Internet-Draft | RGB Segment for MSR6 | October 2021 |
Liu, et al. | Expires 28 April 2022 | [Page] |
This document introduces the RGB (Replication through Global Bitstring) Segment for Multicast Source Routing over IPv6.¶
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 [RFC2119].¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 28 April 2022.¶
Copyright (c) 2021 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.¶
Segment Routing ([RFC8402]) leverages the mechanism of source routing. An ingress node steers a packet through an ordered list of instructions, called "segments". Each one of these instructions represents a function to be implemented at a specific location in the network. A function is locally defined on the node where it is executed. Network Programming combines Segment Routing functions to achieve a networking objective that goes beyond mere packet routing. [RFC8986] defines the SRv6 Network Programming concept and specifies the main Segment Routing behaviors and network programming functions.¶
Bit Index Explicit Replication (BIER) [RFC8279] is an architecture that provides optimal multicast forwarding without requiring a protocol for explicitly building multicast distribution trees or per-flow state maintained by intermediate routers. When a multicast data packet enters BIER forwarding domain, the ingress node encapsulates the packet with a bitstring, each bitposition of which presents the egress nodes. To forward the packet to a given set of egress nodes, the bits corresponding to those egress nodes are set in the bitstring. The intermediate nodes in the BIER domain replicate and forward the packet based on the bitstring.The mechanism of forwarding a packet based on bitstring of BIER are specified in [RFC8279].¶
An IPv6 based multicast source routing (MSR6) solution is defined in [I-D.cheng-spring-ipv6-msr-design-consideration], which leverages the benefits of source routing over IPv6 data plane to provide simplified multicast TE and BE service in an IPv6 network without unnecessary multicast tree status and complex control plane protocols. MSR6 needs to reuse the advantages of SRv6 and BIER to implement source routing.¶
MSR6 has two basic modes of forwarding: one is based on Shortest Path First(SPF), which is called MSR6 BE mode; the other is based on traffic engineered, which is called MSR6 TE mode. This document defines a new type of segment, Replication through Global Bitstring Segment (RGB Segment), and the packet processing procedures over the IPv6 data plane for the MSR6 BE solutions.¶
The following new terms are used throughout this document:¶
MSR6 Domain: a set of nodes participating in the multicast source routing;¶
MSR6 Ingress Node: a node through which a multicast data packet enters an MSR6 domain; The MSR6 Ingress Node could be a host or a network device.¶
MSR6 Egress Node: a node through which a multicast data packet leaves an MSR6 domain; The MSR6 Egress Node could be a host or a network device.¶
MSR6 Root Node: a node which is the beginning point of a multicast tree. It encapsulates the packet with an MSR6 multicast header.¶
MSR6 Leaf Node: a node which is the ending point of a multicast tree. It decapsulates the MSR6 multicast header in the packet.¶
MSR6 Replication Endpoint: the intermediate node of a multicast tree, which replicates packet and forwards the packet to the downstream nodes. For MSR6, the Replication Node is called Replication Endpoint which can be indicated by the MSR6 Segment and replicate packets according to the multicast source routing information encapsulation in the MSR6 header of the packet.¶
MSR6 Transit Node: a node which forwards the MSR6 packet as an IPv6 unicast packet;¶
MSR6 BE needs to indicate egress nodes in the packet to avoid maintaining multicast tree state in the intermediate nodes. Such mechanism that has been defined in IETF, as BIER defined in [RFC8279], could be reused in MSR6 BE with global bitstring indicating egress nodes.¶
A new DoH option called RGB option is introduced in this document to support MSR6 BE. The option includes the bistring representing egress nodes and MUST be encapsulated in the IPv6 Destination Options Header defined in RFC8200.¶
The encoding of RGB(Replication through Global Bitstring) Option Data is defined as follows:¶
0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Next Header | Hdr Ext Len | Option Type | Option Length | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | ~ RGB Option Data ~ | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+¶
RGB Option is defined as follows:¶
0 1 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | BIFT-id | Rsv | TTL | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Rsv | Ver | BSL | Entropy | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |OAM|Rsv| DSCP | Rsv | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | BitString (first 32 bits) ~ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ ~ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ~ BitString (last 32 bits) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+¶
The RGB Option Data reuses some codepoint of Non-MPLS BIER Header defined in [RFC8296] except the fields of Nibble, DSCP and Proto, which are replaced as the Reserved field. The Reserved fields SHOULD be set to 0 and MUST be ignored up reception.¶
The segment defined in [RFC8402] can represent instruction, topological or service based. In an IPv6 domain, a segment could be encoded as an IPv6 address.¶
In the IPv6 data plane, RGB segment is a new type of segment which is used to identify the Replication Endpoint. Replication Endpoint is able to replicate the packet using BIER forwarding mechanism according to the bitstring defined in the RGB Option.¶
RGB segment is used as an IPv6 address, which is 128 bits and follows the SID format defined in [RFC8986], consisting of LOC:FUNCT:ARG. RGB segment is advertised by the RGB replication endpoint.¶
In an MSR6 domain, RGB segment is used as the destination address of the MSR BE packet, when a packet is replicated to the next Replication Endpoint. If there is 1 or more MSR6 transit node between two Replication Endpoints, the packet is forwarded as normal unicast IPv6 packet.¶
In SRv6, a packet processing behavior executed at an SRv6 Segment Endpoint Node ([RFC8986]). In MSR6, a new type of behavior, End.RGB(End. Replication through Global Bitstring), is defined for RGB Segment. The pseudo-code for End.RGB is defined in this section.¶
When an MSR6 Replication Endpoint receives a packet whose IPv6 DA (Destination Address) is a SID and the SID is a local End.RGB SID, the MSR6 Replication Endpoint does the following:¶
1. IF (There is DoH as an IPv6 Extension header and one of the options type is RGB) ; 2. Lookup BIFT(Bit Index Forwarding Table, RFC8279) based on the bitstring inside the RGB Option Data. 3. Forward the packet via the matched entry in the BIFT. 4. ELSE IF NH=ICMPv6 or (NH=RGB Extension Header Type and NH of Extension Header=ICMPv6) ; 5. Send to CPU. 6. ELSE ;Ref 7. Drop the packet.¶
Ref: An ICMPv6 packet using End.RGB as destination address.¶
The Encapsulation of MSR6 BE packet is defined as follows:¶
+---------------+------------------+----------------------+ | IPv6 header | IPv6 DO Header | Client Multicast | | | with RGB Option | Packet or Upper | | | | Layer Encasulations | | Next Hdr = 60 | Nxt Hdr = X | | +---------------+------------------+----------------------+ | | | |<---------MSR6 BE header--------->|<--MSR6 BE payload--->|¶
In the MSR6 BE header, the RGB Segment is used as the IPv6 Destination Address and it indicates the next MSR6 Replication Endpoints in an MSR domain in an MSR6 domain.¶
The MSR6 BE header also includes a DOH containing RGB option defined in section. The RGB Replication Endpoint indicated by the RGB Segment replicate the packet according to the bitstring information in the RGB option.¶
This section defines the general process of MSR6 to transport a multicast service. The corresponding control plane is out of scope of this document and could be discussed in the following work.¶
MSR6 Root Node: Encapsulate the packet with MSR BE encapsulation as defined in section 5. The bitstring in the DoH is determined by the egress nodes the packet is supposed to be replicated to. The IPv6 destination address is the RGB segment which is determined by the next MSR6 RGB Replication Endpoints the packet is supposed to be sent to. The downstream MSR6 Replication Endpoints are determined by the matched entries in BIFT according to the bier forwarding mechanism.¶
MSR6 Replication Endpoint: Replicate the packet and forward it to the next MSR6 Replication Endpoints. When an MSR6 Replication Endpoint receives a packet whose IPv6 Destination Address is A and A is the local RGB segment for the existing MSR6 Replication Endpoint, process the bitstring in the RGB DoH of the packet and look up the corresponding BITF for the next MSR6 Replication Endpoints. Replicate the packet, update the bitstring and DA in each replicated packet based on the lookup result.The RGB processing procedure follows the specification in RGB architecture defined in RFC8279.¶
MSR6 Transit Node: Transit the packet as an ordinary IPv6 packet by looking up FIB until find the next MSR6 Replication Endpoint.¶
MSR6 Leaf Node: Decasulate the MSR BE encapsulation. When an MSR6 Replication Endpoint receives a packet whose IPv6 Destination Address is A and A is the local RGB segment and the one of the bits set to 1 identifies the MSR6 Replication Endpoint itself, this Replication Endpoint is the egress node. If the MSR6 egress node is the edge of a network, copy the packet and send the copy to the multicast flow overlay; If the MSR6 egress node is the host supposed to receive the packet, send the packet to the upper layer.¶
MSR6 Replication Endpoint MSR6 Leaf Node +----+ +----+ +----+ | |-----------| |--------------| | +----+ +----+ +----+ MSR6 Root Node | +----+ +----+ +-----| |-----| | +----+ +----+ MSR6 Transit Node MSR6 Leaf Node¶
+-------------+ +-------------+ | {S=S1,D=P1} | | {S=S1,D=C1} | +-------------+ +-------------+ |[BitStr=0110]| |[BitStr=0100]| +=============+ +=============+ | Upper Layer | >> | Upper Layer | +=============+ +=============+ | Pay Load | | Pay Load | +=============+ +=============+ [Server1]--------------[P1]-------------------[Client1] (MSR6 Ingress) (MSR6 Replication Endpoint) (MSR6 Egress,BFR-id=2) / / +-------------+ +-------------+ | | {S=S1,D=C2} | | {S=S1,D=C2} | | +-------------+ +-------------+ | |[BitStr=0010]| |[BitStr=0010]| | +=============+ +=============+ | >> | Upper Layer | >> | Upper Layer | \ +=============+ +=============+ \ | Pay Load | | Pay Load | \ +=============+ +=============+ +-------------[P2]-----------[Client2] (MSR6 Transit)(MSR6 Egress, BFR-id=3) {S=Server1,D=P1}: Source address and Destination address in IPv6 header. [BitStr=0110]: BitString value in IPv6 Destination Options Header.¶
Server1 generates the packet with an IPv6 Header. Knowing that BFR-ID of Client 1 is 2 and BFR-ID of Client 2 is 3, it follows that when the multicast service is supposed to be transmitted to Client1 and Client2, the bitstring in RGB DoH of the IPv6 header is set as "0110". Look up the BIFT and finds the RGB segment of next MSR6 BFR is P1. The IPv6 DA is set as "P1".¶
P1 receives the packet with DA as "P1", which is the local RGB segment. P1 parses the DoH with RGB Option Data and looks up the BIFT to find the corresponding entry. P1 replicates the packets into 2 copies based on the look up result. DA of one replicated packet is set to "C1" and the bitstring is set to "0100". DA of the other replicated packet is set to "C2" and the bitstring is set to "0010". These 2 packets are forwarded to next hop based on the updated DA.¶
P2 receives the packet and forwards it Client2 based on the DA of "C2".¶
Client1 receives the packet with DA as "C1". "C1" is the local RGB segment and "0100" identifies Client1 itself. The packet is sent to the upper layer.¶
Client2 receives the packet with DA as "C2". "C2" is the local RGB segment and "0010" identifies Client2 itself. The packet is sent to the upper layer.¶
+-------------+ +-------------+ |{S=PE1,D=P2} | |{S=PE1,D=PE2}| +-------------+ +-------------+ |[BitStr=0110]| |[BitStr=0100]| +==========+ +=============+ +=============+ +==========+ |(C-MC Pkt)| >> | (C-MC Pkt) | >> | (C-MC Pkt) | >> |(C-MC Pkt)| +==========+ +=============+ +=============+ +==========+ CE1-----------[PE1]------[P1]------[P2]-----------[PE2]---------CE2 (MSR6 Ingress)(MSR6 Transit)/(MSR6 Replication Endpoint) (MSR6 Egress,BFR-id=2) / / +-------------+ | |{S=PE1,D=PE3}| | +-------------+ | |[BitStr=0010]| \ +=============+ +==========+ \ >> | (C-MC Pkt) | >> |(C-MC Pkt)| \ +=============+ +==========+ +------[P3]------[PE3]----------CE3 (MSR6 Transit)(MSR6 Egress, BFR-id=3) {S=CE1,D=P1}: Source address and Destination address in IPv6 header. [BitStr=0110]: BitString value in IPv6 Destination Options Header. (C-MC Pkt): Customer MultiCast packet.¶
PE1 receives the customer multicast packet from CE1. An MSR BE header is encapsulated as defined in section 3. Knowing that BFR-ID of PE 1 is 2 and BFR-ID of PE 2 is 3, it follows that when the multicast service is supposed to be transmitted to PE2 and PE3, the bitstring in the RGB Options Header of DoH is set as "0110". Look up the corresponding BIFT and finds the RGB segment of next MSR6 BFR is P2. The IPv6 DA is set as "P2".¶
P1 receives the packet and forwards it P2 based on the DA of "P2".¶
P2 receives the packet with DA as "P2", which is the local RGB segment. P2 parses the DoH with RGB Option Data and looks up the BIFT to find the corresponding entry.P2 replicates the packets into 2 copies based on the look up result. DA of one replicated packet is set to "PE2" and the bitstring is set to "0100". DA of the other replicated packet is set to "PE3" and the bitstring is set to "0010". These 2 packets are forwarded to next hop based on the updated DA.¶
P3 receives the packet and forwards it PE3 based on the DA of "PE3".¶
PE2 receives the packet with DA as "PE2". "PE2" is the local RGB segment and "0100" identifies PE2 itself. The packet is sent to the multicast flow overlay.¶
PE3 receives the packet with DA as "PE3". "PE3" is the local RGB segment and "0010" identifies PE3 itself. The packet is sent to the multicast flow overlay.¶
Allocation is expected from IANA for a RGB Option Type codepoint from the "Destination Options and Hop-by-Hop Options" sub-registry of the "Internet Protocol Version 6 (IPv6) Parameters" registry.¶
+-----------+-----+-----+-------+-------------+------------+ | Hex Value | act | chg | rest | Description | Reference | +-----------+-----+-----+-------+-------------+------------+ | TBD | 01 | 1 | TBD | RGB Option | This draft | +-----------+-----+-----+-------+-------------+------------+¶
Allocation is expected from IANA for an End.RGB function codepoint from the "SRv6 Endpoint Behaviors" sub-registry. The value 60 is suggested.¶
+-------+--------+--------------------------+------------+ | Value | Hex | Endpoint function | Reference | +-------+--------+--------------------------+------------+ | TBD | TBD | End.RGB | This draft | +-------+--------+--------------------------+------------+¶
The MSR6 domain can be a single IGP area, an anonymous system (AS) with multiple IGP areas, or multiple anonymous systems (ASes) operated by a network operator.¶
It is expected that all nodes in an MSR6 domain are managed by the same administrative entity. MSR6-encapsulated packets should generally not be accepted from untrusted interfaces or tunnels. For example, an operator may wish to have a policy of accepting MSR6 encapsulated packets only from interfaces to trusted routers, and not from customer-facing interfaces.¶
For applications that require a MSR6 Replication Endpoint to accept a MSR6 encapsulated packet from an interface to a system that is not controlled by the network operator, the security considerations of [RFC8296] apply¶
Generally nodes outside the MSR6 Domain are not trusted: they cannot directly use the End.RGB segment of the domain. This is enforced by two levels of access control lists:¶
1. Any packet entering the MSR6 Domain and destined to an End.RGB Segment within the MSR6 Domain is dropped. This may be realized with the following logic. Other methods with equivalent outcome are considered compliant:¶
* allocate all the End.RGB Segment from a block S/s¶
* configure each external interface of each edge node of the domain with an inbound infrastructure access list (IACL) which drops any incoming packet with a destination address in S/s¶
* Failure to implement this method of ingress filtering may expose the MSR6 Domain to BIER attacks. The security consideration on BIER attacks is as described and referenced in [RFC8296].¶
2. The distributed protection in #1 is complemented with per node protection, dropping packets to End.RGB Segment from source addresses outside the MSR6 Domain. This may be realized with the following logic. Other methods with equivalent outcome are considered compliant:¶
* assign all interface addresses from prefix A/a¶
* assign all the IPv6 addresses used as source address of MSR6 packets from a block B/b¶
* at node k, all End.RGB Segment IPv6 addresses local to k are assigned from prefix Sk/sk¶
* configure each internal interface of each MSR6 node k in the MSR6 Domain with an inbound IACL which drops any incoming packet with a destination address in Sk/sk if the source address is not in A/a or B/b.¶
For simplicity of deployment, a configuration of IACL effective for all interfaces can be provided by a router. Such IACL can be referred to as global IACL(GIACL) .Each MSR6 node k then simply configures a GIACL which drops any incoming packet with a destination address in Sk/sk if the source address is not in A/a or B/b for the intra-domain deployment mode.¶
The MSR6 Replication Endpoint does not send ICMP error messages to the source address of a MSR BE packet, but there is still chance that Non-MSR6 Replication Endpoint routers send ICMP error messages to source nodes within the MSR6 Domain.¶
A large number of ICMP may be elicited and sent to a MSR6 Ingress router, in case when an MSR6 BE packet is filled with wrong Hop Limit, either error or malfeasance. A rate-limiting of ICMP packet should be implemented on each MSR6 Replication Endpoint.¶
The ingress node can take note of the fact that it is getting, in response to MSR6 BE packet, one or more ICMP error packets. By default, the reception of such packet MUST be countered and logged. However, it is possible for such log entries to be "false positives" that generate a lot of "noise" in the log; therefore, implementations SHOULD have a knob to disable this logging.¶
This document introduces a new option used in IPv6 Destination Options Header. An IPv6 packet with a normal IPv6 address of a router (e.g. loopback IPv6 address of the router) as destination address will possibly carry a RGB option.¶
For a router incapable of MSR6 BE, such MSR6 BE packet will not be processed by the procedure described in this document, but be processed as normal IPv6 packet with unknown option, and the existing security considerations for handling IPv6 options apply. Possible way of handling IPv6 packets with RGB option may be send to CPU for slow path processing, with rate-limiting, or be discarded according to the local policy.¶
For a router capable of MSR6 BE, such MSR6 BE packet MUST NOT be forwarded, but should be processed as a normal IPv6 packet with unknown option, or additionally and optionally be countered and logged if the router is capable of doing so.¶