Internet-Draft | BMP TLV | March 2023 |
Lucente & Gu | Expires 28 September 2023 | [Page] |
Most of the message types defined by the BGP Monitoring Protocol (BMP) make provision for data in TLV format. However, Route Monitoring messages (which provide a snapshot of the monitored Routing Information Base) and Peer Down messages (which indicate that a peering session was terminated) do not. Supporting (optional) data in TLV format across all BMP message types allows for a homogeneous and extensible surface that would be useful for the most different use-cases that need to convey additional data to a BMP station. While it is not intended for this document to cover any specific utilization scenario, it defines a simple way to support TLV data in all message types.¶
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 September 2023.¶
Copyright (c) 2023 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 Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
The BGP Monitoring Protocol (BMP) is defined in RFC 7854 [RFC7854].¶
The Route Monitoring message consists of:¶
The Peer Down Notification message consists of:¶
This means that both Route Monitoring and Peer Down messages have a non-extensible format (except for the specific case of Peer Down Reason Code 6 as defined by RFC 9069 [RFC9069]). In the Route Monitoring case, this prevents the transmission of characteristics of transported NLRIs (e.g. to help with stateless parsing) or of vendor-specific data. In the Peer Down case, this prevents matching with TLVs previously sent with the Peer Up message. The proposal of this document is to bump the BMP version, for backward compatibility, and allow all defined BMP message types to make provision for TLV data.¶
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 RFC 2119 [RFC2119] RFC 8174 [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
The TLV data type is already defined in Section 4.4 of [RFC7854] for the Initiation and Peer Up message types. A TLV consists of:¶
TLVs SHOULD be sorted by the sender by their code point. Multiple TLVs of the same type can be repeated as part of the same message, and it is left to the specific use-cases whether all, any, the first or the last TLV should be considered as well as whether ordering matters.¶
Route Monitoring messages may require per-NLRI TLVs, that is, there may be a need to map TLVs to NLRIs contained in the BGP Update message, for example, to express additional characteristics of a specific NLRI. For this purpose specifically, TLVs in Route Monitoring messages MUST be indexed, with the index starting at one (1) to refer to the first NLRI. Index zero (0) specifies that a TLV does apply to all NLRIs contained in the BGP Update message. Multiple TLVs of the same type and with the same index can be repeated as part of the same message. Indexed TLVs are encoded as in the following figure:¶
In indexed TLVs, the reported length refers to the total encoded TLV value (ie. it does exclude the length of the index field).¶
Of the BMP message types defined so far, indexed TLVs apply only to Route Monitoring messages and, for example, they do not apply to Route Mirroring messages because the sender may not be aware of the payload of the transported BGP Update message.¶
Section 4.1 of [RFC7854] defines the Common Header. While the structure remains unaltered, the following two definitions are changed:¶
The Route Monitoring message type is defined in Section 4.6 of [RFC7854]. The consistency model selected to extend this message type with TLVs is with the Route Mirroring type where the Per-peer header is being followed by TLVs.¶
The BGP Update PDU Section 4.3 of [RFC4271] is encoded itself as part of a BGP Message TLV with code point TBD1 and index set to zero. A Route Monitoring message MUST contain one BGP Message TLV which may be preceeded and followed by other optional TLV data.¶
In a Route Monitoring message where the BGP Update PDU carries N NLRIs, indexed TLVs do allow to handle the cases of 1:1 and N:1 relationship among NLRIs and TLVs. The case of N:M relationship does instead require a form of grouping and hence this document defines a Group TLV to achieve this.¶
The TLV MUST have an index set to zero. The value MUST contain:¶
A NLRI index can be listed as part of multiple Group TLVs within the same message. NLRI indexes within a Group TLV SHOULD be sorted by the sender. A Group Index can not reference an NLRI index 0. Finally, a Group TLV MUST NOT include its own or another Group Index. Multiple TLVs can point to the same Group Index, i.e. a group can be reused within the same Route Monitoring message.¶
The Group TLV code point is TBD2¶
The diagram in Figure 3 shows an example of a Route Monitoring message carrying a BGP UPDATE containing 10 NLRIs. The TLVs are comprised of:¶
The Information field contains a UTF-8 string whose value MUST be equal to the value of the VRF or table name (ie. RD instance name) being conveyed. The string size MUST be within the range of 1 to 255 bytes.¶
The VRF/Table Name TLV code point is TBD3¶
This document also defines the following new code points to help stateless parsing of BGP Update PDUs:¶
The Peer Down Notification message type is defined in Section 4.9 of [RFC7854]. The consistency model selected to extend this message type with TLVs is with the Peer Up type where optional TLVs are placed at the end of the message.¶
This means for Reason codes 1 or 3, a BGP Notification PDU follows; the PDU MAY be followed by TLV data. For Reason code 2, a 2-byte field to give additional FSM info follows; this field MAY be followed by TLV data. For all other Reason codes, TLV data MAY follow the Reason field.¶
All other message types defined in RFC7854 [RFC7854] do already provision for TLV data. It is RECOMMENDED that all future BMP message types will also provide for TLV data following a consistency model with existing message types.¶
It is worth nothing that RFC8654 [RFC8654] permits BGP Update and other messages to grow to a length of 65535 octets. This may cause a BMP PDU that attempts to encapsulate such long messages to overflow.¶
It is not believed that this document adds any additional security considerations.¶
In Route Monitoring messages, the number of TLVs can be bound to the amount of NLRIs carried in the BGP Update message. This may degrade the packing of information in such messages and have specific impacts on the memory and CPU used in a BMP implementation. As a result of that it should always be possible to disable such features to mitigate their impact.¶
This document requests the renaming of the "Peer Up TLVs" registry defined by BMP Peer Up Message Namespace [I-D.ietf-grow-bmp-peer-up] into "Peer Up and Peer Down TLVs" and the definition of one new registry "BMP Route Monitoring TLVs". As part of the "BMP Route Monitoring TLVs" registry, the following new TLV types are defined (Section 4.2):¶
The authors would like to thank Jeff Haas, Camilo Cardona, Thomas Graf, Pierre Francois, Ben Maddison, Tim Evens and Luuk Hendriks for their valuable input. The authors would also like to thank Greg Skinner, Zongpeng Du and Maxence Younsi for their review.¶