RFC 6941
MPLS Transport Profile (MPLS-TP) Security Framework
MPLS Transport Profile (MPLS-TP) Security Framework 🦋
RFC 6941 ✨ is an Informational doc, so it's published for information and isn't a Standards Track spec. It's dated April 2013 and has ISSN 2070-1721. The editors are L. Fang (Cisco), B. Niven-Jenkins (Velocix), S. Mansfield (Ericsson), and R. Graveman (RFG Security), each listed as Ed. 💅 Its title block doesn't list any RFCs it obsoletes or updates, so it's standing on its own. 🌸
Part 1 of 3, bestie. More parts are coming, so stay tuned 👀
Abstract 💖
Okay so this doc gives us a security framework for the MPLS Transport Profile (MPLS-TP) 🫶 MPLS-TP extends MPLS, brings in new Operations, Administration, and Maintenance (OAM) capabilities, adds a transport-oriented path protection mechanism, and puts a ton of emphasis on static provisioning that's backed up by network management systems. It's giving very focused energy because it only covers the security stuff that matters for MPLS-TP specifically. It digs into the potential security threats and the mitigation procedures for MPLS-TP networks, and also for MPLS-TP interconnecting with other MPLS and GMPLS networks 🔥
The doc is built on RFC 5920 ("Security Framework for MPLS and GMPLS Networks") and adds extra security considerations that apply to the MPLS-TP extensions. And the big news is that every security consideration from RFC 5920 is assumed to apply too. No cap, it's a whole package deal 🎀
This document is also a product of a joint Internet Engineering Task Force (IETF) / International Telecommunication Union Telecommunication Standardization Sector (ITU-T) effort. The goal was to bring an MPLS Transport Profile into the IETF MPLS and Pseudowire Emulation Edge-to-Edge (PWE3) architectures so a packet transport network can get the capabilities and functionality it needs ✨
Status of This Memo 📜
This document is not an Internet Standards Track specification. It's published for informational purposes only 🍓
It's a product of the Internet Engineering Task Force (IETF) and represents the consensus of the IETF community. It got public review and was approved for publication by the Internet Engineering Steering Group (IESG). Not every IESG-approved doc is a candidate for any level of Internet Standard, so check Section 2 of RFC 5741 if you want the full tea ☕
If you want the current status, any errata, or a way to send feedback, head to http://www.rfc-editor.org/info/rfc6941 🦋
Copyright Notice 🌈
Copyright (c) 2013 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 (http://trustee.ietf.org/license-info) as they stand on the date this document was published. Please read these carefully, because they spell out your rights and restrictions for this document 🙈 Any Code Components pulled out of this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions. They're provided without warranty, as described in the Simplified BSD License.
Table of Contents 🗂️
1. Introduction ....................................................3
1.1. Terminology ................................................3
2. Security Reference Models .......................................4
2.1. Security Reference Model 1 .................................5
2.2. Security Reference Model 2 .................................6
3. Security Threats ................................................9
4. Defensive Techniques ...........................................10
5. Security Considerations ........................................12
6. Acknowledgements ...............................................13
7. References .....................................................13
7.1. Normative References ......................................13
7.2. Informative References ....................................13
Contributors ......................................................141. Introduction 🌸
Welcome to the intro, bestie! This doc is giving us a security framework for MPLS-TP 💅
As defined in "Requirements of an MPLS Transport Profile" [RFC5654] and "A Framework for MPLS in Transport Networks" [RFC5921], MPLS-TP uses a subset of MPLS features and adds extensions that match what transport technology needs. The extra functionality includes in-band OAM, transport-oriented path protection and recovery mechanisms, and new OAM capabilities. Those OAM capabilities were built for MPLS-TP but also apply to MPLS and GMPLS. MPLS-TP really leans into static provisioning through Network Management Systems (NMSs) or Operational Support Systems (OSSs) 🫶
This document is built on [RFC5920], adding security considerations for the MPLS-TP extensions. The security models, threats, and defense techniques that [RFC5920] already defined are assumed to apply to the general aspects of MPLS-TP. Iconic continuity 👑
It's also a product of a joint IETF / ITU-T effort to bring an MPLS Transport Profile into the IETF MPLS and PWE3 architectures, so a packet transport network gets the capabilities and functionality it needs ✨
Need terminology? [RFC5654] and [RFC5921] cover MPLS-TP terms, and [RFC5920] covers the security terms that matter for MPLS and GMPLS 🧚
1.1. Terminology 📖
Here's the glossary. Same terms, same definitions, just vibes 💜
Term Definition
------ -----------------------------------------------
AC Attachment Circuit
BFD Bidirectional Forwarding Detection
CE Customer Edge
DoS Denial of Service
G-ACh Generic Associated Channel
GAL G-ACh Label
GMPLS Generalized Multiprotocol Label Switching
IP Internet Protocol
LDP Label Distribution Protocol
LSP Label Switched Path
NMS Network Management System
MPLS Multiprotocol Label Switching
MPLS-TP MPLS Transport Profile
MS-PW Multi-Segment Pseudowire
OAM Operations, Administration, and Maintenance
PE Provider Edge
PSN Packet-Switched Network
PW Pseudowire
S-PE PW Switching Provider Edge
SP Service Provider
SS-PW Single-Segment Pseudowire
T-PE PW Terminating Provider Edge2. Security Reference Models 🎀
This section is where the doc defines its reference models for security in MPLS-TP networks. It's the blueprint, babe 🦋
The models are built on the MPLS-TP architecture from [RFC5921]. Where you draw the SP boundaries matters a lot, because that's what decides which security model fits a particular deployment 👀
Here's the key idea: a trusted zone is a place where a single SP has total operational control over that part of the network. The main worry is breaches that come from the "outside" of a trusted zone trying to get to the "inside" 🔥
2.1. Security Reference Model 1 💜
In reference model 1, a single SP has total control of the "PE/T-PE to PE/T-PE" part of the MPLS-TP network. One SP, full control, no drama 💅
Security reference model 1(a) shows an MPLS-TP network with a Single-Segment Pseudowire (SS-PW) from PE1 to PE2. The trusted zone is PE1 to PE2, as you can see in Figure 1 🌈
|<-------------- Emulated Service ---------------->|
| |
| |<------- Pseudowire ------->| |
| | | |
| | |<-- PSN Tunnel -->| | |
| v v v v |
v AC +----+ +----+ AC v
+-----+ | | PE1|==================| PE2| | +-----+
| |----------|............PW1.............|----------| |
| CE1 | | | | | | | | CE2 |
| |----------|............PW2.............|----------| |
+-----+ ^ | | |==================| | | ^ +-----+
^ | +----+ +----+ | | ^
| | Provider Edge 1 Provider Edge 2 | |
| | | |
Customer | |Customer
Edge 1 | |Edge 2
| |
Native service Native service
---Untrusted--- >|<------- Trusted Zone ----->|<---Untrusted----
Figure 1. MPLS-TP Security Model 1(a)Security reference model 1(b) shows an MPLS-TP network with a Multi-Segment Pseudowire (MS-PW) from T-PE1 to T-PE2. The trusted zone is T-PE1 to T-PE2, as shown in Figure 2 ✨
Native |<-------------Pseudowire------------>| Native
Service | | Service
(AC) | |<- PSN ->| |<- PSN ->| | (AC)
| v v v v v v |
| +-----+ +-----+ +-----+ |
+----+ | |T-PE1|=========|S-PE1|=========|T-PE2| | +----+
| |------|......PW.Seg't1.......PW.Seg't3......|-------| |
| CE1| | | | | | | | | |CE2 |
| |------|......PW.Seg't2.......PW.Seg't4......|-------| |
+----+ | | |=========| |=========| | | +----+
^ +-----+ ^ +-----+ ^ +-----+ ^
| | | |
| TP LSP TP LSP |
| |
|<----------------- Emulated Service ---------------->|
-Untrusted->|<---------- Trusted Zone ----------->|<-Untrusted--
Figure 2. MPLS-TP Security Model 1(b)2.2. Security Reference Model 2 ✨
Okay bestie, model 2 is where things get messy 😭. A single SP does not have end-to-end control of the segment from PE/T-PE to PE/T-PE. A given S-PE or T-PE might be controlled by another SP, that SP's customers, or its partners. So the MPLS-TP network is not contained within a single trusted zone 🌸.
Security reference model 2(a) is an MPLS-TP network with a Multi-Segment Pseudowire (MS-PW) from T-PE1 to T-PE2. The trusted zone is T-PE1 to S-PE1, as shown in Figure 3 💅.
Native |<-------------Pseudowire------------>| Native
Service | | Service
(AC) | |<--PSN-->| |<--PSN-->| | (AC)
| V V V V V V |
| +-----+ +-----+ +-----+ |
+----+ | |T-PE1|=========|S-PE1|=========|T-PE2| | +----+
| |------|......PW.Seg't1.......PW.Seg't3......|------| |
| CE1| | | | | | | | | |CE2 |
| |------|......PW.Seg't2.......PW.Seg't4......|------| |
+----+ | | |=========| |=========| | | +----+
^ +-----+ ^ +-----+ ^ +-----+ ^
| | | |
| TP LSP TP LSP |
| |
|<---------------- Emulated Service --------------->|
Untrusted-->|<-- Trusted Zone---->|<---------Untrusted--------Figure 3. MPLS-TP Security Model 2(a)
Security reference model 2(b) is an MPLS-TP network with a Multi-Segment Pseudowire (MS-PW) from T-PE1 to T-PE2. This time the trusted zone is only S-PE1, as shown in Figure 4 🦋.
Native |<-------------Pseudowire------------>| Native
Service | | Service
(AC) | |<--PSN-->| |<--PSN-->| | (AC)
| V V V V V V |
| +-----+ +-----+ +-----+ |
+----+ | |T-PE1|=========|S-PE1|=========|T-PE2| | +----+
| |------|......PW.Seg't1.......PW.Seg't3......|------| |
| CE1| | | | | | | | | |CE2 |
| |------|......PW.Seg't2.......PW.Seg't4......|------| |
+----+ | | |=========| |=========| | | +----+
^ +-----+ ^ +-----+ ^ +-----+ ^
| | | |
| TP LSP TP LSP |
| |
|<---------------- Emulated Service --------------->|
--------Untrusted---------->|<--->|<-------Untrusted----------
Trusted
ZoneFigure 4. MPLS-TP Security Model 2(b)
Security reference model 2(c) is an MPLS-TP network with a Multi-Segment Pseudowire (MS-PW) from different SPs, using inter-provider PW connections 🌈. The trusted zone is T-PE1 to S-PE3, as shown in Figure 5.
Native |<--------------------- PW15 ------------------>| Native
Layer | | Layer
Service | |<PSN13>| |<PSN3X>| |<PSNXZ>| | Service
(AC1) V V LSP V V LSP V V LSP V V (AC2)
| +-----+ +-+ +-----+ +-----+ +-+ +-----+ |
+---+ | |T-PE1| | | |S-PE3| |S-PEX| | | |T-PEZ| | +---+
| | | | |=======| |=======| |=======| | | | |
|CE1|----|........PW1........|..PW3..|........PW5........|---|CE2|
| | | | |=======| |=======| |=======| | | | |
+---+ | 1 | |2| | 3 | | X | |Y| | Z | +---+
+-----+ +-+ +-----+ +-----+ +-+ +-----+
|<--Subnetwork 123->| |<--Subnetwork XYZ->|
Untrusted>|<-- Trusted Zone-->|<-------------Untrusted-------------Figure 5. MPLS-TP Security Model 2(c)
In general, the boundaries of a trusted zone must be carefully defined when you analyze the security properties of each individual network 🧐. Those boundaries decide which reference model should be applied to a given network topology.
3. Security Threats 🚨
Here we talk about network security threats that are unique to MPLS-TP and could endanger MPLS-TP networks. Be careful out there 🫣.
Attacks against a GAL or G-ACh may include the following:
- GAL or BFD label manipulation, which includes insertion of false labels and modification, deletion, or replay of messages 🙈.
- DoS attacks through in-band OAM, where someone generates excessive G-ACh/GAL and BFD messages that eat up significant bandwidth and can potentially cause congestion 💀.
These attacks can cause unauthorized protection switchover, an inability to restore one or more LSPs, or loss of network connectivity. Yikes 😭.
When an NMS is used for LSP setup, attacks on the NMS can cause the above effects too. This isn't unique to MPLS-TP, but MPLS-TP networks can be especially vulnerable to NMS attacks, because static provisioning through NMSs is a commonly used model 💜. In the static provisioning model, a compromised NMS can potentially be comparable to a compromised control plane plus a compromised management plane in the dynamic controlled network model.
Attacks on NMSs may come from external attackers or from insiders. Outside attacks are started outside of the trusted zone by unauthorized users of the MPLS-TP NMSs. Insider attacks are started from inside the trusted zone by an entity that has authorized access to the management systems but performs unapproved functions that are harmful to the MPLS-TP networks 👀. These attacks may directly target the NMS, or they may take place via compromised communication channels between the NMS and the network devices being provisioned, or through user access to the provisioning tools. This type of threat may include disclosure of information, generating false OAM messages, taking down MPLS-TP LSPs, connecting to the wrong MPLS-TP tunnel endpoints, and DoS attacks on the MPLS-TP networks 🥲.
There are other, more generic threats too, like unauthorized observation of data traffic (including traffic pattern analysis), modification or deletion of a provider's or user's data, and replay or insertion of inauthentic data into a provider's or user's data stream. These types of attacks apply to MPLS-TP traffic regardless of how the LSP or PW is set up, similar to how they apply to MPLS traffic regardless of how the LSP is set up 🍓. You can read more about the threats above in [RFC5920].
Threats like these may come from malicious behavior or from accidental errors:
- Example 1: Attacks from users: Users of the MPLS-TP network may attack the network infrastructure or attack other users.
- Example 2: Attacks from insiders: Employees of the operators may attack the MPLS-TP network, especially through NMSs.
- Example 3: Attacks from interconnecting SPs or other partners: Other SPs may attack the MPLS-TP network, particularly through the inter-provider connections.
- Example 4: Attacks as the result of operational errors: Operations staff may fail to follow operational procedures or may make operational mistakes 🙃.
4. Defensive Techniques 🛡️💖
The defensive techniques in this document and in [RFC5920] describe methods for addressing some security threats. They are not intended as requirements for all MPLS-TP deployments 🌷. The specific operational environment determines the security requirements for any instance of MPLS-TP. So protocol designers should provide a full set of security capabilities that can be selected and used where appropriate. The MPLS-TP provider should determine how these techniques apply to its specific service offerings, and the end user may want to assess how valuable these techniques are for their service requirements 🧚.
Authentication is the primary defense technique for mitigating the risk of the MPLS-TP security threats discussed in Section 3 (GAL or BFD label manipulation, and DoS attacks through in-band OAM) 🔐. Authentication refers to methods that make sure message sources are properly identified by the MPLS-TP devices they talk to. Authentication includes the following:
- entity authentication for identity verification
- management system authentication
- peer-to-peer authentication
- message integrity and replay detection to ensure the validity of message streams
- network-based access controls such as packet filtering and firewalls
- host-based access controls
- isolation
- aggregation
- protection against denial of service
- event logging
Section 5.2 of [RFC5920] describes these techniques where they apply to MPLS and GMPLS in general 📚.
Besides authentication, the following defense should also be considered to protect MPLS-TP networks ✨:
- Use of isolated infrastructure for MPLS-TP
One way to protect the MPLS-TP infrastructure is to use dedicated network resources to provide MPLS-TP transport services. For example, in security model 2 (Section 2.2), the potential risk of attacks on S-PE1 or T-PE1 in the trusted zone may be reduced by using non-IP-based communication paths, so that the paths in the trusted zone cannot be reached from the outside via IP 🏰.
- Verification of connectivity
To protect against deliberate or accidental misconnection, mechanisms can be put in place to verify both end-to-end connectivity and segment-by-segment resources. These mechanisms can trace the routes of LSPs in both the control plane and the data plane. Also, note that the connectivity verification tools are now developed for general MPLS networks as well 🗺️.
The defense techniques that apply generally to MPLS/GMPLS aren't detailed here, so see [RFC5920] for details about them. For example:
- Authentication, including management system authentication, peer-to-peer authentication, and cryptographic techniques for authenticating identity
- Access control techniques
- Use of aggregated infrastructure
- Mitigation of denial-of-service attacks
- Monitoring, detection, and reporting of security attacks
It's important to point out the following security defense techniques, because they're especially critical for NMSs, given the strong emphasis on static provisioning that NMSs support in MPLS-TP deployments 👑. These techniques include the following:
- entity authentication for identity verification
- encryption for confidentiality
- message integrity and replay detection to ensure the validity of message streams
- user access control and event logging, which must be applied for NMSs and provisioning applications 🔒
5. Security Considerations ✨
okay bestie, this section is literally the whole point of the doc 🔐 so listen up! The security considerations are the sole subject of this document, which is why they're discussed throughout (it's giving main character energy fr 💅).
This document looks at security risks that are specific to MPLS-TP 🦋, plus the mitigation mechanisms you can use to counter the potential threats. All of the techniques covered here are mature and widely implemented technologies that are practical to implement, so no sketchy experimental stuff 🙅♀️. The whole goal is to help equipment vendors and service providers figure out which threats they need to protect against in any given configuration or service offering. That goes for the customer's perspective AND the service provider's perspective, so both sides get a seat at the table 🪑💖.
6. Acknowledgements 🫶
The authors wanted to give a huge shoutout to the following people 📣:
- Joel Halpern and Gregory Mirsky, for their review comments and contributions to this document 🌸
- Mach Chen, for his review and suggestions 🧚
- Adrian Farrel, for his Routing Area Director review and detailed comments 👑
- Loa Andersson, for his continued support and guidance as the MPLS WG co-chair 💜
- Dan Romascanu and Barry Leiba, for their helpful comments during IESG review 🎀
Iconic behavior from everyone, honestly 😭✨.
7. References 📚
7.1. Normative References
[RFC5654]Niven-Jenkins, B., Ed., Brungard, D., Ed., Betts, M., Ed., Sprecher, N., and S. Ueno, "Requirements of an MPLS Transport Profile", RFC 5654, September 2009. 🍓[RFC5920]Fang, L., Ed., "Security Framework for MPLS and GMPLS Networks", RFC 5920, July 2010. 🔥
7.2. Informative References
[RFC5921]Bocci, M., Ed., Bryant, S., Ed., Frost, D., Ed., Levrau, L., and L. Berger, "A Framework for MPLS in Transport Networks", RFC 5921, July 2010. 🌈
Contributors 🧚♀️
Okay the people who helped make this happen, obsessed with all of them 💖:
Raymond Zhang
Alcatel-Lucent
750D Chai Chee Road
Singapore 469004
EMail: raymond.zhang@alcatel-lucent.com
Nabil Bitar
Verizon
40 Sylvan Road
Waltham, MA 02145
US
EMail: nabil.bitar@verizon.com
Masahiro Daikoku
KDDI Corporation
3-11-11 Iidabashi, Chiyodaku, Tokyo
Japan
EMail: ms-daikoku@kddi.com
Lei Wang
Lime Networks
Strandveien 30, 1366 Lysaker
Norway
EMail: lei.wang@limenetworks.no
Henry Yu
TW Telecom
10475 Park Meadow Drive
Littleton, CO 80124
US
EMail: henry.yu@twtelecom.comAuthors' Addresses 📮
The editors, the people you can actually reach 🫶:
Luyuan Fang (editor)
Cisco Systems
111 Wood Ave. South
Iselin, NJ 08830
US
EMail: lufang@cisco.com
Ben Niven-Jenkins (editor)
Velocix
326 Cambridge Science Park
Milton Road
Cambridge CB4 0WG
UK
EMail: ben@niven-jenkins.co.uk
Scott Mansfield (editor)
Ericsson
300 Holger Way
San Jose, CA 95134
US
EMail: scott.mansfield@ericsson.com
Richard F. Graveman (editor)
RFG Security, LLC
15 Park Avenue
Morristown, NJ 07960
US
EMail: rfg@acm.org📦 where this page comes from
- ietf: rfc6941 2013-04
- 💖 girlified by claude-haiku-5-5 on : a language model's rewrite of the original, under its terms