| Internet-Draft | Inter-domain SAVNET Problem Statement | March 2025 |
| Li, et al. | Expires 4 September 2025 | [Page] |
This document provides a gap analysis of existing inter-domain source address validation mechanisms, describes the problem space, and defines the requirements for technical improvements.¶
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 2025.¶
Copyright (c) 2025 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.¶
Source address validation (SAV) is crucial for protecting networks from source address (SA) spoofing attacks [RFC2827] [RFC3704] [RFC8704]. The MANRS initiative advocates deploying SAV as close to the source as possible [manrs], and access networks are the first line of defense against source address spoofing. However, access networks face various challenges in deploying SAV mechanisms due to different network environments, router vendors, and operational preferences. Hence, SAV may not be deployed ubiquitously in access networks. In addition, SA spoofing may also originate in ISP networks at higher levels of heirarchy in the Internet. So, deployment of SAV mechanisms in the edge routers of enterprises as well as the ISP networks (at different heirarchal levels or tiers) is needed to prevent source address spoofing along the data forwarding paths. [RFC5210] highlighted the importance of SAV at various network locations: access, intra-domain, and inter-domain. This document focuses on providing gap analysis and describing the problem space of existing inter-domain SAV solutions, and defining the requirements for a new solution of these problems. Access Control List (ACL) and unicast Reverse Path Forwarding (uRPF) techniques are currently utilized for inter-domain SAV [RFC3704] [RFC8704]. Here only existing IETF RFCs are considered as the state of the art (BCP 38 [RFC2827] and BCP 84 [RFC3704] [RFC8704]); IETF works-in-progress are not included in that.¶
There are several existing mechanisms for inter-domain SAV. This document analyzes them and attempts to answer: i) what are the technical gaps (Section 4), ii) what are the fundamental problems (Section 5), and iii) what are the practical requirements for the solution of these problems (Section 6).¶
The following summarizes the fundamental problems with existing SAV mechanisms, as analyzed in Section 4 and Section 5:¶
Improper block: Existing uRPF-based mechanisms suffer from improper block in two inter-domain scenarios: imited propagation of prefixes and hidden prefixes.¶
Improper permit: Existing uRPF-based mechanisms exhibit improper permit in scenarios involving source address spoofing within a customer cone or from a provider/peer AS.¶
High operational overhead: ACL-based ingress SAV filtering introduces significant operational overhead, as it needs to update ACL rules manually to adapt to prefix or routing changes in a timely manner.¶
To address these problems, in Section 6, this document outlines the following technical requirements for a new solution:¶
Improving validation accuracy over existing mechanisms: A new solution MUST avoid improper block and minimize improper permit.¶
Reducing operational overhead: A new solution MUST have less operational overhead than ACL-based ingress SAV filtering.¶
In addition, this document defines three more requirements to ensure practicality:¶
Working in incremental/partial deployment: A new solution MUST NOT assume pervasive adoption and SHOULD provide effective protection for source addresses when it is partially deployed in the Internet.¶
Providing necessary security guarantee: A new solution SHOULD secure the communicated information between ASes if it requires exchanging specific information between ASes.¶
Guaranteeing convergence: A new solution SHOULD achieve accurate SAV rule convergence in response to prefix or routing changes.¶
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 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.¶
The rule that indicates the validity of a specific source IP address or source IP prefix.¶
The validation results that the packets with legitimate source addresses are blocked improperly due to inaccurate SAV rules.¶
The validation results that the packets with spoofed source addresses are permitted improperly due to inaccurate SAV rules.¶
The paths that the legitimate traffic goes through in the data plane.¶
Inter-domain SAV is typically performed at the AS level (on a per neighbor-AS-interface basis) and can be deployed at AS border routers (ASBRs) to prevent source address spoofing. There are various mechanisms available to implement inter-domain SAV for anti-spoofing ingress filtering [nist] [manrs] [isoc], which are reviewed in this section.¶
ACL-based ingress filtering [RFC3704]: ACL-based ingress SAV filtering is a technique that relies on ACL rules to filter packets based on their source addresses. It can be applied at provider interfaces, peer interfaces, or customer interfaces of an AS, and is recommended for deployment at provider interfaces [manrs]. At the provider interfaces, ACL-based ingress SAV filtering can block source prefixes that are clearly invalid in the inter-domain routing context, such as IANA special purpose or unallocated IPv4/IPv6 prefixes and the AS's internal-only prefixes. However, ACL-based ingress SAV filtering introduces significant operational overhead, as ACL rules need to be updated in a timely manner to reflect prefix or routing changes in the inter-domain routing system. It is also impractical to store a very large and dynamically varying unallocated IPv6 prefixes. At the customer interfaces, ACL-based ingress filtering is less desirable. Other techniques (as described below) are more effective for ingress SAV filtering on customer interfaces. ACL-based ingress SAV filtering has applicability for broadband cable or digital subscriber access loop (DSL) access networks where the service provider has clear knwoledge of IP address prefixes it has allocated to manage those services.¶
uRPF-based mechanisms: A class of SAV mechanisms are based on Unicast Reverse Path Forwarding (uRPF) [RFC3704]. The core idea of uRPF for SAV is to exploit the symmetry of inter-domain routing: in many cases, the best next hop for a destination is also the best previous hop for the source. In other words, if a packet arrives from a certain interface, the source address of that packet should be reachable via the same interface, according to the FIB. However, symmetry in routing does not always holds in practice, and to address cases where it does not hold, many enhancements and modes of uRPF are proposed. Different modes of uRPF have different levels of strictness and flexibility, and network operators can choose from them to suit particular network scenarios. We describe these modes as follows:¶
Strict uRPF [RFC3704]: Strict uRPF is the most stringent mode, and it only permits packets that have a source address that is covered by a prefix in the FIB, and that the next hop for that prefix is the same as the incoming interface. This mode is recommended for deployment at customer interfaces that directly connect to an AS with suballocated address space, as it can prevent spoofing attacks from that AS or its downstream ASes [nist].¶
Loose uRPF [RFC3704]: Loose uRPF verifies that the source address of the packet is routable in the Internet by matching it with one or more prefixes in the FIB, regardless of which interface the packet arrives at. If the source address is not routable, Loose uRPF discards the packet. Loose uRPF is typically deployed at the provider interfaces of an AS to block packets with source addresses that are obviously disallowed, such as non-global prefixes (e.g., private addresses, multicast addresses, etc.) or the prefixes that belong to the customer AS itself [nist].¶
FP-uRPF [RFC3704]: FP-uRPF maintains a reverse path forwarding (RPF) list, which contains the prefixes and all their permissible routes including the optimal and alternative ones. It permits an incoming packet only if the packet's source address is encompassed in the prefixes of the RPF list and its incoming interface is included in the permissible routes of the corresponding prefix. FP-uRPF is recommended to be deployed at customer interfaces or peer interfaces, especially those that are connected to multi-homed customer ASes [nist].¶
Virtual routing and forwarding (VRF) uRPF [RFC4364] [urpf] [manrs]: VRF uRPF uses a separate VRF table for each external BGP peer and is only a way of implementation for a SAV table.¶
EFP-uRPF [RFC8704]: EFP-uRPF consists of two algorithms, algorithm A and algorithm B. EFP-uRPF is based on the idea that an AS can receive BGP updates for multiple prefixes that have the same origin AS at different interfaces. For example, this can happen when the origin AS is multi-homed and advertises the same prefixes to different providers. In this case, EFP-uRPF allows an incoming packet with a source address in any of those prefixes to pass on any of those interfaces. This way, EFP-uRPF can handle asymmetric routing scenarios where the incoming and outgoing interfaces for a packet are different. EFP-uRPF has not been implemented in practical networks yet, but BCP84 [RFC3704] [RFC8704] suggests using EFP-uRPF with algorithm B at customer interfaces of an AS. EFP-uRPF can also be used at peer interfaces of an AS.¶
Carrier Grade NAT (CGN): CGN is a network technology used by service providers to translate between private and public IPv4 addresses within their network. CGN enables service providers to assign private IPv4 addresses to their customer ASes instead of public, globally unique IPv4 addresses. The private side of the CGN faces the customer ASes, and when an incoming packet is received from a customer AS, CGN checks its source address. If the source address is included in the address list of the CGN's private side, CGN performs address translation. Otherwise, it forwards the packet without translation. However, since CGN cannot determine whether the source address of an incoming packet is spoofed or not, additional SAV mechanisms need to be implemented to prevent source address spoofing [manrs].¶
BGP origin validation (BGP-OV) [RFC6811]: Attackers can bypass uRPF-based SAV mechanisms by using prefix hijacking in combination with source address spoofing. By announcing a less-specific prefix that does not have a legitimate announcement, the attacker can deceive existing uRPF-based SAV mechanisms and successfully perform address spoofing. To protect against this type of attack, a combination of BGP-OV and uRPF-based mechanisms like FP-uRPF or EFP-uRPF is recommended [nist]. BGP routers can use ROA information, which is a validated list of {prefix, maximum length, origin AS}, to mitigate the risk of prefix hijacks in advertised routes.¶
Inter-domain SAV is essential in preventing source address spoofing traffic across all AS interfaces, including those of customers, providers, and peers. An ideal inter-domain SAV mechanism MUST block all spoofing traffic while permitting legitimate traffic in all scenarios. However, in some cases, existing SAV mechanisms may unintentionally block legitimate traffic or permit spoofing traffic. This section aims to conduct a gap analysis of existing SAV mechanisms used in the corresponding interfaces of these scenarios to identify their technical limitations.¶
SAV is used at customer interfaces to validate traffic from the customer cone, including both legitimate traffic and spoofing traffic. To prevent the source address spoofing, operators can enable ACL-based ingress filtering and/or uRPF-based mechanisms at customer interfaces, namely Strict uRPF, FP-uRPF, or EFP-uRPF. However, uRPF-based mechanisms may cause improper block problems in two inter-domain scenarios: limited propagation of prefixes and hidden prefixes, or may cause improper permit problems in the scenarios of source address spoofing within a customer cone, while ACL-based SAV ingress filtering needs to update SAV rules in a timely manner and lead to high operational overhead.¶
+--------------------+------------+-----------+-------+--------+ |Traffic & Scenarios | ACL |Strict uRPF|FP-uRPF|EFP-uRPF| +----------+---------+------------+-----------+-------+--------+ |Legitimate| LPP | | | |Traffic +---------+ | Improper Block | | | HP | High | | +----------+---------+Operational +-------------------+--------+ |Spoofing |Spoofing | Overhead | |Improper| |Traffic | within | | Functioning as |Permit | | | a CC | | Expected | | +----------+---------+------------+-------------------+--------+ "LPP" represents a class of scenario called limited propagation of prefixes. "HP" represents a class of scenario called hidden prefixes. "Spoofing within a CC" represents a class of scenario where spoofing traffic occurs within a customer cone (CC) and the spoofed source addresses belong to this customer cone. "Functioning as Expected" represents the inter-domain SAV mechanism does not cause improper block for legitimate traffic or improper permit for spoofing traffic in the corresponding scenarios, and has low operational overhead.
Figure 1 provides an overview of the gaps associated with ACL-based ingress filtering, Strict uRPF, FP-uRPF, and EFP-uRPF for SAV at customer interfaces in the corresponding scenarios. ACL-based ingress filtering has high operational overhead as performing SAV at customer interfaces. Strict uRPF, FP-uRPF, and EFP-uRPF, on the other hand, may incorrectly block legitimate traffic in the scenarios of limited propagation of prefixes or hidden prefixes. Furthermore, in the scenarios of source address spoofing within a customer cone, EFP-uRPF with algorithm B may inadvertently permit the spoofing traffic.¶
In the following, we analyze the gaps of Strict uRPF, FP-uRPF, and EFP-uRPF for SAV at customer interfaces in scenarios of limited propagation of prefixes, hidden prefixes, and source address spoofing within a customer cone, respectively.¶
In inter-domain networks, some prefixes may not be propagated to all domains due to various factors, such as NO_EXPORT or NO_ADVERTISE communities or other route filtering policies. This may cause asymmetric routing in the inter-domain context, which may lead to improper block when performing SAV with existing mechanisms. These mechanisms include EFP-uRPF, which we focus on in the following analysis, as well as Strict uRPF and FP-uRPF. All these mechanisms suffer from the same problem of improper block in this scenario.¶
+----------------+
| AS 3(P3) |
+-+/\------+/\+--+
/ \
/ \
/ \
/ (C2P) \
+------------------+ \
| AS 4(P4) | \
++/\+--+/\+----+/\++ \
/ | \ \
P2[AS 2] / | \ \
/ | \ \
/ (C2P) | \ P5[AS 5] \ P5[AS 5]
+----------------+ | \ \
| AS 2(P2) | | P1[AS 1] \ \
+----------+/\+--+ | P6[AS 1] \ \
\ | NO_EXPORT \ \
P1[AS 1] \ | \ \
NO_EXPORT \ | \ \
\ (C2P) | (C2P/P2P) (C2P) \ (C2P) \
+----------------+ +----------------+
| AS 1(P1, P6) | | AS 5(P5) |
+----------------+ +----------------+
Figure 2 presents a scenario where the limited propagation of prefixes occurs due to the NO_EXPORT community attribute. In this scenario, AS 1 is a customer of AS 2, AS 2 is a customer of AS 4, AS 4 is a customer of AS 3, and AS 5 is a customer of both AS 3 and AS 4. The relationship between AS 1 and AS 4 can be either customer-to-provider (C2P) or peer-to-peer (P2P). AS 1 advertises prefixes P1 to AS 2 and adds the NO_EXPORT community attribute to the BGP advertisement sent to AS 2, preventing AS 2 from further propagating the route for prefix P1 to AS 4. Similarly, AS 1 adds the NO_EXPORT community attribute to the BGP advertisement sent to AS 4, resulting in AS 4 not propagating the route for prefix P6 to AS 3. Consequently, AS 4 only learns the route for prefix P1 from AS 1 in this scenario. Suppose AS 1 and AS 4 have deployed inter-domain SAV while other ASes have not, and AS 4 has deployed EFP-uRPF at its customer interfaces.¶
Assuming that AS 1 is the customer of AS 4, if AS 4 deploys EFP-uRPF with algorithm A at customer interfaces, it will require packets with source addresses in P1 to only arrive from AS 1. When AS 1 sends legitimate packets with source addresses in P1 to AS 4 through AS 2, AS 4 improperly blocks these packets. The same problem applies to Strict uRPF and FP-uRPF. Although EFP-uRPF with algorithm B can avoid improper block in this case, network operators need to first determine whether limited prefix propagation exists before choosing the suitable EFP-uRPF algorithms, which adds more complexity and overhead to network operators. Furthermore, EFP-uRPF with algorithm B is not without its problems. For example, if AS 1 is the peer of AS 4, AS 4 will not learn the route of P1 from its customer interfaces. In such case, both EFP-uRPF with algorithm A and algorithm B have improper block problems.¶
Figure 4 portrays a scenario of source address spoofing within a customer cone and is used to analyze the gaps of uRPF-based mechanisms below.¶
+----------------+
| AS 3(P3) |
+-+/\----+/\+----+
/ \
/ \
/ \
/ (C2P) \
+----------------+ \
| AS 4(P4) | \
++/\+--+/\+--+/\++ \
P6[AS 1, AS 2] / | \ \
P1[AS 1, AS 2] / | \ \
P2[AS 2] / | \ \
/ (C2P) | \ P5[AS 5] \ P5[AS 5]
+----------------+ | \ \
Spoofer(P5')-+ AS 2(P2) | | P1[AS 1] \ \
+----------+/\+--+ | P6[AS 1] \ \
\ | \ \
P6[AS 1] \ | \ \
P1[AS 1] \ | \ \
\ (C2P) | (C2P) (C2P) \ (C2P) \
+----------------+ +----------------+
| AS 1(P1, P6) | | AS 5(P5) |
+----------------+ +----------------+
P5' is the spoofed source prefix P5 by the spoofer which is inside of
AS 2 or connected to AS 2 through other ASes.
In Figure 4, the source address spoofing takes place within AS 4's customer cone, where the spoofer, which is inside of AS 2 or connected to AS 2 through other ASes, sends spoofing traffic with spoofed source addresses in P5 to AS 3 along the path AS 2->AS 4-> AS 3. The arrows in Figure 4 illustrate the commercial relationships between ASes. AS 3 serves as the provider for AS 4 and AS 5, while AS 4 acts as the provider for AS 1, AS 2, and AS 5. Additionally, AS 2 is the provider for AS 1. Suppose AS 1 and AS 4 have deployed inter-domain SAV, while the other ASes have not.¶
If AS 4 deploys EFP-uRPF with algorithm B at its customer interfaces, it will allow packets with source addresses in P5 to originate from AS 1, AS 2, and AS 5. Consequently, when the spoofer which is inside of AS 2 or connected to AS 2 through other ASes sends spoofing packets with spoofed source addresses in P5 to AS 3, AS 4 will improperly permit these packets, thus enabling the spoofing traffic to propagate.¶
In scenarios like these, Strict uRPF, FP-uRPF, VRF uRPF, and EFP-uRPF with algorithm A do not suffer from improper permit problems. This is because these mechanisms enforce strict filtering rules that ensure packets with source addresses in P5 are only permitted to arrive at AS 4's customer interfaces facing AS 5.¶
SAV is used at provider/peer interfaces to validate traffic entering the customer cone, including both legitimate and spoofing traffic. To prevent packets with spoofed source addresses from the provider/peer AS, ACL-based ingress filtering and/or Loose uRPF can be deployed [nist].¶
+------------------------+------------+---------------+ | Traffic & Scenarios | ACL | Loose uRPF | +----------+-------------+------------+---------------+ |Legitimate| Any | | Functioning | |Traffic | Scenarios | High | as Expected | +----------+-------------+Operational +---------------+ |Spoofing | Spoofing | Overhead | | |Traffic | from | |Improper Permit| | |Provider/Peer| | | | | AS | | | +----------+-------------+------------+---------------+ "Spoofing from provider/peer AS" represents a class of scenario where source address spoofing traffic from provider/peer AS occurs and the spoofed source addresses belong to the customer cone which the spoofing traffic enters. "Functioning as Expected" represents the inter-domain SAV mechanism does not cause improper block for legitimate traffic or improper permit for spoofing traffic in the corresponding scenarios, and has low operational overhead.
Figure 5 summarizes the gaps of ACL-based ingress filtering and Loose uRPF for SAV at provider/peer interfaces in the corresponding scenarios. ACL-based ingress filtering effectively blocks spoofing traffic from provider/peer AS, while appropriately allowing legitimate traffic. However, these methods may come with high operational overhead. On the other hand, Loose uRPF correctly permits legitimate traffic, but it can also mistakenly allow spoofing traffic to pass through.¶
In the following, we expose the limitations of ACL-based ingress filtering and Loose uRPF for SAV at provider/peer interfaces in scenarios of source address spoofing from provider/peer AS. The source address spoofing from provider/peer AS represents a class of scenario where spoofing traffic comes from a provider/peer AS and the spoofed source addresses belong to the customer cone which the spoofing traffic enters.¶
Figure 6 depicts the scenario of source address spoofing from provider/peer AS and is used to analyze the gaps of ACL-based ingress filtering and Loose uRPF below.¶
+----------------+
Spoofer(P1')+-+ AS 3(P3) |
+-+/\----+/\+----+
/ \
/ \
/ \
/ (C2P/P2P) \
+----------------+ \
| AS 4(P4) | \
++/\+--+/\+--+/\++ \
P6[AS 1, AS 2] / | \ \
P1[AS 1, AS 2] / | \ \
P2[AS 2] / | \ \
/ (C2P) | \ P5[AS 5] \ P5[AS 5]
+----------------+ | \ \
| AS 2(P2) | | P1[AS 1] \ \
+----------+/\+--+ | P6[AS 1] \ \
\ | \ \
P6[AS 1] \ | \ \
P1[AS 1] \ | \ \
\ (C2P) | (C2P) (C2P) \ (C2P) \
+----------------+ +----------------+
| AS 1(P1, P6) | | AS 5(P5) |
+----------------+ +----------------+
P1' is the spoofed source prefix P1 by the spoofer which is inside of
AS 3 or connected to AS 3 through other ASes.
In the case of Figure 6, the spoofer which is inside of AS 3 or connected to AS 3 through other ASes forges the source addresses in P1 and sends the spoofing traffic to the destination addresses in P2. The arrows in Figure 6 represent the commercial relationships between ASes. AS 3 acts as the provider or lateral peer of AS 4 and the provider for AS 5, while AS 4 serves as the provider for AS 1, AS 2, and AS 5. Additionally, AS 2 is the provider for AS 1. Suppose AS 1 and AS 4 have deployed inter-domain SAV, while the other ASes have not.¶
By applying ACL-based ingress filtering at the provider/peer interface of AS 4, the ACL rules can block any packets with spoofed source addresses from AS 3 in P1. However, this approach incurs heavy operational overhead, as it requires network operators to update the ACL rules promptly based on changes in prefixes or topology of AS 4's customer cone. Otherwise, it may cause improper block of legitimate traffic or improper permit of spoofing traffic.¶
Loose uRPF can greatly reduce the operational overhead because it uses the local FIB as information source, and can adapt to changes in the network. However, it would improperly permit spoofed packets. In Figure 6, Loose uRPF is enabled at AS 4's provider/peer interface, while EFP-uRPF is enabled at AS 4's customer interfaces. A spoofer inside AS 3 or connected to it through other ASes may send packets with source addresses spoofing P1 to AS 2. As AS 3 lacks deployment of inter-domain SAV, the spoofing packets will reach AS 4's provider/peer interface. With Loose uRPF, AS 4 cannot block them at its provider/peer interface facing AS 3, and thus resulting in improper permit.¶
+--------+----------+---------+----------+-------+----------+ |Problems| ACL | Strict | Loose |FP-uRPF|EFP-uRPF | | | | uRPF | uRPF | | | +--------+----------+---------+----------+-------+----------+ |Improper|Not Exist | Exist |Not Exist | Exist | |Block | |(LPP, HP)| | (LPP, HP) | +--------+----------+---------+----------+-------+----------+ |Improper| Not Exist | Exist |Not | Exist | |Permit | | (SPP) |Exist | (SCC) | +--------+----------+---------+----------+-------+----------+ | | Exist | | | HOO | (Any | Not Exist | | |Scenarios)| | +--------+----------+---------------------------------------+ HOO: High Operational Overhead. "LPP" represents a class of scenario called limited propagation of prefixes. "HP" represents a class of scenario called hidden prefixes. "SPP" represents a class of scenario called source address spoofing from provider/peer AS. "SCC" represents a class of scenario called source address spoofing within a customer cone.
Based on the analysis above, we conclude that existing inter-domain SAV mechanisms exhibit limitations in asymmetric routing scenarios, leading to potential issues of improper block or improper permit. Additionally, these mechanisms can result in high operational overhead, especially when network routing undergoes dynamic changes. Figure 7 provides a comprehensive summary of scenarios where existing inter-domain SAV mechanisms may encounter issues, including instances of improper blocking of legitimate traffic, improper permitting of spoofing traffic, or high operational overhead.¶
For ACL-based ingress filtering, network operators need to manually update ACL rules to adapt to network changes. Otherwise, they may cause improper block or improper permit issues. Manual updates induce high operational overhead, especially in networks with frequent policy and route changes.¶
Strict uRPF and Loose uRPF are automatic SAV mechanisms, thus they do not need any manual effort to adapt to network changes. However, they have issues in scenarios with asymmetric routing. Strict uRPF may cause improper block problems when an AS is multi-homed and routes are not symmetrically announced to all its providers. This is because the local FIB may not include the asymmetric routes of the legitimate packets, and Strict uRPF only uses the local FIB to check the source addresses and incoming interfaces of packets. Loose uRPF may cause improper permit problems and fail to prevent source address spoofing. This is because it is oblivious to the incoming interfaces of packets.¶
FP-uRPF improve Strict uRPF in multi-homing scenarios. However, they still have improper block issues in asymmetric routing scenarios. For example, they may not handle the cases of limited propagation of prefixes. These mechanisms use the local RIB to learn the source prefixes and their valid incoming interfaces. But the RIB may not have all the prefixes with limited propagation and their permissible incoming interfaces.¶
EFP-uRPF allows the prefixes from the same customer cone at all customer interfaces. This solves the improper block problems of FP-uRPF in multi-homing scenarios. However, this approach also compromises partial protection against spoofing from the customer cone. EFP-uRPF may still have improper block problems when it does not learn legitimate source prefixes. For example, hidden prefixes are not learned by EFP-uRPF.¶
Finally, existing inter-domain SAV mechanisms cannot work in all directions (i.e. interfaces) of ASes to achieve effective SAV. Network operators need to carefully analyze the network environment and choose appropriate SAV mechanism for each interface. This leads to additional operational and cognitive overhead, which hinders the rate of adoption of inter-domain SAV.¶
This section lists the requirements which can help bridge the technical gaps of existing inter-domain SAV mechanisms. These requirements serve as the practical guidelines that can be met, in part or in full, by proposing new techniques.¶
The new inter-domain SAV mechanism MUST improve the validation accuracy in all directions of ASes over existing inter-domain SAV mechanisms, while working in incremental/partial deployment and providing necessary security guarantee.¶
It MUST avoid improper block and permit less spoofing traffic than existing inter-domain SAV mechanisms. To avoid improper block, ASes that deploy the new inter-domain SAV mechanism SHOULD be able to acquire all the real data plane forwarding paths, which are the paths that the legitimate traffic goes through in the data plane.¶
However, it may be hard to learn the real forwarding paths of prefixes exactly under some scenarios, such as asymmetric routing scenario and DSR scenario. For such scenarios, it is crucial to minimize the set of acceptable paths while ensuring the inclusion of all real forwarding paths, thereby preventing improper block and minimizing improper permit. Note that the acceptable paths are all the possible paths that the legitimate traffic may go through in the data plane, cover all the links at each level of customer-provider hierarchy, and MUST include all the real forwarding paths. Reducing the set of acceptable paths means eliminating the paths that are not the real forwarding paths of the prefixes from the set.¶
+----------------+
| AS 3(P3) |
+-+/\----+/\+----+
/ \
/ \
/ \
/ (C2P) \
+----------------+ \
| AS 4(P4) | \
++/\+--+/\+--+/\++ \
P6[AS 1, AS 2] / | \ \
P2[AS 2] / | \ \
/ | \ \
/ (C2P) | \ P5[AS 5] \ P5[AS 5]
+----------------+ | \ \
| AS 2(P2) | | P1[AS 1] \ \
+----------+/\+--+ | P6[AS 1] \ \
P6[AS 1] \ | NO_EXPORT \ \
P1[AS 1] \ | \ \
NO_EXPORT \ | \ \
\ (C2P) | (C2P) (C2P) \ (C2P) \
+----------------+ +----------------+
| AS 1(P1, P6) | | AS 5(P5) |
+----------------+ +----------------+
Multiple sources of SAV-related information, such as RPKI ROA objects and ASPA objects, and SAV-specific information from other ASes, can assist in reducing the set of acceptable paths. Figure 8 is used as an example to illustrate how to avoid improper block and minimize improper permit in all directions of an AS based on different SAV information sources. AS 3 is the provider of AS 4 and AS 5, while AS 4 is the provider of AS 1, AS 2, and AS 5, and AS 2 is the provider of AS 1. Assuming prefixes P1, P2, P3, P4, P5, and P6 are all the prefixes in the network. Inter-domain SAV has been deployed by AS 1 and AS 4, but not by other ASes. Here, the focus is on how to conduct SAV in all directions of AS 4 when different SAV information sources are available to use.¶
Since the source prefix ranges of the traffic entering the customer cone of AS 4 are not fully learned in the partial deployment scenario, SAV at provider/peer interfaces can use a blocklist. For example, as shown in Figure 8, the traffic with source addresses in P5 may come from AS 5 or AS 3. In contrast, SAV at customer interfaces for traffic going out of the customer cone can use an allowlist to allow the known prefixes of the customer cone at the corresponding customer interfaces and other unknown prefixes at all the customer interfaces.¶
The followings show how to generate SAV rules based on the SAV-related information from different SAV information sources to avoid improper block and reduce as much improper permit as possible.¶
If only the RIB is available, AS 4 can conduct SAV towards its neighboring ASes as follows like [RFC8704]: SAV towards AS 1 permits the prefixes P1 and P6, SAV towards AS 2 permits the prefixes P1, P2, and P6, SAV towards AS 5 permits the prefix P5, and SAV towards AS 3 does not block any prefix.¶
When both RPKI ROA objects and ASPA objects are deployed by AS 1 and AS 4, AS 4 can conduct SAV towards its neighboring ASes as follows like [bar-sav]: SAV towards AS 1 permits the prefixes P1 and P6, SAV towards AS 2 permits the prefixes P1, P2, and P6, SAV towards AS 5 permits the prefix P5, and SAV towards AS 3 blocks the prefixes P1, P2, and P6.¶
Moreover, if SAV-specific information that exactly contains all the real data plane forwarding paths of prefixes is accessible, SAV rules can be refined. AS 4 can conduct SAV towards its neighboring ASes as follows: SAV towards AS 1 permits only P1. SAV towards AS 2 permits the prefixes P2 and P6, while SAV towards AS 5 permits the prefix P5 and SAV towards AS 3 blocks the prefixes P1, P2, and P6.¶
It is evident that, in a partial deployment scenario, more accurate SAV-related information can effectively achieve 0% improper block and significantly minimize improper permit.¶
The new inter-domain SAV mechanism MUST NOT assume pervasive adoption and SHOULD provide effective protection for source addresses when it is partially deployed in the Internet. Not all AS border routers can support the new SAV mechanism at once, due to various constraints such as capabilities, versions, or vendors. The new SAV mechanism should not be less effective in protecting all directions of ASes under partial deployment than existing mechanisms.¶
The new inter-domain SAV mechanism SHOULD secure the communicated SAV-specific information between ASes and prevent malicious ASes from generating forged information.¶
The new inter-domain SAV mechanism SHOULD update SAV rules and detect the changes of SAV-specific information automatically while guaranteeing convergence.¶
The new inter-domain SAV mechanism MUST be able to adapt to dynamic networks and asymmetric routing scenarios automatically, instead of relying on manual update. At least, it MUST have less operational overhead than ACL-based ingress filtering.¶
The new inter-domain SAV mechanism SHOULD promptly detect the network changes and launch the convergence process quickly. It is essential that the new inter-domain SAV mechanism converges towards accurate SAV rules in a proper manner, effectively reducing improper block and improper permit throughout the whole convergence process.¶
The new inter-domain SAV mechanisms should work in the same scenarios as existing ones. Generally, it includes all IP-encapsulated scenarios:¶
Native IP forwarding: This includes both global routing table forwarding and CE site forwarding of VPN.¶
IP-encapsulated Tunnel (IPsec, GRE, SRv6, etc.): In this scenario, we focus on the validation of the outer layer IP address.¶
Both IPv4 and IPv6 addresses.¶
Scope does not include:¶
Non-IP packets: This includes MPLS label-based forwarding and other non-IP-based forwarding.¶
In addition, the new inter-domain SAV mechanisms should not modify data plane packets. Existing architectures or protocols or mechanisms can be inherited by the new SAV mechanism to achieve better SAV effectiveness.¶
SAV rules can be generated based on route information (FIB/RIB) or non-route information. If the information is poisoned by attackers, the SAV rules will be false. Legitimate packets may be dropped improperly or malicious traffic with spoofed source addresses may be permitted improperly. Route security should be considered by routing protocols. Non-route information, such as RPKI ASPA objects, should also be protected by corresponding mechanisms or infrastructure. If SAV mechanisms or protocols require exchanging specific information between ASes, some considerations on the avoidance of message alteration or message injection are needed to propose.¶
The SAV procedure referred in this document modifies no field of packets. So, security considerations on the data plane are not in the scope of this document.¶
This document does not request any IANA allocations.¶
Lancheng Qin
Zhongguancun Laboratory
Beijing,
China
Email: qinlc@zgclab.edu.cn¶
Nan Geng
Huawei
Beijing,
China
Email: gengnan@huawei.com¶
Many thanks to Jared Mauch, Barry Greene, Fang Gao, Anthony Somerset, Yuanyuan Zhang, Igor Lubashev, Alvaro Retana, Joel Halpern, Aijun Wang, Michael Richardson, Li Chen, Gert Doering, Mingxing Liu, John O'Brien, Roland Dobbins, etc. for their valuable comments on this document.¶