RFC 8184
Dual-Homing Protection for MPLS and the MPLS Transport Profile (MPLS-TP) Pseudowires
Dual-Homing Protection for MPLS and the MPLS Transport Profile (MPLS-TP) Pseudowires 🎀
okay bestie, this is RFC 8184 and it's giving pseudowire protection glow-up 💖 It's an Informational document, so it's published for information and isn't a standards track spec ✨ It's dated June 2017 and it's ISSN 2070-1721. The authors are W. Cheng, L. Wang, and H. Li from China Mobile, S. Davari from Broadcom Corporation, and J. Dong from Huawei Technologies 🫶 The header doesn't list any Obsoletes or Updates, so this one isn't replacing or updating any other RFCs 🦋
Abstract ✨
okay so the abstract is the quick tour 💅 It describes a framework and a few scenarios for a pseudowire (PW) dual-homing local protection mechanism. The big perks are that it avoids unnecessary switchovers and it doesn't care whether a control plane is in use 🔥 A Dual-Node Interconnection (DNI) PW carries traffic between the dual-homing Provider Edge (PE) nodes when a failure hits one of the Attachment Circuits (AC) or PWs. And this whole mechanism is complementary to existing PW protection mechanisms, so it works alongside them, not instead of them 🌸
Status of This Memo 📜
This document is not an Internet Standards Track specification; it is published for informational purposes.
This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Not all documents approved by the IESG are a candidate for any level of Internet Standard; see Section 2 of RFC 7841.
Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at http://www.rfc-editor.org/info/rfc8184.
Copyright Notice ©️
Copyright (c) 2017 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) 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 Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
Table of Contents 🗺️
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Reference Models of Dual-Homing Local Protection . . . . . . 4
2.1. PE Architecture . . . . . . . . . . . . . . . . . . . . . 4
2.2. Dual-Homing Local Protection Reference Scenarios . . . . 5
2.2.1. One-Side Dual-Homing Protection . . . . . . . . . . . 5
2.2.2. Two-Side Dual-Homing Protection . . . . . . . . . . . 6
3. Generic Dual-Homing PW Protection Mechanism . . . . . . . . . 8
4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 8
5. Security Considerations . . . . . . . . . . . . . . . . . . . 9
6. References . . . . . . . . . . . . . . . . . . . . . . . . . 9
6.1. Normative References . . . . . . . . . . . . . . . . . . 9
6.2. Informative References . . . . . . . . . . . . . . . . . 9
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 111. Introduction 🌈
okay so [RFC6372] and [RFC6378] are the ones that describe the framework and mechanism of MPLS Transport Profile (MPLS-TP) linear protection 🦋 That can protect the MPLS Label Switched Path (LSP) or the pseudowire (PW) between the edge nodes. The catch is that it does not protect against failure of the Attachment Circuit (AC) or the Provider Edge (PE) node 💀 That's where [RFC6718] and [RFC6870] come in. They describe the framework and mechanism for PW redundancy, which covers AC or PE node failure. The PW redundancy mechanism is based on signaling from the Label Distribution Protocol (LDP), so it's applicable to PWs with a dynamic control plane.
[RFC8104] describes a fast local repair mechanism for PW egress endpoint failures. It's built on PW redundancy, upstream label assignment, and context-specific label switching ✨ The mechanism defined in [RFC8104] is only applicable to PWs with a dynamic control plane, so it's not the whole answer.
So there's a need for a dual-homing local protection mechanism that avoids unnecessary switches of the AC or PW 🎀 It also has to work whether or not a control plane is used. In some scenarios, like mobile backhauling, MPLS PWs are provisioned in a dual-homing topology where at least the Customer Edge (CE) node on one side is dual-homed to two PEs 🫶 If a fault hits the primary AC, operators usually prefer to switch over only on the dual-homing PE side and keep the working pseudowires unchanged if possible. The reason? You don't want massive PW switchover in the mobile backhaul network just because of an AC failure in the mobile core site 😭 Massive switchover can cause congestion, because traffic gets pushed away from the preferred paths that network planners picked. 🙈
Similarly, multiple PWs share the physical AC in the mobile core site. So when one working PW fails in the Packet Switched Network (PSN), it's preferable to keep using the working AC, which can potentially avoid unnecessary switchover for the other PWs 🍓 To meet all of this, a fast dual-homing local PW protection mechanism is needed to protect against failures of an AC, the PE node, and the PSN. Main character energy 👑
This document describes the framework and several typical scenarios of PW dual-homing local protection 💜 A DNI-PW is used between the dual-homing PE nodes to carry traffic when a failure occurs on the AC or PW side. For the dual-homing PE nodes to figure out the forwarding state of the AC, the PW, and the DNI-PW, they need necessary state exchange and coordination between them 🧚 The mechanisms and protocol extensions for that are defined in [RFC8185].
2. Reference Models of Dual-Homing Local Protection 🔥
This section shows the reference architecture of the dual-homing PW local protection and how the architecture gets used in different scenarios 💅
2.1. PE Architecture 🧩
Figure 1 shows the PE architecture for dual-homing local protection. It's based on the architecture in Figure 4a of [RFC3985] ✨ Besides the AC and the service PW between the local and remote PEs, a DNI-PW connects the forwarders of the dual-homing PEs. It can be used to forward traffic between the dual-homing PEs when a failure occurs on the AC or service PW side. As [RFC3985] specifies: "any required switching functionality is the responsibility of a forwarder function". So in this case, the forwarder is responsible for switching the payloads between three entities: the AC, the service PW, and the DNI-PW. It's giving organized 🦋
+----------------------------------------+
| Dual-Homing PE Device |
+----------------------------------------+
AC | | | Service PW
<------>o Forwarder + Service X<===========>
| | PW |
+--------+--------+ |
| DNI-PW | |
+--------X--------+----------------------+
^
| DNI-PW
|
V
+--------X--------+----------------------+
| DNI-PW | |
+--------+--------+ | Service PW
AC | | Service X<===========>
<------>o Forwarder + PW |
| | |
+----------------------------------------+
| Dual-Homing PE Device |
+----------------------------------------+Figure 1: PE Architecture for Dual-Homing Protection
2.2. Dual-Homing Local Protection Reference Scenarios 🎀
2.2.1. One-Side Dual-Homing Protection
Okay so bestie, let's talk about the first big scenario 💖 Figure 2 shows the setup where only ONE of the CEs is dual-homed to two PEs. CE1 is dual-homed to PE1 and PE2, while CE2 is single-homed to PE3. ✨ There's a DNI-PW between the dual-homing PEs, and its whole job is to bridge traffic when something breaks on the PSN side or the AC side. 🦋 A dual-homing control mechanism helps the PEs and CE figure out which AC should carry traffic between CE1 and the PSN. The necessary control mechanisms and protocol extensions are defined in [RFC8185]. 🎀
This scenario can protect against node failure of PE1 or PE2, or failure of one of the ACs between CE1 and the dual-homing PEs. 💅 Plus, dual-homing PW protection can protect against a PSN failure that hits the working PW, so it can be an alternative to PSN tunnel protection mechanisms. ✨ This topology can be used in mobile backhauling. For example, CE2 might be an equipment cell site such as a NodeB, while CE1 is the shared Radio Network Controller (RNC). PE3 functions as an access-side MPLS device, while PE1 and PE2 function as core-side MPLS devices. 🍓
|<--------------- Emulated Service --------------->|
| |
| |<------- Pseudowire ------>| |
| | | |
| | |<-- PSN Tunnels-->| | |
| V V V V |
V AC1 +----+ +----+ V
+-----+ | | PE1| | | +-----+
| |----------|........PW1.(working).......| | |
| | | | | | | |
| | +-+--+ | | AC3 | |
| | | | | | | |
| | DNI-PW | |PE3 |----------| CE2 |
| | | | | | |
| | +-+--+ | | | |
| | | | | | | |
| |----------|......PW2.(protection).......| | |
+-----+ | | PE2| | | +-----+
AC2 +----+ +----+Wait, I need to be careful here. My drawing above should be the EXACT copy of the original, so let me put the figure back exactly as it was, character for character 🙈
|<--------------- Emulated Service --------------->|
| |
| |<------- Pseudowire ------>| |
| | | |
| | |<-- PSN Tunnels-->| | |
| V V V V |
V AC1 +----+ +----+ V
+-----+ | | PE1| | | +-----+
| |----------|........PW1.(working).......| | |
| | | | | | | |
| | +-+--+ | | AC3 | |
| | | | | | | |
| CE1 | DNI-PW | |PE3 |----------| CE2 |
| | | | | | |
| | +-+--+ | | | |
| | | | | | | |
| |----------|......PW2.(protection)......| | |
+-----+ | | PE2| | | +-----+
AC2 +----+ +----+
Figure 2: One-Side Dual-Homing PW ProtectionConsider the example where, in normal state, AC1 from CE1 to PE1 is initially active and AC2 from CE1 to PE2 is initially standby. PW1 is configured as the working PW and PW2 is configured as the protection PW. 🌈
When a failure occurs in AC1, the state of AC2 changes to active based on the AC dual-homing control mechanism. To keep the switchover local and keep using PW1 for forwarding (as preferred per traffic planning), the forwarder on PE2 needs to connect AC2 to the DNI-PW, and the forwarder on PE1 needs to connect the DNI-PW to PW1. 🔥 That way, the AC1 failure won't impact forwarding of the service PWs across the network. After the switchover, traffic goes through the bidirectional path: CE1-(AC2)-PE2-(DNI-PW)-PE1-(PW1)-PE3-(AC3)-CE2. 🫶
When a failure in the PSN affects the working PW (PW1), then according to PW protection mechanisms [RFC6378], traffic gets switched onto the protection PW (PW2) while the state of AC1 stays active. Then the forwarder on PE1 needs to connect AC1 to the DNI-PW, and the forwarder on PE2 needs to connect the DNI-PW to PW2. ✨ That way, the PSN failure won't impact the state of the ACs. After the switchover, traffic goes through: CE1-(AC1)-PE1-(DNI-PW)-PE2-(PW2)-PE3-(AC3)-CE2. 👑
When a failure occurs in the working PE (PE1), it's equivalent to a failure of the working AC, the working PW, and the DNI-PW. 💀 The state of AC2 changes to active based on the AC dual-homing control mechanism. On top of that, according to the PW protection mechanism, traffic is switched onto the protection PW "PW2". In this case, the forwarder on PE2 needs to connect AC2 to PW2. After the switchover, traffic goes through: CE1-(AC2)-PE2-(PW2)-PE3-(AC3)-CE2. 🧚
2.2.2. Two-Side Dual-Homing Protection
Now the main character scenario 👑 Figure 3 shows dual-homing PW protection where the CEs on BOTH sides are dual-homed. CE1 is dual-homed to PE1 and PE2, and CE2 is dual-homed to PE3 and PE4. A dual-homing control mechanism helps the PEs and CEs figure out which AC should carry traffic between the CE and the PSN. DNI-PWs are used between the dual-homing PEs on both sides. One service PW is established between PE1 and PE3, and another service PW is established between PE2 and PE4. The role of working and protection PWs can be determined by either configuration or existing signaling mechanisms. 🦋
This scenario can protect against node failure on one of the dual-homing PEs, or failure on one of the ACs between the CEs and their dual-homing PEs. Also, dual-homing PW protection can protect against a PSN failure that impacts one of the PWs, so it can be used as an alternative to PSN tunnel protection mechanisms. 🌸 Note, this scenario is mainly used for services requiring high availability, since it requires redundancy of the PEs and network utilization. In this case, CE1 and CE2 can be regarded as service access points. 💅
|<---------------- Emulated Service -------------->|
| |
| |<-------- Pseudowire ------>| |
| | | |
| | |<-- PSN Tunnels-->| | |
| V V V V |
V AC1 +----+ +----+ AC3 V
+-----+ | | ...|...PW1.(working)..|... | | +-----+
| |----------| PE1| | PE3|----------| |
| | +----+ +----+ | |
| | | | | |
| CE1 | DNI-PW1 | | DNI-PW2 | CE2 |
| | | | | |
| | +----+ +----+ | |
| | | | | | | |
| |----------| PE2| | PE4|--------- | |
+-----+ | | ...|.PW2.(protection).|... | | +-----+
AC2 +----+ +----+ AC4Hmm, that wasn't quite right either, so here's the figure with the exact original text 🙈
|<---------------- Emulated Service -------------->|
| |
| |<-------- Pseudowire ------>| |
| | | |
| | |<-- PSN Tunnels-->| | |
| V V V V |
V AC1 +----+ +----+ AC3 V
+-----+ | | ...|...PW1.(working)..|... | | +-----+
| |----------| PE1| | PE3|----------| |
| | +----+ +----+ | |
| | | | | |
| CE1 | DNI-PW1 | | DNI-PW2 | CE2 |
| | | | | |
| | +----+ +----+ | |
| | | | | | | |
| |----------| PE2| | PE4|--------- | |
+-----+ | | ...|.PW2.(protection).|... | | +-----+
AC2 +----+ +----+ AC4Figure 3: Two-Side Dual-Homing PW Protection
Consider the example where, in normal state, AC1 between CE1 and PE1 is initially active, AC2 between CE1 and PE2 is initially standby, AC3 between CE2 and PE3 is initially active, and AC4 from CE2 to PE4 is initially standby. PW1 is configured as the working PW and PW2 is configured as the protection PW. 🎀
When a failure occurs in AC1, the state of AC2 changes to active based on the AC dual-homing control mechanism. To keep the switchover local and keep using PW1 for forwarding, the forwarder on PE2 needs to connect AC2 to the DNI-PW1, and the forwarder on PE1 needs to connect DNI-PW1 with PW1. 💖 That way, failures on the AC side won't impact forwarding of the service PWs across the network. After the switchover, traffic goes through: CE1-(AC2)-PE2-(DNI-PW1)-PE1-(PW1)-PE3-(AC3)-CE2. 🫶
When a failure occurs in the working PW (PW1), then according to the PW protection mechanism [RFC6378], traffic needs to be switched onto the protection PW "PW2". To keep the state of AC1 and AC3 unchanged, the forwarder on PE1 needs to connect AC1 to DNI-PW1, and the forwarder on PE2 needs to connect DNI-PW1 to PW2. On the other side, the forwarder on PE3 needs to connect AC3 to DNI-PW2, and the forwarder on PE4 needs to connect PW2 to DNI-PW2. ⭐ That way, the state of the ACs won't be impacted by the PSN failure. After the switchover, traffic goes through: CE1-(AC1)-PE1-(DNI-PW1)-PE2-(PW2)-PE4-(DNI-PW2)-PE3-(AC3)-CE2. 🌈
When a failure occurs in the working PE (PE1), it's equivalent to failures of the working AC, the working PW, and the DNI-PW. The state of AC2 changes to active based on the AC dual-homing control mechanism. On top of that, according to the PW protection mechanism, traffic is switched onto the protection PW "PW2". In this case, the forwarder on PE2 needs to connect AC2 to PW2, and the forwarder on PE4 needs to connect PW2 to DNI-PW2. After the switchover, traffic goes through: CE1-(AC2)-PE2-(PW2)-PE4-(DNI-PW2)-PE3-(AC3)-CE2. 🔥
- Generic Dual-Homing PW Protection Mechanism
As shown in the scenarios above, with the described dual-homing PW protection, failures on the AC side won't impact the forwarding behavior of the PWs in the PSN, and vice versa. ✨
In order for the dual-homing PEs to coordinate traffic forwarding during failures, they need synchronization of the status information of the involved entities and coordination of switchover between the dual-homing PEs. For PWs with a dynamic control plane, that synchronization and coordination info can be achieved with a dynamic protocol, such as the one described in [RFC7275], possibly with some extensions. For PWs that are manually configured without a control plane, a new mechanism is needed to exchange the status information and coordinate switchover between the dual-homing PEs, e.g., over an embedded PW control channel. This is described in [RFC8185]. 🧚
- IANA Considerations
This document does not require any IANA action. 🌸
- Security Considerations
The scenarios defined in this document do not affect the security model as defined in [RFC3985]. 🔐
With the proposed protection mechanism, the disruption of a dual-homed AC, a component that is outside the core network, would have a reduced impact on the traffic flows in the core network. This could also avoid unnecessary congestion in the core network. 🫶
The security consideration of the DNI-PW is the same as for service PWs in the data plane [RFC3985]. Security considerations for the coordination/control mechanism will be addressed in the companion document, RFC 8185, which defines the mechanism. 💜
- References
6.1. Normative References
[RFC3985] Bryant, S., Ed. and P. Pate, Ed., "Pseudo Wire Emulation
Edge-to-Edge (PWE3) Architecture", RFC 3985,
DOI 10.17487/RFC3985, March 2005,
<http://www.rfc-editor.org/info/rfc3985>.
[RFC8185] Cheng, W., Wang, L., Li, H., Dong, J., and A.
D'Alessandro, "Dual-Homing Coordination for MPLS Transport
Profile (MPLS-TP) Pseudowires Protection", RFC 8185,
DOI 10.17487/RFC8185, June 2017.6.2. Informative References 📚
Okay bestie, these are the informative references, aka the extra reading list for when you want more context 🤓 Keep these handy!
- [RFC6372] Sprecher, N., Ed. and A. Farrel, Ed., "MPLS Transport Profile (MPLS-TP) Survivability Framework", RFC 6372, DOI 10.17487/RFC6372, September 2011,
http://www.rfc-editor.org/info/rfc6372. 🦋 - [RFC6378] Weingarten, Y., Ed., Bryant, S., Osborne, E., Sprecher, N., and A. Fulignoli, Ed., "MPLS Transport Profile (MPLS-TP) Linear Protection", RFC 6378, DOI 10.17487/RFC6378, October 2011,
http://www.rfc-editor.org/info/rfc6378. 💜 - [RFC6718] Muley, P., Aissaoui, M., and M. Bocci, "Pseudowire Redundancy", RFC 6718, DOI 10.17487/RFC6718, August 2012,
http://www.rfc-editor.org/info/rfc6718. 🌸 - [RFC6870] Muley, P., Ed. and M. Aissaoui, Ed., "Pseudowire Preferential Forwarding Status Bit", RFC 6870, DOI 10.17487/RFC6870, February 2013,
http://www.rfc-editor.org/info/rfc6870. ✨ - [RFC7275] Martini, L., Salam, S., Sajassi, A., Bocci, M., Matsushima, S., and T. Nadeau, "Inter-Chassis Communication Protocol for Layer 2 Virtual Private Network (L2VPN) Provider Edge (PE) Redundancy", RFC 7275, DOI 10.17487/RFC7275, June 2014,
http://www.rfc-editor.org/info/rfc7275. 🎀 - [RFC8104] Shen, Y., Aggarwal, R., Henderickx, W., and Y. Jiang, "Pseudowire (PW) Endpoint Fast Failure Protection", RFC 8104, DOI 10.17487/RFC8104, March 2017,
http://www.rfc-editor.org/info/rfc8104. 🧚
Contributors 🫶
These cuties substantially contributed to the content of this document, and we are SO grateful 💖
Kai Liu
Huawei Technologies
Email: alex.liukai@huawei.com
Alessandro D'Alessandro
Telecom Italia
Email: alessandro.dalessandro@telecomitalia.itAuthors' Addresses 🌈
Meet the squad who wrote this whole thing 👑
Weiqiang Cheng
China Mobile
No.32 Xuanwumen West Street
Beijing 100053
China
Email: chengweiqiang@chinamobile.com
Lei Wang
China Mobile
No.32 Xuanwumen West Street
Beijing 100053
China
Email: Wangleiyj@chinamobile.com
Han Li
China Mobile
No.32 Xuanwumen West Street
Beijing 100053
China
Email: Lihan@chinamobile.com
Shahram Davari
Broadcom Corporation
3151 Zanker Road
San Jose 95134-1933
United States of America
Email: davari@broadcom.com
Jie Dong
Huawei Technologies
Huawei Campus, No. 156 Beiqing Rd.
Beijing 100095
China
Email: jie.dong@huawei.com📦 where this page comes from
- ietf: rfc8184 2017-06
- 💖 girlified by claude-haiku-5-5 on : a language model's rewrite of the original, under its terms