Internet-Draft Juniper Networks, Inc.
Intended status: Standards Track A. Gulko
Expires: September 14, 2017 Thomson Reuters
March 13, 2017

Separating Routing Planes using Segment Routing


Many network deployments arrange the network topologies in two or more planes. The traffic generally uses one of the planes and fails over to the other plane when there are link or node failure. Certain applications require the traffic to be strictly restricted to a particular plane and should not failover to the other plane. This document proposes a solution for the strict planar routing using Segment Routing.

Requirements Language

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].

Status of This Memo

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 http://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 September 14, 2017.

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

                  / |                      /|
                 /  |                     / |
                /   |                    /  |
               A1---|-------------------A2  |
               |    B3------------------|---B4
               |   /                    |   /
               |  /                     |  / 
               | /                      | /

Figure 1: Network Planes

The above Figure 1 represents a network topology in two planes.Nodes A1, A2, A3, A4 are in plane A and B1, B2, B3, B4 are in plane B. A1->B1, A2->B2, A3->B3, A4->B4 represent the "shunt links" which connect the two planes. Certain applications require that the traffic follows plane A and remains in plane A in case of failures and does not cross over to plane B. Strict routing plane requirements can be met using multiple techniques. This draft focuses on solutions using Segment Routing technology.

2. Motivation

The motivation of this document is to provide solutions to strict routing plane requirements using Segment Routing. The following objectives help to accomplish this in a range of deployment scenarios.

  1. Maintain strict routing within routing planes.
  2. Allow traffic to failover within routing plane and do not allow traffic to failover to other planes
  3. Achieve ease of configuration and operational management

3. Solutions

There are multiple ways to address the strict routing plane requirements. Section 3.1 describes a mechanism by using diferrent SIDs for each plane.Another option is to use Multi-topology SIDs as defined in Section 3.2.

3.1. Routing-plane SID

A new SID called Routing-Plane SID is defined and associated with new algorithm values. This document proposes 4 new algorithm values which represent SPF in routing-planes. When the network topology is organized into two different planes, each node in plane A associates a new Routing-Plane SID to one of it's loopback addresses and advertises the SID in the segment routing extension defined in [SR-OSPF] section 2.1 and [SR-IS-IS] section 5. The prefix-SID sub-TLV carries the new algorithm values corresponding to the routing-plane. The traffic which needs to be restricted to a certain routing-plane,should use the Routing-Plane SID corresponding to that plane to forward the traffic. The paths through the Routing planes MAY use single Routing Plane SID or a stack of multiple Routing Plane SIDs. Adjacency-SIDs can also be used build paths across routing planes. Adjacency-SIDs are not scoped per-algorithm and it is possible that the protection path for the adjacency SIDs uses a path going over a different routing-plane.It is recommended to use non-protected adjacency-SIDs to avoid backup traffic flowing through different plane.

3.1.1. Elements of procedure

All the nodes that belong to a certain routing-plane MUST advertise the algorithm corresponding to that routing-plane in the algorithm sub-TLV as defined in [SR-OSPF] and [SR-IS-IS]. The nodes SHOULD also advertise Routing-Plane SID corresponding to that algorithm in the prefix-SID Sub-TLV.

A node that receives the algorithm sub-TLV with new algorithm value specified in Section 6 MUST compute SPF with all the nodes that advertised the new algorithm. The next-hops and RIB entries for the Routing-Plane SIDs MUST be computed from the routing-plane SPF. Certain nodes MAY belong to multiple routing-planes. Such nodes MUST compute SPF corresponding to each plane and compute the next-hops for the SIDs corresponding to each plane.

Each router MAY assign different IP address corresponding to each plane or MAY use the same IP address to assign the node-SIDs and Routing-Plane SIDs. The ingress routers MAY advertise binding-SIDs as defined in [SR-ARCH]section 3.5.2, for the label stacks that are constructed using routing-Plane SIDs. The ingress routers MAY map the incoming IP traffic onto the Routing-Plane SIDs, the mechanisms to do so is implementation dependant and outside the scope of this document.

When the network topology is organized into multiple IGP levels or areas, the Routing Plane SIDs MAY be re-originated from one IGP domain into the other domain by the border routers. The border IGP routers MUST re-advertise the Routing-Plane SIDs if they belong to the corresponding Routing plane and has advertised the algorithm corresponding to the routing-plane.

3.2. Multi-topology SID

Multi topology Routing defines mechanisms to support multiple topologies in a single physical network. ISIS and OSPF extensions to support multi-topology routing is defined in [RFC5120] and [RFC4915] respectively. Multi-topology routing defined in [RFC5120] and [RFC4915] describes mechanisms to separate topologies in ISIS and OSPF by advertising separate MT-TLVs in ISIS and multi-topology scoped Router LSA in OSPF. The different routing planes in customer network can be assigned with different MT-ID for each routing-plane and the multi-topology SIDs can be advertised for each MT-ID as described in [SR-OSPF] and [SR-IS-IS]. Multi-topology SIDs are associated with algorithm 0 and no new algorithm definition is required. All the nodes in the network MUST also support multi-topology routing as defined in [RFC5120] and [RFC4915]. All the nodes in the network compute separate SPF per MT-ID and program the forwarding planes with MT-SIDs accordingly. Multi-topology SIDs are used to build the explicit paths through the network. Multi-topology based solution has an advantage of possibility of assigning different IGP costs to links for different MTs. For deployments that do not need separate IGP costs and topologies for each routing plane, it comes with an additional operational overhead of maintaining multi-topology configurations.

4. Backward compatibility

The mechanism described in the document is fully backward compatible. If a node does not support the extensions defined in this document, it will not advertise the additional algorithm values in the algorithm sub-TLV. All the computing nodes will not consider the node in the SPF computation if it has not advertised the specific algorithm. For the multi-topology based solution backward compatibility mechanism described in [RFC5120] and [RFC4915] are applicable.

5. Security Considerations

This document does not introduce any further security issues other than those discussed in [SR-OSPF] and [SR-IS-IS].

6. IANA Considerations

This specification updates OSPF and ISIS registry:

OSPF Router Information (RI) TLVs Registry

8 (IANA Preallocated) - SR-Algorithm TLV

Algorithm 2 -5 : SPF in routing plane

ISIS Sub TLVs for Type 242

Type: TBD (suggested value 19)

Description: Segment Routing Algorithm

Algorithm 2-5 : SPF in Routing Plane

7. Acknowledgements

8. References

8.1. Normative References

, ", ", "
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997.
[RFC4915] Psenak, P., Mirtorabi, S., Roy, A., Nguyen, L. and P. Pillay-Esnault, "Multi-Topology (MT) Routing in OSPF", RFC 4915, DOI 10.17487/RFC4915, June 2007.
[RFC5120] Przygienda, T., Shen, N. and N. Sheth, M-ISIS: Multi Topology (MT) Routing in Intermediate System to Intermediate Systems (IS-ISs)", RFC 5120, DOI 10.17487/RFC5120, February 2008.
[SR-IS-IS]IS-IS Extensions for Segment Routing, draft-ietf-isis-segment-routing-extensions-11(work in progress)", October 2016.
[SR-OSPF]OSPF Extensions for Segment Routing, draft-ietf-ospf-segment-routing-extensions-11(work in progress)", July 2016.
[SR-OSPFv3]OSPFv3 Extensions for Segment Routing, draft-ietf-ospf-ospfv3-segment-routing-extensions-06(work in progress)", July 2016.

8.2. Informative References

, "
[RFC7684] Psenak, P., Gredler, H., Shakir, R., Henderickx, W., Tantsura, J. and A. Lindem, OSPFv2 Prefix/Link Attribute Advertisement", RFC 7684, DOI 10.17487/RFC7684, November 2015.
[SR-ARCH]Segment Routing Architecture, draft-ietf-spring-segment-routing-09(work in progress)", July 2016.

Authors' Addresses

Shraddha Hegde Juniper Networks, Inc. Embassy Business Park Bangalore, KA 560093 India EMail: shraddha@juniper.net
Arkadiy Gulko Thomson Reuters EMail: arkadiy.gulko@thomsonreuters.com