Signaling MSD (Maximum SID Depth) using Border Gateway Protocol

When Segment Routing (SR) paths are computed by a centralized controller, it is critical that the controller learn the Maximum SID Depth (MSD) that can be imposed at each node/link on a given SR path. This ensures that the Segment Identifier (SID) stack depth of a computed path doesn't exceed the number of SIDs the node is capable of imposing. defines how to signal MSD in the Path Computation Element Protocol (PCEP). The OSPF and IS-IS extensions for signaling of MSD are defined in and respectively. However, if PCEP is not supported/configured on the head-end of a SR tunnel or a Binding-SID anchor node, and controller does not participate in IGP routing, it has no way of learning the MSD of nodes and links. BGP-LS defines a way to expose topology and associated attributes and capabilities of the nodes in that topology to a centralized controller. This document defines extensions to BGP-LS to advertise one or more types of MSDs at node and/or link granularity. Other types of MSD are known to be useful. For example, and define Readable Label Depth Capability (RLDC) that is used by a head-end to insert an Entropy Label (EL) at a depth that can be read by transit nodes. In the future, it is expected that new MSD-Types will be defined to signal additional capabilities, e.g., ELs, SIDs that can be imposed through recirculation, or SIDs associated with another data plane such as IPv6. MSD advertisements may be useful even if SR itself is not enabled. For example, in a non-SR MPLS network, MSD defines the maximum label depth.

MSD: Maximum SID Depth - the number of SIDs supported by a node or a link on a node PCC: Path Computation Client PCE: Path Computation Element PCEP: Path Computation Element Protocol SID: Segment Identifier as defined in SR: Segment Routing Label Imposition: Imposition is the act of modifying and/or adding labels to the outgoing label stack associated with a packet. This includes: replacing the label at the top of the label stack with a new label. pushing one or more new labels onto the label stack. The number of labels imposed is then the sum of the number of labels that are replaced and the number of labels that are pushed. See for further details.

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here .

This document describes extensions that enable BGP-LS speakers to signal the MSD capabilities ( ) of nodes and their links in a network to a BGP-LS consumer of network topology such as a centralized controller. The centralized controller can leverage this information in computation of SR paths based on their MSD capabilities. When a BGP-LS speaker is originating the topology learnt via link-state routing protocols like OSPF or IS-IS, the MSD information for the nodes and their links is sourced from the underlying extensions as defined in and respectively. The BGP-LS speaker may also advertise the MSD information for the local node and its links when not running any link-state IGP protocol e.g. when running BGP as the only routing protocol. The Protocol-ID field should be set to BGP since the link and node attributes have BGP based identifiers. Deployment model for such case would be: a limited number (meeting resiliecy requirements) of BGP-LS speakers exposing the topology to the controller, full-mesh/RouteReflectors for iBGP(Internal Border Gateway Protocol) or regular eBGP(External Border Gateway Protocol) connectivity between nodes in the topology. The extensions introduced in this document allow for advertisement of different MSD-Types, which are defined elsewhere and were introduced in . This enables sharing of MSD-Types that may be defined in the future by the IGPs in BGP-LS.

The Node MSD ( ) is encoded in a new Node Attribute TLV to carry the provisioned SID depth of the router identified by the corresponding Router-ID. Node MSD is the smallest MSD supported by the node on the set of interfaces configured for use. MSD values may be learned via a hardware API or may be provisioned. The following format is used:

Where: Type: 266 Length: variable (multiple of 2); represents the total length of the value field in octets. Value : consists of one or more pairs of a 1-octet MSD-Type and 1-octet MSD-Value. MSD-Type : one of the values defined in the "IGP MSD-Types" registry defined in . MSD-Value : a number in the range of 0-255. For all MSD-Types, 0 represents the lack of ability to impose an MSD stack of any depth; any other value represents that of the node. This value MUST represent the lowest value supported by any link configured for use by the advertising protocol instance.

The Link MSD ( ) is defined to carry the MSD of the interface associated with the link. It is encoded in a new Link Attribute TLV using the following format:

Where: Type: 267 Length: variable (multiple of 2); represents the total length of the value field in octets. Value : consists of one or more pairs of a 1-octet MSD-Type and 1-octet MSD-Value. MSD-Type : MSD-Type : one of the values defined in the "IGP MSD-Types" registry defined in . MSD-Value : a number in the range of 0-255. For all MSD-Types, 0 represents the lack of ability to impose an MSD stack of any depth; any other value represents that of the link when used as an outgoing interface.

When Link MSD is present for a given MSD-type, the value of the Link MSD MUST take precedence over the Node MSD. When a Link MSD-type is not signaled but the Node MSD-type is, then the Node MSD-type value MUST be considered as the MSD value for that link. In order to increase flooding efficiency, it is RECOMMENDED that routers with homogenous link MSD values advertise just the Node MSD value. The meaning of the absence of both Node and Link MSD advertisements for a given MSD-type is specific to the MSD-type. Generally it can only be inferred that the advertising node does not support advertisement of that MSD-type. However, in some cases the lack of advertisement might imply that the functionality associated with the MSD-type is not supported. The correct interpretation MUST be specified when an MSD-type is defined in .

This document requests assigning code-points from the registry "BGP-LS Node Descriptor, Link Descriptor, Prefix Descriptor, and Attribute TLVs" based on table below. Early allocation for these code-points have been done by IANA.

The new protocol extensions introduced in this document augment the existing IGP topology information that is distributed via . Procedures and protocol extensions defined in this document do not affect the BGP protocol operations and management other than as discussed in the Manageability Considerations section of . Specifically, the malformed attribute tests for syntactic checks in the Fault Management section of now encompass the new BGP-LS Attribute TLVs defined in this document. The semantic or content checking for the TLVs specified in this document and their association with the BGP-LS NLRI types or their BGP-LS Attribute is left to the consumer of the BGP-LS information (e.g. an application or a controller) and not the BGP protocol. A consumer of the BGP-LS information retrieves this information over a BGP-LS session (refer Section 1 and 2 of ). This document only introduces new Attribute TLVs and any syntactic error in them would result in the BGP-LS Attribute being discarded . The MSD information introduced in BGP-LS by this specification, may be used by BGP-LS consumer applications like a SR path computation engine (PCE) to learn the SR SID stack handling capabilities of the nodes in the topology. This can enable the SR PCE to perform path computations taking into consideration the size of SID stack that the specific head-end node may be able to impose. Errors in the encoding or decoding of the MSD information may result in the unavailability of such information to the SR PCE or incorrect information being made available to it. This may result in the head-end node not being able to instantiate the desired SR path in its forwarding and provide the SR based optimization functionality. The handling of such errors by applications like SR PCE may be implementation specific and out of scope of this document. The extensions specified in this document, do not specify any new configuration or monitoring aspects in BGP or BGP-LS. The specification of BGP models is an ongoing work based on the .

The advertisement of an incorrect MSD value may have negative consequences. If the value is smaller than supported, path computation may fail to compute a viable path. If the value is larger than supported, an attempt to instantiate a path that can't be supported by the head-end (the node performing the SID imposition) may occur. The presence of this information may also inform an attacker of how to induce any of the aforementioned conditions. The procedures and protocol extensions defined in this document do not affect the BGP security model. See the "Security Considerations" section of for a discussion of BGP security. Also, refer to and for analyses of security issues for BGP. Security considerations for acquiring and distributing BGP-LS information are discussed in . The TLVs introduced in this document are used to propagate the MSD IGP extensions defined in . It is assumed that the IGP instances originating these TLVs will support all the required security (as described in ) in order to prevent any security issues when propagating the TLVs into BGP-LS. The advertisement of the node and link attribute information defined in this document presents no additional risk beyond that associated with the existing node and link attribute information already supported in .

We like to thank Acee Lindem, Stephane Litkowski, Bruno Decraene and Alvaro Retana for their reviews and valuable comments.