| PCE Working Group | D. Dhody |
| Internet-Draft | U. Palle |
| Intended status: Experimental | Huawei Technologies |
| Expires: May 22, 2016 | R. Casellas |
| CTTC | |
| November 19, 2015 |
Domain Subobjects for Path Computation Element (PCE) Communication Protocol (PCEP).
draft-ietf-pce-pcep-domain-sequence-10
The ability to compute shortest constrained Traffic Engineering Label Switched Paths (TE LSPs) in Multiprotocol Label Switching (MPLS) and Generalized MPLS (GMPLS) networks across multiple domains has been identified as a key requirement. In this context, a domain is a collection of network elements within a common sphere of address management or path computational responsibility such as an Interior Gateway Protocol (IGP) area or an Autonomous System (AS). This document specifies a representation and encoding of a Domain-Sequence, which is defined as an ordered sequence of domains traversed to reach the destination domain to be used by Path Computation Elements (PCEs) to compute inter-domain constrained shortest paths across a predetermined sequence of domains . This document also defines new subobjects to be used to encode domain identifiers.
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 May 22, 2016.
Copyright (c) 2015 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.
A Path Computation Element (PCE) may be used to compute end-to-end paths across multi-domain environments using a per-domain path computation technique [RFC5152]. The backward recursive path computation (BRPC) mechanism [RFC5441] also defines a PCE-based path computation procedure to compute inter-domain constrained path for (G)MPLS TE LSPs. However, both per-domain and BRPC techniques assume that the sequence of domains to be crossed from source to destination is known, either fixed by the network operator or obtained by other means. Also for inter-domain point-to-multi-point (P2MP) tree computation, [RFC7334] assumes the domain-tree is known in priori.
The list of domains (Domain-Sequence) in point-to-point (P2P) or a domain tree in point-to-multipoint (P2MP) is usually a constraint in inter-domain path computation procedure.
The Domain-Sequence (the set of domains traversed to reach the destination domain) is either administratively predetermined or discovered by some means like H-PCE.
[RFC5440] defines the Include Route Object (IRO) and the Explicit Route Object (ERO). [RFC5521] defines the Exclude Route Object (XRO) and the Explicit Exclusion Route Subobject (EXRS). The use of Autonomous System (AS) (albeit with a 2-Byte AS number) as an abstract node representing a domain is defined in [RFC3209]. In the current document, we specify new subobjects to include or exclude domains including IGP area or an Autonomous Systems (4-Byte as per [RFC6793]).
Further, the domain identifier may simply act as delimiter to specify where the domain boundary starts and ends in some cases.
This is a companion document to Resource ReserVation Protocol - Traffic Engineering (RSVP-TE) extensions for the domain identifiers [DOMAIN-SUBOBJ].
The procedures described in this document are experimental. The experiment is intended to enable research for the usage of Domain-Sequence at the PCEs for inter-domain paths. For this purpose this document specifies new domain subobjects as well as how they incorporate with existing subobjects to represent a Domain-Sequence.
The experiment will end two years after the RFC is published. At that point, the RFC authors will attempt to determine how widely this has been implemented and deployed.
This document does not change the procedures for handling existing subobjects in PCEP.
The new subobjects introduced by this document will not be understood by legacy implementations. If a legacy implementation receives one of the subobjects that it does not understand in a PCEP object, the legacy implementation will behave as described in Section 3.4.3. Therefore, it is assumed that this experiment will be conducted only when both the PCE and the PCC form part of the experiment. It is possible that a PCC or PCE can operate with peers some of which form part of the experiment and some that do not. In this case, since no capabilities exchange is used to identify which nodes can use these extensions, manual configuration should be used to determine which peerings form part of the experiment.
When the result of implementation and deployment are available, this document will be updated and refined, and then be moved from Experimental to Standard Track.
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 [RFC2119].
The following terminology is used in this document.
[RFC4726] and [RFC4655] define domain as a separate administrative or geographic environment within the network. A domain could be further defined as a zone of routing or computational ability. Under these definitions a domain might be categorized as an AS or an IGP area. Each AS can be made of several IGP areas. In order to encode a Domain-Sequence, it is required to uniquely identify a domain in the Domain-Sequence. A domain can be uniquely identified by area-id or AS number or both.
A Domain-Sequence is an ordered sequence of domains traversed to reach the destination domain.
A Domain-Sequence can be applied as a constraint and carried in a path computation request to PCE(s). A Domain-Sequence can also be the result of a path computation. For example, in the case of Hierarchical PCE (H-PCE) [RFC6805], Parent PCE could send the Domain-Sequence as a result in a path computation reply.
In a P2P path, the domains listed appear in the order that they are crossed. In a P2MP path, the domain tree is represented as a list of Domain-Sequences.
A Domain-Sequence enables a PCE to select the next domain and the PCE serving that domain to forward the path computation request based on the domain information.
Domain-Sequence can include Boundary Nodes (ABR or ASBR) or Border links (Inter-AS-links) to be traversed as an additional constraint.
Thus a Domain-Sequence can be made up of one or more of -
These are encoded in the new subobjects defined in this document as well as the existing subobjects to represent a Domain-Sequence.
Consequently, a Domain-Sequence can be used:
Domain-Sequence appears in PCEP messages, notably in -
As per [RFC5440], IRO (Include Route Object) can be used to specify that the computed path needs to traverse a set of specified network elements or abstract nodes.
Some subobjects are defined in [RFC3209], [RFC3473], [RFC3477] and [RFC4874], but new subobjects related to Domain-Sequence are needed.
This document extends the support for 4-Byte AS numbers and IGP Areas.
Type Subobject
TBD1 Autonomous system number (4 Byte)
TBD2 OSPF Area id
TBD3 ISIS Area id
Note: The twins of these subobjects are carried in RSVP-TE messages as defined in [DOMAIN-SUBOBJ].
[RFC3209] already defines 2 byte AS number.
To support 4 byte AS number as per [RFC6793] following subobject is defined:
0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |L| Type | Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AS-ID (4 bytes) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Since the length and format of Area-id is different for OSPF and ISIS, following two subobjects are defined:
For OSPF, the area-id is a 32 bit number. The subobject is encoded as follows:
0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |L| Type | Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | OSPF Area Id (4 bytes) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
For IS-IS, the area-id is of variable length and thus the length of the Subobject is variable. The Area-id is as described in IS-IS by ISO standard [ISO10589]. The subobject is encoded as follows:
0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |L| Type | Length | Area-Len | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | // IS-IS Area ID // | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
[RFC5440] describes IRO as an optional object used to specify network elements to be traversed by the computed path. It further state that the L bit of such subobject has no meaning within an IRO. It also did not mention if IRO is an ordered or un-ordered list of subobjects.
An update to IRO specification [IRO-UPDATE] makes IRO as an ordered list, as well as support for loose bit (L-bit) is added.
The use of IRO for Domain-Sequence, assumes the updated specification for IRO, as per [IRO-UPDATE].
The subobject type for IPv4, IPv6, and unnumbered Interface ID can be used to specify Boundary Nodes (ABR/ASBR) and Inter-AS-Links. The subobject type for the AS Number (2 or 4 Byte) and the IGP Area are used to specify the domain identifiers in the Domain-Sequence.
The IRO can incorporate the new domain subobjects with the existing subobjects in a sequence of traversal.
Thus an IRO, comprising subobjects, that represents a Domain-Sequence, define the domains involved in an inter-domain path computation, typically involving two or more collaborative PCEs.
A Domain-Sequence can have varying degrees of granularity. It is possible to have a Domain-Sequence composed of, uniquely, AS identifiers. It is also possible to list the involved IGP areas for a given AS.
In any case, the mapping between domains and responsible PCEs is not defined in this document. It is assumed that a PCE that needs to obtain a "next PCE" from a Domain-Sequence is able to do so (e.g. via administrative configuration, or discovery).
A PCC builds an IRO to encode the Domain-Sequence, so that the cooperating PCEs could compute an inter-domain shortest constrained path across the specified sequence of domains.
A PCC may intersperse Area and AS subobjects with other subobjects without change to the previously specified processing of those subobjects in the IRO.
If a PCE receives an IRO in a Path Computation request (PCReq) message that contains the subobjects defined in this document, that it does not recognize, it will respond according to the rules for a malformed object as per [RFC5440]. The PCE MAY also include the IRO in the PCErr message as per [RFC5440].
The interpretation of Loose bit (L bit) is as per section 4.3.3.1 of [RFC3209] (as per [IRO-UPDATE]).
In a Path Computation reply (PCRep), PCE MAY also supply IRO (with Domain-Sequence information) with the NO-PATH object indicating that the set of elements (domains) of the request's IRO prevented the PCEs from finding a path.
The following processing rules apply for Domain-Sequence in IRO -
Note that it is advised that, PCC should use AS and Area subobject while building the domain-sequence in IRO and avoid using other mechanism to change the "current AS" and "current Area" as described above.
The Exclude Route Object (XRO) [RFC5521] is an optional object used to specify exclusion of certain abstract nodes or resources from the whole path.
Some subobjects to be used in XRO as defined in [RFC3209], [RFC3477], [RFC4874], and [RFC5520], but new subobjects related to Domain-Sequence are needed.
This document extends the support for 4-Byte AS numbers and IGP Areas.
Type Subobject
TBD1 Autonomous system number (4 Byte)
TBD2 OSPF Area id
TBD3 ISIS Area id
The new subobjects to support 4 byte AS and IGP (OSPF / ISIS) Area MAY also be used in the XRO to specify exclusion of certain domains in the path computation procedure.
0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |X| Type | Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | AS-ID (4 bytes) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
The X-bit indicates whether the exclusion is mandatory or desired.
All other fields are consistent with the definition in Section 3.4.
Since the length and format of Area-id is different for OSPF and ISIS, following two subobjects are defined:
For OSPF, the area-id is a 32 bit number. The subobject is encoded as follows:
0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |X| Type | Length | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | OSPF Area Id (4 bytes) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
The X-bit indicates whether the exclusion is mandatory or desired.
All other fields are consistent with the definition in Section 3.4.
For IS-IS, the area-id is of variable length and thus the length of the subobject is variable. The Area-id is as described in IS-IS by ISO standard [ISO10589]. The subobject is encoded as follows:
0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |X| Type | Length | Area-Len | Reserved | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | | // IS-IS Area ID // | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
The X-bit indicates whether the exclusion is mandatory or desired.
All other fields are consistent with the definition in Section 3.4.
All the processing rules are as per [RFC5521].
Note that, if a PCE receives an XRO in a PCReq message that contains subobjects defined in this document, that it does not recognize, it will respond according to the rules for a malformed object as per [RFC5440].
IGP Area subobjects in the XRO are local to the current AS. In case of multi-AS path computation to exclude an IGP area in a different AS, IGP Area subobject should be part of Explicit Exclusion Route Subobject (EXRS) in the IRO to specify the AS in which the IGP area is to be excluded. Further policy may be applied to prune/ignore Area subobjects in XRO after "current AS" change during path computation.
EXRS [RFC5521] is used to specify exclusion of certain abstract nodes between a specific pair of nodes.
The EXRS subobject can carry any of the subobjects defined for inclusion in the XRO, thus the new subobjects to support 4 byte AS and IGP (OSPF / ISIS) Area can also be used in the EXRS. The meanings of the fields of the new XRO subobjects are unchanged when the subobjects are included in an EXRS, except that scope of the exclusion is limited to the single hop between the previous and subsequent elements in the IRO.
The EXRS subobject should be interpreted in the context of the current AS and current Area of the preceding subobject in the IRO. The EXRS subobject does not change the current AS or current Area. All other processing rules are as per [RFC5521].
Note that, if a PCE that supports the EXRS in an IRO, parses an IRO, and encounters an EXRS that contains subobjects defined in this document, that it does not recognize, it will act according to the setting of the X-bit in the subobject as per [RFC5521].
The Explicit Route Object (ERO) [RFC5440] is used to specify a computed path in the network. PCEP ERO subobject types correspond to RSVP-TE ERO subobject types as defined in [RFC3209], [RFC3473], [RFC3477], [RFC4873], [RFC4874], and [RFC5520]. The subobjects related to Domain-Sequence are further defined in [DOMAIN-SUBOBJ].
The new subobjects to support 4 byte AS and IGP (OSPF / ISIS) Area can also be used in the ERO to specify an abstract node (a group of nodes whose internal topology is opaque to the ingress node of the LSP). Using this concept of abstraction, an explicitly routed LSP can be specified as a sequence of domains.
In case of Hierarchical PCE [RFC6805], a Parent PCE can be requested to find the Domain-Sequence. Refer example in Section 4.6. The ERO in reply from parent PCE can then be used in Per-Domain path computation or BRPC.
If a PCC receives an ERO in a PCRep message that contains subobject defined in this document, that it does not recognize, it will respond according to the rules for a malformed object as per [RFC5440].
The examples in this section are for illustration purposes only; to highlight how the new subobjects could be encoded. They are not meant to be an exhaustive list of all possible usecases and combinations.
In an inter-area path computation where the ingress and the egress nodes belong to different IGP areas within the same AS, the Domain-Sequence could be represented using a ordered list of Area subobjects.
----------------- -----------------
| | | |
| +--+ | | +--+ |
| +--+ | | | | | | |
| | | +--+ | | +--+ +--+ |
| +--+ | | | | |
| | | +--+ |
| +--+ | | |
| | | | | +--+ |
| +--+ | | | | |
| | -------------------------- | +--+ |
| +--+ +--+ |
| | | +--+ | | |
|Area 2 +--+ | | +--+ Area 4 |
----------------- | +--+ | -----------------
| |
| +--+ |
| +--+ | | |
| | | +--+ |
| +--+ |
| |
| |
| |
| |
| +--+ |
| | | |
| +--+ |
----------------- | | ------------------
| +--+ +--+ |
| | | | | |
| +--+ Area 0 +--+ |
| | -------------------------- | +--+ |
| +--+ | | | | |
| | | | | +--+ |
| +--+ +--+ | | |
| | | | | +--+ |
| +--+ | | | | |
| | | +--+ |
| +--+ | | |
| | | | | +--+ |
| +--+ | | | | |
| | | +--+ |
| | | |
| Area 1 | | Area 5 |
----------------- ------------------
Figure 1: Inter-Area Path Computation
AS Number is 100.
If the ingress is in Area 2, egress in Area 4 and transit through Area 0. Some possible way a PCC can encode the IRO:
+---------+ +---------+ +---------+ |IRO | |Sub | |Sub | |Object | |Object | |Object | |Header | |Area 0 | |Area 4 | | | | | | | | | | | | | +---------+ +---------+ +---------+ or +---------+ +---------+ +---------+ +---------+ |IRO | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Header | |Area 2 | |Area 0 | |Area 4 | | | | | | | | | | | | | | | | | +---------+ +---------+ +---------+ +---------+ or +---------+ +---------+ +---------+ +---------+ +---------+ |IRO | |Sub | |Sub | |Sub | |Sub | |Object | |Object AS| |Object | |Object | |Object | |Header | |100 | |Area 2 | |Area 0 | |Area 4 | | | | | | | | | | | | | | | | | | | | | +---------+ +---------+ +---------+ +---------+ +---------+
The Domain-Sequence can further include encompassing AS information in the AS subobject.
In inter-AS path computation, where ingress and egress belong to different AS, the Domain-Sequence could be represented using an ordered list of AS subobjects. The Domain-Sequence can further include decomposed area information in the Area subobject.
As shown in Figure 2, where AS has a single area, AS subobject in the domain-sequence can uniquely identify the next domain and PCE.
AS A AS E AS C
<-------------> <----------> <------------->
A4----------E1---E2---E3---------C4
/ / \
/ / \
/ / AS B \
/ / <----------> \
Ingress------A1---A2------B1---B2---B3------C1---C2------Egress
\ / /
\ / /
\ / /
\ / /
A3----------D1---D2---D3---------C3
<---------->
AS D
* All AS have one area (area 0)
Figure 2: Inter-AS Path Computation
If the ingress is in AS A, egress in AS C and transit through AS B. Some possible way a PCC can encode the IRO:
+-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Object | |Object | |Object | |Header | |AS B | |AS C | | | | | | | +-------+ +-------+ +-------+ or +-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Header | |AS A | |AS B | |AS C | | | | | | | | | +-------+ +-------+ +-------+ +-------+ or +-------+ +-------+ +-------+ +-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Object | |Object | |Object | |Header | |AS A | |Area 0 | |AS B | |Area 0 | |AS C | |Area 0 | | | | | | | | | | | | | | | +-------+ +-------+ +-------+ +-------+ +-------+ +-------+ +-------+
Note that to get a domain disjoint path, the ingress could also request the backup path with -
+-------+ +-------+ |XRO | |Sub | |Object | |Object | |Header | |AS B | | | | | +-------+ +-------+
As described in Section 3.4.3, domain subobject in IRO changes the domain information associated with the next set of subobjects; till you encounter a subobject that changes the domain too. Consider the following IRO:
+-------+ +-------+ +-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Object | |Object | |Header | |AS B | |IP | |IP | |AS C | |IP | | | | | |B1 | |B3 | | | |C1 | +-------+ +-------+ +-------+ +-------+ +-------+ +-------+
On processing subobject "AS B", it changes the AS of the subsequent subobjects till we encounter another subobject "AS C" which changes the AS for its subsequent subobjects.
Consider another IRO:
+-------+ +-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Object | |Header | |AS D | |IP | |IP | |IP | | | | | |D1 | |D3 | |C3 | +-------+ +-------+ +-------+ +-------+ +-------+
Here as well, on processing "AS D", it changes the AS of the subsequent subobjects till you encounter another subobject "C3" which belong in another AS and changes the AS for its subsequent subobjects.
Further description for the Boundary Node and Inter-AS-Link can be found in Section 4.3.
In Figure 3, AS 200 is made up of multiple areas.
|
| +-------------+ +----------------+
| |Area 2 | |Area 4 |
| | +--+| | +--+ |
| | | || | | B| |
| | +--+ +--+| | +--+ +--+ |
| | | | | | | | |
| | +--+ | | +--+ |
| | +--+ | | +--+ |
| | | | | | | | |
| | +--+ | | +--+ +--+ |
| | +--+ |+--------------+| | | |
| | | | +--+ +--+ +--+ |
+-------------+| | +--+ | | | | |
| || | +--+ +--+ |
| +--+|| +-------------+| |+----------------+
| | ||| | +--+ |
| +--+|| | | | |
| +--+ || | +--+ |
| | | +---+ +--+ |
| +--+ | |----------------| | |
| +---+ Inter-AS +--+ +--+ |
|+--+ || Links | | | |
||A | +---+ +--+ +--+ |
|+--+ | |----------------| | |
| +---+ +--+ +--+ |
| +--+ || +------------+ | | | |+----------------+
| | | || |Area 3 +--+ +--+ +--+ Area 5 |
| +--+ || | | | | | |
| || | +--+ +--+ |
| +--+|| | +--+ | | Area 0 || +--+ |
| | ||| | | | | +--------------+| | | |
| +--+|| | +--+ | | +--+ |
| || | | | +--+ |
|Area 0 || | +--+ | | +--+ | | |
+-------------+| | | | | | | | +--+ |
| | +--+ +--+ | +--+ |
| | | | | |
| | +--+ | +--+ |
| | +--+ | | | C| |
| | | | | | +--+ |
| | +--+ | | |
| | | | |
| +------------+ +----------------+
|
|
AS 100 | AS 200
|
Figure 3: Inter-AS Path Computation
For LSP (A-B), where ingress A is in (AS 100, Area 0), egress B in (AS 200, Area 4) and transit through (AS 200, Area 0). Some possible way a PCC can encode the IRO:
+-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Header | |AS 200 | |Area 0 | |Area 4 | | | | | | | | | +-------+ +-------+ +-------+ +-------+ or +-------+ +-------+ +-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Object | |Object | |Header | |AS 100 | |Area 0 | |AS 200 | |Area 0 | |Area 4 | | | | | | | | | | | | | +-------+ +-------+ +-------+ +-------+ +-------+ +-------+
For LSP (A-C), where ingress A is in (AS 100, Area 0), egress C in (AS 200, Area 5) and transit through (AS 200, Area 0). Some possible way a PCC can encode the IRO:
+-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Header | |AS 200 | |Area 0 | |Area 5 | | | | | | | | | +-------+ +-------+ +-------+ +-------+ or +-------+ +-------+ +-------+ +-------+ +-------+ +-------+ |IRO | |Sub | |Sub | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Object | |Object | |Header | |AS 100 | |Area 0 | |AS 200 | |Area 0 | |Area 5 | | | | | | | | | | | | | +-------+ +-------+ +-------+ +-------+ +-------+ +-------+
A PCC or PCE can include additional constraints covering which Boundary Nodes (ABR or ASBR) or Border links (Inter-AS-link) to be traversed while defining a Domain-Sequence. In which case the Boundary Node or Link can be encoded as a part of the Domain-Sequence.
Boundary Nodes (ABR / ASBR) can be encoded using the IPv4 or IPv6 prefix subobjects usually the loopback address of 32 and 128 prefix length respectively. An Inter-AS link can be encoded using the IPv4 or IPv6 prefix subobjects or unnumbered interface subobjects.
For Figure 1, an ABR (say 203.0.113.1) to be traversed can be specified in IRO as:
+---------+ +---------+ +---------++---------+ +---------+
|IRO | |Sub | |Sub ||Sub | |Sub |
|Object | |Object | |Object ||Object | |Object |
|Header | |Area 2 | |IPv4 ||Area 0 | |Area 4 |
| | | | |203.0. || | | |
| | | | |112.1 || | | |
+---------+ +---------+ +---------++---------+ +---------+
For Figure 3, an inter-AS-link (say 198.51.100.1 - 198.51.100.2) to be traversed can be specified as:
+---------+ +---------+ +---------+ +---------+
|IRO | |Sub | |Sub | |Sub |
|Object | |Object AS| |Object | |Object AS|
|Header | |100 | |IPv4 | |200 |
| | | | |198.51. | | |
| | | | |100.2 | | |
+---------+ +---------+ +---------+ +---------+
A single PCE can be responsible for multiple domains; for example PCE function deployed on an ABR could be responsible for multiple areas. A PCE which can support adjacent domains can internally handle those domains in the Domain-Sequence without any impact on the other domains in the Domain-Sequence.
[RFC7334] describes an experimental inter-domain P2MP path computation mechanism where the path domain tree is described as a series of Domain-Sequences, an example is shown in the below figure:
+----------------+
| |Domain D1
| R |
| |
| A |
| |
+-B------------C-+
/ \
/ \
/ \
Domain D2 / \ Domain D3
+-------------D--+ +-----E----------+
| | | |
| F | | |
| G | | H |
| | | |
| | | |
+-I--------------+ +-J------------K-+
/\ / \
/ \ / \
/ \ / \
/ \ / \
/ \ / \
/ \ / \
/ Domain D4 \ Domain D5 / Domain D6 \
+-L-------------W+ +------P---------+ +-----------T----+
| | | | | |
| | | Q | | U |
| M O | | S | | |
| | | | | V |
| N | | R | | |
+----------------+ +----------------+ +----------------+
The domain tree can be represented as a series of domain-sequence -
The domain sequence handling described in this document could be applied to P2MP path domain tree.
In case of H-PCE [RFC6805], the parent PCE can be requested to determine the Domain-Sequence and return it in the path computation reply, using the ERO. . For the example in section 4.6 of [RFC6805], the Domain-Sequence can possibly appear as:
+---------+ +---------+ +---------+ +---------+ |ERO | |Sub | |Sub | |Sub | |Object | |Object | |Object | |Object | |Header | |Domain 1 | |Domain 2 | |Domain 3 | | | | | | | | | | | | | | | | | +---------+ +---------+ +---------+ +---------+ or +---------+ +---------+ +---------+ |ERO | |Sub | |Sub | |Object | |Object | |Object | |Header | |BN 21 | |Domain 3 | | | | | | | | | | | | | +---------+ +---------+ +---------+
Instead of a Domain-Sequence, a sequence of PCEs MAY be enforced by policy on the PCC, and this constraint can be carried in the PCReq message (as defined in [RFC5886]).
Note that PCE-Sequence can be used along with Domain-Sequence in which case PCE-Sequence MUST have higher precedence in selecting the next PCE in the inter-domain path computation procedures.
[RFC3209] already describes the notion of abstract nodes, where an abstract node is a group of nodes whose internal topology is opaque to the ingress node of the LSP. It further defines a subobject for AS but with a 2-Byte AS Number.
[DOMAIN-SUBOBJ] extends the notion of abstract nodes by adding new subobjects for IGP Areas and 4-byte AS numbers. These subobjects can be included in Explicit Route Object (ERO), Exclude Route object (XRO) or Explicit Exclusion Route Subobject (EXRS) in RSVP-TE.
In any case subobject type defined in RSVP-TE are identical to the subobject type defined in the related documents in PCEP.
IANA maintains the "Path Computation Element Protocol (PCEP) Numbers" at <http://www.iana.org/assignments/pcep>. Within this registry IANA maintains two sub-registries:
Upon approval of this document, IANA is requested to make identical additions to these registries as follows:
Subobject Type Reference TBD1 4 byte AS number [This I.D.][DOMAIN-SUBOBJ] TBD2 OSPF Area ID [This I.D.][DOMAIN-SUBOBJ] TBD3 IS-IS Area ID [This I.D.][DOMAIN-SUBOBJ]
Further upon approval of this document, IANA is requested to add a reference to this document to the new RSVP numbers that are registered by [DOMAIN-SUBOBJ].
The protocol extensions defined in this document do not substantially change the nature of PCEP. Therefore, the security considerations set out in [RFC5440] apply unchanged. Note that further security considerations for the use of PCEP over TCP are presented in [RFC6952].
This document specifies a representation of Domain-Sequence and new subobjects, which could be used in inter-domain PCE scenarios as explained in [RFC5152], [RFC5441], [RFC6805], [RFC7334] etc. The security considerations set out in each of these mechanisms remain unchanged by the new subobjects and Domain-Sequence representation in this document.
But the new subobjects do allow finer and more specific control of the path computed by a cooperating PCE(s). Such control increases the risk if a PCEP message is intercepted, modified, or spoofed because it allows the attacker to exert control over the path that the PCE will compute or to make the path computation impossible. Consequently, it is important that implementations conform to the relevant security requirements of [RFC5440]. These mechanisms include:
Note, however, that the Domain-Sequence mechanisms also provide the operator with the ability to route around vulnerable parts of the network and may be used to increase overall network security.
The exact behaviour with regards to desired inclusion and exclusion of domains MUST be available for examination by an operator and MAY be configurable. Manual configurations is needed to identify which PCEP peers understand the new domain subobjects defined in this document.
A MIB module for management of the PCEP is being specified in a separate document [RFC7420]. This document does not imply any new extension to the current MIB module.
Mechanisms defined in this document do not imply any new liveness detection and monitoring requirements in addition to those already listed in [RFC5440].
Mechanisms defined in this document do not imply any new operation verification requirements in addition to those already listed in [RFC5440].
In case of per-domain path computation [RFC5152], where the full path of an inter-domain TE LSP cannot be, or is not determined at the ingress node, a signaling message can use the domain identifiers. The Subobjects defined in this document SHOULD be supported by RSVP-TE. [DOMAIN-SUBOBJ] extends the notion of abstract nodes by adding new subobjects for IGP Areas and 4-byte AS numbers.
Apart from this, mechanisms defined in this document do not imply any requirements on other protocols in addition to those already listed in [RFC5440].
The mechanisms described in this document can provide the operator with the ability to exert finer and more specific control of the path computation by inclusion or exclusion of domain subobjects. There may be some scaling benefit when a single domain subobject may substitute for many subobjects and can reduce the overall message size and processing.
Backward compatibility issues associated with the new subobjects arise when a PCE does not recognize them, in which case PCE responds according to the rules for a malformed object as per [RFC5440]. For successful operations the PCEs in the network would need to be upgraded.
Authors would like to especially thank Adrian Farrel for his detailed reviews as well as providing text to be included in the document.
Further, we would like to thank Pradeep Shastry, Suresh Babu, Quintin Zhao, Fatai Zhang, Daniel King, Oscar Gonzalez, Chen Huaimo, Venugopal Reddy, Reeja Paul, Sandeep Boina, Avantika Sergio Belotti and Jonathan Hardwick for their useful comments and suggestions.
Thanks to Jonathan Hardwick for shepherding this document.
Thanks to Joel Halpern for Gen-ART Review.
Thanks to Klaas Wierenga for SecDir Review.
Thanks to Spencer Dawkins and Barry Leiba for comments during the IESG Review.