Network Working Group Luca Martini
Internet Draft Samer Salam
Expiration Date: August 2009 Ali Sajassi
Intended status: Standards Track Cisco
Satoru Matsushima Thomas D. Nadeau
Softbank BT
February 17, 2009
Inter-Chassis Communication Protocol for L2VPN PE Redundancy
draft-martini-pwe3-iccp-01.txt
Status of this Memo
This Internet-Draft is submitted to IETF in full conformance with the
provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet-
Drafts.
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."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html.
This Internet-Draft will expire on August 17, 2009
Abstract
This document specifies an inter-chassis communication protocol
(ICCP) that enables PE redundancy for Virtual Private Wire Service
(VPWS) and Virtual Private LAN Service (VPLS) applications. The
protocol runs within a set of two or more PEs, forming a redundancy
group, for the purpose of synchronizing data amongst the systems. It
accommodates multi-chassis attachment circuit as well as pseudowire
redundancy mechanisms.
Martini, et al. [Page 1]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
Table of Contents
1 Specification of Requirements ........................ 3
2 Acknowledgments ...................................... 4
3 Introduction ......................................... 4
4 ICCP Overview ........................................ 4
4.1 Redundancy Model & Topology .......................... 4
4.2 ICCP Interconnect Scenarios .......................... 6
4.2.1 Co-located Dedicated Interconnect .................... 6
4.2.2 Co-located Shared Interconnect ....................... 7
4.2.3 Geo-redundant Dedicated Interconnect ................. 7
4.2.4 Geo-redundant Shared Interconnect .................... 8
4.3 ICCP Requirements .................................... 9
5 ICC LDP Protocol Extension Specification ............. 10
5.1 LDP ICCP Capability Advertisement .................... 11
5.2 RG Membership Management ............................. 11
5.3 Application Connection Management .................... 12
5.4 Application Versioning ............................... 13
5.5 Application Data Transfer ............................ 14
5.6 Dedicated Redundancy Group LDP session ............... 14
6 ICCP PE Node Failure Detection Mechanism ............. 14
7 ICCP Message Formats ................................. 15
7.1 Encoding ICC into LDP Messages ...................... 15
7.1.1 ICC Header ........................................... 16
7.1.2 Message Encoding ..................................... 17
7.2 RG Connect Message ................................... 19
7.2.1 Sender Name TLV ...................................... 19
7.3 RG Disconnect Message ................................ 20
7.4 RG Notification Message .............................. 23
7.4.1 Notification Message TLVs ............................ 23
7.5 RG Application Data Message .......................... 26
8 Application TLVs ..................................... 27
8.1 Pseudowire Redundancy (PW-RED) Application TLVs ...... 27
8.1.1 PW-RED Connect TLV ................................... 27
8.1.2 PW-RED Disconnect TLV ................................ 28
8.1.3 PW-RED Config TLV .................................... 29
8.1.4 Service Name TLV ..................................... 29
8.1.5 PW ID TLV ............................................ 29
8.1.6 Generalized PW ID TLV ................................ 30
8.2 Multi-chassis LACP (mLACP) Application TLVs .......... 31
8.2.1 mLACP Connect TLV .................................... 32
8.2.2 mLACP Disconnect TLV ................................. 33
8.2.3 mLACP System Config TLV .............................. 33
8.2.4 mLACP Port Config TLV ................................ 35
8.2.5 mLACP Change Port Priority TLV ....................... 36
8.2.6 mLACP Port State TLV ................................. 37
Martini, et al. [Page 2]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
9 LDP Capability Negotiation ........................... 40
10 Client Applications .................................. 41
10.1 Pseudowire Redundancy Application Procedures ......... 41
10.1.1 Initial Setup ........................................ 41
10.1.2 Pseudowire Configuration ............................. 41
10.1.3 Pseudowire Status Synchronization .................... 42
10.1.4 PE Node Failure ...................................... 42
10.2 Attachment Circuit Redundancy Application Procedures . 42
10.2.1 Common AC Procedures ................................. 42
10.2.2 AC Failure ........................................... 43
10.2.3 PE Node Failure ...................................... 43
10.2.4 PE Isolation ......................................... 43
10.2.5 ATM AC Procedures .................................... 43
10.2.6 Frame Relay AC Procedures ............................ 43
10.2.7 Ethernet AC Procedures ............................... 43
10.2.8 Multi-chassis LACP (mLACP) Application Procedures .... 43
10.2.8.1 Initial Setup ........................................ 44
10.2.8.2 mLACP Port Configuration ............................. 44
10.2.8.3 mLACP Port Status Synchronization .................... 45
10.2.8.4 Triggering Failover .................................. 45
11 Security Considerations .............................. 45
12 IANA Considerations .................................. 46
12.1 MESSAGE TYPE NAME SPACE .............................. 46
12.2 TLV TYPE NAME SPACE .................................. 46
12.3 ICC RG Parameter Type Space .......................... 46
12.4 STATUS CODE NAME SPACE ............................... 47
13 Normative References ................................. 48
14 Informative References ............................... 48
15 Author's Addresses ................................... 48
1. Specification of Requirements
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.
Martini, et al. [Page 3]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
2. Acknowledgments
The authors wish to acknowledge the important contributions of Dennis
Cai, Neil McGill and Amir Maleki.
3. Introduction
Network availability is a critical metric for service providers as it
has a direct bearing on their profitability. Outages translate not
only to lost revenue but also to potential penalties mandated by
contractual agreements with customers running mission-critical
applications that require tight SLAs. This is true for any carrier
network, and networks employing Layer2 Virtual Private Network
(L2VPN) technology are no exception. Network high-availability can
be achieved by employing intra and inter-chassis redundancy
mechanisms. The focus of this document is on the latter. The document
defines an Inter-Chassis Communication Protocol (ICCP) that allows
synchronization of state and configuration data between a set of two
or more PEs forming a Redundancy Group (RG). The protocol supports
multi-chassis redundancy mechanisms that can be employed on either
the attachment circuit or pseudowire front.
4. ICCP Overview
4.1. Redundancy Model & Topology
The focus of this document is on PE node redundancy. It is assumed
that a set of two or more PE nodes are designated by the operator to
form a Redundancy Group (RG). Members of a Redundancy Group fall
under a single administration (e.g. service provider) and employ a
common redundancy mechanism towards the access (attachment circuits
or access pseudowires) and/or towards the core (pseudowires) for any
given service instance. It is possible, however, for members of a RG
to make use of disparate redundancy mechanisms for disjoint services.
The PE devices may be offering any type of L2VPN service, i.e. VPWS
or VPLS. As a matter of fact, the use of ICCP may even be applicable
for Layer 3 service redundancy, but this is considered to be outside
the scope of this document.
The PEs in a RG offer multi-homed connectivity to either individual
devices (e.g. CE, DSLAM, etc...) or entire networks (e.g. access
network). Figure 1 below depicts the model.
Martini, et al. [Page 4]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
+=================+
| |
Mutli-homed +----+ | +-----+ |
Node ------------> | CE |-------|--| PE1 ||<------|---Pseudowire-->|
| |--+ -|--| ||<------|---Pseudowire-->|
+----+ | / | +-----+ |
| / | || |
|/ | || ICCP |--> Towards Core
+-------------+ / | || |
| | /| | +-----+ |
| Access |/ +----|--| PE2 ||<------|---Pseudowire-->|
| Network |-------|--| ||<------|---Pseudowire-->|
| | | +-----+ |
| | | |
+-------------+ | Redundancy |
^ | Group |
| +=================+
|
Multi-homed Network
Figure 1: Generic Multi-chassis Redundancy Model
In the topology of Figure 1, the redundancy mechanism employed
towards the access node/network can be one of a multitude of
technologies, e.g. it could be IEEE 802.3ad Link Aggregation Groups
with Link Aggregation Control Protocol (LACP), or SONET APS. The
specifics of the mechanism are out of the scope of this document.
However, it is assumed that the PEs in the RG are required to
communicate amongst each other in order for the access redundancy
mechanism to operate correctly. As such, it is required to run an
inter-chassis communication protocol among the PEs in the RG in order
to synchronize configuration and/or running state data.
Furthermore, the presence of the inter-chassis communication channel
allows simplification of the pseudowire redundancy mechanism. This is
primarily because it allows the PEs within a RG to run some
arbitration algorithm to elect which pseudowire(s) should be in
active or standby mode for a given service instance. The PEs can then
advertise the outcome of the arbitration to the remote-end PE(s), as
opposed to having to embed a hand-shake procedure into the pseudowire
redundancy status communication mechanism, and every other possible
Layer 2 status communication mechanism.
Martini, et al. [Page 5]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
4.2. ICCP Interconnect Scenarios
When referring to 'interconnect' in this section, we are concerned
with the links or networks over which Inter-Chassis Communication
Protocol messages are transported, and not normal data traffic
between PEs. The PEs which are members of a RG may be either
physically co-located or geo-redundant. Furthermore, the physical
interconnect between the PEs over which ICCP is to run may comprise
of either dedicated back-to-back links or a shared connection through
the PSN network (e.g., core). This gives rise to a matrix of four
interconnect scenarios, described next.
4.2.1. Co-located Dedicated Interconnect
In this scenario, the PEs within a RG are co-located in the same
physical location (POP, CO). Furthermore, dedicated links provide the
interconnect for ICCP among the PEs.
+=================+ +-----------------+
|CO | | |
| +-----+ | | |
| | PE1 |________|_____| |
| | | | | |
| +-----+ | | |
| || | | |
| || ICCP | | Core |
| || | | Network |
| +-----+ | | |
| | PE2 |________|_____| |
| | | | | |
| +-----+ | | |
| | | |
+=================+ +-----------------+
Figure 2: ICCP Co-located PEs Dedicated Interconnect Scenario
Given that the PEs are connected back-to-back in this case, it is
possible to rely on Layer 2 redundancy mechanisms to guarantee the
robustness of the links carrying the ICCP. For example, if the
interconnect comprises of IEEE 802.3 Ethernet links, it is possible
to provide redundant interconnect by means of IEEE 802.3ad Link
Aggregation Groups.
Martini, et al. [Page 6]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
4.2.2. Co-located Shared Interconnect
In this scenario, the PEs within a RG are co-located in the same
physical location (POP, CO). However, unlike the previous scenario,
there are no dedicated links between the PEs. The interconnect for
ICCP is provided through the core network to which the PEs are
connected. Figure 3 depicts this model.
+=================+ +-----------------+
|CO | | |
| +-----+ | | |
| | PE1 |________|_____| |
| | |<=================+ |
| +-----+ ICCP | | || |
| | | || |
| | | || Core |
| | | || Network |
| +-----+ | | || |
| | PE2 |________|_____| || |
| | |<=================+ |
| +-----+ | | |
| | | |
+=================+ +-----------------+
Figure 3: ICCP Co-located PEs Shared Interconnect Scenario
Given that the PEs in the RG are connected over the Packet Switched
Network (PSN), then PSN Layer mechanisms can be leveraged to ensure
the resiliency of the interconnect against connectivity failures. For
example, it is possible to employ RSVP LSPs with Fast ReRoute (FRR)
and/or end-to-end backup LSPs.
4.2.3. Geo-redundant Dedicated Interconnect
In this variation, the PEs within a Redundancy Group are located in
different physical locations to provide geographic redundancy. This
may be desirable, for example, to protect against natural disasters
or the like. A dedicated interconnect is provided to link the PEs,
which is a costly option, especially when considering the possibility
of providing multiple such links for interconnect robustness. The
resiliency mechanisms for the interconnect are similar to those
highlighted in the co-located interconnect counterpart.
Martini, et al. [Page 7]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
+=================+ +-----------------+
|CO 1 | | |
| +-----+ | | |
| | PE1 |________|_____| |
| | | | | |
| +-----+ | | |
+=====||==========+ | |
|| ICCP | Core |
+=====||==========+ | Network |
| +-----+ | | |
| | PE2 |________|_____| |
| | | | | |
| +-----+ | | |
|CO 2 | | |
+=================+ +-----------------+
Figure 4: ICCP Geo-redundant PEs Dedicated Interconnect Scenario
4.2.4. Geo-redundant Shared Interconnect
In this scenario, the PEs of a RG are located in different physical
locations and the interconnect for ICCP is provided over the PSN
network to which the PEs are connected. This interconnect option is
more likely to be the one used for geo-redundancy as it is more
economically appealing compared to the geo-redundant dedicated
interconnect. The resiliency mechanisms that can be employed to
guarantee the robustness of the ICCP transport are PSN Layer
mechanisms as has been described in a previous section.
+=================+ +-----------------+
|CO 1 | | |
| +-----+ | | |
| | PE1 |________|_____| |
| | |<=================+ |
| +-----+ ICCP | | || |
+=================+ | || |
| || Core |
+=================+ | || Network |
| +-----+ | | || |
| | PE2 |________|_____| || |
| | |<=================+ |
| +-----+ | | |
|CO 2 | | |
+=================+ +-----------------+
Figure 5: ICCP Geo-redundant PEs Shared Interconnect Scenario
Martini, et al. [Page 8]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
4.3. ICCP Requirements
The Inter-chassis Communication Protocol should satisfy the following
requirements:
-i. Provide a control channel for communication between PEs in
Redundancy Group (RG). Nodes maybe co-located or remote
(refer to Interconnect Scenarios section above). It is
expected that client applications which make use of ICCP
services will only use this channel to communicate control
information and not data-traffic. As such the protocol
should cater for low-bandwidth, low-delay and highly
reliable message transfer.
-ii. Accommodate multiple client applications (e.g. multi-chassis
LACP, PW redundancy, SONET APS, etc...). This implies that
the messages should be extensible (e.g. TLV-based) and the
protocol should provide a robust application registration
and versioning scheme.
-iii. Provide reliable message transport and in-order delivery
between nodes in a RG with secure authentication mechanisms
built into the protocol. The redundancy applications that
are clients of ICCP expect reliable message transfer, and as
such will assume that the protocol takes care of flow-
control and retransmissions. Furthermore, given that the
applications will rely on ICCP to communicate data used to
synchronize state-machines on disparate nodes, it is
critical that ICCP guarantees in-order message delivery.
Loss of messages or out-of-sequence messages would have
adverse side-effects to the operation of the client
applications.
-iv. Provide a common mechanism to actively monitor the health of
PEs in a RG. This mechanism will be used to detect PE node
failure and inform the client applications. The applications
require this to trigger failover according to the procedures
of the employed redundancy protocol on the AC and PW. It is
desired to achieve sub-second detection of loss of remote
node (~ 50 - 150 msec) in order to give the client
applications (redundancy mechanisms) enough reaction time to
achieve sub-second service restoration time.
-v. Provide asynchronous event-driven state update, independent
of periodic messages, for immediate notification of client
applications' state changes. In other words, the
transmission of messages carrying application data should be
on-demand rather than timer-based to minimize inter-chassis
Martini, et al. [Page 9]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
state synchronization delay.
-vi. Accommodate multi-link and multi-hop interconnect between
nodes. When the devices within a RG are located in different
physical locations, the physical interconnect between them
will comprise of a network rather than a link. As such, ICCP
should accommodate the case where the interconnect involves
multiple hops. Furthermore, it is possible to have multiple
(redundant) paths or interconnects between a given pair of
devices. This is true for both the co-located and geo-
redundant scenarios. ICCP should handle this as well.
-vii. Ensure transport security between devices in a RG. This is
especially important in the scenario where the members of a
RG are located in different physical locations and connected
over a shared (e.g. PSN) network.
-viii. Must allow operator to statically configure members of RG.
Auto-discovery may be considered in the future.
-ix. Allow for flexible RG membership. It is expected that only
two nodes per an RG will cover most of the redundancy
applications for common deployments. However, ICCP should
not preclude supporting more than two nodes in a RG by
virtue of design. Furthermore, it is required to allow a
single node to be member of multiple RGs simultaneously.
5. ICC LDP Protocol Extension Specification
To address the requirements identified in the previous section, ICCP
is modeled to comprise of three layers:
-i. Application Layer: This provides the interface to the
various redundancy applications that make use of the
services of ICCP. ICCP is concerned with defining common
connection management procedures and the formats of the
messages exchanged at this layer; however, beyond that, it
does not impose any restrictions on the procedures or
state-machines of the clients, as these are deemed
application-specific and lie outside the scope of ICCP.
This guarantees implementation inter-operability without
placing any unnecessary constraints on internal design
specifics.
Martini, et al. [Page 10]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
-ii. Inter Chassis Communication (ICC) Layer: This layer
implements the common set of services which ICCP offers to
the client applications. It handles protocol versioning, RG
membership, PE node identification and ICCP connection
management.
-iii. Transport Layer: This layer provides the actual ICCP message
transport. It is responsible for addressing, route
resolution, flow-control, reliable and in-order message
delivery, connectivity resiliency/redundancy and finally PE
node failure detection. The Transport layer may differ
depending on the Physical Layer of the inter-connect.
5.1. LDP ICCP Capability Advertisement
When a RG is enabled on a particular PE, the capability of supporting
ICCP must be advertised to all LDP peers in that RG. This is achieved
by using the methods in [LDP-CAP] and advertising the ICCP LDP
capability TLV. If an LDP peer supports the dynamic capability
advertisement, this can be done by sending a new capability message
with the S bit set for the ICCP capability TLV when the first RG is
enabled on the PE. If the peer does not support dynamic capability
advertisement, then the ICCP TLV MUST be included in the LDP
initialization procedures in the capability parameter [LDP-CAP].
5.2. RG Membership Management
ICCP defines a mechanism that enables PE nodes to manage their RG
membership. When a PE is configured to be a member of a RG, it will
first advertise the ICCP capability to its peers. Subsequently, the
PE sends a RG Connect message to the peers that have also advertised
ICCP capability. The PE then waits for the peers to send their own RG
Connect messages, if they haven't already. For a given RG, the ICCP
connection between two devices is considered to be operational only
when both have sent and received ICCP RG Connect messages for that
RG.
If a PE that has sent a particular RG Connect message doesn't receive
a corresponding RG Connect (or a Notification message with NAK) from
a destination, it will remain in a state expecting the corresponding
RG Connect message (or Notification message). The RG will not become
operational until the corresponding RG Connect Message has been
received. If a PE that has sent an RG Connect message receives a
Notification message with a NAK, it will stop attempting to bring up
the ICCP connection immediately. The PE MUST resume bringing up the
connection after it receives a RG Connect message from the peer PE
Martini, et al. [Page 11]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
for the RG in question. This is achieved by responding to the
incoming RG Connect message with an appropriate RG Connect.
A device MUST send a NAK for a RG Connect message if at least one of
the following conditions is satisfied:
-i. the PE is not a member of the RG;
-ii. the maximum number of simultaneous ICCP connections that the
PE can handle is exceeded.
A PE sends a RG Disconnect message to tear down the ICCP connection
for a given RG. This is a unilateral operation and doesn't require
any acknowledgement from the other PEs. Note that the ICCP connection
for a RG MUST be operational before any client application can make
use of ICCP services in that RG.
5.3. Application Connection Management
ICCP provides a common set of procedures by which applications on one
PE can connect to their counterparts on another PE, for purpose of
inter-chassis communication in the context of a given RG. The
prerequisite for establishing an application connection is to have an
operational ICCP RG connection between the two endpoints. It is
assumed that the association of applications with RGs is known
apriori, e.g. by means of device configuration. ICCP then sends an
Application-specific Connect TLV (carried in RG Connect message), on
behalf of each client application, to each remote PE within the RG.
The client may piggyback application-specific information in that
Connect TLV, which for example can be used to negotiate parameters or
attributes prior to bringing up the actual application connection.
The procedures for bringing up the application connection are similar
to those of the ICCP connection: An application connection between
two nodes is up only when both nodes have sent and received RG
Connect Messages with the proper Application-specific Connect TLVs. A
PE MUST send a Notification Message to NAK an application connection
request if one of the following conditions is encountered:
-i. the application doesn't exist or is not configured for that
RG;
-ii. the application connection count exceeds the PE's
capabilities.
When a PE receives such a NAK notification, it should stop attempting
to bring up the application connection until it receives a new
application connection request from the remote PE. This is done by
Martini, et al. [Page 12]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
responding to the incoming RG Connect message (carrying an
Application-specific Connect TLV) with an appropriate RG Connect
message (carrying a corresponding Application-specific Connect TLV).
When an application is stopped on a device or it is no longer
associated with a RG, it should signal ICCP to trigger sending an
Application-specific Disconnect TLV (in RG Disconnect message). This
is a unilateral notification to the other PEs within a RG, and as
such doesn't trigger any response.
5.4. Application Versioning
During application connection setup time, a given application on one
PE can negotiate with its counterpart on a peer PE the proper
application version to use for communication. If no common version is
agreed upon, then the application connection is not brought up. This
is achieved through the following simple rules:
- If an application receives an Application-specific Connect TLV
with a version that is higher than its own, it MUST send a
Notification message with a NAK TLV indicating status code
"Incompatible Protocol Version" and supplying the version that is
locally supported by the PE.
- If an application receives an Application-specific Connect TLV
with a version that is lower than its own, it MAY respond with a
RG Connect that has an Application-specific Connect TLV using the
same version that was received. Alternatively, the application
MAY respond with a Notification message to NAK the request using
the "Incompatible Protocol Version" code, and supplying the
version that is supported. The above allows an application to
operate in either backwards compatible or incompatible modes.
- If an application receives an Application-specific Connect TLV
with a version that is equal to its own, then the application
MUST honor or reject the request based on whether the application
is configured for the RG in question, and whether or not the
application connection count has been exceeded.
Martini, et al. [Page 13]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
5.5. Application Data Transfer
When an application has information to transfer over ICCP it triggers
the transmission of an Application Data message. ICCP guarantees in-
order and loss-less delivery of data. An application may NAK a
message or a set of one or more TLVs within a message by using the
Notification Message with NAK TLV. Furthermore, an application may
implement its own ACK mechanism, if deemed required, by defining an
application-specific TLV to be transported in an Application Data
message.
5.6. Dedicated Redundancy Group LDP session
In some ICCP applications there is a requirement to exchange a fairly
large amount of RG information in a very short period of time. In
order to better distribute the load in a multiple processor system,
and to avoid head of line blocking to other LDP appications, it may
be required to inititate a separate TCP/IP session between the two
LDP speakers.
This procedure is OPTIONAL, and does not change the operation of LDP
or ICCP.
An LSR that requires a separate LDP session will advertise a separate
LDP adjacency with a non-zero label space identifier. This will cause
the remote peer to open a separate LDP session for this label space.
No labels need be advertised in in this label space , as it is only
used for a particular , or particular set of ICCP RGs. All relevant
LDP , and ICCP procedures still apply as described in the relevant
documents.
6. ICCP PE Node Failure Detection Mechanism
ICCP provides its client applications a notification when a remote PE
that is member of the RG fails. This is used by the client
applications to trigger failover according to the procedures of the
employed redundancy protocol on the AC and PW. To that end, ICCP does
not define its own KeepAlive mechanism for purpose of monitoring the
health of remote PE nodes, but rather reuses existing fault detection
mechanisms. The following mechanisms may be used by ICCP to detect PE
node failure:
- BFD
Run a BFD session [BFD] between the PEs that are members of a
given RG, and use that to detect PE node failure. This assumes
Martini, et al. [Page 14]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
that resiliency mechanisms are in place to protect connectivity
to the remote PE nodes, and hence loss of BFD periodic messages
from a given PE node can only mean that the node itself has
failed.
- IP Reachability Monitoring
It is possible for a PE to monitor IP layer connectivity to other
members of a RG that are participating in IGP/BGP. When
connectivity to a given PE is lost, the local PE interprets that
to mean loss of the remote PE node. This assumes that resiliency
mechanisms are in place to protect the route to the remote PE
nodes, and hence loss of IP reachability to a given node can only
mean that the node itself has failed.
It is worth noting here that loss of the LDP session with a PE in a
RG is not a reliable indicator that the remote PE itself is down. It
is possible, for e.g. that the remote PE encounters a local event
that leads to resetting the LDP session, while the PE node remains
operational for purpose of traffic forwarding.
7. ICCP Message Formats
This section defines the messages exchanged at the Application and
ICC layers.
7.1. Encoding ICC into LDP Messages
ICCP requires reliable, in order, state-full message delivery, as
well as capability negotiation between PEs. The LDP protocol offers
all these features, and is already in wide use in the applications
that would also require the ICCP protocol extensions. For these
reasons, ICCP takes advantage of the already defined LDP protocol
infrastructure. [RFC5036] Section 3.5 defines a generic LDP message
structure. A new set of LDP message types is defined to communicate
the ICCP information. LDP message types in the range of 0x700 to
0x7ff will be used for ICCP.
Message types are allocated by IANA, and requested in the IANA
section below.
Martini, et al. [Page 15]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
7.1.1. ICC Header
Every ICCP message comprises of an ICC specific LDP Header followed
by an ICCP message. The format of the ICC Header is 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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U| Message Type | Message Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Message ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type=0x0005 (ICC RG ID) | Length=4 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ICC RG ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| |
+ +
| Mandatory Parameters |
~ ~
+ +
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| |
+ +
| Optional Parameters |
~ ~
+ +
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U-bit
Unknown message bit. Upon receipt of an unknown message, if U is
clear (=0), a notification is returned to the message originator;
if U is set (=1), the unknown message is silently ignored. The
following sections which define messages specify a value for the
U-bit.
- Message Type
Identifies the type of the ICCP message, must be in the range of
0x0700 to 0x07ff.
Martini, et al. [Page 16]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
- Message Length
Two octet integer specifying the total length of this message in
octets, excluding the U-bit, Message Type and Length fields.
- Message ID
Four octet value used to identify this message. Used by the
sending PE to facilitate identifying RG Notification messages
that may apply to this message. A PE sending a RG Notification
message in response to this message SHOULD include this Message
ID in the "NAK TLV" of the RG Notification message; see Section
"RG Notification Message".
- ICC RG ID
A TLV of type 0x0005, length 4, containing 4 octects unsigned
integer designating the Redundancy Group which the sending device
is member of. RG ID value 0x00000000 is reserved by the protocol.
- Mandatory Parameters
Variable length set of required message parameters. Some
messages have no required parameters.
For messages that have required parameters, the required
parameters MUST appear in the order specified by the individual
message specifications in the sections that follow.
- Optional Parameters
Variable length set of optional message parameters. Many
messages have no optional parameters.
For messages that have optional parameters, the optional
parameters may appear in any order.
7.1.2. Message Encoding
The generic format of an ICC parameter is:
Martini, et al. [Page 17]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| TLV(s) |
~ ~
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U-bit
Unknown TLV bit. Upon receipt of an unknown TLV, if U is clear
(=0), a notification MUST be returned to the message originator
and the entire message MUST be ignored; if U is set (=1), the
unknown TLV MUST be silently ignored and the rest of the message
processed as if the unknown TLV did not exist. The sections
following that define TLVs specify a value for the U-bit.
- F-bit
Forward unknown TLV bit. This bit applies only when the U-bit is
set and the LDP message containing the unknown TLV is to be
forwarded. If F is clear (=0), the unknown TLV is not forwarded
with the containing message; if F is set (=1), the unknown TLV is
forwarded with the containing message. The sections following
that define TLVs specify a value for the F-bit. By setting both
the U- and F-bits, a TLV can be propagated as opaque data through
nodes that do not recognize the TLV.
- Type
Fourteen bits indicating the parameter type.
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- TLV(s): A set of 0 or more TLVs, that vary according to the
message type.
Martini, et al. [Page 18]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
7.2. RG Connect Message
The RG Connect Message is used to establish ICCP connection in
addition to individual Application connections between PEs in a RG: a
RG Connect message with no "Application-specific connect TLVs"
signals establishment of the base ICCP connection. RG Connect
messages with appropriate "Application-specific connect TLVs" signal
the establishment of Application connections, in addition to the base
ICCP connection (if not already established). A PE sends a RG
Connect Message to declare its membership in a Redundancy Group. One
such message should be sent to each PE that is member of the same RG.
The set of PEs to which RG Connect Messages should be transmitted is
known via configuration or an auto-discovery mechanism that is
outside the scope of this specification. If a device is member of
multiple RGs, it must send separate RG Connect Messages for each RG
even if the receiving device(s) happen to be the same.
The format of the RG Connect Message is as follows:
-i. ICC header with Message type = "RG Connect Message" (0x0700)
-ii. "Sender Name TLV"
-iii. "Application specific connect TLV"
The currently defined Application-specific connect TLVs are:
- PW Redundancy Connect TLV
- mLACP Connect TLV
The details of these TLVs are discussed in the "Application TLVs"
section.
The RG Connect message can contain zero, one or more Application-
specific connect TLVs. Multiple application connect TLVs can be sent
in a single message, or multiple messages can be sent containing
different application connect TLVs, but no application connect TLV
can be sent more than once.
7.2.1. Sender Name TLV
A TLV that carries the hostname of the sender encoded in UTF-8. This
is used primarily for purpose of management of the RG and easing
network operations. The specific format is shown below:
Martini, et al. [Page 19]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Sender Name |
+ +-+-+-+-+-+-+-+-+-+
~ ~
| ... |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U=F=0
- Type set to "ICC sender name" Parameter type (from ICC parameter
name space).
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Sender Name
Hostname of sending device encoded in UTF-8, and SHOULD NOT
exceed 80 characters.
7.3. RG Disconnect Message
The RG Disconnect Message serves dual-purpose: to signal that a
particular Application connection is being closed within a RG, or
that the ICCP connection itself is being disconnected because the PE
wishes to leave the RG. The format of this message is:
Martini, et al. [Page 20]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U| Message Type=0x0701 | Message Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Message ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type=0x0005 (ICC RG ID) | Length=4 |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ICC RG ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Disconnect Code TLV |
+ +
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Optional Application-specific Disconnect TLVs |
~ ~
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Optional Parameter TLVs |
+ +
| |
~ ~
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U-bit
U=0
- Message Type
The message type for RG Disconnect Message is set to (0x0701)
- Length
Length of the TLV in octets excluding the U-bit, Message Type,
and Message Length fields.
- Message ID
Defined in the "ICC Header" section above.
- ICC RG ID
Defined in the "ICC Header" section above.
Martini, et al. [Page 21]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
- Disconnect Code TLV
The format of this TLV is 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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0004 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ICCP Status Code |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U,F Bits
both U and F are set to 0.
- Type
set to "Disconnect Code TLV" (0x0004)
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- ICCP Status Code
A status code that reflects the reason for the disconnect
message. Allowed values are "ICCP RG Removed" and "ICCP
Application Removed from RG".
- Optional Application-specific Disconnect TLVs
Zero, one or more Application-specific Disconnect TLVs which are
defined later in the document. If the RG Disconnect message has a
status code of "RG Removed", then it should not contain any
Application-specific Disconnect TLVs, as the sending PE is
signaling that it has left the RG and, thus, is disconnecting the
entire ICCP connection, with all associated client application
connections. If the message has a status code of "Application
Removed from RG", then it should contain one or more
Application-specific Disconnect TLVs, as the sending PE is only
tearing down the connection for the specified applications. Other
applications, and the base ICCP connection are not to be
affected.
Martini, et al. [Page 22]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
- Optional Parameter TLVs
None are defined for this message in this document.
7.4. RG Notification Message
A PE sends a RG Notification Message to indicate one of the
following: to reject an ICCP connection, to reject an application
connection, to NAK an entire message or to NAK one or more TLV(s)
within a message. The Notification message can only be sent to a PE
that is already part of a RG.
The format of the Notification Message is:
-i. ICC header with Message type = "RG Notification Message"
(0x0702)
-ii. Notification Message TLVs.
The currently defined Notification message TLVs are:
-i. Sender Name TLV
-ii. NAK TLV.
7.4.1. Notification Message TLVs
The Sender Name TLV uses the same format as in the RG Connect
message, and was described above.
The NAK TLV is defined 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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0002 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ICCP Status Code |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Rejected Message ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Optional TLV(s) |
+ +
| |
~ ~
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Martini, et al. [Page 23]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
- U,F Bits
both U and F are set to 0.
- Type
set to "NAK TLV" (0x0002)
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- ICCP Status Code
A status code the reflects the reason for the NAK TLV. Allowed
values are:
-i. Unknown RG (0x00010001)
This code is used to reject a new incoming ICCP
connection for a RG that is not configured on the local
PE. When this code is used, the Rejected Message ID
field must contain the message ID of the rejected "RG
Connect" message.
-ii. ICCP Connection Count Exceeded (0x00010002)
This is used to reject a new incoming ICCP connection
that would cause the local PE's ICCP connection count to
exceed its capabilities. When this code is used, the
Rejected Message ID field must contain the message ID of
the rejected "RG Connect" message.
-iii. Application Connection Count Exceeded (0x00010003)
This is used to reject a new incoming application
connection that would cause the local PE's ICCP
connection count to exceed its capabilities. When this
code is used, the Rejected Message ID field must contain
the message ID of the rejected "RG Connect" message and
the corresponding Application Connect TLV must be
included in the "Optional TLV".
-iv. Application not in RG (0x00010004)
This is used to reject a new incoming application
connection when the local PE doesn't support the
application, or the application is not configured in the
Martini, et al. [Page 24]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
RG. When this code is used, the Rejected Message ID
field must contain the message ID of the rejected "RG
Connect" message and the corresponding Application
Connect TLV must be included in the "Optional TLV".
-v. Incompatible Protocol Version (0x00010005)
This is used to reject a new incoming application
connection when the local PE has an incompatible version
of the application. When this code is used, the Rejected
Message ID field must contain the message ID of the
rejected "RG Connect" message and the corresponding
Application Connect TLV must be included in the
"Optional TLV".
-vi. Rejected Message (0x00010006)
This is used to reject a RG Application Data message, or
one or more TLV(s) within the message. When this code
is used, the Rejected Message ID field must contain the
message ID of the rejected "RG Application Data"
message.
-vii. ICCP Administratively Disabled (0x00010007)
This is used to reject any ICCP messages from a peer
from which the PE is not allowed to exchenge ICCP
messages due to local administrative policy.
- Rejected Message ID
If non-zero, 32-bit value that identifies the peer message to
which the NAK TLV refers. If zero, no specific peer message is
being identified.
- Optional TLV(s)
A set of one or more optional TLVs. If the status code is
"Rejected Message" then this field contains the TLV(s) that were
rejected. If the entire message is rejected, all its TLVs MUST be
present in this field; otherwise, the subset of TLVs that were
rejected MUST be echoed in this field.
If the status code is "Incompatible Protocol Version" then this
field contains the "Requested Protocol Version TLV" defined as
follows:
Martini, et al. [Page 25]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0003 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Connection Reference | Requested Version |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0003 for "Requested Protocol Version TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Connection Reference
This field is set to the Type field of the Application specific
Connect TLV that was rejected because of incompatible version.
- Requested Version
The version of the application supported by the transmitting
device. For this version of the protocol it is set to 0x0001.
7.5. RG Application Data Message
The RG Application Data Message is used to transport application data
between PEs within a RG. A single message can be used to carry data
from multiple applications, as long as all these applications are
part of the same RG. Such multiplexing is possible because the
transported TLVs are application specific which allows for
identifying the target application for each TLV at the receiving
side. The format of the Application Data Message is:
-i. ICC header with Message type = "RG Application Data Message"
(0x703)
Martini, et al. [Page 26]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
-ii. "Application specific TLVs"
The details of these TLVs are discussed in the "Application TLVs"
section.
8. Application TLVs
8.1. Pseudowire Redundancy (PW-RED) Application TLVs
This section discusses the ICCP TLVs for the Pseudowire Redundancy
application.
8.1.1. PW-RED Connect TLV
This TLV is included in the RG Connect message to signal the
establishment of PW-RED application connection.
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0010 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Protocol Version | Optional Sub-TLVs |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +
~ ~
| |
+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ... |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0010 for "PW-RED Connect TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
Martini, et al. [Page 27]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
- Protocol Version
The version of this particular protocol for the purposes of ICCP.
This is set to 0x0001.
- Optional Sub-TLVs
There are no optional Sub-TLVs defined for this version of the
protocol.
8.1.2. PW-RED Disconnect TLV
This TLV is used in a RG Disconnect Message to indicate that the
connection for the PW-RED application is to be terminated.
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0011 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Optional Sub-TLVs |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0011 for "PW-RED Disconnect TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Optional Sub-TLVs
There are no optional Sub-TLVs defined for this version of the
protocol.
Martini, et al. [Page 28]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
8.1.3. PW-RED Config TLV
The PW-RED Config TLV is used in RG Application Data message and is
composed of the following TLVs in the following order:
-i. Service Name TLV
-ii. PW ID TLV or Generalized PW ID TLV
In the PW-RED Config TLV the U and F Bits are both are set to 0, and
the TLV type is set to 0x0012.
8.1.4. Service Name TLV
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Service Name |
~ ~
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0013 for "Service Name TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Service Name
The name of the L2VPN service instance encoded in UTF-8 format
and up to 80 character in length.
8.1.5. PW ID TLV
This TLV is used to communicate the configuration of PWs for VPWS.
Martini, et al. [Page 29]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Peer ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Group ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| PW ID |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0014 for "PW ID TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Peer ID
Four octet LDP Router ID of the peer at the far end of the PW.
- Group ID
Same as Group ID in [RFC4447] section 5.2.
- PW ID
Same as PW ID in [RFC4447] section 5.2.
8.1.6. Generalized PW ID TLV
This TLV is used to communicate the configuration of PWs for VPLS.
Martini, et al. [Page 30]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type = 0x0015 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| AGI Type | Length | Value |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
~ AGI Value (contd.) ~
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| AII Type | Length | Value |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
~ SAII Value (contd.) ~
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| AII Type | Length | Value |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
~ TAII Value (contd.) ~
| |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F bits
both set to 0.
- Type
set to 0x0015
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- AGI, AII, SAII and TAII
defined in [RFC4447] section 5.3.2.
8.2. Multi-chassis LACP (mLACP) Application TLVs
This section discusses the ICCP TLVs for Ethernet attachment circuit
redundancy using the multi-chassis LACP (mLACP) application.
Martini, et al. [Page 31]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
8.2.1. mLACP Connect TLV
This TLV is included in the RG Connect message to signal the
establishment of mLACP application connection.
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0030 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Protocol Version | Optional Sub-TLVs |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +
~ ~
| |
+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| ... |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0030 for "mLACP Connect TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Protocol Version
The version of this particular protocol for the purposes of ICCP.
This is set to 0x0001.
- Optional Sub-TLVs
There are no optional Sub-TLVs defined for this version of the
protocol.
Martini, et al. [Page 32]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
8.2.2. mLACP Disconnect TLV
This TLV is used in a RG Disconnect Message to indicate that the
connection for the mLACP application is to be terminated.
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0031 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Optional Sub-TLVs |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0031 for "mLACP Disconnect TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Optional Sub-TLVs
There are no optional Sub-TLVs defined for this version of the
protocol.
8.2.3. mLACP System Config TLV
The mLACP System Config TLV is sent in the RG Application Data
message. This TLV announces the local node's LACP System Parameters
to the RG peers.
Martini, et al. [Page 33]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0032 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| System ID |
+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| | System Priority |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Node ID |
+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0032 for "mLACP System Config TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- System ID
6 octets field encoding the System ID used by LACP as specified
in [IEEE-802.3] section 43.3.2.
- System Priority
2 octets encoding the LACP System Priority as defined in [IEEE-
802.3] section 43.3.2.
- Node ID
One octet, LACP node ID. Used to ensure that the LACP Port IDs
are unique across all devices in a RG. Valid values are in the
range 0 - 7.
Martini, et al. [Page 34]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
8.2.4. mLACP Port Config TLV
The mLACP Port Config TLV is sent in the RG Application Data message.
This TLV is used to notify RG peers about the local configuration
state of a port.
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0033 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Port Number | MAC Address |
+-------------------------------+ +
| |
+---------------------------------------------------------------+
| Admin Key | Port Priority |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Flags | IF Name Len | Interface Name |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +
~ ~
| ... |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0033 for "mLACP Port Config TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Port Number
Two octets, LACP Port Number for the corresponding interface as
specified in [IEEE-802.3] section 43.3.4. When the value of this
field is 0, it denotes the Aggregator whose key is specified in
the "Admin Key" field.
- MAC Address
Six octets encoding the port MAC address.
Martini, et al. [Page 35]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
- Admin Key
Two octets, LACP Admin key for the corresponding interface, as
specified in [IEEE-802.3] section 43.3.5.
- Port Priority
Two octets, LACP port priority for the corresponding interface,
as specified in [IEEE-802.3] section 43.3.4. This field is valid
only when the "Flags" field has "Priority Set" asserted.
- Flags
Valid values are:
-i. Synchronized (0x01)
Indicates that the sender has concluded transmitting all
member link port configurations for a given Aggregator.
Also, indicates to the receiving device that its local
port priorities will not be overridden.
-ii. Purge Configuration (0x02)
Indicates that the port is no longer configured for
mLACP operation.
-iii. Priority Set (0x04)
Indicates that the "Port Priority" field is valid.
- IF Name Len
One octet, length of the "Interface Name" field in octets.
- Interface Name
Interface name encoded in UTF-8 format, up to a maximum of 20
characters.
8.2.5. mLACP Change Port Priority TLV
The mLACP Port State TLV is used in RG Application Data message. This
TLV is used by a device to authoritatively request that a particular
member of a RG change its port priority.
Martini, et al. [Page 36]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0034 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Port Number | Port Key |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Port Priority |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0034 for "mLACP Change Port Priority TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- Port Number
2 octets field representing the LACP Port Number as specified in
[IEEE-802.3] section 43.3.4. When the value of this field is 0,
it denotes all ports bound to the Aggregator whose key is
specified in the "Port Key" field.
- Port Key
Two octets, LACP port key for the corresponding interface, as
specified in [IEEE-802.3] section 43.3.5.
- Port Priority
Two octets, LACP port priority for the corresponding interface,
as specified in [IEEE-802.3] section 43.3.4.
8.2.6. mLACP Port State TLV
The mLACP Port State TLV is used in RG Application Data message. This
TLV is used by a device to report its LACP port status to other
members in the RG.
Martini, et al. [Page 37]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| Type=0x0035 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Partner System ID |
+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| | Partner System Priority |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Partner Port Number | Partner Port Priority |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Partner Key | Partner State | Actor State |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Actor Port Number | Actor Key |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Port State |
+-+-+-+-+-+-+-+-+
- U and F Bits
Both are set to 0.
- Type
set to 0x0035 for "mLACP Port State TLV"
- Length
Length of the TLV in octets excluding the U-bit, F-bit, Type, and
Length fields.
- LACP Partner System ID
6 octets, the LACP Partner System ID for the corresponding
interface, encoded as a MAC address as specified in [IEEE-802.3]
section 43.4.2.2 item r.
- LACP Partner System Priority
2 octets field specifying the LACP Partner System Priority as
specified in [IEEE-802.3] section 43.4.2.2 item q.
- LACP Partner Port Number
2 octets encoding the LACP Partner Port Number as specified in
[IEEE-802.3] section 43.4.2.2 item u.
Martini, et al. [Page 38]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
- LACP Partner Port Priority
2 octets field encoding the LACP Partner Port Priority as
specified in [IEEE-802.3] section 43.4.2.2 item t.
- LACP Partner Oper Key
2 octets field representing the LACP Partner Key as defined in
[IEEE-802.3] section 43.4.2.2 item s.
- LACP Partner Oper State
1 octet field encoding the LACP Partner State Variable as defined
in [IEEE-802.3] section 43.4.2.2 item v.
- LACP Actor Oper State
1 octet encoding the LACP Actor's State Variable for the port as
specified in [IEEE-802.3] section 43.4.2.2 item m.
- LACP Actor Port Number
2 octets field representing the LACP Actor Port Number as
specified in [IEEE-802.3] section 43.3.4. When the value of this
field is 0, it denotes the Aggregator whose key is specified in
the "Actor Operational Key" field.
- LACP Actor Oper Key
2 octet field encoding the LACP Actor Operational Key as
specified in [IEEE-802.3] section 43.3.5.
- Port State
1 octet encoding the operational state of the port as follows:
0x00 Up
0x01 Down
0x02 Administrative Down
0x03 Test (e.g. IEEE 802.3ah OAM Intrusive Loopback mode)
Martini, et al. [Page 39]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
9. LDP Capability Negotiation
As requited in [LDP-CAP] the following TLV is defined to indicate the
ICCP capability:
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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|U|F| TLV Code Point=0x405 | Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|S| Reserved | Reserved | VER/Maj | Ver/Min |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
where:
- U-bit
SHOULD be 1 (ignore if not understood).
- F-bit
SHOULD be 0 (don't forward if not understood).
- TLV Code Point
The TLV type, which identifies a specific capability. For the
ICCP code point is requested in the IANA allocation section
below.
- S-bit The State Bit indicates whether the sender is advertising
or withdrawing the ICCP capability. The State bit is used as
follows:
1 - The TLV is advertising the capability specified by the
TLV Code Point.
0 - The TLV is withdrawing the capability specified by the
TLV Code Point.
- Ver/Maj
The major version revision of the ICCP protocol, this document
specifies 1.0. This field is then set to 1
- Ver/Min
The minor version revision of the ICCP protocol, this document
specifies 1.0. This field is then set to 0
Martini, et al. [Page 40]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
ICCP capability is advertised to a LDP peer if there is at least one
RG enabled on the local PE.
10. Client Applications
10.1. Pseudowire Redundancy Application Procedures
This section defines the procedures for the Pseudowire Redundancy
(PW-RED) Application.
10.1.1. Initial Setup
When a RG is configured on a system and multi-chassis pseudowire
redundancy is enabled in that RG, the PW-RED application should send
an "RG Connect" message with "PW-RED Connect TLV" to each PE that is
member of the same RG. When the system receives similar "RG Connect"
messages from a PE, the two devices can start exchanging "RG
Application Data" messages for the PW-RED application.
If a system receives an "RG Connect" message with "PW-RED Connect
TLV" that has a differing Protocol Version, it must follow the
procedures outlined in the "Application Versioning" section above.
When the PW-RED application is disabled on the device, or is
unconfigured for the RG in question, the system should send an "RG
Disconnect" message with "PW-RED Disconnect TLV".
10.1.2. Pseudowire Configuration
A system should advertise its local PW configuration to other PEs
that are members of the same RG. This allows the PEs to build a view
of the redundant nodes and pseudowires that are protecting the same
service instances. The advertisement should be initiated when the
PW-RED application connection first comes up, as well as upon any
subsequent PW configuration change. To that end, the system should
send "RG Application Data" messages with "PW-RED Config TLV". It is
possible to send configuration information for multiple PWs in a
single "RG Application Data" message.
The "Service Name TLV" is used on the receiving system for the
purpose of associating PW information advertised by some PE with the
corresponding AC information received over ICCP from that PE's AC
redundancy application. The Service Name has a global context in a
RG, so redundant PWs for the same service on disparate member PEs
should share the same Service Name, in order to be correlated.
Martini, et al. [Page 41]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
10.1.3. Pseudowire Status Synchronization
On a given PE, the forwarding status of the PW (Active or Standby) is
derived from the state of the associated AC(s). This simplifies the
operation of the multi-chassis redundancy solution (Figure 1) and
eliminates the possibility of deadlock conditions between the AC and
PW redundancy mechanisms. The rules by which the PW state is derived
from the AC state are as follows:
- VPWS
For VPWS, there's a single AC per service instance. If the AC is
Active, then the PW status should be Active. If the AC is
Standby, then the PW status should be Standby.
- VPLS
For VPLS, there could be multiple ACs per service instance (i.e.
VFI). If AT LEAST ONE AC is Active, then the PW status should be
Active. If ALL ACs are Standby, then the PW status should be
Standby.
The PW-RED application does not synchronize PW status across chassis,
per se. Rather, the AC Redundancy application should synchronize AC
status between chassis, in order to determine which AC (and
subsequently which PE) is Active or Standby for a given service. When
that is determined, each PE will then adjust its local PWs state
according to the rules described above.
10.1.4. PE Node Failure
When a PE node detects that a remote PE, that is member of the same
RG, has gone down, the local PE examines if it has redundant PWs for
the affected services. If the local PE has the highest priority
(after the failed PE) then it becomes the active node for the
services in question, and subsequently activates its associated PWs.
10.2. Attachment Circuit Redundancy Application Procedures
10.2.1. Common AC Procedures
This section describes generic procedures for AC Redundancy
applications, independent of the type of the AC (ATM, FR or
Ethernet).
Martini, et al. [Page 42]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
10.2.2. AC Failure
When the AC Redundancy mechanism on the Active PE detects a failure
of the AC, it should send an ICCP Application Data message to inform
the redundant PEs of the need to take over. The AC failures can be
categorized into the following scenarios:
- Failure of CE interface connecting to PE
- Failure of CE uplink to PE
- Failure of PE interface connecting to CE
10.2.3. PE Node Failure
When a PE node detects that a remote PE, that is member of the same
RG, has gone down, the local PE examines if it has redundant ACs for
the affected services. If the local PE has the highest priority
(after the failed PE) then it becomes the active node for the
services in question, and subsequently activates its associated ACs.
10.2.4. PE Isolation
When a PE node detects that is has been isolated from the core
network (i.e. all core facing interfaces/links are not operational),
then it should instruct its AC Redundancy mechanism to change the
status of any active ACs to Standby. The AC Redundancy application
should then send ICCP Application Data messages in order to trigger
failover to a standby PE.
10.2.5. ATM AC Procedures
10.2.6. Frame Relay AC Procedures
10.2.7. Ethernet AC Procedures
10.2.8. Multi-chassis LACP (mLACP) Application Procedures
This section defines the procedures that are specific to the multi-
chassis LACP (mLACP) application.
Martini, et al. [Page 43]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
10.2.8.1. Initial Setup
When a RG is configured on a system and mLACP is enabled in that RG,
the mLACP application should send an "RG Connect" message with "mLACP
Connect TLV" to each PE that is member of the same RG. When the
system receives similar "RG Connect" message from a PE, the two
devices can start exchanging "RG Application Data" messages for the
mLACP application.
If a system receives an "RG Connect" message with "mLACP Connect TLV"
that has a differing Protocol Version, it must follow the procedures
outlined in the "Application Versioning" section above.
After the mLACP application connection has been established, every PE
must communicate its system level configuration to its peers via the
use of "mLACP System Config TLV". This allows every PE to discover
the Node ID and the locally configured System ID and System Priority
values of its peers. It is necessary for all PEs in a RG to agree
upon the System ID and System Priority values to be used
ubiquitously. To achieve this, every PE MUST use the numerically
lowest value (among RG members) for each of the two parameters. This
guarantees that the PEs always agree on uniform values, which yield
the highest System Priority.
When the mLACP application is disabled on the device, or is
unconfigured for the RG in question, the system should send an "RG
Disconnect" message with "mLACP Disconnect TLV".
10.2.8.2. mLACP Port Configuration
A system must synchronize the configuration of its mLACP operating
ports with other RG members. To that end, a system must use the "Port
Config TLVs". An implementation must advertise the configuration of
Aggregators prior to advertising the configuration of any of their
associated member links. Aggregators are identified by using the Port
Number 0 (which is not a valid LACP port number) and the associated
Key. If the "Priority Set" flag is asserted in such TLV, it indicates
that the same Port Priority applies to all member links that are
attached to the Aggregator in question. When the configuration of all
ports for member links associated with a given Aggregator has been
sent by a device, it asserts that fact by setting the "Synchronized"
flag in the last port's "Port Config TLV". This also serves as a cue
for the receiving system that its local port priorities will not be
remotely overridden by the sending PE.
Furthermore, for a given port, an implementation must advertise the
port's configuration prior to advertising its state (via the "mLACP
Martini, et al. [Page 44]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
Port State TLV").
When mLACP is unconfigured on a port, a PE must send a "Port Config
TLV" with the "Purge Configuration" flag asserted. This allows
receiving PEs to purge any state maintained for the decommissioned
port.
10.2.8.3. mLACP Port Status Synchronization
PEs within a RG need to synchronize their state-machines for proper
mLACP operation with a multi-homed device. This is achieved by having
each system advertise its ports' running state in "mLACP Port State
TLVs". Whenever any port parameter, whether on the Partner (i.e.
multi-homed device) or the Actor (i.e. PE) side, is changed a system
MUST transmit an updated "mLACP Port State TLV" for the affected
port.
10.2.8.4. Triggering Failover
A PE MAY trigger a failover to a redundant PE within the RG by
sending an "mLACP Change Port Priority TLV" specifying the affected
Aggregator and a priority value that causes the remote PE to have a
higher Port Priority thereby moving to active forwarding state.
A PE MAY assume active role within the RG by sending an "mLACP Change
Port Priority TLV" to the currently active PE, specifying the
affected Aggregator and a port priority value that is less than its
local port priority for the links associated with that Aggregator.
11. Security Considerations
The security considerations described in [RFC5036] and [RFC4447] that
apply to the base LDP specification, and to the PW LDP control
protocol extensions apply to the capability mechanism described in
this document.
The ICCP protocol is not intended to be applicable when the
redundancy group spans PE in different administrative domains.
Furthermore, implementations MUST provide a mechanism to select to
which LDP peers the ICCP capability will be advertised, and from wich
LDP peers the ICCP messages will be accepted.
Martini, et al. [Page 45]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
12. IANA Considerations
12.1. MESSAGE TYPE NAME SPACE
This document uses several new LDP message types, IANA already
maintains a registry of name "MESSAGE TYPE NAME SPACE" defined by
[RFC5036]. The following values are suggested for assignment:
Message type Description
0x0700 RG Connect Message
0x0701 RG Disconnect Message
0x0702 RG Notification Message
0x0703 RG Application Data Message
12.2. TLV TYPE NAME SPACE
This document use a new LDP TLV type, IANA already maintains a
registry of name "TLV TYPE NAME SPACE" defined by [RFC5036]. The
following value is suggested for assignment:
TLV Type Description
0x700 ICCP capability TLV.
0x701 LDP TCP/IP Port TLV.
12.3. ICC RG Parameter Type Space
IANA needs to set up a registry of "ICC RG parameter type". These are
14-bit values. Parameter Type values 1 through 0x000F are specified
in this document, Parameter Type values 0x0010 through 0x1FFF are to
be assigned by IANA, using the "Expert Review" policy defined in
[RFC5226]. Parameter Type values 0x2000 through 0x2FFF, 0x3FFF, and 0
are to be allocated using the IETF consensus policy defined in
[RFC5226]. Parameter Type values 0x3000 through 0x3FFE are reserved
for vendor proprietary extensions and are to be assigned by IANA,
using the "First Come First Served" policy defined in [RFC5226]. A
Parameter Type description is required for any assignment from this
registry. Additionally, for the vendor proprietary extensions range a
citation of a person or company name is also required. A document
reference should also be provided.
Initial ICC RG parameter type space value allocations are specified
below:
Parameter Type Description Reference
-------------- --------------------------------- ---------
0x0001 ICC sender name [RFCxxxx]
0x0002 NAK TLV [RFCxxxx]
Martini, et al. [Page 46]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
0x0003 Requested Protocol Version TLV [RFCxxxx]
0x0004 Disconnect Code TLV [RFCxxxx]
0x0005 ICC RG ID TLV [RFCxxxx]
0x0010 PW-RED Connect TLV [RFCxxxx]
0x0011 PW-RED Disconnect TLV [RFCxxxx]
0x0012 PW-RED Config TLV [RFCxxxx]
0x0013 Service Name TLV [RFCxxxx]
0x0014 PW ID TLV [RFCxxxx]
0x0015 Generalized PW ID TLV [RFCxxxx]
0x0030 mLACP Connect TLV [RFCxxxx]
0x0031 mLACP Disconnect TLV [RFCxxxx]
0x0032 mLACP System Config TLV [RFCxxxx]
0x0033 mLACP Port Config TLV [RFCxxxx]
0x0034 mLACP Change Port Priority TLV [RFCxxxx]
0x0035 mLACP Port State TLV [RFCxxxx]
12.4. STATUS CODE NAME SPACE
This document use several new Status codes, IANA already maintains a
registry of name "STATUS CODE NAME SPACE" defined by [RFC5036]. The
following values is suggested for assignment: The "E" column is the
required setting of the Status Code E-bit.
Range/Value E Description Reference
------------- ----- ---------------------- ---------
0x00010001 0 Unknown ICCP RG
0x00010002 0 ICCP Connection Count Exceeded
0x00010003 0 ICCP Application Connection
Count Exceeded
0x00010004 0 ICCP Application not in RG
0x00010005 0 Incompatible ICCP Protocol Version
0x00010006 0 ICCP Rejected Message
0x00010007 0 ICCP Administratively Disabled
0x00010010 0 ICCP RG Removed
0x00010011 0 ICCP Application Removed from RG
Martini, et al. [Page 47]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
13. Normative References
[RFC5036] L. Andersson et al, "LDP Specification", RFC 5036,
October 2007.
[LDP-CAP] "LDP Capabilities", draft-ietf-mpls-ldp-capabilities-02.txt
April 2008, (Work in Progress)
[RFC4447] "Transport of Layer 2 Frames Over MPLS", Martini, L.,
et al., rfc4447 April 2006.
[IEEE-802.3] IEEE Std. 802.3-2005, "Part 3: Carrier Sense Multiple
Access with Collision Detection (CSMA/CD) Access Method and
Physical Layer Specifications", IEEE Computer Society, December
2005.
14. Informative References
[BFD] D. Katz, D. Ward, "Bidirectional Forwarding Detection",
draft-ietf-bfd-base-09.txt, February 2009 (Work in Progress)
[RFC5226] Narten, T. and H. Alvestrand, "Guidelines for Writing an
IANA Considerations section in RFCs", BCP 26, RFC 5226, May 2008
15. Author's Addresses
Luca Martini
Cisco Systems, Inc.
9155 East Nichols Avenue, Suite 400
Englewood, CO, 80112
e-mail: lmartini@cisco.com
Samer Salam
Cisco Systems, Inc.
595 Burrard Street, Suite 2123
Vancouver, BC V7X 1J1
Canada
e-mail: ssalam@cisco.com
Martini, et al. [Page 48]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
Ali Sajassi
Cisco Systems, Inc.
170 West Tasman Drive
San Jose, CA 95134
e-mail: sajassi@cisco.com
Satoru Matsushima
Softbank Telecom
1-9-1, Higashi-Shinbashi, Minato-ku
Tokyo 105-7313, JAPAN
e-mail: satoru.matsushima@tm.softbank.co.jp
Thomas D. Nadeau
BT
BT Centre
81 Newgate Street
London, EC1A 7AJ
United Kingdom
e-mail: tom.nadeau@bt.com
Full Copyright Statement
Copyright (c) 2009 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 in effect on the date of
publication of this document (http://trustee.ietf.org/licenseinfo).
Please review these documents carefully, as they describe your rights
and restrictions with respect to this document.
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.
Martini, et al. [Page 49]
Internet Draft draft-martini-pwe3-iccp-01.txt February 17, 2009
Acknowledgments
The authors wish to acknowledge the contributions of Satoru
Matsushima, Wei Luo, Neil Mcgill, Skip Booth, Neil Hart, Michael Hua,
and Tiberiu Grigoriu.
Expiration Date: August 2009
Martini, et al. [Page 50]