| Internet-Draft | Route Leak Detection | March 2024 |
| Gu, et al. | Expires 4 September 2024 | [Page] |
According to Route-Leak Problem Definition [RFC7908], Route leaks refer to the case that the delivery range of route advertisements is beyond the expected range. For many current security protection solutions, the ISPs (Internet Service Providers) are focusing on finding ways to prevent the happening of BGP [RFC4271] route leaks. However, the real-time route leak detection if any occurs is important as well, and serves as the basis for leak mitigation. This document extends the BGP Monitoring Protocol (BMP) [RFC7854] to provide a routing security scheme suitable for ISPs to detect BGP route leaks at the prefix level.¶
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 4 September 2024.¶
Copyright (c) 2024 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.¶
BMP: BGP Monitoring Protocol¶
BMS: BGP Monitoring Station¶
C2P: Customer to Provider¶
ISP: Internet Service Provider¶
P2C: Provider to Customer¶
P2P: Peer to Peer¶
RIB: Routing Information Base¶
RLP: Route Leak Protection¶
RLD: Route Leak Detection¶
RFC7908 [RFC7908] defines "Route Leak" as: A route leak is the propagation of routing announcement(s) beyond their intended scope, which can result in possible situations such as eavesdropping, device overload, routing black hole and so on. More specifically, the intended scope of route announcements is usually defined by local route filtering/distribution policies within devices. These policies are designed to realise the pair-wise peering business relationships between ASes (autonomous systems), which include Customer to Provider (C2P), Peer to Peer (Peer to Peer), and Provider to Customer (P2C). In a C2P relationship, the customer pays the transit provider for traffic sent between the two ASes. In return, the customer gains access to the ASes that the transit provider can reach, including those which the transit provider reaches through its own transit providers. In a P2P relationship, the peering ASes gain access to each other's customers, typically without either AS paying the other AS Relationships, Customer Cones, and Validation [Luckie].¶
More precisely, the route leaks we discuss in this draft, referring to Type 1, 2, 3, and 4 Route Leaks defined in RFC7908 [RFC7908], can be summarized as: a route leak occurs when a route received from a transit provider or a lateral peer is propagated to another transit provider or a lateral peer.¶
There are several types of approaches against route leak from different perspectives. In this draft, we mainly discuss the following three types:¶
Route leak prevention: The approach to prevent route leak from happening. Commonly used methods include inbound/outbound prefix/peer/AS filtering policies configured at the ingress/egress nodes of ASes per the propagation of BGP routes.¶
Route leak detection: The approach to detect the existence of route leaks that happen at either the local AS, or upstream AS per the propagation of BGP routes. An intuitive way of detecting route leak is by checking the business relationship information at both the ingress and egress nodes of the local AS along the BGP route propagation path with the route leak violation rules defined in RFC7908 [RFC7908]. Thus, it requires the knowledge of the actual route propagation trace, as well as the resulting business relationship information at the ingress and egress nodes. With the above information collected, the analysis can be done by the routing device or a centralized server. This draft specifies one such method.¶
Route leak mitigation: The approach to mitigate route leaks that already happened at either the local AS, or upstream AS per the propagation of BGP routes. Commonly used methods include reject, drop or stop propagating the invalid routes once detected the existence of leaks.¶
The above mentioned actions can be used alone or in combination, depending on the entities (routing devices, network manager) that execute the actions, and the relative positions of the executing entities from the leaking point (local or downstream).¶
Route Leak Prevention [I-D.ietf-idr-bgp-open-policy] updates the BGP Open negotiation process with a new BGP capability to exchange the BGP Roles between two BGP speakers, and also proposes to use a new BGP attribute, called the iOTC (Internal Only To Customer) Path attribute to mark routes according to the BGP Roles established in Open Message. The iOTC attribute of the incoming route is set at the ingress node of the local AS, and is conveyed with the BGP Update to the egress node of the local AS for outbound filtering to prevent route leaks in the local AS. This attribute is removed at the egress node before the BGP Update is sent to eBGP neighbors. For representation simplification, we use iOTC to refer to the method specified in Route Leak Prevention [I-D.ietf-idr-bgp-open-policy].¶
Route-Leak Detection and Mitigation [I-D.ietf-grow-route-leak-detection-mitigation] describes a route leak detection and mitigation solution based on conveying route-leak protection (RLP) information in a well-known transitive BGP community, called the RLP community. The RLP community helps with detection and mitigation of route leaks that happen at the upstream AS (per the BGP routes propagation), as an Inter-AS solution. For representation simplification, we use RLP to refer to the method specified in Route-Leak Detection and Mitigation [I-D.ietf-grow-route-leak-detection-mitigation].¶
The above two drafts provide solutions for route leak prevention, detection and mitigation. To summarize:¶
iOTC is used for route leak prevention of the local AS. It does not provide the detection or mitigation of route leaks of either local As or upstream AS per the BGP routes propagation.¶
iOTC is peer/AS-level route leak prevention, due to the fact the BGP Role negotiation is peer-level. It does not provide prefix-level route leak prevention.¶
RLP is used for route leak detection and mitigation of route leak that happens in the upstream AS (per the BGP Update distribution). It is prefix-level detection and mitigation.¶
Thus, there lacks method for local AS route leak detection.¶
Considering the challenges facing the existing approaches, this draft proposes a method called Route Leak Detection (RLD). It utilizes the BGP Monitoring Protocol (BMP) to convey the RLD information from to the BMP server to realize centralized leak detection. BMP is currently deployed by OTT and carriers to monitor the BGP routes, such as monitoring BGP Adj-RIB-In using the process defined in RFC7854 [RFC7854], and monitoring BGP Adj-RIB-Out using the process defined in RFC8671 [RFC8671]. On the other hand, the RLD information is in fact a representation of the business relationships between the local AS and its neighboring AS. It does not involve any information disclosure issue regarding third parties. Thus, a single ISP can deploy RLD without relying on any information from either other ISPs or other third parties.¶
A RLD TLV is defined for the BMP Route Monitoring Message. Considering that the AS relationships are sometimes per route based instead of per peer/AS based, this TLV is appended to each route, following the BGP Update Message. The order of placing the RLD TLV among other BMP supported TLVs is out of the scope of this draft. The TLV format is defined as follows:¶
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type (2 octets) | Length (2 octets) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Value(1 octet)|
+-+-+-+-+-+-+-+-+
Figure 1: RLD TLV
¶
Type (2 octets) = TBD1, the RLD TLV represents the prefix-level business relationship between the transmitter AS and the receiver AS. The local AS is a transmitter or a receiver, depending on if the route is an incoming route from a neighbor AS or an outgoing route to a neighbor AS.¶
Length (2 octets): Defines the length of the Value filed. It SHOULD be set to 0x01, considering the Value field is of 1 octet fixed length.¶
Value (1 octet): Currently 4 values are defined to represent the business relationships, which are specified in Table 1.¶
+-------+-------------+ | Value | Business | | | Relationship| +---------------------+ | 0 | P2C | | 1 | C2P | | 2 | P2P | | 3 | I2I | +-------+-------------+ Table 1: Business relationship value¶
The RLD TLV, presenting the business relationship between the neighbor AS and the local AS of the incoming route, SHOULD be prepended to the Adj-RIB-In at the ingress node of the local AS. The RLD TLV, representing the business relationship between the local AS and the neighbor AS of the outgoing route, SHOULD also be prepended to the Adj-RIB-Out at the egress node of the local AS. The BMP server, by analyzing the above two RLD TLVs of the same route, can use the rules defined in RFC7908 [RFC7908] to detect the existence of any route leak. As example is shown in Figure 2.¶
+------------+
| BMP server |
+------> + +<-------+
| | RLD server | |
+ +------------+ +
BMP Adj_RIB_In: BMP Adj_RIB_Out:
(AS2 --> AS1) (AS1 --> AS4)
P2C C2P
| +
***|********************* |
* | * "Send Route
Route A * | AS1 +*+---> A to AS3"
+--> * | + * | +-----+
+-----+ * +---+ +---+ *+P2C+-----+ AS3 +----+ ...
+--+ AS2 +---+P2C+*+-+ R1+---------+ R2| * | +-----+
+-----+ * +-+-+\ +---+ * |
* | \\ // |\ * |
* | \\// | \ * "Do not send
* | //\ | \+-----> Route A to AS4"
* | // \\ | * | +-----+
* | / \ | *+C2P+-----+ AS4 +----+ ...
* +-+-+ +-+-+ * | +-----+
+--+ R3+---------+ R4+----------+
* +---+ +---+ *
* *
*************************
Figure 2: RLD depolyment by a single ISP
¶
As shown in Figure 2, with the RLD TLV attached to each Route Monitoring Message, the RLD server (also working as the BMP server) combines the BMP Adj_RIB_In message collected from R1 and the BMP Adj_RIB_Out message collected from R4 to decide if there's a route leak. For example, if the RLD TLV in R1's Adj_RIB_In message indicates a value of "0", and the RLD TLV in R4's Adj_RIB_Out message indicates a value of "1", then the RLD server knows there exists a route leak.¶
RLD can be used as a complementary method to the existing methods against route leaks. More specifically, RLD can coordination with both iOTC and RLP.¶
With the settlement of the iOTC draft, the iOTC attribute is naturally included in the BGP Update and can be collected to the BMP server without BMP extension. With the RLD TLV collected also by BMP (more specifically, the iBGP Adj-RIB-In of the ingress node), the BMP server can do validate the consistency of the iOTC attribute with the RLD. If contradiction detected, the BMP server may further check the bussiness contract for the actual business relationship.¶
For special prefixes that does not obey the peer/AS level business relationship negotiated through BGP Open Message, the BMP server can use the RLD TLV to detect such routes since the RLD TLV is set at prefix level.¶
For devices that do not support RLP, using RLD to collect the BGP routes, which conveys the RLD information from upstream ASes, allows the BMP server to detect and mitigate the route leaks that happen in the upstream AS. In other words, the detection and mitigation process can be also done in the BMP server, should the BMP server collects the BGP Update messages at the ingress or egress nodes.¶
Haibo Wang¶
Huawei¶
Email: rainsword.wang@huawei.com¶
This document defines the following new BMP Route Monitoring message TLV type (Section 4.1):¶
Type = TBD1, the RLD TLV represents the prefix-level business relationship between the transmitter AS and the receiver AS. The local AS is a transmitter or a receiver, depending on if the route is an incoming route from a neighbor AS or an outgoing route to a neighbor AS.¶
It is not believed that this document adds any additional security considerations.¶