Timos 5.0 MC Lag PW Red

TiMOS-5.
0 workshop
MC-LAG and PW Redundancy
[email protected]
April, 2007
All Rights Reserved © Alcatel-Lucent 2006, #####

Agenda
1. 5.0 Redundancy features overview
2. LAG & LACP Basics
3. MC-LAG
4. Dual homing in TPSDA – L2-CO model
5. Pseudo-Wire redundancy
6. PW Switching
7. Convergence time
2 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

1
5.0 Redundancy features overview

Access Redundancy 4.0 Capabilities Overview
Business services
 Link redundancy for access to single PE
 VLL service  LAG
 VPLS service  LAG or STP
 L3 services  LAG
 Dual homing to two PEs
 VLL service  not supported
 VPLS service  STP
 L3 services  VRRP
TPSDA
 Link redundancy to single node
 L2-CO model  LAG
 Dual homing to two nodes
 L2-CO model  not supported
 L3-CO model  not supported

Access Redundancy 5.0
Business services
 Link redundancy for access to single PE
 VLL service  LAG
 VPLS service  LAG or STP
 L3 services  LAG
 Dual homing to two PEs
 VLL service  Multi-chassis LAG (MC-LAG) + active/standby PW
 VPLS service  STP or Multi-chassis LAG (MC-LAG)
 L3 services  VRRP
TPSDA
 Link redundancy to single node
 Dual homing to two nodes
 L2-CO model  Multi-chassis LAG (MC-LAG) + MCS
 L3-CO model  Subscriber-routed-redundancy-protocol (SRRP)

Access Redundancy 5.0 VLL Service: MC-LAG + active/standby PW
Redundant
PWs
MC-LAG MC-LAG
near-end- SAP near-end-
SAP
redundancy redundancy
Access
Node Access
Node
Optional
ICB Spoke-SDP
7x50 7x50
EPIPE is point-to-point service

 Building e2e redundancy requires 2 independent e2e paths
Main building blocks:
 MC-LAG
 Provides access redundancy
 Active/standby PW
 Provides point-to-point (but redundant) path between set of PEs
 Inter-Chassis Back-up (ICB) spoke-sdp
 Provides faster recovery for “in-flight” packets

Access Redundancy 5.0 VPLS Service: MC-LAG
Active
Standby
Standby Active
Access Mesh
Node
PWs
Access
VPLS is point-to-multipoint service Node
 BSA link and node protection  MC-LAG

 VPLS path protection  MPLS with FRR
 Provides access redundancy
 No need for Active/standby PW
 No Inter-Chassis Back-up (ICB) spoke-SDPs

Access Redundancy 5.0 TPSDA L2-CO Model: MC-LAG + MCS
BSR
BSA
Spoke-sdp
DSLAM
Mesh-sdp
SAP VRRP
or VPLS
Spoke-sdp
LAG
Spoke-sdp
BSA
BSR

 No load balancing across different BSA nodes
 BSR protection  VRRP
 Multi-Chassis Synchronization (MCS) performs synchronization between redundant-pair
chassis
 IGMP-snooping, DHCP-Lease and ESM states are synced

Access Redundancy 5.0 TPSDA L3-CO Model: SRRP
BSR
SAP-A
DSLAM
Redundant-
SRRP Interface
SAP-B
 SRRP is Alcatel proprietary extension to VRRP BSR
 main goal is to provide redundancy for subscriber-interfaces

 SRRP operates on per group-interface basis
 load balancing across different BSR nodes is possible
 Detect link and BSR node failures and provide fast switch-over
 Keep-alive intervals configurable in the range of 100ms - 10s (100ms increments)
 Required (e.g. QoS enforcement) that all downstream traffic is passing through active BSR
 Redundant-interface (PW based) is used to shortcut the downstream traffic
 Multi-Chassis Synchronization (MCS) performs synchronization between redundant-pair
chassis
 IGMP, DHCP-Lease, BSM and ESM states are synced

Access Redundancy 5.0 TPSDA L3-CO Model: SRRP
configure service ies subscriber-interface miro

BSR address 100.100.0.0/16 gw-ip-address 100.100.100.100
group-interface “dslam-1”
SAP-A arp-populate
DSLAM
sap-a create
srrp 10
message-path sap-a
SRRP
keep-alive-interval 1
gtw-mac aa:aa:aa:aa:aa:aa
redundant-interface “back-up”
spoke-sdp 2:20
SAP-B address 20.20.20.20/24 remote-ip 30.30.30.30
BSR
No MC-LAG involved!!!
Following configuration steps are required:
 Create subscriber-interface and group interfaces on both nodes
 Parameters such as gw-address and gw-mac needs to be consistent
 Configure SRRP instances on all group-interfaces
 Create redundant-interface
 L3 interface based on spoke-sdp
 Configure MCS peering and enable sub-mgmt and srrp synchronization
 Configure subscriber management features
 Dhcp-snooping, arp-populate, and etc….

Access Redundancy TPSDA Remarks
At this moment (pre-5.0R4):

 ies/vprn interfaces are compatible with single chassis LAG
 vll/vpls services are compatible with both single chassis and multi-chassis LAG
TPSDA dual homing to 2 nodes:

 Dual homing only supported for MC-LAG / ethernet
 TPSDA dual homing to 2 nodes not supported for STP
 Ring of DSLAMs not supported (planned for 6.0)
TPSDA dual homing to 1 node:

 Dual homing only supported for LAG / ethernet
 TPSDA not supported for dual homing with STP
ESM not supported on ATM interfaces!

2
LAG Basics

LAG Basics
Link Aggregation Group:
 Aggregates up to 8 Ethernet ports in 1 LAG

 LAG forms 1 logical interface
 Used for:
 BW increase
 redundancy
 Done in firmware -> line rate
 Static or signalled via LACP
 Auto-negotiation must be disabled or set limited mode for ports that are part of a
LAG to guarantee a specific port speed. Full-Duplex required.
 Packet sequencing maintained -> hashing algorithm
 See other docs

LAG Basics
notes
1/1/2 1/1/2
1/1/1 1/1/1
port 1/1/1
ethernet
mode access
port 1/1/1
no autonegotiate
ethernet
exit
no autonegotiate
no shutdown
exit
exit
no shutdown
port 1/1/2
exit
ethernet
port 1/1/2
mode access
ethernet
no autonegotiate
no autonegotiate
exit
exit
no shutdown
no shutdown
exit
exit
-----------------------------
-----------------------------
lag 2
lag 2
mode access
port 1/1/1
port 1/1/1
port 1/1/2
port 1/1/2
no shutdown
no shutdown
exit
exit
-----------------------------
-----------------------------
router
vpls 5 customer 1 create
interface "toPE1"
sap lag-2 create
address 10.1.1.1/24
exit
port lag-2
no shutdown
exit
exit

LAG Basics
notes
*A:PE2# show lag 2 detail
===============================================================================
LAG Details
===============================================================================
Description:
-------------------------------------------------------------------------------
Details
-------------------------------------------------------------------------------
Lag-id : 2 Mode : access
Adm : up Opr : up
Thres. Exceeded Cnt : 2 Port Threshold : 0
Thres. Last Cleared : 01/26/2007 23:35:18 Threshold Action : down
Dynamic Cost : false Encap Type : null
Configured Address : 1e:2f:ff:00:01:42 Lag-IfIndex : 1342177282
Hardware Address : 1e:2f:ff:00:01:42 Adapt Qos : distribute
Hold-time Down : 0.0 sec
LACP : disabled
-------------------------------------------------------------------------------
Port-id Adm Act/Stdby Opr Primary Sub-group Forced Prio
-------------------------------------------------------------------------------
1/1/1 up active up yes 1 - 32768
1/1/2 up active up 1 - 32768
===============================================================================
*A:PE1#

LAG CLI Tree
configure [no] lag lag-id

access [no] adopt-qos [link |distribute]
[no] description description-string
[no] dynamic-cost
[no] encap-type {dot1Q | null | qinq}
[no] hold-time down hold-down-time
[no] mac ieee-address
[no] mode {access | network}
[no] port port-id [port-id …up to 8 total] [priority priority] [sub-group sub-group-id]
[no] port-threshold value [action {dynamic-cost | down}]
[no] shutdown
exit

LAG Basics: port-threshold and dynamic-cost
*A:PE1>config>lag# info
LAG example with 3 ports configured: ----------------------------------------------
port 1/1/1
port 1/1/2
 No port-threshold, no dynamic-cost: port 1/1/3
 3 ports up: OSPF cost 333 dynamic-cost
 2 ports up: OSPF cost 333 port-threshold 1 action down
no shutdown
 1 port up: OSPF cost 333
----------------------------------------------
 No port-threshold, dynamic-cost
 3 ports up: OSPF cost 333
 Port-threshold 1 action down, no dynamic-cost  Port-threshold 1 action down, dynamic-cost:

 3 ports up: OSPF cost 333  3 ports up: OSPF cost 333
 2 ports up: OSPF cost 333  2 ports up: OSPF cost 500
 1 port up: LAG down
 1 port up: LAG down
 Port-threshold 1 action dynamic-cost, no dynamic-cost
 Port-threshold 1 action dynamic-cost, dynamic-cost

 Same as “No port-threshold, dynamic-cost”

LACP basics
notes
Link Aggregation Control Protocol:
 Based on IEEE 802.3ad

 Free download:
 http://standards.ieee.org/getieee802/download/802.3-2005_section3.pdf, clause 43
 LACP messages sent over each link every 1s (fast) or 30s (slow)
 Uses Slow Protocols destination MAC (01:80:C2:00:00:02)
 3 parameters uniquely identify a LAG instance:
 Lacp-key (default 32768 for first LAG, step up by 1 for next LAGs)
 System-id (derived from base MAC address; show chassis)
 System-priority (default 32768; configure system lacp-system-priority x)
 Actor (self) and Partner (other side of the link) info in every LACP packet
 Marker Protocol optional, not supported
 Churn Detection optional, not supported
 Max 200 LAGs on SR/ESS 7/12, Max 64 LAGs on SR/ESS1 and SRc

LAG Basics: Architectural positioning of Link Aggregation sublayer
LACP
Block diagram

LAG Basics: LACP
notes
1/1/2 1/1/2
1/1/1 1/1/1
*A:PE2>config>lag# info detail

----------------------------------------------
no mac
mode access
... encap-type null
lag 2 port 1/1/1 priority 32768 sub-group 1
port 1/1/1 port 1/1/2 priority 32768 sub-group 1
port 1/1/2 no dynamic-cost
lacp active administrative-key 32768 lacp passive administrative-key 32768
no shutdown port-threshold 0 action down
exit lacp-xmit-interval fast
... lacp-xmit-stdby
no selection-criteria
no hold-time
access
adapt-qos distribute
exit
no shutdown

LACP Basics
*A:PE2# show lag 2 detail notes
===============================================================================
LAG Details
===============================================================================
Description:
-------------------------------------------------------------------------------
Details
-------------------------------------------------------------------------------
Adm : up Opr : up
LACP : enabled Mode : passive
LACP Transmit Intvl : fast LACP xmit stdby : enabled
Selection Criteria : highest-count Slave-to-partner : disabled lacpdu
Number of sub-groups: 1 Forced : -
System Id : 1e:2f:ff:00:00:00 System Priority : 32768 structure
Admin Key : 32768 Oper Key : 32768
Prtr System Id : 1e:2d:ff:00:00:00 Prtr System Priority : 32768
Prtr Oper Key : 32768
------------------------------------------------------------------------------- ethereal
------------------------------------------------------------------------------- capture
1/1/2 up active up 1 - 32768
-------------------------------------------------------------------------------
Port-id Role Exp Def Dist Col Syn Aggr Timeout Activity
------------------------------------------------------------------------------- Actor/Partner
1/1/1 actor No No Yes Yes Yes Yes Yes No Bit Encoding
1/1/1 partner No No Yes Yes Yes Yes Yes Yes
1/1/2 actor No No Yes Yes Yes Yes Yes No
===============================================================================

LACP LAG: CLI Tree
configure [no] lag lag-id

[no] lacp [mode] [administrative-key admin-key]
[no] lacp-xmit-interval {slow | fast}
[no] lacp-xmit-stdby
[no] selection-criteria [highest-count | highest-weight] [slave-to-partner]
exit

LACP Basics: sub-groups and selection-criteria
Sub-groups: notes
 Maximum 8 sub-groups in 1 LAG, only 1 group active when LACP is configured
 By default LAG ports belong to sub-group 1
 Sub-groups should only be configured on 1 side of a LAG for consistent selection
 Selection-criteria decide which sub-group is active
 ‘auto-mda’/’auto-iom’: all ports of the same IOM/MDA are assigned to the same subgroup
*A:PE1# configure lag 1 port 1/1/1 sub-group
- no port <port-id> [<port-id>... (up to 8 total)]
- port <port-id> [<port-id>... (up to 8 total)] [priority <priority>]
[sub-group <sub-group-id>]
<port-id> : [slot/mda/port]
<priority> : [1..65535]
<sub-group-id> : [1..8]|auto-iom|auto-mda
Selection-criteria:
 ‘highest-count’ (default): highest number of eligible members. If same # members -> highest weight
 ‘highest-weight’: highest weight (65535 – priority)
 Optional parameter: ‘slave-to-partner’: considers configured selection-criteria + members should be
operationally up for use by the remote side, i.e. partner sync bit
 Each time configuration changes / status of any link in a lag is changed, selection algorithm is re-run.
*A:PE1# configure lag 1 selection-criteria
- selection-criteria [{highest-count|highest-weight}] [slave-to-partner]
- no selection-criteria
<highest-count|hig*> : keywords
<slave-to-partner> : keyword

LACP: Tools dump lag lag-id
Port 1/1/1 LACP info Port 1/1/2 LACP info
RX sm state......CURRENT RX sm state......CURRENT
PT sm state......FAST_PERIODIC PT sm state......FAST_PERIODIC
MUX sm state.....COLLECTING_DISTRIBUTING MUX sm state.....COLLECTING_DISTRIBUTING
begin............FALSE begin............FALSE
NTT..............FALSE NTT..............FALSE
lacpEnbld........TRUE lacpEnbld........TRUE
ready_N..........FALSE ready_N..........FALSE
selected.........SELECTED selected.........SELECTED
portMoved........FALSE portMoved........FALSE
portEnabled......TRUE portEnabled......TRUE
txTimestamp 0....17408 txTimestamp 0....17408
*A:CE5# tools dump lag lag-id 1 txTimestamp 1....17508 txTimestamp 1....17508
txTimestamp 2....17608 txTimestamp 2....17608
Port state : Up nextTxTimestamp..0 nextTxTimestamp..0
Selected subgrp : 1 pLacpPdu.........0x0 pLacpPdu.........0x0
NumActivePorts : 2 subGrpId.........1 subGrpId.........1
ThresholdRising : 6 isCollDistr......TRUE isCollDistr......TRUE
ThresholdFalling: 3
IOM bitmask : 2 currentWhileTimerExpired..FALSE currentWhileTimerExpired..FALSE
Config MTU : 1514 periodicTimerExpired......FALSE periodicTimerExpired......FALSE
Oper. MTU : 1514
Bandwidth : 200000 PrtrOprSysPri : 32768 PrtrOprSysPri : 32768
PrtrOprSysId : 1e:2f:ff:00:00:00 PrtrOprSysId : 1e:2f:ff:00:00:00
multi-chassis : NO PrtrOprkey : 32768 PrtrOprkey : 32768
------------------------------------------------------------------------------------
Indx PortId RX pkts TX pkts State Active Port Cfg Oper Speed BW AP CS
Pri Mtu Mtu
------------------------------------------------------------------------------------
0 1/1/1 469 751 Up yes 32768 1514 1514 100 100000 0 2
1 1/1/2 469 749 Up yes 32768 1514 1514 100 100000 0 2

LACP: Debug commands
*A:CE5# debug lag

- lag [lag-id <lag-id> [port <port-id>]] [all]
- lag [lag-id <lag-id> [port <port-id>]] [sm] [pkt] [cfg] [red] [iom-upd]
[port-state] [timers] [sel-logic] [mc] [mc-pkt]
- no lag [lag-id <lag-id>]
<lag-id> : [1..200]
<all> : keyword
<sm> : keyword
<pkt> : keyword
<cfg> : keyword
sm — Specifies to display trace LACP state machine.
<red> : keyword
<iom-upd> : keyword pkt — Specifies to display trace LACP packets.
<port-state> : keyword cfg — Specifies to display trace LAB configuration.
<timers> : keyword
<sel-logic> : keyword
red — Specifies to display trace LAG high availability.
<port-id> : slot/mda/port iom-upd — Specifies to display trace LAG IOM updates.
<mc> : keyword port-state — Specifies to display trace LAG port state transitions.
<mc-pkt> : keyword
timers — Specifies to display trace LAG timers.
sel-logic — Specifies to display trace LACP selection logic.
mc — Specifies to display multi-chassis parameters.
mc-packet — Specifies to display the MC-LAG control packets with valid
authentication were received on this system.

3
MC-LAG

MC-LAG
 Provides access link and PE node level redundancy to the access node
 Redundancy to max 2 PE routers
 Redundant PE routers appear as 1 and same LACP partner -> can interop with any
access node that supports LACP
 Configuration and states signaled between redundant PE routers via proprietary
protocol. Protocol is routed and uses UDP port 1025
 PE router can have max 4 MC peers (R5.3, 2 MC peers before R5.3)
 Max 200 LAGs on SR/ESS 7/12, Max 64 LAGs on SR/ESS1 and SRc
 LAG on access + Ethernet MDA only
 SRc: R5.3
PE1
LAG
MC-LAG
Control
CE1 connection
PE2

MC-LAG configuration: Access node
1/1/1
MC-LAG
 Access node is unaware about MC-LAG
 Configuration same as normal LACP LAG 1/1/2
1/1/1
1/1/2
CE1
1/1/3
A:CE1>config>lag# info 1/1/4

----------------------------------------------
port 1/1/1
port 1/1/2 1/1/1
port 1/1/3
port 1/1/4
lacp active administrative-key 32768
no shutdown 1/1/2
----------------------------------------------

MC-LAG configuration: PE nodes LAG configuration
A:PE1>config>lag# info
----------------------------------------------
PE1 port 1/1/1
port 1/1/2
1/1/1 no shutdown
MC-LAG ----------------------------------------------
*A:PE1>config>redundancy>multi-chassis# info
----------------------------------------------
peer 10.0.0.2 create
1/1/2 mc-lag
1/1/1 lag 1 lacp-key 1 system-id 00:00:00:00:00:01 system-priority 100
no shutdown
1/1/2 exit
exit
----------------------------------------------
1/1/3 *A:PE2>config>lag# info

----------------------------------------------
1/1/4 port 1/1/1
port 1/1/2
1/1/1 no shutdown
----------------------------------------------
*A:PE2>config>redundancy>multi-chassis# info
----------------------------------------------
1/1/2 peer 10.0.0.1 create
mc-lag
lag 1 lacp-key 1 system-id 00:00:00:00:00:01 system-priority 100
no shutdown
PE2 exit
exit
----------------------------------------------

MC-LAG detailed info on PE1
*A:PE1>config>redundancy>multi-chassis# show lag 1 detail
... notes
Adm : up Opr : up
LACP : enabled Mode : active
Selection Criteria : highest-count Slave-to-partner : disabled
System Id : 1e:2f:ff:00:00:00 System Priority : 32768
Prtr System Id : 1e:2e:ff:00:00:00 Prtr System Priority : 32768
MC Peer Address : 10.0.0.2 MC Peer Lag-id : 1

MC System Id : 00:00:00:00:00:01 MC System Priority : 100
MC Admin Key : 1 MC Active/Standby : active
MC Lacp ID in use : true MC extended timeout : false
MC Selection Logic : peer decided
MC Config Mismatch : no mismatch
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
1/1/2 up active up 1 - 32768
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
1/1/1 actor No No Yes Yes Yes Yes Yes Yes
===============================================================================

MC-LAG detailed info on PE2
...
Adm : up Opr : down
Configured Address : 1e:2d:ff:00:01:41 Lag-IfIndex : 1342177281
Hardware Address : 1e:2d:ff:00:01:41 Adapt Qos : distribute
System Id : 1e:2d:ff:00:00:00 System Priority : 32768
Prtr System Id : 1e:2e:ff:00:00:00 Prtr System Priority : 32768

MC Admin Key : 1 MC Active/Standby : standby
MC Selection Logic : local master decided
MC Config Mismatch : no mismatch
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
1/1/1 up standby down yes 1 - 32768
1/1/2 up standby down 1 - 32768
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
1/1/1 actor No No No No No Yes Yes Yes
1/1/1 partner No No No No Yes Yes Yes Yes
1/1/2 actor No No No No No Yes Yes Yes
1/1/2 partner No No No No Yes Yes Yes Yes
===============================================================================

MC-LAG detailed info on CE1
*A:CE5# show lag 1 detail
...
Lag-id : 1 Mode : network
Adm : up Opr : up
Configured Address : 1e:2e:ff:00:01:41 Lag-IfIndex : 1342177281
Hardware Address : 1e:2e:ff:00:01:41
System Id : 1e:2e:ff:00:00:00 System Priority : 32768
Prtr System Id : 00:00:00:00:00:01 Prtr System Priority : 100
Prtr Oper Key : 1
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
1/1/2 up active up 1 - 32768
1/1/3 up active down 1 - 32768
1/1/4 up active down 1 - 32768
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
1/1/3 actor No No No No Yes Yes Yes Yes
1/1/3 partner No No No No No Yes Yes Yes
1/1/4 actor No No No No Yes Yes Yes Yes
1/1/4 partner No No No No No Yes Yes Yes
===============================================================================

MC-LAG: inter-chassis LAG control protocol
 Should run over protected path to avoid single point of failure !!!!!!
 Verifies configuration on both chassis (lacp-key, system-id and system-priority,
selection criteria must be the same)
 Signals active/standby selection - all active links must be on the same chassis
 Signals link failure and detects node failures
 This communication channel needs to fulfill following requirements:
 Support of connection at IP level which does not require direct link between two nodes. The IP
address configured at the neighbor system must be one of the addresses of the system
(interface or loop-back IP address)
 heartbeat mechanism to enhance robustness of the MC-LAG operation and detecting node
failures. In addition to this "keep-alive" mechanism MC-LAG SW should use all available triggers
to detect remote-node failure.
 Support for operator actions on any node that either forces an operational change in operation
(tools perform lag force/clear-force).
 LAG group-ids do not need to match between neighbour systems. At the same time, there can
be multiple LAG groups between the same pair of neighbours.
 Authentication

MC-LAG: configuration check

...
MC Admin Key : 1 MC Active/Standby : standby
MC Selection Logic : config mismatch, local subgroups put to standby
MC Config Mismatch : LAG selection criteria mismatch with peer
-------------------------------------------------------------------------------
-------------------------------------------------------------------------------
1/1/1 up standby down yes 1 - 32768
...
 Other checks:
 ‘MC lacp-key mismatch with peer’
 ‘MC system-priority mismatch with peer’
 ‘MC system-id mismatch with peer’

MC-LAG: MC-LAG Redundancy CLI tree
notes
configure redundancy multi-chassis [no] peer ip-address

[no] authentication-key [authentication-key | hash-key] [hash | hash2]
[no] description description-string
[no] mc-lag
[no] hold-on-neighbor-failure multiplier
[no] keep-alive-interval interval
[no] lag lag-id lacp-key admin-key system-id system-id [remotelag lag-id]
system-priority system-priority
[no] shutdown
source-address ip-address
exit

MC-LAG: extra features
 Authentication
*A:PE2# configure redundancy multi-chassis peer 10.0.0.1 authentication-key
- authentication-key <authentication-key|hash-key> [hash|hash2]
- no authentication-key
<authentication-ke*> : authentication-key - [20 chars max]

hash1-key - [33 chars max]
hash2-key - [55 chars max]
<hash|hash2> : keywords - specify hashing scheme
 Different LAG-id
*A:PE2# configure redundancy multi-chassis peer 10.0.0.1 mc-lag lag 1 remote-lag 2

system-id 00:00:00:00:00:01 system-priority 100 lacp-key 1
 Keep-alive-interval (5-500 deci-seconds) and hold-on-neighbor (2-25)

*A:PE1>config>redundancy>mc>peer>mc-lag# info detail
----------------------------------------------
hold-on-neighbor-failure 3
keep-alive-interval 10
lag 1 lacp-key 1 system-id 00:00:00:00:00:01 remote-lag 1 system-priority 100
no shutdown
----------------------------------------------

MC-LAG: show commands
 Overview of MC-peers notes
A:PE2# show redundancy multi-chassis all
===============================================================================
Multi-Chassis Peers
===============================================================================
Peer IP Src IP Auth Peer Admin
MCS Admin MCS Oper MCS State MC-LAG Admin MC-LAG Oper
-------------------------------------------------------------------------------
10.0.0.1 10.0.0.2 hash2 Enabled
-- -- -- Enabled Enabled
===============================================================================
 Detailed view of peer (more info shown when MCS is used/see later)
A:PE1# show redundancy multi-chassis sync peer 10.0.0.2 detail
===============================================================================
Multi-chassis Peer Table
===============================================================================
Peer
-------------------------------------------------------------------------------
Peer IP Address : 10.0.0.2
Authentication : Enabled
Source IP Address : 10.0.0.1
Admin State : Enabled
===============================================================================
Ports synced on peer 10.0.0.2
===============================================================================
Port/Encap Tag
-------------------------------------------------------------------------------
===============================================================================
===============================================================================

 MC-LAG setup
A:PE1# show redundancy multi-chassis mc-lag peer 10.0.0.2
===============================================================================
Multi-Chassis MC-Lag Peer 10.0.0.2
===============================================================================
Last Changed : 03/18/2007 23:04:25
Admin State : Up Oper State : Up
KeepAlive : 10 deci-seconds Hold On Ngbr Failure : 3
-------------------------------------------------------------------------------
Lag Id Lacp Key Remote Lag Id System Id Sys Prio Last Changed
-------------------------------------------------------------------------------
1 1 1 00:00:00:00:00:01 100 03/18/2007 23:04:25
-------------------------------------------------------------------------------
Number of LAGs : 1
===============================================================================

 Statistics
A:PE1# show redundancy multi-chassis mc-lag peer 10.0.0.2 statistics
===============================================================================
Multi-Chassis Statistics, Peer 10.0.0.2
===============================================================================
Packets Rx : 61793
Packets Rx Keepalive : 61779
Packets Rx Config : 2
Packets Rx Peer Config : 1
Packets Rx State : 11
Packets Dropped State Disabled : 0
Packets Dropped Packets Too Short : 0
Packets Dropped Tlv Invalid Size : 0
Packets Dropped Tlv Invalid LagId : 0
Packets Dropped Out of Seq : 0
Packets Dropped Unknown Tlv : 0
Packets Dropped MD5 : 0
Packets Tx : 61865
Packets Tx Keepalive : 61778
Packets Tx Peer Config : 72
Packets Tx Failed : 0
===============================================================================
 Statistics per LAG

A:PE1# show redundancy multi-chassis mc-lag peer 10.0.0.2 lag 1 statistics
===============================================================================
Multi-Chassis Statistics, Peer 10.0.0.2 Lag 1
===============================================================================
Packets Rx Config : 2
Packets Rx State : 11
Packets Tx Config : 1
Packets Tx State : 14
Packets Tx Failed : 0
===============================================================================

MC-LAG: tools
notes
 Force
*A:PE1# tools perform lag force
- force all-mc {active|standby}
- force lag-id <lag-id> [sub-group <sub-group-id>] {active|standby}
- force peer-mc <peer-ip-address> {active|standby}
<lag-id> : [1..200]
<sub-group-id> : [1..16]
<all-mc> : keyword
<peer-ip-address> : a.b.c.d
<active|standby> : keywords
 Clear-force
*A:PE1# tools perform lag clear-force

- clear-force all-mc
- clear-force lag-id <lag-id> [sub-group <sub-group-id>]
- clear-force peer-mc <ip-address>
<lag-id> : [1..200]
<sub-group-id> : [1..16]
<all-mc> : keyword
<ip-address> : a.b.c.d

LAB Exercise 1: MC-LAG configuration
PE1 PE3
SIM IP Addresses 10.0.0.1 10.1.3/24 10.0.0.3
CE1 (138.203.15.80) 1/1/4 1/1/3 1/1/1
CE2 (138.203.15.45) 1/1/1
CE3 (138.203.15.43) 1/1/1
CE4 (138.203.15.44) 1/1/2 1/1/1
1/1/2
1/1/3
PE1 (138.203.15.81) 1/1/2
10.1.2/24 10.3.4/24
PE2 (138.203.15.76) 1/1/3
1/1/3
PE3 (138.203.15.77) 1/1/2
CE1 1/1/1 1/1/1 1/1/2 CE2
PE4 (138.203.15.78) 1/1/4
1/1/4 1/1/3
1/1/2
10.2.4/24
 Team and : 10.0.0.2 10.0.0.4
 Lag-id 1 PE2 PE4
 LACP-key 1
 System-id 00:00:00:00:00:01 PE3 PE4
10.3.4/24 10.0.0.4
 System-priority 100 10.0.0.3
1/1/2 1/1/2 1/1/4
 Team and : 1/1/4
1/1/4
 Lag-id 2 1/1/4
1/1/3 1/1/3
 LACP-key 2 10.1.3/24 10.3.4/24
 System-id 00:00:00:00:00:02 1/1/4 1/1/4
CE3 1/1/5 1/1/3 CE4
 System-priority 100 1/1/3
1/1/3 1/1/3
1/1/5
 Authentication-key: Alcatel 10.1.2/24
10.0.0.1 10.0.0.2
 Use show commands and force tools PE1 PE2

4
LAG redundancy in VPLS
Dual homing in TPSDA – L2-CO model

LAG Redundancy in non-TPSDA VPLS:
2
Sends LDP address withdraw message to all peers
(so called – “forget me” MAC-FLUSH)
Active
Standby
Standby
Active
Active
Access
Node
1
MC-LAG performs switch over
Access
3 Node
Relearning process starts again
 Works with both Mesh and Spoke-SDPs

LAG Redundancy in non-TPSDA VPLS: mac-flush message
 Sent on all SDPs configured in the VPLS

Dual homing in TPSDA: setup
BSR
BSA
Spoke-sdp
DSLAM
SAP Mesh/Spoke VRRP

-sdp VPLS
LAG
Spoke-sdp
BSA
BSR
Levels of redundancy
 BSR protection  VRRP

Dual homing in TPSDA: Multi-Chassis Synchronization (MCS)
notes
MCS Highlights
•TCP connection set-up between IP@ on two nodes

DSLAM -- Single chassis can peer with 4 other peers
•It supports dual-homing only
-- every state can be synced only with 1 peer
•Add/delete operation on database is synchronized automatically
-- active chassis always performs “Global delete”
-- standby chassis always performs “Local delete”
• tolerant to asymmetrical configuration on two chassis
-- allows different migration scenarios (upgrade to ESM)
• routed
MCS maintains a distributed database of entries entered by applications

 Subscriber-management
 Dynamic DHCP host table including ESM attributes
 IGMP at L3 interfaces or IGMP-snooping at VPLS sap level
 Synchronizing IGMP FIB allows almost hitless switch-over of BTV service
 SRRP

Dual homing in TPSDA: configuration
configure>redundancy>multi-chassis
source-address 10.10.10.10
sync
sub-mgmt
BSA
igmp-snooping
DSLAM port lag-2 create
range 0-4095 “dslam-1”
configure>redundancy>multi-chassis
source-address 20.20.20.20
sync
BSA sub-mgmt
igmp-snooping
Following configuration steps required: port lag-1 create
range 0-4095 “dslam-1”
 Create peer
 Defines destination and source-address
 Optionally: enable authentication by defining authentication-key
 Enable individual client applications (e.g. sub-mgmt, igmp-snooping)
 Define pairing of ports and VLANs based on sync-tag
 Port-id on different chassis does not need to be the same
 VLAN-id on different chassis for the same sync-tag is the same
 Configure NTP (time should be same)

Dual homing in TPSDA: setup
 Separate VPLS service for Unicast traffic and Mcast Traffic

 see RED-7 Configuration Notes
 If ring of BSAs are used:
 BTV: spoke-SDPs between all BSAs
 Unicast: spoke or mesh between redundant BSAs, spoke-SDP to BSR, service/redundant pair
 Should also be possible with spoke-SDPs between all BSAs (not tested)

Dual homing in TPSDA: MCS CLI tree
configure redundancy multi-chassis [no] peer ip-address

[no] sync
[no] igmp
[no] igmp-snooping
[no] port [port-id | lag-id] [sync-tag sync-tag]
[no] range encap-range [sync-tag sync-tag]
[no] shutdown
[no] srrp
[no] sub-mgmt
 SRRP in 5.0R4
 IGMP on IES/VPRN interface
 Don’t forget the sync-tag !!!

Dual homing in TPSDA: show commands
A:PE2# show redundancy multi-chassis all
===============================================================================
Multi-Chassis Peers
===============================================================================
Peer IP Src IP Auth Peer Admin
MCS Admin MCS Oper MCS State MC-LAG Admin MC-LAG Oper
-------------------------------------------------------------------------------
10.0.0.1 10.0.0.2 hash2 Enabled
Enabled Enabled inSync Enabled Enabled
===============================================================================
 MCS Admin ‘enabled’ as soon as ‘sync’ is no shut

 MCS Oper ‘enabled’ as soon as any of the sync parms (igmp, sub-mgmt,..) is configured
 MCS State ‘inSync ‘ as soon as databases are synchronized

A:PE2# show redundancy multi-chassis sync peer 10.0.0.1 detail

======================================
===============================================================================
MCS Application Stats
Multi-chassis Peer Table
======================================
===============================================================================
Application : igmp
Peer
Num Entries : 0
-------------------------------------------------------------------------------
Lcl Deleted Entries : 0
Alarm Entries : 0
Authentication : Enabled
--------------------------------------
Source IP Address : 10.0.0.2
Rem Num Entries : 0
Admin State : Enabled
Rem Lcl Deleted Entries : 0
-------------------------------------------------------------------------------
Rem Alarm Entries : 0
Sync-status
--------------------------------------
-------------------------------------------------------------------------------
Application : igmpSnooping
Client Applications : IGMPSnooping SUBMGMT
Num Entries : 1
Sync Admin State : Up
Sync Oper State : Up
Alarm Entries : 0
DB Sync State : inSync
--------------------------------------
Num Entries : 2 ---------------------------------
Rem Num Entries : 1
Lcl Deleted Entries : 1 Rem Num Entries : 0
Alarm Entries : 0 Rem Lcl Deleted Entries : 0
Rem Alarm Entries : 0
Rem Num Entries : 2 Rem Alarm Entries : 0
--------------------------------------
Application : subMgmt=================================
Rem Alarm Entries : 0 =================================
Num Entries : 1
... Ports synced on peer 10.0.0.1
Alarm Entries : 0 =================================
Port/Encap
-------------------------------------- Tag
Rem Num Entries : 1 ---------------------------------
Rem Lcl Deleted Entries : 0 lag-1
Rem Alarm Entries : 0 11-20 submngt
51-51
-------------------------------------- btv
Application : srrp =================================
... ======================================

A:PE2# show redundancy multi-chassis sync peer 10.0.0.1 statistics
===============================================================================
Multi-chassis Peer Sync Stats
===============================================================================
Packets Tx Total : 99489
Packets Tx Hello : 99471
Packets Tx Data : 6
Packets Tx Other : 12
Packets Tx Error : 0
Packets Rx Total : 99486
Packets Rx Hello : 99468
Packets Rx Data : 11
Packets Rx Other : 7
Packets Rx Error : 0
Packets Rx Header Err : 0
Packets Rx Body Err : 0
Packets Rx Seq Num Err : 0
===============================================================================
 Check if TX/RX Data increases

Dual homing in TPSDA: tools commands
*A:PE1# tools dump redundancy multi-chassis sync-database peer 10.0.0.2 detail
If no entries are present for an application, no detail will be displayed.
FLAGS LEGEND: ld - local delete; da - delete alarm
Peer Ip 10.0.0.2
Application IGMPSnooping
Sap-id Client Key
SyncTag DLen Flags timeStamp
deleteReason code and description
-------------------------------------------------------------------------------
lag-1:51 Group=225.1.1.1
btv 5 -- -- 04/11/2007 09:58:52
0x0
Application SUBMGMT
Sap-id Client Key
SyncTag DLen Flags timeStamp
deleteReason code and description
-------------------------------------------------------------------------------
lag-1:11 192.168.1.100
submngt 97 -- -- 04/11/2007 13:47:45
0x0
The following totals are for:

peer ip 10.0.0.2, port/lag ALL, sync-tag ALL, application ALL
Valid Entries: 2
Locally Deleted Entries: 0
Locally Deleted Alarmed Entries: 0

5
Pseudo-Wire Redundancy

Pseudo-Wire Redundancy
 Provides the ability to protect a PW with a pre-provisioned PW

 Normally taken care of by SDP redundancy but SDP redundancy falls short when:
 2 different destinations for same VLL service
– Dual homed CPEs
– Active/Standby BRAS nodes
 PW is switched in middle of network and PW switching node fails
– Handled in separate chapter
 Supported for Epipe, Apipe, Fpipe and Ipipe
 SDPs can be other type (GRE, MPLS, LDP)
 Maximum 4 SDPs
 Revertive or non-revertive

Pseudo-Wire Redundancy: Redundant VLL Service Model
VLL
SAP Spoke-SDP
ICB Spoke-SDP
spoke-sdp
Spoke-SDP
Spoke-SDP
Endpoint X Endpoint Y
 VLL service supports by default 2 implicit endpoints managed internally

 Types of endpoint objects:
 SAP: max 1 per endpoint
 Primary spoke-SDP: max 1 per endpoint
 Secondary spoke-SDP: max 4 per endpoint
 Inter-Chassis Backup (ICB) spoke-SDP
 Endpoint can have max 1 SAP + 1 ICB (SAP must be part of LAG if ICB is already present)
 Maximum 4 spoke-SDPs, any combination of:
 Primary spoke-SDP (max 1, revertive)
 Secondary spoke-SDP (max 4, not revertive, precedence) Object Selection rules
 ICB spoke-SDP (max 1)
 Forwarding blocked on same node between ICBs
 Endpoint transmit: only 1 active object; Endpoint receive: from all endpoint objects
 See Appendix B for Object Selection Rules
 Traffic cannot be received and forwarded in the same endpoint: traffic has to cross 2 endpoints

Pseudo-Wire Redundancy: T-LDP Status Bits
T-LDP Status Bits:

0x00000000 - Pseudo Wire forwarding ( clear all failures )
0x00000001 - Pseudo Wire Not Forwarding (MTU problems, etc)
0x00000002 - SAP ( ingress ) Receive Fault
0x00000004 - SAP ( egress ) Transmit Fault
0x00000008 - SDP-binding ( ingress ) Receive Fault
0x00000010 - SDP-binding ( egress ) Transmit Fault
0x00000020 - PW forwarding Standby <=== NEW Ethereal capture
Rules for processing endpoint SAP Active/Standby status bits:

1. Advertised forwarding status of “Active/Standby” reflects status of the local LAG SAP in MC-LAG
application.
2. When SAP in endpoint ‘X’ is part of a MC-LAG instance, node must send T-LDP forwarding status bit of
"SAP Active/Standby" over all ‘Y’ endpoint spoke-sdp's, except the ICB spoke-sdp, whenever this status
changes. Status bit sent over the ICB is always zero (“Active” by default).
3. When the SAP in endpoint ‘X’ is not part of a MC-LAG instance, then the forwarding status sent over
all ‘Y’ endpoint spoke-sdp's should always be set to zero ("Active" by default).
4. The received SAP “Active/Standby” status is saved and used for selecting the active transmit
endpoint object (Appendix B)
Status Bits rules

Pseudo-Wire Redundancy: Active/Standby concept
TLDP
Aggregation Active Standby Aggregation
Node Node
A C
Active Standby
Active Standby
Inter-Chassis PW Inter-Chassis PW
MSAN for VLL for VLL MSAN
Standby Active
Standby Active
Aggregation Aggregation
Node Node
B Standby Active D
TLDP
Node Node
A C
Active Standby
Inter-Chassis PW Inter-Chassis PW
MSAN for VLL for VLL MSAN
Standby Active
Node Node
B D

Pseudo-Wire Redundancy: Endpoint Design 4 PE, 2 * dual homed
PE1 PE3
epipe epipe
LAG LAG
X Y SDP X Y
SAP SDP SAP
SDP SDP SDP SDP SDP SDP

MC-LAG MC-LAG
CE5 CE6
ICB spoke-sdp ICB spoke-sdp
SAP SDP SDP SAP

X Y X Y
LAG LAG
epipe epipe
PE2 PE4

Pseudo-Wire Redundancy: Configuration
A:PE1>config>service>epipe# info
----------------------------------------------
PE1 PE3
endpoint "x" create
exit
endpoint "y" create
epipe epipe
13:50 exit
LAG sap lag-1 endpoint
LAG "x" create
X Y SDP X Y
exit
SAP SDP SAP
spoke-sdp 13:50 endpoint "y" create
exit
SDP SDP SDP SDP SDP SDP 14:50 endpoint "y" create
spoke-sdp
MC-LAG exit MC-LAG
14:50 spoke-sdp 12:50 endpoint "x" icb create
exit
spoke-sdp 12:60 endpoint "y" icb create
CE5 12:50 12:60 exit CE6
no shutdown
ICB spoke-sdp A:PE2>config>service>epipe#
ICB spoke-sdp info
----------------------------------------------
21:60 21:50 endpoint "x" create
exit
endpoint "y" create
23:50 exit
sap lag-1 endpoint "x" create
SDP SDP SDP SDP SDP exitSDP
24:50 exit
SAP SDP SDP SAP
X Y X Y
LAG exit LAG
epipe epipe
exit
spoke-sdp 21:60 endpoint "x" icb create
exit
PE2 PE4
no shutdown

Pseudo-Wire Redundancy: CLI Tree
configure service epipe service-id [vc-switching]

endpoint endpoint-name [create] Handled
[no] active-hold-delay hundred-milliseconds
[no] description description-string later
[no] revert-time [seconds | infinite]
exit
sap sap-id [endpoint endpoint-name] [create]
spoke-sdp sdp-id[:vc-id] [endpoint endpoint-name] [icb] [create]
precedence [prec-value | primary]
exit
exit
 Adding the primary keyword to an SDP binding indicates that the service should revert to that SDP
whenever possible
• revert-timer specifies reversion delay interval
• default is zero i.e. immediate reversion
 A backup spoke-sdp is one specified without the primary keyword
• Configure all spoke-sdp bindings as backup if reversion is not required
• No reversion between backup spoke-SDPs
 Precedence used as a tie breaker if multiple backup SDPs have status active-active
 Explicit configuration of ICB required as an ICB is always the last forwarding resort
• Traffic never forwarded between ICBs an a 7750
 Endpoint-name can be any name

LAB Exercise 2: PW-Redundancy configuration 1 (builds on Exercise 1)
PE1 PE3
SIM IP Addresses 10.0.0.1 10.1.3/24 10.0.0.3
CE1 (138.203.15.80) 1/1/4 1/1/3 1/1/1
CE2 (138.203.15.45) 1/1/1
CE3 (138.203.15.43) 1/1/1
CE4 (138.203.15.44) 1/1/2 1/1/1
1/1/2
1/1/3
PE1 (138.203.15.81) 1/1/2
10.1.2/24 10.3.4/24
PE2 (138.203.15.76) 1/1/3
1/1/3
PE3 (138.203.15.77) 1/1/2
CE1 1/1/1 1/1/1 1/1/2 CE2
PE4 (138.203.15.78) 1/1/4
1/1/4 1/1/3
1/1/2
 Team and : 10.0.0.2
10.2.4/24
10.0.0.4
 Configure Epipe 100 PE2 PE4
 PW/Redundancy + ICB (vc 100/101)

 Use show/tools force commands PE3 PE4
 Only fail Access-links!!! 10.0.0.3 10.3.4/24 10.0.0.4
1/1/2 1/1/2 1/1/4
 Team and : 1/1/4
1/1/4
 Configure Epipe 200 1/1/3

1/1/4
1/1/3
 PW/Redundancy + ICB (vc 200/201) 10.1.3/24 10.3.4/24
 Use show/tools force commands 1/1/4 1/1/4
CE3 1/1/5 1/1/3 CE4
1/1/3
 Only fail Access-links!!! 1/1/3 1/1/3
1/1/5
10.1.2/24
10.0.0.1 10.0.0.2
PE1 PE2

Pseudo-Wire Redundancy: Supported MC-LAG/PW-Redundancy configurations
Access
Node
4 PE, 2 * Dual homed 3 PE, 1 * Dual homed
Active Standby Access

Node EPIPE
Active Active Standby
Standby
Standby
Standby Active
Acces Active
s
Node Active
Standby
T-LDP
3 PE, 2 * Dual homed EPIPE
Acces
Redundant BRAS nodes s
Standby Node
Active
Active 2 PE, 2 * Dual homed
Access Access
Standby Node Node
Acces Standby
s
EPIPE
Node
Active

PE1
epipe
LAG X Y
LAG
SAP SAP
SDP SDP
MC-LAG
MC-LAG
CE1 CE2
ICB spoke-sdp
SDP SDP
SAP SAP
X Y
LAG LAG
epipe
PE2

PE1
epipe
MC-LAG
LAG X Y NOTE: icb parameter not
SAP
required on PE1 and PE3 but
SDP SDP used for consistency
CE1
ICB spoke-sdp
PE2
epipe
LAG
SDP X Y
SAP MC-LAG
SAP SDP CE2

LAG
ICB spoke-sdp
epipe
LAG
X Y
SAP
SDP SDP
PE3

PE1 PE3
epipe epipe CE2
MC-LAG
LAG X Y X Y
SAP
CE1 SAP
SDP SDP SDP SDP SDP
ICB spoke-sdp PE2

epipe
SDP SDP
LAG SDP
SAP
X Y
PE1 PE3
epipe epipe CE2
MC-LAG
LAG X Y X Y
SAP
CE1 SAP
SDP SDP SDP SDP
ICB spoke-sdp PE2

epipe
NOTE: icb parameter not
SDP SDP required on PE2 but
LAG used for consistency
SAP
X Y

Pseudo-Wire Redundancy: Show commands
 Endpoint information
A:PE1# show service id 50 endpoint
===============================================================================
Service 50 endpoints
===============================================================================
Endpoint name : x
Revert time : 0
Act Hold Delay : 0
Tx Active : lag-1
-------------------------------------------------------------------------------
Members
-------------------------------------------------------------------------------
SAP : lag-1
Spoke-sdp : 12:50 Precedence:4 (icb)
===============================================================================
Endpoint name : y
Revert time : 0
Act Hold Delay : 0
Tx Active : 14:50
-------------------------------------------------------------------------------
Members
-------------------------------------------------------------------------------
Spoke-sdp : 12:60 Precedence:4 (icb)
Spoke-sdp : 13:50 Precedence:4
Spoke-sdp : 14:50 Precedence:4
===============================================================================
===============================================================================

 detailed spoke-SDP information

A:PE1# show service id 50 sdp 13:50 detail
===============================================================================
Service Destination Point (Sdp Id : 13:50) Details
===============================================================================
-------------------------------------------------------------------------------
Sdp Id 13:50 -(10.0.0.3)
-------------------------------------------------------------------------------
SDP Id : 13:50 Type : Spoke
VC Type : Ether VC Tag : n/a
Admin Path MTU : 0 Oper Path MTU : 9190
Far End : 10.0.0.3 Delivery : LDP

Acct. Pol : None Collect Stats : Disabled
Ingress Label : 131068 Egress Label : 131065
Ing mac Fltr : n/a Egr mac Fltr : n/a
Ing ip Fltr : n/a Egr ip Fltr : n/a
Ing ipv6 Fltr : n/a Egr ipv6 Fltr : n/a
Admin ControlWord : Not Preferred Oper ControlWord : False
Last Status Change : 03/18/2007 23:05:03 Signaling : TLDP
Last Mgmt Change : 03/18/2007 23:04:25
Endpoint : y Precedence : 4
Flags : None
Peer Pw Bits : pwFwdingStandby Local status
Peer Fault Ip : None
Peer Vccv CV Bits : lspPing
Peer Vccv CC Bits : mplsRouterAlertLabel PW Status bits signaled by T-LDP notification message
MAC Pinning : Disabled
KeepAlive Information :
Admin State : Disabled Oper State : Disabled
Hello Time : 10 Hello Msg Len : 0
Max Drop Count : 3 Hold Down Time : 10
...

 service label information

A:PE1# show router ldp bindings service-id 50
===============================================================================
LDP LSR ID: 10.0.0.1
===============================================================================
Legend: U - Label In Use, N - Label Not In Use, W - Label Withdrawn
S - Status Signaled Up, D - Status Signaled Down
E - Epipe Service, V - VPLS Service, M - Mirror Service
A - Apipe Service, F - Fpipe Service, I - IES Service, R - VPRN service
P - Ipipe Service
===============================================================================
LDP Service Bindings
===============================================================================
Type VCId SvcId SDPId Peer IngLbl EgrLbl LMTU RMTU
-------------------------------------------------------------------------------
E-Eth 50 50 12 10.0.0.2 131065U 131066D 1500 1500
E-Eth 50 50 13 10.0.0.3 131068U 131065D 1500 1500
E-Eth 50 50 14 10.0.0.4 131067U 131065S 1500 1500
Signaled Down:
E-Eth 60 50 12 10.0.0.2 131064U 131065S 1500 1500 reflects SAP status
------------------------------------------------------------------------------- on PE2
No. of VC Labels: 4
===============================================================================
Signaled Down:
PE1 PE3
reflects SAP status
on PE3
Active Standby
Active Active Standby

12:50 Standby
12:60
Access Standby Standby Active

Active Access
Node
Node
Standby Active
PE2
T-LDP PE4

Pseudo-Wire Redundancy: Convergence -> Normal situation
PE1 PE3 Flags

Peer PW Bits
None
None
Active Object
epipe epipe
LAGSAP X Y SDP TX-ACTIVE X Y
LAG
SDP SAP
SDP SDP SDP None
TX-ACTIVE SDP SDP SDP
pwFwdingStandby
MC-LAG MC-LAG
None TX-ACTIVE TX-ACTIVE
None None SapOperDown
lacIngressFault None None None
None None
lacEgressFault
None
CE5 pwFwdingStandby CE6
None
None pwFwdingStandby None
None None lacIngressFault
SapOperDown lacEgressFault
TX-ACTIVE None None
None
None
TX-ACTIVE TX-ACTIVE
SDP SDP SDP None SDP SDP SDP
None
Note: Endpoint X sends SAP SDP TX-ACTIVE SDP SAP
status its about Y and LAG X Y X Y LAG
None
vice versa epipe pwFwdingStandby epipe
PE2 PE4

Pseudo-Wire Redundancy: Convergence -> Access Link Failure
PE1 PE3
None
epipe None epipe
LAGSAP SDP TX-ACTIVE LAG
X Y SDP X Y SAP
TX-ACTIVE
TX-ACTIVE SDP SDP SDP SDP SDP SDP
None
None TX-ACTIVE
MC-LAG MC-LAG
None None None
None None
lacIngressFault None pwFwdingStandby
None lacIngressFault
lacEgressFault lacEgressFault
None
CE5 pwFwdingStandby CE6
None
None None None
None SapOperDown TX-ACTIVE SapOperDown
None
TX-ACTIVE None None
None TX-ACTIVE
None
TX-ACTIVE
None
pwFwdingStandby SAP
SAP SDP SDP
LAG X Y X Y LAG
None
epipe pwFwdingStandby epipe
PE2 PE4

Pseudo-Wire Redundancy: Convergence -> Access Node Failure
PE1 PE3
None
epipe None epipe
LAGSAP TX-ACTIVE LAG
X Y SDP SDP X Y SAP
TX-ACTIVE
TX-ACTIVE SDP SDP SDP SDP SDP SDP
None
None TX-ACTIVE
MC-LAG SdpOperDown MC-LAG
None SdpOperDown SdpOperDown
None NoIngVCLabel
lacIngressFault NoIngVCLabel NoIngVCLabel
None NoEgrVCLabel
lacEgressFault NoEgrVCLabel NoEgrVCLabel
None None
CE5 None None CE6
pwFwdingStandby
None
None None
None SapOperDown TX-ACTIVE
TX-ACTIVE None
SdpOperDown
SDP SDP SDP NoIngVCLabel SDP SDP SDP
NoEgrVCLabel
SAP SDP None SDP SAP
LAG X Y X Y LAG
epipe epipe
Convergence slides
In Appendix C
PE2 PE4

Pseudo-Wire Redundancy: Dual BRAS Topology
Redundant BRAS nodes
Standby
Active
Active
Standby Standby
Acces
s
Node
Active
 More than 1 spoke-SDP Active/Active

 Election based on precedence
 Revertive or non-Revertive behaviour

Pseudo-Wire Redundancy: Dual BRAS Topology
PE1
epipe
CONFIGURATION NEXT SLIDE
LAG MC-LAG
SDP X Y SAP
PE5
epipe SDP SDP SDP primary
precedence 1
X Y
SDP
ICB spoke-sdp
SAP SDP
precedence 2 epipe
SDP
SDP SDP SDP
SDP
MC-LAG precedence 3 LAG
SDP SAP
X Y
precedence 4
PE2
PE3
epipe
precedence 1
SDP LAG MC-LAG
X Y SAP
epipe
precedence 2 SDP SDP SDP secondary
SDP
SAP SDP precedence 3 ICB spoke-sdp

SDP
epipe
X Y SDP
precedence 4 SDP SDP SDP
LAG
PE6 SAP
SDP
X Y
PE4

Pseudo-Wire Redundancy: Dual BRAS Topology non-revertive
*A:PE5>config>service>epipe# info
----------------------------------------------
PE5 endpoint "x" create
epipe precedence 1 exit
endpoint "y" create
X Y 51:90
exit
SDP
sap 1/1/1 endpoint "x" create
exit
SDP
SAP 52:90 spoke-sdp 51:90 endpoint "y" create
precedence 1
SDP precedence 2 exit
SDP spoke-sdp 52:90 endpoint "y" create
53:90 precedence 2
exit
54:90 precedence 3 precedence 3
exit
precedence 4 exit
no shutdown
----------------------------------------------
 By default spoke-SDPs are secondary

 Election based on precedence
 Non-revertive behaviour

Pseudo-Wire Redundancy: Dual BRAS Topology revertive example
----------------------------------------------
PE5 endpoint "x" create
primary exit
epipe
endpoint "y" create
X Y 51:90 revert-time 10
SDP exit
sap 1/1/1 endpoint "x" create
SDP exit
SAP 52:90 spoke-sdp 51:90 endpoint "y" create
SDP precedence primary
precedence 2 exit
SDP 53:90 spoke-sdp 52:90 endpoint "y" create
precedence 2
exit
54:90 precedence 3 precedence 3
exit
precedence 4 exit
no shutdown
----------------------------------------------
 Maximum 1 spoke-SDP can be primary

 Revertive behaviour
 Revert-time [0..600]|infinite

Pseudo-Wire Redundancy: Dual BRAS Topology configuration
A:PE1>config>service>epipe# info
----------------------------------------------
endpoint "x" create PE1
exit epipe
endpoint "y" create
active-hold-delay 10 15:90 LAG
SDP X Y SAP
exit
sap lag-1 endpoint "y" create SDP SDP SDP
exit
spoke-sdp 15:90 endpoint "x" create
exit
spoke-sdp 16:90 endpoint "x" create 12:100 12:90
exit 16:90
exit ICB spoke-sdp
spoke-sdp 12:100 endpoint "x" icb create
exit
no shutdown
----------------------------------------------
 active-hold-delay specifies that the node will delay sending the

change in the T-LDP status bits
 Use explained in next slide
 [1..60] deci-seconds
 ICBs forward traffic during convergence

Pseudo-Wire Redundancy: active-hold-delay
No active-hold-delay configured PE1
epipe
1. Normal flow LAG
SDP X Y SAP MC-LAG
2. Access link failure primary
SDP SDP
3. PE1 PW Stdby sent + MC-LAG convergence
4. PE5: spoke-SDP selection process -> PE3 ICB spoke-sdp
5. PE2: PW Active status bits sent epipe

PE5 SDP SDP SDP
epipe LAG
SAP
X Y X Y
SDP
PE2
SAP SDP
SDP PE3
epipe
SDP
SDP LAG MC-LAG
X Y SAP
active-hold-delay configured SDP SDP SDP secondary
1. Normal flow
2. Access link failure ICB spoke-sdp
3. Active-hold-delay timer started + MC-LAG convergence epipe

4. PE2: PW Active status bits sent SDP SDP SDP
LAG
5. PE5: spoke-SDP selection process -> PE2 SAP
X Y
6. PE1 PW Stdby sent PE4

LAB Exercise 3: PW-Redundancy configuration 2
PE3 CE2
SIM IP Addresses 1/1/1
1/1/1
CE1 (138.203.15.80) PE1
CE2 (138.203.15.45) PE4 1/1/1 1/1/2
PE1 (138.203.15.81)
CE1
PE2 (138.203.15.76)
PE2 1/1/1 1/1/3
PE3 (138.203.15.77)
PE4 (138.203.15.78)
PE5 (138.203.15.43)
PE6 (138.203.15.44) CE2
PE5 1/1/3
PE7 (138.203.15.46) 1/1/1
PE2
PE8 (138.203.15.47)
PE6 1/1/1 1/1/4
CE1
 Team : Epipe 1001, lag 1, lacp-key 1, sys-id 1 PE1 1/1/1 1/1/1
 Team : Epipe 1002, lag 2, lacp-key 2, sys-id 2

CE1
 Team : Epipe 1003, lag 3, lacp-key 3, sys-id 3 PE3
PE1
1/1/2
1/1/2
 Team : Epipe 1004, lag 4, lacp-key 3, sys-id 4 PE2 1/1/2 1/1/4

CE2
 All secondary, precedence 1,2,3 PE7 1/1/4 1/1/5
 With/without active-hold-delay
 Primary and revert-time PE5
1/1/6
1/1/5
CE1
PE4
 Only fail access-links!!!! PE6 1/1/6 1/1/6
 Use tools commands specifically per LAG! PE8 1/1/4 1/1/6

CE2

7
Pseudo-Wire Switching

Pseudo-Wire Switching
T-PE1 S-PE3 S-PE5 T-PE7 notes
MC-LAG MC-LAG
OSPF / LDP ISIS / RSVP OSPF / GRE
T-PE2 S-PE4 S-PE6 T-PE8
 Method for scaling a large network of VLL or VPLS services

 Removes need for a full mesh of T-LDP sessions between the 7x50 PE nodes
 Switching PE (S-PE) node cross-connects 2 spoke SDPs
 S-PE nodes represent a single point of failure -> standby PW switched path
 SAPs configured on Termination PE nodes (T-PE)
 All VLL service types, i.e,.apipe, epipe, fpipe, and ipipe are supported
 S-PE can bind two spoke SDPs of different types (LDP/RSVP/GRE)
 In R5.0 switching of dynamic PW – dynamic PW and static PW – static PW supported

Pseudo-Wire Switching: CLI Tree
configure service epipe service-id vc-switching
spoke-sdp sdp-id[:vc-id] [create]

exit
exit
 ‘vc-switching’ parameter required, otherwise not possible to add 2 spoke-SDPs:

MINOR: SVCMGR #1854 The service cannot support any more SDP bindings
 Only 2 spoke-SDPs can be configured:
MINOR: SVCMGR #1312 Only two Spoke SDPs are allowed for PW switching service
 Spoke-SDPs cannot explicitly be assigned to an endpoint:
MINOR: SVCMGR #1313 Explicit endpoint is not allowed for PW switching service
 vc-ids can be different within the same Epipe configuration

Pseudo-Wire Switching: Label-Mapping and Notification Messages
Label-mapping message:
 S-PEs act as slaves for the purpose of PW signaling, T-PEs act as masters
 S-PE waits for LDP label-mapping message from T-PE (needs SAP-interface parms like MTU)
 S-PE appends optional PW switching point TLV to the FEC TLV in which it records its
system address, then relays label-mapping message
Notification message:
 PW status messages can be generated by the T-PE nodes and/or the S-PE nodes
 PW status messages received by S-PE are processed and then passed to the next hop
 S-PE appends optional PW switching TLV only if it originated the message or the
message was received with the TLV in it
 If local status for both spokes id UP, S-PE passes status notification TLV unchanged
 VCid in the FEC TLV is set to value of the PW segment to the next hop
 If local status for any of the spokes is down, S-PE sends SDP-binding down bits
 regardless of the received status bits from the remote node

Pseudo-Wire Switching: PW-Switching 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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|1|0| PW sw TLV (0x096D) | PW sw TLV Length |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Type | Length | Variable Length Value |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Variable Length Value |
| " |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Details specific to PW Switching Point Sub-TLVs described:

 PW ID of last PW segment traversed
This sub-TLV type conains a PW ID in the format of the PWID described in [PWE3-MPLS]
 PW Switching Point description string
An optional description string of text up to 80 characters long
 IP address of PW Switching Point
The IP V4 or V6 address of the PW Switching Point. This is an OPTIONAL Sub-TLV.

Pseudo-Wire Switching: S-PE configuration
ONLY S-PEs configured

epipe epipe epipe epipe
X Y SDP SDP X Y SDP SDP X Y SDP SDP X Y
SAP SAP
LAG LAG
SDP SDP SDP SDP
MC-LAG
CE9 epipe 100 customer 1 vc-switching create

CE10
spoke-sdp 31:100 create
exit
exit
no shutdown
exit

SDP SDP SDP SDP
LAG LAG
SAP
SAP X Y SDP SDP X Y SDP SDP X Y SDP SDP X Y

Pseudo-Wire Switching: S-PEs are slaves

SAP SAP
LAG LAG
SDP SDP SDP SDP
MC-LAG
CE9 A:PE5# show router ldp bindings service-id 100 CE10

===============================================================================
===============================================================================
P - Ipipe Service
===============================================================================
epipe epipe epipe
=============================================================================== epipe
SDP SDP -------------------------------------------------------------------------------
SDP SDP
E-Eth 100 100 53 10.0.0.3 131070U -- 1500 0
LAG E-Eth 100 100 57 10.0.0.7 131069U -- 1500 0 LAG
SAP
SAP X Y SDP -------------------------------------------------------------------------------
SDP X Y SDP SDP X Y SDP SDP X Y
No. of VC Labels: 2
T-PE2 S-PE4 S-PE6
=============================================================================== T-PE8
S-PE waits for LDP label-mapping message from T-PE before sending label-mapping

Pseudo-Wire Switching: T-PE1 configured
T-PE1 S-PE3 ON PE5: S-PE5 T-PE7

1 2007/03/24 21:51:37.72 UTC MINOR: DEBUG #2001 Base LDP
epipe epipe "LDP: LDP
epipe epipe
X Y SDP SDP X Y SDP
Recv
SDPLabel Mapping packet (msgId = SDP
X Y SDP
2148) Xfrom
Y
10.0.0.3:0
SAP Label 131070 advertised for the following FECs SAP
LAG Service FEC Epipe/100 Group ID = 0 cBit = 0 LAG
SDP SDP Interface parameter Mtu = 1500 SDP SDP
Interface parameter VCCV = 0x2
PW status bits = 0x0 MC-LAG
Switching hop: System = 10.0.0.3, previous segment PW ID 100
CE9*A:PE1# configure service epipe 100 " CE10
---------------------------------------------- 2 2007/03/24 21:51:37.72 UTC MINOR: DEBUG #2001 Base LDP
endpoint "x" create "LDP: LDP
exit ICB spoke-sdp Recv Notification packet (msgId = 2149) from 10.0.0.3:0
ICB spoke-sdp
endpoint "y" create Status Code = Null Non-fatal
exit Causing message Id = 0
sap lag-1 endpoint "x" create Causing message type = NULL
exit Service FEC Epipe/100 Group ID = 0 cBit = 0
spoke-sdp 13:100 endpoint "y" create PW status bits = 0x18
exit Switching hop: System = 10.0.0.3, previous segment PW ID 100
spoke-sdp 12:100 endpoint "y" icb create "
exit
SDPspoke-sdp
SDP 12:101 endpoint "x" icb create 3 2007/03/24 21:51:37.72 UTC MINOR: SDP
DEBUG #2001
SDP Base LDP
exit "LDP: LDP
no shutdown Recv Notification packet (msgId = 2150) from 10.0.0.3:0
LAG
LAG Status SAP
SAP X Y SDP SDP X Y SDP
---------------------------------------------- SDPCode
X =YNull Non-fatal
SDP SDP X Y
Causing message Id = 0
T-PE2 S-PE4 Causing message
S-PE6 type = NULL
Service FEC Epipe/100 Group ID = 0 cBit = 0T-PE8
PW status bits = 0x0
"
After service creation PE1 sends label-mapping and notification. S-PE3 relays.

Pseudo-Wire Switching: T-PE1 configured

SAP SAP
LAG LAG
SDP SDP SDP SDP
MC-LAG
CE9 CE10
*A:PE5# show router ldp bindings service-id 100
===============================================================================
===============================================================================
P - Ipipe Service
===============================================================================
epipe epipe
epipe epipe
===============================================================================
SDP SDP SDP SDP
------------------------------------------------------------------------------- LAG
LAG SAP
SAP X Y SDP E-Eth 100 SDP X100 Y SDP 53 10.0.0.3
SDP X Y SDP 131070U 131070S 1500
SDP X 1500
Y
E-Eth 100 100 57 10.0.0.7 131069U -- 1500 0
T-PE2 -------------------------------------------------------------------------------
S-PE4 S-PE6
No. of VC Labels: 2 T-PE8
===============================================================================
After service creation PE1 sends label-mapping and notification. S-PE3 relays.

Pseudo-Wire Switching convergence: Normal behaviour
TX-ACTIVE
TX-ACTIVE
T-PE1 None S-PE3 None S-PE5 None T-PE7
epipe None epipe None epipe None epipe
TX-ACTIVE TX-ACTIVE
SAP SAP
LAG None None None LAG
None None None
SDP SDP SDP SDP
None None MC-LAG
lacIngressFault None None lacIngressFault
lacEgressFault None None lacEgressFault
CE9 CE10
None SapOperDown SapOperDown None

None None None None
TX-ACTIVE TX-ACTIVE TX-ACTIVE TX-ACTIVE
SDP SDP SDP SDP
None None None
pwFwdingStandby pwFwdingStandby pwFwdingStandby LAG
LAG SAP
None None None
T-PE2 pwFwdingStandby S-PE4 pwFwdingStandby S-PE6 pwFwdingStandby T-PE8

Pseudo-Wire Switching convergence: Acces link failure
TX-ACTIVE
T-PE1 None
S-PE3 S-PE5 T-PE7
None None
epipe pwFwdingStandby epipe pwFwdingStandby
epipe pwFwdingStandby
epipe
TX-ACTIVE
e.g. ONLAGSAP
PE3: SAP
None None None LAG
3 2007/03/24 22:07:26.20 UTC MINOR: DEBUG
#2001 Base LDP
None None None
"LDP: LDP SDP SDP SDP SDP
Recv Notification
None packet (msgId = 2296) from 10.0.0.5:0 MC-LAG
TX-ACTIVE
Status Code = Null Non-fatal
lacIngressFault SapOperDown TX-ACTIVE
lacEgressFault None None
CE9
Causing message type = NULLNone
None CE10
None
Service FEC Epipe/100 Group ID = 0 cBit = 0
"
"LDP: LDP
Recv Notification packet (msgId = 2297) from 10.0.0.5:0
None
Null Non-fatal SapOperDown
Status Code =None None
None
None
None
Causing message Id = 0 lacIngressFault
TX-ACTIVE TX-ACTIVE lacEgressFault
Causing message type = NULL
TX-ACTIVE
" SDP SDP None
None
SDP SDP
None None
None None
5 2007/03/24
LAG 22:07:26.20 UTC MINOR: DEBUG #2001 Base LDP LAG
SAP
"LDP: LDP SAP X Y SDP SDP X Y SDP SDP X Y SDP SDP X Y
Recv Notification packet (msgId = 2298) fromNone
10.0.0.5:0 None None
T-PE2
Status Code = Null Non-fatal pwFwdingStandby S-PE4 pwFwdingStandby S-PE6 pwFwdingStandby T-PE8
Causing message Id = 0 TX-ACTIVE
"

Pseudo-Wire Switching convergence: Acces node failure
SdpOperDown
NoIngVCLabel
T-PE1 NoEgrVCLabel
S-PE3 NoEgrVCLabel
S-PE5 NoEgrVCLabel T-PE7
epipe None epipe None epipe None epipe
TX-ACTIVE
e.g. ON SAP
PE3: SAP
LAG None None LAG
7 2007/03/24 22:13:18.35 UTC MINOR: DEBUG
#2001 Base LDP
None None
"LDP: LDP SDP SDP SDP SDP
Recv Address
NoneWithdraw packet (msgId = 2339) from 10.0.0.5:0 MC-LAG
TX-ACTIVE
Address Family = 1 Number of addresses = 1
lacIngressFault
Address 1 = 10.5.7.5
lacEgressFault None
CE9
" None CE10
"LDP: LDP
Recv Label Withdraw packet (msgId = 2340) from 10.0.0.5:0
Label 131070 withdrawn for the following FECs
Service FEC Epipe/100
None
Group ID = 0 cBit = 0
" SdpOperDown SdpOperDown
None SapOperDown NoIngVCLabel NoIngVCLabel
TX-ACTIVE psnIngressFault NoEgrVCLabel NoEgrVCLabel
9 2007/03/24 22:13:19.33 UTC MINOR: DEBUG
#2001 Base LDP
psnEgressFault None None
"LDP: LDP
Send Label Release packet (msgId 2341) to 10.0.0.5:0
Label 131070 released for the following FECs
Service FEC Epipe/100
SDP Group
SDP ID =None
0 cBit = 0 None
SDP SDP TX-ACTIVE
" None None
None None
LAG LAG
SAP
None None None
TX-ACTIVE TX-ACTIVE

Pseudo-Wire Switching convergence: SDP failure
SdpOperDown SdpOperDown
NoIngVCLabel NoIngVCLabel
NoEgrVCLabel NoEgrVCLabel
T-PE1 NoEgrVCLabel
S-PE3 NoEgrVCLabel
S-PE5
None NoneT-PE7
epipe None epipe None epipe epipe
TX-ACTIVE SDP shut
X Y SDP SDP X Y SDP SDP X Y SDP On PE5 SDP X Y
SAP SAP
LAG None None LAG
None None
SDP SDP SDP SDP TX-ACTIVE
None TX-ACTIVE MC-LAG
lacIngressFault TX-ACTIVE
lacEgressFault None None None
CE9 None None CE10
lacIngressFault
lacEgressFault
None SdpOperDown None

None SapOperDown psnIngressFault None
TX-ACTIVE psnIngressFault psnEgressFault TX-ACTIVE
psnEgressFault

SDP SDP None None SDP SDP
None pwFwdingStandby pwFwdingStandby
pwFwdingStandby LAG
LAG SAP
None None None
TX-ACTIVE TX-ACTIVE

Nonr
psnIngressFault
Nonr psnEgressFault
psnIngressFault 10.0.0.5
T-PE1 psnEgressFault
S-PE3 ServiceAdminDown S-PE5 T-PE7
epipe epipe None epipe epipe
10.0.0.5
TX-ACTIVE
X Y SDP SDP X Y SDP SDPShut
X epipe
Y SDP SDP X Y
SAP on PE5 SAP
LAG None Nonr LAG
None psnIngressFault ServiceAdminDown
SDP SDP SDP SDP TX-ACTIVE
psnEgressFault None
None TX-ACTIVE 10.0.0.5 MC-LAG
TX-ACTIVE
lacIngressFault
None
ON PE3: None None
lacEgressFault
CE9 None 1 2007/03/24 21:58:32.54 UTC MINOR: DEBUG #2001 Base LDPNone CE10
lacIngressFault
"LDP: LDP lacEgressFault
Recv Notification packet (msgId = 2236) from 10.0.0.5:0
None PW status bits = 0x18 SdpOperDown None
None SapOperDown Switching hop: System = 10.0.0.5, previous segment PW ID 100
psnIngressFault None
TX-ACTIVE psnIngressFault " psnEgressFault TX-ACTIVE
psnEgressFault
ON PE1:
"LDP: LDP
SDP SDP None SDP SDP
None
Recv Notification packet
None (msgId = 28205) from 10.0.0.3:0
pwFwdingStandby pwFwdingStandby
LAG
pwFwdingStandby LAG
SAP
Causing
SAP Xmessage
Y SDP Id = 0 SDP X Y SDP SDP X Y SDP SDP X Y
Causing message type = NULL None None None
Service T-PE2
FEC Epipe/100pwFwdingStandby
Group ID = 0 cBit =S-PE4
0 pwFwdingStandby S-PE6 pwFwdingStandby T-PE8
TX-ACTIVE TX-ACTIVE
Switching hop: System = 10.0.0.5, previous segment PW ID 100
"

*A:PE1# show service id 100 sdp 13:100 detail

===============================================================================
Service Destination Point (Sdp Id : 13:100) Details
===============================================================================
-------------------------------------------------------------------------------
Sdp Id 13:100 -(10.0.0.3)
-------------------------------------------------------------------------------
Admin Path MTU : 0 Oper Path MTU : 9190
Far End : 10.0.0.3 Delivery : LDP

Acct. Pol : None Collect Stats : Disabled
Ingress Label : 131067 Egress Label : 131063
Ing mac Fltr : n/a Egr mac Fltr : n/a
Ing ip Fltr : n/a Egr ip Fltr : n/a
Ing ipv6 Fltr : n/a Egr ipv6 Fltr : n/a
Admin ControlWord : Not Preferred Oper ControlWord : False
Last Status Change : 03/24/2007 14:16:20 Signaling : TLDP
Last Mgmt Change : 03/24/2007 13:46:10
Endpoint : y Precedence : 4
Flags : None
Peer Pw Bits : psnIngressFault psnEgressFault IP address of node with
Peer Fault Ip : 10.0.0.5 problem
Peer Vccv CV Bits : lspPing From PW Switching TLV
Peer Vccv CC Bits : None
...

Pseudo-Wire Switching: VCCV
*A:PE1# oam vccv-ping 13:100

MINOR: OAM #3302 Could not send VCCV ping because the far-end peer does not support it
 Control Word must be configured !!!

----------------------------------------------
control-word
exit
control-word
exit
no shutdown
----------------------------------------------
*A:PE1# oam vccv-ping 13:100

VCCV-PING 13:100 88 bytes MPLS payload
Seq=1, reply from 10.0.0.3 via Control Channel
udp-data-len=32 rtt<10ms rc=8 (DSRtrMatchLabel)
---- VCCV PING 13:100 Statistics ----

1 packets sent, 1 packets received, 0.00% packet loss
round-trip min < 10ms, avg < 10ms, max < 10ms, stddev < 10ms

*A:PE1# oam vccv-ping 13:100 ttl 1

udp-data-len=32 rtt<10ms rc=8 (DSRtrMatchLabel)


udp-data-len=32 rtt<10ms rc=4 (NoFECMapping)


udp-data-len=32 rtt<10ms rc=4 (NoFECMapping)


*A:PE3# show service id 100 sdp 31 detail
===============================================================================
Service Destination Point (Sdp Id : 31) Details
===============================================================================
-------------------------------------------------------------------------------
Sdp Id 31:100 -(10.0.0.1)
-------------------------------------------------------------------------------
...
Peer Pw Bits : None
Peer Fault Ip : None
Peer Vccv CV Bits : lspPing
Peer Vccv CC Bits : pwe3ControlWord mplsRouterAlertLabel
...

Pseudo-Wire Switching: VCCV trace

LAB Exercise 4: PW-Switching
SIM IP Addresses PE1 PE3 PE7

CE1 (138.203.15.80) 1/1/1 1/1/4
CE2 (138.203.15.45) CE1 1/1/1 1/1/5 CE2
PE1 (138.203.15.81)
PE2 (138.203.15.76)
PE3 (138.203.15.77) 1/1/6
PE4 (138.203.15.78) 1/1/3
1/1/1 1/1/4
PE5 (138.203.15.43)
PE6 (138.203.15.44) PE2 PE4 PE8
PE7 (138.203.15.46)
PE8 (138.203.15.47)
PE1 PE3 PE5
1/1/2 1/1/1
 Team : Epipe 501 CE1 1/1/2 1/1/3 CE2
 lag 1, LACP-key 1, sys-id 1

1/1/4
 Team : Epipe 502 1/1/4
1/1/2 1/1/1
 lag 2, LACP-key 2, sys-id 2 PE2 PE4 PE6
 Team : Epipe 503

PE5 PE1 PE3
 lag 3, LACP-key 2, sys-id 3
1/1/6 1/1/1
 Only fail access-links!!!! CE1 1/1/5 1/1/1 CE2
 Debug commands 1/1/2

1/1/6
1/1/6 1/1/1
PE6 PE2 PE4

8
Convergence Time
Convergence Time: MC-LAG
Dut-D - 138.203.22.171/172 IOM :sfm-400g

10.10.10.2/32 MDA : m60-10/100eth-tx
SR12
MC-LAG
Mesh_sdp
1/2/1
Ldp sigbnalling
3/2/1
MC-LAG link Dut-A - 138.203.22.116/117
MDA : m60-10/100eth-tx
connection 10.10.10.4
1/1/1
established via SR7
OSPF routing
2/1/58
1/2/1
SR12 Mesh_sdp
10.10.10.6/32 Ldp sigbnalling
2/2/1
Dut-C - 138.203.22.107/108
1/2/1
IOM :sfm-200g
SR7
10.10.10.1/32
Dut-B - 138.203.70.230/231
IXIA Test Tool
 Traffic: 64 Bytes, 100000 pps, 51,2 Mbps, bidirectional Dut-A – Dut-B

 SW: 1433+beta
 Dut-D has 4 peers connected and in total 200 lags active
 Lag-1 between Dut-D and Dut-B used with 1 bidirectional stream for this test
 Node D is local master decided node (lacp-system-prio 1)
 Action performed on the active MC-LAG path unless specified otherwise
Convergence Time: MC-LAG
Traffic from : Traffic from : Traffic from : Traffic from :

Node C =-> Node A => Node C =-> Node A =>
Node A Node C Node A Node C
Link Cut 10ms / 31ms 11ms / 20 ms 9ms / 8 ms 7ms / 6ms
MDA removal 137ms / 113ms / 167ms / 142ms /

132ms 103ms 175ms 130ms
IOM removal 227ms / 168ms / 590ms / 400ms /

222ms 160ms 234ms 132ms
Active CPMa removal – lag stayes 319ms / 214ms / 0,03ms / 0ms 0,03ms / 0ms
active 778ms 522ms
Active CPMb removal – lag stayes 0ms / 0ms 0ms / 0ms 0ms / 0ms 0,11ms / 0ms
active
Active CPM of stdb MC-LAG path 0ms / 0ms 0ms / 0ms 0ms / 0ms 0ms / 0ms
removal – lag stays active.
Active CPM removal when no stdb 2309ms 1549ms 2233ms 1498ms

CPM is there – lag switchover.
Active node boot – lag switchover 1566ms 1051ms 1450ms 973ms
Convergence Time: MC-LAG + MCS
 subscriber-management enabled and syncing between the MC-LAG nodes

 Three streams : HSI, TV and VOIP
 Lag consists out of 4 active and 4 stdb members.
 3000 frames/s
BSM ESM NON SM
Forced switchover via tools command 20ms / 19ms / 25ms / 26ms / 25ms / 14ms /
from ‘master decided’ – active MC-LAG 23ms / 27ms 33ms / 21 ms 12ms / 29ms
node to ‘peer decided’ standby MC-LAG
node.LAG’s over FE.
Switchover when pulling our SFP. HSI : 399ms /

LAG’s over electrical SFP. 405ms
TV : 72ms / 50ms
VOIP : 71ms / 49ms
Convergence Time:
More to come
Appendices
Appendix A: Link Aggregation sub-layer block diagram
Appendix A: LACPDU structure
Appendix A LACP: Bit encoding of the Actor_State and Partner_State fields
1) LACP_Activity: indicates the Activity control value. Active LACP is encoded as a 1; Passive LACP is encoded as a 0.
2) LACP_Timeout: indicates the Timeout control value. Short Timeout is encoded as a 1; Long Timeout is encoded as a
0.
3) Aggregation: If TRUE (encoded as 1), the System considers this link to be Aggregateable; i.e., a potential candidate
for aggregation. If FALSE (encoded as 0), the link is considered to be Individual; i.e., this link can be operated only as an
individual link.
4) Synchronization: If TRUE (encoded as a 1), the System considers this link to be IN_SYNC; i.e., it has been allocated
to the correct Link Aggregation Group, the group has been associated with a compatible Aggregator, and the identity of
the Link Aggregation Group is consistent with the System ID and operational Key information transmitted. If FALSE
(encoded as a 0), then this link is currently OUT_OF_SYNC; i.e., it is not in the right Aggregation.
5) Collecting: TRUE (encoded as a 1) means collection of incoming frames on this link is definitely enabled; i.e., collection
is currently enabled and is not expected to be disabled in the absence of administrative changes or changes in received
protocol information. Its value is otherwise FALSE (encoded as a 0);
6) Distributing: FALSE (encoded as a 0) means distribution of outgoing frames on this link is definitely disabled; i.e.,
distribution is currently disabled and is not expected to be enabled in the absence of administrative changes or changes
in received protocol information.Its value is otherwise TRUE (encoded as a 1);
7) Defaulted is encoded in bit 6. If TRUE (encoded as a 1), this flag indicates that the Actor’s Receive machine is using
Defaulted operational Partner information, administratively configured for the Partner. If FALSE (encoded as a 0), the
operational Partner information in use has been received in a LACPDU;
8) Expired is encoded in bit 7. If TRUE (encoded as a 1), this flag indicates that the Actor’s Receive machine is in the
EXPIRED state; if FALSE (encoded as a 0), this flag indicates that the Actor’s Receive machine is not in the EXPIRED
state.
NOTE—The received values of Defaulted and Expired state are not used by LACP; however, knowing
108 |their values
TiMOS-5.0 can
workshop be 2007
| Januari useful when diagnosing protocol problems.
All Rights Reserved © Alcatel-Lucent 2007
Appendix A: LACPDU ethereal capture
Appendix B: PW-Redundancy Rules
VLL Endpoint Active Transmit Object Selection Rules
1. ICB on ‘X’ endpoint is a backup for the MC-LAG SAP. VLL path active transmit object is the SAP if it is UP and Active,
or ICB if it is UP and did not receive either sdp-binding down status bits or “PW Not Forwarding” status bits.
2. ICB in the ‘Y’ endpoint is a backup for active spoke-sdp. VLL path active transmit object is the spoke-sdp which is UP and shows
remote Active. If active spoke-sdp goes locally down or there is no active spoke-sdp (i.e., all spokes are locally down or showing
remote Standby), we will switch to the ICB spoke-sdp provided it is UP and did not receive sdp-binding down status bits or “PW
not forwarding” status bits. Otherwise, the algorithm selects the best available spoke-sdp among those which are not locally
down based on the following priority:
a. Spoke-sdp with all received T-LDP status bits clear (0x00000000). This means the remote spoke-sdp is active and
both remote SAP and spoke-sdp are operationally UP.
b. Spoke-sdp with received T-LDP status bits indicating remote spoke-sdp is in standby (0x00000020) but both remote SAP and
spoke-sdp are operationally UP.
c. Spoke-sdp with received T-LDP status bits indicating remote spoke-sdp is active and remote SAP is down (0x00000006).
d. Spoke-sdp with with received T-LDP status bits indicating remote spoke-sdp is in standby and remote SAP is down
(0x00000026).
e. Spoke-sdp with received T-LDP status bits indicating remote spoke-sdp is active and in “not forwarding” (0x00000001) state.
f. Spoke-sdp with received T-LDP status bits indicating remote spoke-sdp is in standby and in “not forwarding” (0x00000021)
state.
g. Spoke-sdp with received T-LDP status bits indicating remote soke-sdp is active and is operationally down (0x00000018).
h. Spoke-sdp with received T-LDP status bits indicating remote spoke-sdp is in standby and is operationally down (0x00000038).
i. If multiple objects remain after running steps a-h, the selected active transmit object is the one with the lowest local
precedence value. If multiple objects remain, the selected active transmit object is the one with the lowest spoke-sdp identifier
j. If an active transmit object is found in steps (a-i) but its received T-LDP status bits indicate values other than Active and
operationally UP (0x00000000) and “SAP down” (0x00000006), select ICB spoke-sdp. The ICB must be operationally UP and
did not receive “sdp-binding down status bits or “PW not forwarding” status bits.
k. If an active transmit object is not found in steps (a-i), select ICB spoke-sdp if it is operationally UP and did not receive “sdp-
binding down status bits or “PW not forwarding” status bits.
3. The only two cases where traffic is discarded locally are when all objects on the endpoint are locally down and when the selection
of the active transmit object results in an attempt to forward between two ICB spoke-sdp’s.
Appendix B: PW-Redundancy Rules
 Rules for processing and merging local and received endpoint object Up/Down operational
status:
1. Endpoint ‘X’ is operationally UP if at least one of its objects is operationally UP. It is Down if all its objects are operationally Down.
2. If the SAP in endpoint ‘X’ transitions locally to the Down state, or received a “SAP Down” notification via SAP specific OAM signal, the node must
send T-LDP “SAP Down” status bits on the ‘Y’ endpoint ICB spoke-sdp only. Note that Ethernet SAP does not support SAP OAM protocol in R5.0.
All other SAP types cannot exist on the same endpoint as an ICB spoke-sdp since non Ethernet SAP cannot be part of a MC-LAG instance.
3. If the ICB spoke-sdp in endpoint ‘X’ transitions locally to Down state, the node must send T-LDP “SDP-binding Down” status bits on this spoke-
sdp.
4. If the ICB spoke-sdp in endpoint ‘X’ received T-LDP “SDP-binding Down” status bits or “PW not forwarding” status bits, the node saves this status
and takes no further action. The saved status is used for selecting the active transmit endpoint object.
5. If all objects in endpoint ‘X’ transition locally to Down state, and/or received a “SAP Down” notification via remote T-LDP status bits or via SAP
specific OAM signal, and/or received status bits of “SDP-binding Down”, and/or received status bits of “PW not forwarding”, the node must send
status bits of “SAP Down” over all ‘Y’ endpoint spoke-sdp’s, including the ICB.
6. Endpoint ‘Y’ is operationally UP if at least one of its objects is operationally UP. It is Down if all its objects are operationally Down.
7. If a spoke-sdp in endpoint ‘Y’, including the ICB spoke-sdp, transitions locally to Down state, the node must send T-LDP “SDP-binding Down”
status bits on this spoke-sdp.
8. If a spoke-sdp in endpoint ‘Y’, including the ICB spoke-sdp, received T-LDP “SAP Down” status bits, and/or received T-LDP “SDP-binding Down”
status bits, and/or received status bits of “PW not forwarding”, the node saves this status and takes no further action. The saved status is used
for selecting the active transmit endpoint object.
9. If all objects in endpoint ‘Y’, except the ICB spoke-sdp, transition locally to Down state, and/or received T-LDP “SAP Down” status bits, and/or
received T-LDP “SDP-binding Down” status bits, and/or received status bits of “PW not forwarding”, the node must send status bits of “SDP-
binding Down” over the ‘X’ endpoint ICB spoke-sdp only.
10. If all objects in endpoint ‘Y’ transition locally to Down state, and/or received T-LDP “SAP Down” status bits, and/or received T-LDP “SDP-binding Down”
status bits, and/or received status bits of “PW not forwarding”, the node must send status bits of “SDP-binding Down” over the ‘X’ endpoint ICB spoke-sdp, and
must send a “SAP Down” notification on the ‘X’ endpoint SAP via the SAP specific OAM signal if applicable. An Ethernet SAP does not support signaling status
notifications in R5.0.
Appendix B: PW-Redundancy: Standby signaling capture
Appendix C: Convergence -> Normal situation
PE1 PE3
None
lacIngressFault
epipe lacEgressFault epipe
pwFwdingStandby LAG
LAG SAP X Y SDP SDP X Y SAP
TX-ACTIVE SapOperDown
SDP SDP
None
MC-LAG
MC-LAG
SapOperDown
TX-ACTIVE
lacIngressFault
None lacEgressFault
None pwFwdingStandby
CE5 CE6
SapOperDown
lacIngressFault
lacEgressFault None
pwFwdingStandby None
TX-ACTIVE
TX-ACTIVE
None
SDP lacIngressFault SDP
lacEgressFault
SAP SDP pwFwdingStandby SDP SAP
LAG X Y X Y LAG
epipe SapOperDown epipe
None
PE2 PE4
Appendix C: Convergence -> Access Link Failure
Appendix C: Convergence -> Access Node Failure
PE1 PE3
None
epipe None None epipe
TX-ACTIVE None LAG
LAG SAP X Y SDP SDP X Y SAP
TX-ACTIVE
SDP TX-ACTIVE SDP TX-ACTIVE
MC-LAG MC-LAG
SdpOperDown None
NoIngVCLabel lacIngressFault
NoEgrVCLabel lacEgressFault CE6
CE5
None pwFwdingStandby
SapOperDown
None
SDP SDP
SAP SDP SDP SAP

LAG X Y X Y LAG
epipe SdpOperDown epipe
NoIngVCLabel
NoEgrVCLabel
PE2 None PE4
PE1
epipe
TX-ACTIVE LAG SAP LAG
X Y SAP TX-ACTIVE
SDP SDP
MC-LAG None None

lacIngressFault MC-LAG
lacIngressFault
lacEgressFault lacEgressFault
CE3 CE4
ICB spoke-sdp
TX-ACTIVE TX-ACTIVE
SapOperDown SapOperDown
None None
SDP SDP
SAP SAP
LAG X Y LAG
epipe
PE2
PE1
epipe
TX-ACTIVE LAG SAP X Y LAG
SAP
SDP SDP
MC-LAG SapOperDown TX-ACTIVE

lacIngressFault None MC-LAG
lacEgressFault None
CE3 CE4
ICB spoke-sdp
None SapOperDown
TX-ACTIVE lacEgressFault
SDP SDP
SAP SAP
LAG X Y LAG TX-ACTIVE
epipe
PE2
PE1
epipe
LAG SAP X Y LAG
SAP
SDP SDP
MC-LAG
MC-LAG
CE3 CE4
ICB spoke-sdp
SdpOperDown SdpOperDown
NoEgrVCLabel NoEgrVCLabel
None None
SDP SDP
SAP SAP
TX-ACTIVE LAG X Y LAG TX-ACTIVE
epipe
PE2
PE1
epipe
LAG SAP X Y
MC-LAG
SDP SDP
CE4
TX-ACTIVE SapOperDown
SapOperDown lacIngressFault
None lacEgressFault
ICB spoke-sdp pwFwdingStandby
None
lacIngressFault
lacEgressFault PE2
epipe
TX-ACTIVE
LAG
SDP X Y SAP
SAP SDP MC-LAG

TX-ACTIVE
LAG
None CE5
lacIngressFault
lacEgressFault
SapOperDown ICB spoke-sdp
lacIngressFault SapOperDown
lacEgressFault None
pwFwdingStandby TX-ACTIVE
epipe
LAG
X Y SAP
SDP SDP
PE3
PE1
epipe
LAGSAP X Y
MC-LAG
SDP SDP
CE4
TX-ACTIVE
SapOperDown SapOperDown
lacIngressFault None
lacEgressFault
ICB spoke-sdp
SapOperDown
lacIngressFault
lacEgressFault PE2
epipe
LAG
SDP X Y SAP
TX-ACTIVE SAP SDP MC-LAG

LAG
TX-ACTIVE CE5
None
None
None ICB spoke-sdp
lacIngressFault None
lacEgressFault None
pwFwdingStandby TX-ACTIVE
epipe
LAG
X Y SAP
TX-ACTIVE
SDP SDP
PE3
Appendix C: Convergence -> Double Access Link Failure
PE1
epipe
TX-ACTIVE
LAG SAP X Y
MC-LAG
SDP SDP
CE4 None TX-ACTIVE
lacIngressFault
None
lacEgressFault
None
ICB spoke-sdp
SapOperDown
None
TX-ACTIVE PE2
epipe
LAG
SDP X Y SAP
SAP SDP MC-LAG

LAG
TX-ACTIVE CE5
SapOperDown
None
ICB spoke-sdp
None None
TX-ACTIVE lacEgressFault
epipe
LAG
X Y SAP
SDP SDP TX-ACTIVE
PE3
PE1
epipe
TX-ACTIVE
LAG
X Y SAP MC-LAG
TX-ACTIVE SDP
None SDP SDP
None CE6
None None
ICB spoke-sdp lacEgressFault
None
SapOperDown None
None TX-ACTIVE
None epipe
None
None
TX-ACTIVE SDP SDP
None
PE5
epipe SDP LAG
SAP
X Y TX-ACTIVE X Y
SDP
PE2
None
SDP PE3
SAP pwFwdingStandby
epipe
TX-ACTIVE SDP
None LAG
None None X Y SAP TX-ACTIVE
SDP SDP MC-LAG
None
SDP SDP
TX-ACTIVE
None CE7
None None
pwFwdingStandby
None
SapOperDown None
None TX-ACTIVE
epipe
None
None SDP SDP
SDP LAG
TX-ACTIVE SAP
X Y
PE4
PE1
epipe
LAG
X Y SAP MC-LAG
TX-ACTIVE SDP
None
SDP SDP
None CE6
SapOperDown TX-ACTIVE
None None
ICB spoke-sdp None
None
None lacEgressFault
None epipe
pwFwdingStandby
None
None SDP SDP TX-ACTIVE
PE5 SDP
epipe LAG
TX-ACTIVE SAP
X Y TX-ACTIVE X Y
SDP
PE2
None
SDP None PE3
SAP
epipe
SDP
TX-ACTIVE None LAG TX-ACTIVE
None None X Y SAP
SDP SDP MC-LAG
None
SDP SDP
TX-ACTIVE
None
None None CE7
pwFwdingStandby
None
SapOperDown None
None None TX-ACTIVE
epipe
None
SDP SDP
TX-ACTIVE
SDP LAG
SAP
X Y
PE4
PE1
epipe
LAG
X Y SAP MC-LAG
SDP
SDP SDP
CE6
SdpOperDown ICB spoke-sdp SdpOperDown

SdpOperDown NoEgrVCLabel NoEgrVCLabel
NoIngVCLabel None epipe None
NoEgrVCLabel None
None None SDP SDP TX-ACTIVE
PE5 SDP
epipe LAG
SAP
TX-ACTIVE
X Y TX-ACTIVE X Y
SDP
PE2
None
SDP None PE3
SAP
epipe
TX-ACTIVE SDP
None LAG
None None X Y SAP TX-ACTIVE
SDP SDP MC-LAG
None
TX-ACTIVE SDP SDP
None CE7
None None
pwFwdingStandby
None
SapOperDown
None
None TX-ACTIVE
None
epipe
None
TX-ACTIVE SDP SDP

SDP LAG
SAP
X Y
PE4
Appendix C: Convergence -> BRAS Node Failure
PE1
epipe
LAG
X Y SAP MC-LAG
TX-ACTIVE SDP
None SDP SDP
None CE6
SapOperDown None TX-ACTIVE
None lacEgressFault
lacIngressFault epipe
lacEgressFault None
None SDP SDP
PE5 SDP
epipe LAG
SAP
X Y TX-ACTIVE X Y
SDP
None
lacIngressFault PE2
SDP
SAP lacEgressFault PE3
pwFwdingStandby epipe
TX-ACTIVE SDP
None LAG
None X Y TX-ACTIVE
None SAP
SDP SDP MC-LAG
None
TX-ACTIVE SDP SDP
None TX-ACTIVE
None None CE7
pwFwdingStandby
None
SapOperDown None
None None TX-ACTIVE
None epipe
TX-ACTIVE SDP SDP

SDP LAG
SAP
X Y
PE4
Appendix C: Convergence -> BRAS Node Failure + Access Link Failure
PE1
epipe
LAG
X Y SAP MC-LAG
TX-ACTIVE SDP
None
SDP SDP
None CE6
None
lacIngressFault epipe lacEgressFault
lacEgressFault
pwFwdingStandby None
None SDP SDP
PE5 SDP
epipe LAG
SAP
X Y X Y
SDP TX-ACTIVE
PE2
None
SDP lacIngressFault PE3
SAP
lacEgressFault epipe
TX-ACTIVE SDP
None LAG
None X Y SAP
SDP None MC-LAG
SDP
pwFwdingStandby
SDP SDP
TX-ACTIVE
None CE7
None SapOperDown TX-ACTIVE
None None
TX-ACTIVE ICB spoke-sdp None
None
None
lacIngressFault
None None
epipe lacEgressFault
None
SDP SDP
TX-ACTIVE
SDP LAG TX-ACTIVE
SAP
X Y
PE4
www.alcatel-lucent.com

Timos 5.0 MC Lag PW Red

Uploaded by

Copyright:

Available Formats

Timos 5.0 MC Lag PW Red

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Timos 5.0 MC Lag PW Red

Uploaded by

Copyright:

Available Formats

TiMOS-5.

All Rights Reserved © Alcatel-Lucent 2006, #####

1. 5.0 Redundancy features overview

2. LAG & LACP Basics

4. Dual homing in TPSDA – L2-CO model

2 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

3 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

4 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

5 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

EPIPE is point-to-point service

6 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

 BSA link and node protection  MC-LAG

7 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

 BSA link and node protection  MC-LAG

8 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

 SRRP is Alcatel proprietary extension to VRRP BSR

 main goal is to provide redundancy for subscriber-interfaces

9 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

configure service ies subscriber-interface miro

10 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

At this moment (pre-5.0R4):

TPSDA dual homing to 2 nodes:

TPSDA dual homing to 1 node:

ESM not supported on ATM interfaces!

11 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

12 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

Link Aggregation Group:

 Aggregates up to 8 Ethernet ports in 1 LAG

13 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

14 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

*A:PE2# show lag 2 detail

15 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

configure [no] lag lag-id

16 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

 Port-threshold 1 action down, no dynamic-cost  Port-threshold 1 action down, dynamic-cost:

 Port-threshold 1 action dynamic-cost, dynamic-cost

17 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

 Based on IEEE 802.3ad

18 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

19 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

*A:PE2>config>lag# info detail

20 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

21 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

configure [no] lag lag-id

22 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

23 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

24 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

*A:CE5# debug lag

25 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

26 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

27 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

A:CE1>config>lag# info 1/1/4

28 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

1/1/3 *A:PE2>config>lag# info

29 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

MC Peer Address : 10.0.0.2 MC Peer Lag-id : 1

30 | TiMOS-5.0 workshop | Januari 2007 All Rights Reserved © Alcatel-Lucent 2007

MC Peer Address : 10.0.0.1 MC Peer Lag-id : 1