.ietf-dmm-fpc-cpdp]. - The configuration for a hierarchical network is shown in Figures 1(c) - and 1(d) in Section 3.1. With centralized control plane, CPA and - CPN, with the associated LM and FM-CP are all co-located. There are - multiple DPAs (each with FM-DP) in distributed mobility anchoring. - In the data plane, there are multiple DPNs (each with FM-DP) - hierarchically below each DPA. The DPA at each AR supports - forwarding to the DPN at each of a number of forwarding switches - (FW's). A mobility event in this configuration belonging to +5.3. Hierarchical Network + + The configuration for a hierarchical network has been shown in + Figures 2(a) and 2(b) in Section 3.1.2. With centralized control + plane, CPA and CPN, with the associated LM and FM-CP are all co- + located. There are multiple DPAs (each with FM-DP) in distributed + mobility anchoring. In the data plane, there are multiple DPNs (each + with FM-DP) hierarchically below each DPA. The DPA at each AR + supports forwarding to the DPN at each of a number of forwarding + switches (FW's). A mobility event in this configuration belonging to distributed mobility management will be deferred to Section 5.4. In this distributed mobility configuration, a mobility event involving change of FW only but not of AR as shown in Figure 11 may still belong to centralized mobility management and may be supported using PMIPv6. This configuration of network-based mobility is also applicable to host-based mobility with the modification for the MN directly taking the role of DPN and CPN, and the corresponding centralized mobility event may be supported using MIPv6. @@ -1618,162 +1755,296 @@ | CPA,CPN: | | LM:IP1<-->IPn2 | | FM-CP | +----------------------------------------------------------------------+ +---------------+ |AR1 | +---------------+ |DPA(IPa1): | |anchors IP1 | - |FM:DHCPv6-PD | + |FM-DP | +---------------+ +---------------+ +---------------+ |FW1 | |FW2 | -+---------------+ move +---------------+ -|DPN(IPn1): | =======> |DPN(IPn2): | ++---------------+ FW is changed +---------------+ +|DPN(IPn1): | -------> |DPN(IPn2): | +|FM-DP | |FM-DP | +---------------+ +---------------+ +...............+ +---------------+ -.MN(IP1) . move |MN(IP2) | +.MN(IP1) . MN moves |MN(IP2) | .flow(IP1,...) . =======> |flow(IP1,...) | +...............+ +---------------+ Figure 11. Mobility without involving change of IP anchoring in a network in which the IP prefix allocated to the MN is anchored at an AR which is hierarchically above multiple FWs to which the MN may connect. -5.3.1. Additional Guidelines for IPv6 Nodes: No Anchoring Change with a - Hierarchical Network +5.3.1. Additional Guidelines for IPv6 Nodes: Hierarchical Network with + No Anchor Relocation - The configuration guideline ( ) for a hierarchical network or network + The configuration guideline for a hierarchical network or network slice with centralized control plane and supporting a mix of flows requiring and not requiring IP mobility support is: GL-cfg:5 Multiple instances of DPAs (at access routers) which are providing IP prefix to the MNs are needed to provide distributed mobility anchoring according to Figure 2(a) or Figure 2(b)in Section 3.1.2 with centralized control plane for a hierarchical network. The appropriate IPv6 nodes (CPA, DPA) are to be implemented the mobility functions LM and FM as described respectively in LM-cfg:3 or LM-cfg:4 and FM-cfg:2 in Section 3.2. Even when the mobility event does not involve change of anchor, it is still necessary to distinguish whether a flow needs IP mobility support. The GL-mix guidelines in Section 4.1.1 and in Section 4.2.1 for the IPv6 nodes for a network or network slice supporting a mix of flows - requiring and not requiring IP mobility support apply here. The - guidelines (GL-switch) in Section 5.1.1 and in Section 5.2.1 also - apply here. In addition, the following are required. + requiring and not requiring IP mobility support apply here. In + addition, the following are required. - GL-switch:7 Here, the LM operations and mobility message parameters + GL-switch:8 Here, the LM operations and mobility message parameters described in Section 3.2.1 provides information of which IP prefix from its FW needs to be used by a flow using which new FW. The anchor operations to properly forward the packets of a flow described in the FM operations and mobility message parameters (FM-path, FM-path-ind, FM- cpdp in Section 3.2.2) may be realized with PMIPv6 - protocol ([I-D.korhonen-dmm-local-prefix]) or with AERO - protocol ([I-D.templin-aerolink]) to tunnel between the - AR and the FW. + protocol [I-D.korhonen-dmm-local-prefix] or with AERO + protocol [I-D.templin-aerolink] to tunnel between the AR + and the FW. 5.4. IP Prefix/Address Anchor Switching for a Hierarchical Network - The configuration for the hierarchical network is again shown in - Figures 1(c) and 1(d) in Section 3.1. Again, with centralized + The configuration for the hierarchical network has been shown in + Figures 2(a) and 2(b) in Section 3.1.2. Again, with centralized control plane, CPA and CPN, with the associated LM and FM-CP are all co-located. There are multiple DPAs (each with FM-DP) in distributed mobility anchoring. In the data plane, there are multiple DPNs (each with FM-DP) hierarchically below each DPA. The DPA at each AR supports forwarding to the DPN at each of a number of forwarding switches (FW's). A distributed mobility event in this configuration involves change from a previous DPN which is hierarchically under the previous DPA to a new DPN which is hierarchically under a new DPA. Such an event involving change of both DPA and DPN is shown in Figure 12. Net1 Net2 +----------------------------------------------------------------------+ -| CPA,CPN: | +| CPA,CPN,Aggregate Router: | | LM:IP1<-->IPa2,IPn2 | | FM-CP | +----------------------------------------------------------------------+ - +---------------+ - |Aggregate Point| - |---------------| - |FM, LM | - +---------------+ + +-----------------+ + |Aggregate Router | + +-----------------+ + |FM-DP | + +-----------------+ +---------------+ +---------------+ |AR1 | |AR2 | +---------------+ +---------------+ -|DPA(IPa1): | |DPA(IPa2): | -|anchors IP1 | move |anchors IP2,IP1| -|FM:DHCPv6-PD | =======> |FM:DHCPv6-PD | +|DPA(IPa1): | anchoring of IP1 is moved |DPA(IPa2): | +|anchored IP1 | =======> |anchors IP2,IP1| +|FM:DHCPv6-PD | |FM:DHCPv6-PD | +---------------+ +---------------+ +---------------+ +---------------+ |FW1 | |FW2 | -+---------------+ move +---------------+ -|DPN(IPn1): | =======> |DPN(IPn2): | ++---------------+ FW is changed +---------------+ +|DPN(IPn1): | -------> |DPN(IPn2): | +|FM-DP | |FM-DP | +---------------+ +---------------+ +...............+ +---------------+ -.MN(IP1) . move |MN(IP2,IP1) | +.MN(IP1) . MN moves |MN(IP2,IP1) | .flow(IP1,...) . =======> |flow(IP1,...) | +...............+ +---------------+ Figure 12. Mobility involving change of IP anchoring in a network with hierarchy in which the IP prefix allocated to the MN is anchored at an Edge Router supporting multiple access routers to which the MN may connect. This deployment case involves both a change of anchor from AR1 to AR2 and a network hierarchy AR-FW. It can be realized by a combination - of changing the IP prefix/address anchoring from AR1 to AR2 with the - mechanism as described in Section 5.2 and then forwarding the packets - with network hierarchy AR-FW as described in Section 5.3. + of relocating the IP prefix/address anchoring from AR1 to AR2 with + the mechanism as described in Section 5.2 and then forwarding the + packets with network hierarchy AR-FW as described in Section 5.3. - To change AR, AR1 acting as a DHCPv6-PD client may exchange message - with the DHCPv6 server to release the prefix IP1. Meanwhile, AR2 - acting as a DHCPv6-PD client may exchange message with the DHCPv6 - server to delegate the prefix IP1 to AR2. + To change the anchoring of IP1, AR1 acting as a DHCPv6-PD client may + exchange message with the DHCPv6 server to release the prefix IP1. + Meanwhile, AR2 acting as a DHCPv6-PD client may exchange message with + the DHCPv6 server to delegate the prefix IP1 to AR2. 5.4.1. Additional Guidelines for IPv6 Nodes: Switching Anchor with Hierarchical Network The configuration guideline (GL-cfg) for a hierarchical network or network slice with centralized control plane described in Section 5.3.1 apply here. The GL-mix guidelines in Section 4.1.1 and in Section 4.2.1 for the IPv6 nodes for a network or network slice supporting a mix of flows requiring and not requiring IP mobility support apply here. - The guidelines (GL-switch) in Section 5.1.1 and in Section 5.2.1 also - apply here to change the anchoring of the IP prefix/address with a - centralized control plane. + The guidelines (GL-switch) in Section 5.1.1 for anchoring relocation + and in Section 5.2.1 for a centralized control plane also apply here. - In addition, the guideline for indirection between the new DPA and - the new DPN as described in Section 5.3.1 apply here. + In addition, the guidelines for indirection between the new DPA and + the new DPN as described in Section 5.3.1 apply as well. + +5.5. Network Mobility + + The configuration for network mobility has been shown in Figures 4(a) + and 4(b) in Section 3.1.4. Again, with centralized control plane, + CPA, with the associated LM and FM-CP are all co-located. There are + multiple DPAs (each with FM-DP) in the data plane in distributed + mobility anchoring. The MR possesses the mobility functions FM and + LMc. The IP prefix IPn1 is delegated to the MR, to which a MNN is + attached and is allocated with an IP address from IPn1. + + Figure 13 shows a distributed mobility event in a hierarchical + network with a centralized control plane involving a change of + attachment of the MR from a previous DPA to a new DPA while the MNN + is attached to and therefore moves with the MR. + +Net1 Net2 ++----------------------------------------------------------------------+ +| CPA,Aggregate Router: | +| LM:IP1<-->IPa2; IPn1<-->IP1 | +| FM-CP, LM | ++----------------------------------------------------------------------+ + + +-----------------+ + |Aggregate Router | + +-----------------+ + |FM-DP | + +-----------------+ + ++---------------+ +---------------+ +|AR1 | |AR2 | ++---------------+ +---------------+ +|DPA(IPa1): | anchoring of IP1 is moved |DPA(IPa2): | +|anchored IP1 | =======> |anchors IP2,IP1| +|DHCPv6-PD IPn1 | | | +|FM-DP | |FM-DP | ++---------------+ +---------------+ + ++...............+ +---------------+ +.MR(IP1) . MR moves |MR(IP2,IP1) | ++...............+ =======> +---------------+ +.FM, LMc . |FM, LMc | +.anchors IPn1 . |anchors IPn1 | ++...............+ +---------------+ + ++...............+ +---------------+ +.MNN(IPn1) . MNN moves with MR |MNN(IPn1) | +.flow(IPn1,...) . =======> |flow(IPn1,...) | ++...............+ +---------------+ + + Figure 13. Mobility involving change of IP anchoring for a MR to + which a MNN is attached. + + As the MR with source IP prefix IP1 moves from AR1 to AR2, mobility + support may be provided by moving the anchoring of IP1 from AR1 to + AR2 using the mechanism described in Section 5.2. + + The forwarding table updates will take place at AR1, AR2, the + aggregate router, and other affected routers such that the packet + from the CN to the MNN will traverse from the aggregate router + towards AR2 instead of towards AR1. + +5.5.1. Additional Guidelines for IPv6 Nodes: Network mobility + + The configuration guideline for a network or network slice with + centralized control plane to provide network mobility is: + + GL-cfg:6 Multiple instances of DPAs (at access routers) which are + providing IP prefix of the MRs are needed to provide + distributed mobility anchoring according to Figure 4(a) or + Figure 4(b) in Section 3.1. + + The appropriate IPv6 nodes (CPA, DPA) are to be implemented + the mobility functions LM and FM as described respectively + in LM-cfg:3 or LM-cfg:4 and FM-cfg:4 in Section 3.2. + + The GL-mix guidelines in Section 4.1.1 and in Section 4.2.1 for the + IPv6 nodes for a network or network slice supporting a mix of flows + requiring and not requiring IP mobility support apply here. + + Here, because the MN is a MR, the following guideline is added: + + GL-mix:11 There are no flows requiring network mobility support when + there are no MNN attaching to the MR. Here there are also + no MNN using a prefix delegated to the MR. Therefore the + anchor of the MR may change to a new AR. The new AR may + delegate new IP prefix to the AR, so that the MR may + support potential MNN to attach to it. On the other hand + the delegation of IP prefix to the MR from the old AR may + be deleted. + + The guidelines (GL-switch) in Section 5.1.1 for anchoring relocation + and in Section 5.2.1 for a centralized control plane also apply here. + + Again because the MN is a MR, the following guidelines are added: + + GL-switch:9 Network mobility may be provided using the FM operations + and mobility message parameters as described in FM-mr in + Section 3.2.2. + + GL-switch:10 The following changes to forwarding table entries are + needed: + + New entries to the forwarding tables are added between + AR2, the aggregate router and other affected routers so + that packets from the CN to the MNN destined to IPn1 + will traverse towards AR2. Meanwhile, changes to the + forwarding table will also occur between AR1, the + aggregate router and other affected routers so that such + packets ever reaches any of them, the packet will not + traverse towards AR1 but will traverse towards AR2. + + GL-switch:11 The security management function in the anchor node at a + new network must allow to assign the original IP prefix/ + address allocated to the MR and used by the MNN at the + previous (original) network. As the assigned original + IP prefix/address is to be used in the new network, the + security management function in the anchor node must + allow to advertise the prefix of the original IP address + and also allow the MNN to send and receive data packets + with the original IP address. + + GL-switch:12 The security management function in the mobile router + must allow to configure the original IP prefix/address + delegated to the MR from the previous (original) network + when the original IP prefix/address is being delegated + to the MR in the new network. The security management + function in the mobile router also allows to use the + original IP address by the MNNs for the previous flow in + the new network. 6. Security Considerations - TBD + The security considerations are already described in different + sessions through this document. They are described in terms of + integrity support, privacy support etc. in describing the mobility + functions in Section 3.2. They are also described in the guidelines + for IPv6 nodes in various subsections Section 4 and Section 5. 7. IANA Considerations This document presents no IANA considerations. 8. Contributors This document has benefited from other work on mobility solutions using BGP update, on mobility support in SDN network, on providing mobility support only when needed, and on mobility support in @@ -1789,29 +2060,29 @@ 9. References 9.1. Normative References [I-D.ietf-dmm-deployment-models] Gundavelli, S. and S. Jeon, "DMM Deployment Models and Architectural Considerations", draft-ietf-dmm-deployment- models-00 (work in progress), August 2016. [I-D.ietf-dmm-fpc-cpdp] - Liebsch, M., Matsushima, S., Gundavelli, S., Moses, D., - and L. Bertz, "Protocol for Forwarding Policy - Configuration (FPC) in DMM", draft-ietf-dmm-fpc-cpdp-03 - (work in progress), March 2016. + Matsushima, S., Bertz, L., Liebsch, M., Gundavelli, S., + and D. Moses, "Protocol for Forwarding Policy + Configuration (FPC) in DMM", draft-ietf-dmm-fpc-cpdp-05 + (work in progress), October 2016. [I-D.ietf-dmm-ondemand-mobility] - Yegin, A., Moses, D., Kweon, K., Lee, J., and J. Park, "On - Demand Mobility Management", draft-ietf-dmm-ondemand- - mobility-07 (work in progress), July 2016. + Yegin, A., Moses, D., Kweon, K., Lee, J., Park, J., and S. + Jeon, "On Demand Mobility Management", draft-ietf-dmm- + ondemand-mobility-09 (work in progress), December 2016. [I-D.jhlee-dmm-dnpp] Lee, J. and Z. Yan, "Deprecated Network Prefix Provision", draft-jhlee-dmm-dnpp-01 (work in progress), April 2016. [I-D.korhonen-dmm-local-prefix] Korhonen, J., Savolainen, T., and S. Gundavelli, "Local Prefix Lifetime Management for Proxy Mobile IPv6", draft- korhonen-dmm-local-prefix-01 (work in progress), July 2013. @@ -1836,29 +2107,24 @@ [I-D.mccann-dmm-prefixcost] McCann, P. and J. Kaippallimalil, "Communicating Prefix Cost to Mobile Nodes", draft-mccann-dmm-prefixcost-03 (work in progress), April 2016. [I-D.seite-dmm-dma] Seite, P., Bertin, P., and J. Lee, "Distributed Mobility Anchoring", draft-seite-dmm-dma-07 (work in progress), February 2014. - [I-D.sijeon-dmm-deployment-models] - Jeon, S. and Y. Kim, "Deployment Models for Distributed - Mobility Management", draft-sijeon-dmm-deployment- - models-03 (work in progress), July 2016. - [I-D.templin-aerolink] Templin, F., "Asymmetric Extended Route Optimization - (AERO)", draft-templin-aerolink-71 (work in progress), - September 2016. + (AERO)", draft-templin-aerolink-74 (work in progress), + November 2016. [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, . [RFC3753] Manner, J., Ed. and M. Kojo, Ed., "Mobility Related Terminology", RFC 3753, DOI 10.17487/RFC3753, June 2004, .