Internet Engineering Task Force (IETF) D. Fedyk
Request for Comments: 6060 Alcatel-Lucent
Category: Standards Track H. Shah
ISSN: 2070-1721 Ciena
N. Bitar
Verizon
A. Takacs
Ericsson
March 2011
Generalized Multiprotocol Label Switching (GMPLS) Control of
Ethernet Provider Backbone Traffic Engineering (PBB-TE)
Abstract
This specification is complementary to the GMPLS Ethernet Label
Switching Architecture and Framework and describes the technology-
specific aspects of GMPLS control for Provider Backbone Bridge
Traffic Engineering (PBB-TE). The necessary GMPLS extensions and
mechanisms are described to establish Ethernet PBB-TE point-to-point
(P2P) and point-to-multipoint (P2MP) connections. This document
supports, but does not modify, the standard IEEE data plane.
Status of This Memo
This is an Internet Standards Track document.
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). Further information on
Internet Standards is available in Section 2 of RFC 5741.
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/rfc6060.
Copyright Notice
Copyright (c) 2011 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
Fedyk, et al. Standards Track [Page 1]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
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.
This document may contain material from IETF Documents or IETF
Contributions published or made publicly available before November
10, 2008. The person(s) controlling the copyright in some of this
material may not have granted the IETF Trust the right to allow
modifications of such material outside the IETF Standards Process.
Without obtaining an adequate license from the person(s) controlling
the copyright in such materials, this document may not be modified
outside the IETF Standards Process, and derivative works of it may
not be created outside the IETF Standards Process, except to format
it for publication as an RFC or to translate it into languages other
than English.
Table of Contents
1. Introduction ....................................................3
1.1. Co-Authors .................................................3
2. Terminology .....................................................4
2.1. PBB-TE and GMPLS Terminology ...............................5
2.2. Conventions Used in This Document ..........................6
3. Creation and Maintenance of PBB-TE Paths Using GMPLS ............6
3.1. Shared Forwarding ..........................................9
3.2. P2P Connections Procedures for Shared Forwarding ..........10
4. Specific Procedures ............................................10
4.1. P2P Ethernet LSPs .........................................10
4.1.1. P2P Path Maintenance ...............................11
4.2. P2MP Ethernet-LSPs ........................................12
4.3. PBB-TE Ethernet Label .....................................12
4.4. Protection Paths ..........................................13
4.5. Service Instance Identification ...........................13
5. Error Conditions ...............................................15
5.1. ESP-VID-Related Errors ....................................15
5.1.1. Invalid ESP-VID Value in the PBB-TE
Ethernet Label .....................................15
5.1.2. Allocated ESP-VID Range is Exhausted ...............16
5.2. Invalid MAC Address .......................................16
6. Security Considerations ........................................16
7. IANA Considerations ............................................17
8. References .....................................................17
8.1. Normative References ......................................17
8.2. Informative References ....................................19
9. Acknowledgments ................................................19
Fedyk, et al. Standards Track [Page 2]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
1. Introduction
The IEEE 802.1 Provider Backbone Bridge Traffic Engineering (PBB-TE)
[IEEE802.1Qay] standard supports the establishment of explicitly
routed traffic engineered paths within Provider Backbone Bridged
(PBB) networks. PBB-TE allows the disabling of:
- the Spanning Tree Protocol
- unknown destination address forwarding
- source address learning
for administratively selected VLAN Identifiers. With PBB-TE an
external provisioning system or control plane can be used to
configure static entries in the managed objects of bridges and so
establish traffic engineered paths in the network.
Generalized MPLS (GMPLS) [RFC3945] is a family of control plane
protocols designed to operate in connection oriented and traffic
engineering transport networks. GMPLS is applicable to a range of
network technologies including L2SC networks (Layer 2 Switching
Capable). The purpose of this document is to specify extensions for
a GMPLS-based control plane to manage PBB-TE explicitly routed
traffic engineered paths. This specification is complementary to the
GMPLS Ethernet Label Switching Architecture and Framework document
[RFC5828].
1.1. Co-Authors
This document is the result of a large team of authors and
contributors. The following is a list of the co-authors:
David Allan
Ericsson
EMail: david.i.allan@ericsson.com
Diego Caviglia
Ericsson
Via Negrone 1/A
Genoa, Italy 16153
EMail: diego.caviglia@ericsson.com
Alan McGuire
BT Group PLC
OP6 Polaris House,
Adastral Park, Martlesham Heath,
Ipswich, Suffolk, IP5 3RE, UK
EMail: alan.mcguire@bt.com
Fedyk, et al. Standards Track [Page 3]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
Nurit Sprecher
Nokia Siemens Networks,
GmbH & Co. KG
COO RTP IE Fixed
3 Hanagar St. Neve Ne'eman B,
45241 Hod Hasharon, Israel
EMail: nurit.sprecher@nsn.com
Lou Berger
LabN Consulting, L.L.C.
Phone: +1-301-468-9228
EMail: lberger@labn.net
2. Terminology
In addition to well-understood GMPLS terms, this memo uses the
following terminology from IEEE 802.1 [IEEE802.1ah] [IEEE802.1Qay]:
- BCB Backbone Core Bridge
- BEB Backbone Edge Bridge
- B-MAC Backbone MAC
- B-VID Backbone VLAN ID
- B-VLAN Backbone VLAN
- CBP Customer Backbone Port
- CCM Continuity Check Message
- CNP Customer Network Port
- C-MAC Customer MAC
- C-VID Customer VLAN ID
- C-VLAN Customer VLAN
- ESP Ethernet Switched Path
- ESP-MAC SA ESP Source MAC Address
- ESP-MAC DA ESP Destination MAC Address
- ESP-VID ESP VLAN ID
- Eth-LSP Ethernet Label Switched Path
- IB-BEB A BEB comprised of both I- and B-components
- I-SID Ethernet Service Instance Identifier
- TAG An Ethernet Header Field with Type and Values
- MAC Media Access Control
- PBB Provider Backbone Bridges
- PBB-TE Provider Backbone Bridges Traffic Engineering
- PIP Provider Instance Port
- PNP Provider Network Port
- PS Protection Switching
- P2P Point-to-Point
- P2MP Point-to-Multipoint
- SVL Shared VLAN Learning
Fedyk, et al. Standards Track [Page 4]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
- TESI Traffic Engineering Service Instance
- VID VLAN ID
- VIP Virtual Instance Port
- VLAN Virtual LAN
2.1. PBB-TE and GMPLS Terminology
The PBB-TE specification [IEEE802.1Qay] defines some additional
terminology to clarify the PBB-TE functions. We repeat these here in
expanded context to translate from IEEE to GMPLS terminology. The
terms "bridge" and "switch" are used interchangeably in this
document. The signaling extensions described here apply equally well
to a PBB-TE-capable bridge supporting GMPLS signaling or to a GMPLS-
capable switch supporting Ethernet PBB-TE forwarding.
- Ethernet Switched Path (ESP):
A provisioned traffic engineered unidirectional connectivity
path between two or more Customer Backbone Ports (CBPs) that
extends over a Provider Backbone Bridge Network (PBBN). The
path is identified by the 3-tuple <ESP-MAC DA, ESP-MAC SA, ESP-
VID>. An ESP is point-to-point (P2P) or point-to-multipoint
(P2MP). An ESP is analogous to a (unidirectional) point-to-
point or point-to-multipoint LSP. We use the term Ethernet-LSP
(Eth-LSP) for GMPLS established ESPs.
- Point-to-Point ESP:
An ESP between two CBPs. The ESP-DA and the ESP-SA in the ESP's
3-tuple identifier are the individual MAC addresses of the two
CBPs.
- Point-to-Multipoint ESP:
An ESP among one root CBP and n leaf CBPs. The ESP-DA in the
ESP's 3-tuple identifier is a group MAC address identifying the
n leaf CBPs, and the ESP-SA is the individual MAC address of the
root.
- Point-to-Point PBB-TE Service Instance (P2P TESI):
A service instance supported by two point-to-point ESPs where
the ESPs' endpoints have the same CBP MAC addresses. The two
unidirectional ESPs are forming a bidirectional service. The
PBB-TE standard [IEEE802.1Qay] notes the following: for reasons
relating to TE service monitoring diagnostics, operational
simplicity, etc., the IEEE PBB-TE standard assumes that the
point-to-point ESPs associated with a point-to-point TESI are
Fedyk, et al. Standards Track [Page 5]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
co-routed. Support for a point-to-point TE services that
comprises non-co-routed ESPs is problematic, and is not defined
in this standard. Hence, a GMPLS bidirectional LSP is analogous
to a P2P TE Service Instance. We use the term "bidirectional
Ethernet-LSP" for GMPLS-established P2P PBB-TE Service
Instances.
2.2. Conventions Used in This Document
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].
3. Creation and Maintenance of PBB-TE Paths Using GMPLS
IEEE PBB-TE is a connection-oriented Ethernet technology. PBB-TE
ESPs are created bridge by bridge (or switch by switch) by simple
configuration of Ethernet forwarding entries. This document
describes the use of GMPLS as a valid control plane for the setup,
teardown, protection, and recovery of ESPs and TESIs and specifies
the required RSVP-TE extensions for the control of PBB-TE Service
Instances.
PBB-TE ESP and services are always originated and terminated on
IB-Backbone Edge Bridges (IB-BEBs). IB-BEBs are constituted of I and
B components, this is illustrated in Figure 1. A B-component refers
to the structure and mechanisms that support the relaying of frames
identified by Backbone VLANs in a Provider Backbone Bridge. An
I-component refers to the structure and mechanisms that support the
relaying of frames identified by service instances (I-SIDs) in a
Provider Backbone Bridge. PBB and PBB-TE relay frames with added
I-Component TAGs in the I-component and VLAN TAGs in the B-component.
PBB and PBB-TE forward frames based on VLAN ID in the VLAN TAG (in
the PBB case a B-VID) until the destination MAC address is supported
locally by a B-component on this bridge indicating the destination
has been reached. At that point, the B-VLAN tag is removed and
processing or forwarding on the next TAG begins (in the PBB case an
I-Component TAG) until the I-component identified by the I-SID is
reached. At the I-component, the I-Component TAG is removed and the
next Ethernet type identifies the TAG, etc.
An Ethernet service supported by a PBB-TE TESI is always attached to
a Customer Network Port (CNP) of the I-component. A Service Instance
Identifier (I-SID) is assigned for the service. I-SIDs are only
looked at by source and destination (edge) bridges, so I-SIDs are
transparent to path operations and MAY be signaled. The I- and
B-components have internal ports that are connected via an internal
LAN. These internal ports are the Provider Instance Ports (PIPs) and
Fedyk, et al. Standards Track [Page 6]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
Customer Backbone Ports (CBPs). PIPs and CBPs are not visible
outside the IB-BEB. ESPs are always originated and terminated on CBP
ports and use the MAC address of that port. The I-component
encapsulates the service frames arriving from the CNP by adding an
I-SID and a complete Ethernet MAC header with an ESP-MAC DA and
ESP-MAC SA. The B-component adds the ESP-VID.
This document defines extensions to GMPLS to establish ESPs and
TESIs. As can be seen from the above, this requires configuration of
both the I- and B-components of the IB-BEBs connected by the ESPs.
In the GMPLS control plane, TE Router IDs are used to identify the
IB-BEBs and Backbone Core Bridges (BCBs), and TE Links describe links
connected to PNPs and CNPs. TE Links are not associated with CBPs or
PIPs.
Note that since multiple internal CBPs may exist, an IB-BEB receiving
a PATH message MUST be able to determine the appropriate CBP that is
the termination point of the Eth-LSP. To this end, IB-BEBs SHOULD
advertise the CNP TE Links in the GMPLS control plane and RSVP-TE
signaling SHOULD use the CNP TE Links to identify the termination
point of Eth-LSPs. An IB-BEB receiving a PATH message specifying one
of its CNPs can locally determine which CBPs have internal
connectivity to the I-component supporting the given CNP. In the
case that there is more than one suitable CBP, and no I-SID
information is provided in the PATH message or previously in the
associated Call setup, then the IB-BEB can decide freely which CBP to
assign to the requested connection. On the other hand, if there is
information on the service (I-SID) that the given ESP will support,
then the IB-BEB MUST first determine which PIP and associated CBP is
configured with the I-SID and MUST assign that CBP to the ESP.
Fedyk, et al. Standards Track [Page 7]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
Backbone Edge Bridge (BEB)
+------------------------------------------------------+
| <TE - Router ID > |
| |
| I-Component Relay B-Component Relay |
| +-----------------------+ +---------------------+ |
| | +---+ | | B-VID | |
| | |VIP| | | +---+ +---+ | | <TE Link>
| | +---+ | +---|CBP| |PNP|------
| | | | | +---+ +---+ | |
| | +---+ +---+ | | | | |
------|CNP| |PIP|----+ | | |
| | +---+ +---+ | | | |
| +-----------------------+ +---------------------+ |
| |
| PBB Edge Bridge |
+------------------------------------------------------+
^--------Configured--------------^
^-----------GMPLS or Configured------^
Figure 1: IB-BEBs and GMPLS Identifiers
Control TE Router ID TE Router ID
Plane | (TE Link) |
V | V
+----+ | +-----+
Data | | | | |
Plane | | V label=ESP:VID/MAC DA | |
-----N N----------------------------N N----------
| | PBB-TE | | \ Network
| | / | Or
+----+ /+-----+ Customer
BCB ESP:MAC IB-BEB Facing
Ethernet
Ports
Figure 2: Ethernet/GMPLS Addressing and Label Space
PBB-TE defines the tuple of <ESP-MAC DA, ESP-MAC SA, ESP-VID> as a
unique connection identifier in the data plane, but the forwarding
operation only uses the ESP-MAC DA and the ESP-VID in each direction.
The ESP-VID typically comes from a small number of VIDs dedicated to
PBB-TE. ESP-VIDs can be reused across ESPs. There is no requirement
that ESP-VIDs for two ESPs that form a P2P TESI be the same.
Fedyk, et al. Standards Track [Page 8]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
When configuring an ESP with GMPLS, the ESP-MAC DA and ESP-VID are
carried in a generalized label object and are assigned hop by hop,
but are invariant within a domain. This invariance is similar to
GMPLS operation in transparent optical networks. As is typical with
other technologies controlled by GMPLS, the data plane receiver MUST
accept, and usually assigns, labels from its available label pool.
This, together with the label invariance requirement mentioned above,
result in each PBB-TE Ethernet Label being a domain-wide unique
label, with a unique ESP-VID + ESP-MAC DA, for each direction.
The following illustrates PBB-TE Ethernet Labels and ESPs for a P2P
TESI.
GMPLS Upstream Label <ESP:MAC1(DA), VID1> (60 bits)
GMPLS Downstream Label <ESP:MAC2(DA), VID2> (60 bits)
Upstream PBB-TE ESP 3-tuple <ESP:MAC1, MAC2, VID1> (108 bits)
Downstream PBB-TE ESP 3-tuple <ESP:MAC2, MAC1, VID2> (108 bits)
Table 1: Labels and ESPs
3.1. Shared Forwarding
One capability of a connectionless Ethernet data plane is to reuse
destination forwarding entries for packets from any source within a
VLAN to a destination. When setting up P2P PBB-TE connections for
multiple sources sharing a common destination, this capability MAY be
preserved provided certain requirements are met. We refer to this
capability as "shared forwarding". Shared forwarding is invoked
based on policy when conditions are met. It is a local decision by
label allocation at each end plus the path constraints. Shared
forwarding has no impact on the actual paths that are set up, but it
allows the reduction of forwarding entries. Shared forwarding paths
are identical in function to independently routed paths that share a
path from an intersecting bridge or link except they share a single
forwarding entry.
The forwarding memory savings from shared forwarding can be quite
dramatic in some topologies where a high degree of meshing is
required; however, it is typically easier to achieve when the
connectivity is known in advance. Normally, the originating GMPLS
switch will not have knowledge of the set of shared forwarding paths
rooted on the source or destination switch.
Use of a Path Computation Element [RFC4655] or other planning style
of tool with more complete knowledge of the network configuration is
a way to impose pre-selection of shared forwarding with multiple
paths using a single forwarding entry and optimizing for both
Fedyk, et al. Standards Track [Page 9]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
directions. In this scenario, the originating bridge uses the
LABEL_SET and UPSTREAM_LABEL objects to indicate the selection of the
shared forwarding labels at both ends.
3.2. P2P Connections Procedures for Shared Forwarding
The ESP-VID/ESP-MAC DA can be considered to be a shared forwarding
identifier or label consisting of some number of P2P connections
distinctly identified by the <ESP-MAC DA, ESP-MAC SA, ESP-VID> tuple.
This is analogous to an LDP label merge, but in the shared forwarding
case, the ESP header contains sufficient information to identify the
flow to which a packet belongs. Resources can continue to be
allocated per LSP with shared forwarding.
VLAN-tagged Ethernet packets include priority marking. Priority bits
MAY be used to indicate Class of Service (COS) and drop priority.
Thus, traffic from multiple COSs could be multiplexed on the same
Eth-LSP (i.e., similar to E-LSPs) and queuing and drop decisions are
made based on the p-bits. This means that the queue selection can be
done based on a per-flow basis (i.e., Eth-LSP + priority) and is
decoupled from the actual steering of the packet at any given bridge.
A bridge terminating an Eth-LSP will frequently have more than one
suitable candidate for sharing a forwarding entry (common
ESP-VID/ESP-MAC DA, unique ESP-MAC SA). It is a local decision of
how this is performed but a good choice is a path that reduces the
requirement for new forwarding entries by reusing common existing
paths.
The concept of bandwidth management still applies equally well with
shared forwarding.
4. Specific Procedures
4.1. P2P Ethernet LSPs
PBB-TE is designed to be bidirectional and symmetrically routed just
like Ethernet. That is, complete and proper functionality of
Ethernet protocols is only guaranteed for bidirectional Eth-LSPs. In
this section, we discuss the establishment of bidirectional Eth-LSPs.
Note, however, that it is also possible to use RSVP-TE to configure
unidirectional ESPs, if the UPSTREAM_LABEL is not included in the
PATH message.
Fedyk, et al. Standards Track [Page 10]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
To initiate a bidirectional Eth-LSP, the initiator of the PATH
message MUST use the procedures outlined in [RFC3473] with the
following specifics:
1) it MUST set the LSP encoding type to Ethernet (2) [RFC3471].
2) it MUST set the LSP switching type to "802_1 PBB-TE", value 40.
3) it SHOULD set the Generalized Payload Identifier (G-PID) to
Ethernet (33) [RFC3471].
4) it MUST set the UPSTREAM_LABEL to the ESP-VID1/ESP-MAC1 tuple
where the ESP-VID1 is administered locally for the local MAC
address: MAC1.
5) it SHOULD set the LABEL_SET or SUGGESTED_LABEL if it chooses to
influence the choice of ESP-VID/ESP-MAC DA.
6) it MAY carry an I-SID via Call/Connection ID [RFC4974].
Intermediate and egress bridge processing is not modified by this
document, i.e., is per [RFC3473]. However, as previously stated,
intermediate bridges supporting the 802_1 PBB-TE switching type MUST
NOT modify LABEL values.
The ESP-VID1/ESP-MAC1 tuple contained in the UPSTREAM_LABEL is used
to create a static forwarding entry in the Filtering Database of
bridges at each hop for the upstream direction. This behavior is
inferred from the switching type, which is 802_1 PBB-TE. The port
derived from the RSVP_HOP object and the ESP-VID1 and ESP-MAC1
included in the PBB-TE Ethernet Label constitute the static entry.
At the destination, an ESP-VID (ESP-VID2) is allocated for the local
MAC address: MAC2, the ESP-VID2/ESP-MAC2 tuple is passed in the LABEL
object in the RESV message. As with the PATH message, intermediate
bridge processing is per [RFC3473], and the LABEL object MUST be
passed on unchanged, upstream. The ESP-VID2/ESP-MAC2 tuple contained
in the LABEL object is installed in the forwarding table as a static
forwarding entry at each hop. This creates a bidirectional Eth-LSP
as the PATH and RESV messages follow the same path.
4.1.1. P2P Path Maintenance
Make-before-break procedures can be employed to modify the
characteristics of a P2P Eth-LSP. As described in [RFC3209], the LSP
ID in the sender template is updated as the new path is signaled.
The procedures (including those for shared forwarding) are identical
to those employed in establishing a new LSP, with the extended tunnel
Fedyk, et al. Standards Track [Page 11]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
ID in the signaling exchange ensuring that double booking of an
associated resource does not occur.
Where individual paths in a protection group are modified, signaling
procedures MAY be combined with Protection Switching (PS)
coordination to administratively force PS operations such that
modification is only ever performed on the protection path. PS is a
native capability of PBB-TE [IEEE802.1Qay] that can operate when two
paths are set up between two common endpoints.
4.2. P2MP Ethernet-LSPs
PBB-TE supports P2MP VID/Multicast MAC (MMAC) forwarding. In this
case, the PBB-TE Ethernet Label consists of a VID and a Group MAC
address. The procedures outlined in [RFC3473] and [RFC4875] could be
adapted to signal P2MP LSPs for the source (point) to destination
(multipoint) direction. Each one of the branches of the P2MP Eth-LSP
would be associated with a reverse-path symmetric and congruent P2P
Eth-LSP.
Complete procedures for signaling bidirectional P2MP E-LSPs are out
of scope for this document.
4.3. PBB-TE Ethernet Label
The PBB-TE Ethernet Label is a new generalized label with the
following format:
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0| ESP VID | ESP MAC (highest 2 bytes) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ESP MAC |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 3: PBB-TE Ethernet Label
This format MUST be used for both P2P and P2MP Eth-LSPs. For P2P
Eth-LSPs, the fields specify a VID and a unicast MAC address;
whereas, for P2MP Eth-LSPs, a VID and a group MAC address is carried
in the label. The PBB-TE Ethernet Label is a domain-wide unique
label and MUST be passed unchanged at each hop. This has similarity
to the way in which a wavelength label is handled at an intermediate
bridge that cannot perform wavelength conversion, and is described in
[RFC3473].
Fedyk, et al. Standards Track [Page 12]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
4.4. Protection Paths
When protection is used for path recovery, it is required to
associate the working and protection paths into a protection group.
This is achieved as defined in [RFC4872] and [RFC4873] using the
ASSOCIATION and PROTECTION objects.
4.5. Service Instance Identification
The I-SID is used to uniquely identify services within the network.
Unambiguous identification is achieved by ensuring global uniqueness
of the I-SIDs within the network or at least between any pair of edge
bridges. On IB-BEBs, the Backbone Service Instance Table is used to
configure the mapping between I-SIDs and ESPs. This configuration
can be either manual or semi-automated by signaling described here.
RSVP-TE Signaling MAY be used to automate I-SID to ESP mapping. By
relying on signaling, it is ensured that the same I-SID is assigned
to the service and mapped to the same ESP. Note, by signaling the
I-SID associated to the ESP, one can ensure that IB-BEBs select the
appropriate CBP port.
CALL signaling [RFC4974] MAY be used to create an association between
the Eth-LSP endpoints prior to establishment of the LSP. The
CALL_ATTRIBUTES object can be used during CALL signaling, as
described in [RFC4974], to indicate properties of the CALL. The
Service ID TLV, defined below, can be carried in the CALL_ATTRIBUTES
object to indicate the I-SID to ESP mapping for the Eth-LSP that will
be set up in association with the CALL.
Alternatively, the GMPLS RSVP-TE PATH message can carry the I-SID
association using the Service ID TLV in the LSP_ATTRIBUTES object
[RFC5420] at the time of Eth-LSP signaling. Using this mechanism, it
is possible to create the I-SID association, either when the path is
set up or at a later time using a PATH refresh.
A new Service ID TLV is defined for the CALL_ATTRIBUTES and
LSP_ATTRIBUTES objects. The type value is 3 when carried in the
CALL_ATTRIBUTES object and the type value is 2 when carried in the
LSP_ATTRIBUTES object. The format is depicted below.
Fedyk, et al. Standards Track [Page 13]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Length (variable) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| I-SID Set Object 1 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
: : :
: : :
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| I-SID Set Object n |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 4: Service ID TLV
- I-SID Set Object: is used to define a list or range of I-SIDs.
Multiple I-SID Set Objects can be present. At least one I-SID
Set Object MUST be present. In most of the cases, a single
I-SID Set Object with a single I-SID value is used. The I-SID
Set Object is used to define a list or range of I-SIDs. The
format of the I-SID Set Object is based on the LABEL_SET Object:
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Action | Reserved | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Reserved | I-SID 1 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
: : :
: : :
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Reserved | I-SID n |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 5: I-SID Set Object
- Action: 8 bits
The following actions are defined: list (0), range (1). When a
range is defined, there are only two I-SIDs that follow the
beginning I-SID and the end of the range I-SID. When list is
defined, a number of I-SIDs may be defined.
- Length: 16 bits
This indicates the length of the I-SID Set object.
Fedyk, et al. Standards Track [Page 14]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
- I-SID: 24 bits
The I-SID value identifies a particular backbone service
instance.
5. Error Conditions
The following errors identify Eth-LSP-specific problems.
In PBB-TE, a set of ESP-VIDs allocated to PBB-TE must be configured.
Therefore, it is possible in some situations that the configuration
of a bridge is not the same as other bridges. If the ESP-VIDs of
various bridges have some ESP-VIDs in common, it is possible some
paths may be set up before encountering issues. This is a management
issue since all bridges should have the same ESP-VID range.
Configuration should be consistent.
5.1. ESP-VID-Related Errors
The network operator administratively selects a set of VLAN
Identifiers that can be used to set up ESPs. Consequently, any VID
outside the allocated range is invalid, and an error MUST be
generated where the mismatch is discovered. The Error indication is
carried in the PathErr message from any intermediate bridge that does
not support the signaled source VID or optionally the destination
VID. The Error MAY be indicated in the ResvErr if the allocation
error happens on the RESV message. In this case, a bridge that does
not support the signaled destination VID MUST signal the error.
5.1.1. Invalid ESP-VID Value in the PBB-TE Ethernet Label
If a bridge is not configured to use the ESP-VID value, carried in
the Label object, for PBB-TE ESPs, it MUST immediately generate an
error: Routing problem (24) / Unacceptable label value (6). Handling
of this error is according to [RFC3209].
Note that an originating bridge can reuse an ESP-VID with a different
source or destination B-MAC address. By allocating a number of
B-MACs and a number of ESP-VIDs, a large number of PBB-TE connections
may be supported.
Note, this error may be originated by any bridge along the path.
Fedyk, et al. Standards Track [Page 15]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
5.1.2. Allocated ESP-VID Range is Exhausted
The destination bridge, after receiving the PATH message, has to
assign a VID, which, together with its MAC address, will constitute
the PBB-TE Ethernet Label. An existing VID may be reused when shared
forwarding is used or when there are no path conflicts; otherwise,
the bridge has to allocate a VID.
Depending on the size of the allocated VLAN range and the number of
Eth-LSPs terminated on a particular bridge, it is possible that the
available VIDs are exhausted; hence, no PBB-TE Ethernet Label can be
allocated. In this case, the destination bridge SHOULD generate a
PathErr message with error code: Routing problem (24) and error
value: MPLS Label allocation failure (9).
5.2. Invalid MAC Address
IEEE defines a set of reserved MAC addresses from 01-80-C2-00-00-00
to 01-80-C2-00-00-0F as explained in [IEEE802.1Q] that have special
meaning, processing, and follow specific forwarding rules. These
addresses cannot be used for PBB-TE ESPs. In the case the PBB-TE
Ethernet Label refers to such a MAC address, a bridge encountering
the mismatch MUST immediately generate an error: Routing problem (24)
/ Unacceptable label value (6). Handling of this error is according
to [RFC3209].
6. Security Considerations
This document does not introduce new security issues; the
considerations in [RFC4872] and [RFC4873] apply.
A GMPLS-controlled Ethernet PBB-TE system assumes that users and
devices attached to User-to-Network Interfaces (UNIs) may behave
maliciously, negligently, or incorrectly. Intra-provider control
traffic is trusted not to be malicious. In general, these
requirements are no different from the security requirements for
operating any GMPLS network. Access to the trusted network will only
occur through the protocols defined for the UNI or Network-to-Network
Interface (NNI) or through protected management interfaces.
When in-band GMPLS signaling is used for the control plane, the
security of the control plane and the data plane may affect each
other. When out-of-band GMPLS signaling is used for the control
plane, the data-plane security is decoupled from the control plane;
therefore, the security of the data plane has less impact on overall
security.
Fedyk, et al. Standards Track [Page 16]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
Where GMPLS is applied to the control of VLAN only, the commonly
known techniques for mitigation of Ethernet denial-of-service (DoS)
attacks may be required on UNI ports. PBB-TE has been designed to
interwork with legacy VLANs and the VLANs provide isolation from
Ethernet legacy control planes.
Where control-plane communications are point-to-point over links that
employ 802.1AE Media Access Control Security [MACSEC], it may
reasonably be determined that no further security measures are used.
In other cases, it is appropriate to use control-plane security where
it is deemed necessary to secure the signaling messages. GMPLS
signaling security measures are described in [RFC3471] and [RFC3473],
and they inherit security techniques applicable to RSVP-TE, as
described in [RFC3209] and [RFC2205]. For a fuller overview of GMPLS
security techniques, see [RFC5920].
7. IANA Considerations
A new Switching Type, "802_1 PBB-TE" (40), has been assigned in the
Switching Types registry of the GMPLS Signaling Parameters registry.
The Service ID TLV has been assigned in the Attributes TLV Space in
the RSVP-TE Parameters registry. It is carried in the LSP_ATTRIBUTES
object (class = 197, C-Type = 1) [RFC5420]. This new type has been
registered as follows:
Type: 2
Name: Service ID TLV
Allowed on LSP_ATTRIBUTES: Yes
Allowed on LSP_REQUIRED_ATTRIBUTES: No
The Service ID TLV has been assigned value 3 in the Call Attributes
TLV registry in the RSVP Parameters registry. It is carried in the
CALL_ATTRIBUTES object (class = 202, C-Type = 1) defined by
[RFC6001].
8. References
8.1. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
[RFC2205] Braden, R., Ed., Zhang, L., Berson, S., Herzog, S., and S.
Jamin, "Resource ReSerVation Protocol (RSVP) -- Version 1
Functional Specification", RFC 2205, September 1997.
Fedyk, et al. Standards Track [Page 17]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
[RFC3209] Awduche, D., Berger, L., Gan, D., Li, T., Srinivasan, V.,
and G. Swallow, "RSVP-TE: Extensions to RSVP for LSP
Tunnels", RFC 3209, December 2001.
[RFC3471] Berger, L., Ed., "Generalized Multi-Protocol Label
Swicthing (GMPLS) Signaling Functional Description", RFC
3471, January 2003.
[RFC3473] Berger, L., Ed., "Generalized Multi-Protocol Label
Switching (GMPLS) Signaling Resource ReserVation Protocol-
Traffic Engineering (RSVP-TE) Extensions", RFC 3473,
January 2003.
[RFC3945] Mannie, E., Ed., "Generalized Multi-Protocol Label
Switching (GMPLS) Architecture", RFC 3945, October 2004.
[RFC4872] Lang, J., Ed., Rekhter, Y., Ed., and D. Papadimitriou,
Ed., "RSVP-TE Extensions in Support of End-to-End
Generalized Multi-Protocol Label Switching (GMPLS)
Recovery", RFC 4872, May 2007.
[RFC4873] Berger, L., Bryskin, I., Papadimitriou, D., and A. Farrel,
"GMPLS Segment Recovery", RFC 4873, May 2007.
[RFC4974] Papadimitriou, D. and A. Farrel, "Generalized MPLS (GMPLS)
RSVP-TE Signaling Extensions in Support of Calls", RFC
4974, August 2007.
[RFC5420] Farrel, A., Ed., Papadimitriou, D., Vasseur, JP., and A.
Ayyangarps, "Encoding of Attributes for MPLS LSP
Establishment Using Resource Reservation Protocol Traffic
Engineering (RSVP-TE)", RFC 5420, February 2009.
[RFC6001] Papadimitriou, D., Vigoureux, M., Shiomoto, K., Brungard,
D., and JL. Le Roux, "Generalized MPLS (GMPLS) Protocol
Extensions for Multi-Layer and Multi-Region Networks
(MLN/MRN)", RFC 6001, October 2010.
Fedyk, et al. Standards Track [Page 18]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
8.2. Informative References
[IEEE802.1ah]
"IEEE Standard for Local and Metropolitan Area Networks -
Virtual Bridged Local Area Networks - Amendment 6:
Provider Backbone Bridges", (2008)
[IEEE802.1Q]
"IEEE Standard for Local and Metropolitan Area Networks -
Virtual Bridged Local Area Networks", IEEE Std
802.1Q-2005, May 19, 2006.
[IEEE802.1Qay]
"IEEE Standard for Local and Metropolitan Area Networks -
Virtual Bridged Local Area Networks - Amendment : Provider
Backbone Bridges Traffic Engineering", 2009.
[MACSEC] "IEEE Standard for Local and metropolitan area networks
Media Access Control (MAC) Security", IEEE 802.1AE-2006,
August 18, 2006.
[RFC4875] Aggarwal, R., Ed., Papadimitriou, D., Ed., and S.
Yasukawa, Ed., "Extensions to Resource Reservation
Protocol - Traffic Engineering (RSVP-TE) for Point-to-
Multipoint TE Label Switched Paths (LSPs)", RFC 4875, May
2007.
[RFC4655] Farrel, A., Vasseur, J.-P., and J. Ash, "A Path
Computation Element (PCE)-Based Architecture", RFC 4655,
August 2006.
[RFC5828] Fedyk, D., Berger, L., and L. Andersson, "Generalized
Multiprotocol Label Switching (GMPLS) Ethernet Label
Switching Architecture and Framework", RFC 5828, March
2010.
[RFC5920] Fang, L., Ed., "Security Framework for MPLS and GMPLS
Networks", RFC 5920, July 2010.
9. Acknowledgments
The authors would like to thank Dinesh Mohan, Nigel Bragg, Stephen
Shew, Dave Martin and Sandra Ballarte for their contributions to this
document. The authors thank Deborah Brungard and Adrian Farrel for
their review and suggestions to this document.
Fedyk, et al. Standards Track [Page 19]
RFC 6060 GMPLS Control of Ethernet PBB-TE March 2011
Authors' Addresses
Don Fedyk
Alcatel-Lucent
Groton, MA 01450
Phone: +1-978-467-5645
EMail: donald.fedyk@alcatel-lucent.com
Himanshu Shah
Ciena
1741 Technology Dr, #400
San Jose, CA 95110
Phone: 508-435-0448
EMail: hshah@ciena.com
Nabil Bitar
Verizon
40 Sylvan Rd.
Waltham, MA 02451
EMail: nabil.n.bitar@verizon.com
Attila Takacs
Ericsson
1. Laborc u.
Budapest, HUNGARY 1037
EMail: attila.takacs@ericsson.com
Fedyk, et al. Standards Track [Page 20]