Friday, March 2, 2012

MC-LAG


This provides node and link level redundancy to the CE.

This uses UDP port 1025 and can be routed.

The CE device is unaware of the MC lag only Lag needs to be configured on the CE device.

When configuring  mc-lag one side should be active while the other side should be in passive mode.

On the CE all the links should be active however the links going to the standby PE should be in an operational down state.

MC-LAG connection is made at the ip level so it is not required for both the nodes to be directly connected.

A keep-alive mechanism is present between the active and the passive nodes which can detect the failure pretty easily.

Authentication can also be configured between MC lag peers but the configuration is not mandatory.

Lag id configured in both the boxes are generally same if different lag-id are configured then they should be notified by a remote-lag id command.

Given below are the examples:-




When configuring  mc-lag one side should be active while the other side should be in passive mode.

On the CE all the links should be active however the links going to the standby PE should be in an operational down state.

MC-LAG connection is made at the ip level so it is not required for both the nodes to be directly connected.

A keep-alive mechanism is present between the active and the passive nodes which can detect the failure pretty easily.

Authentication can also be configured between MC lag peers but the configuration is not mandatory.

Lag id configured in both the boxes are generally same if different lag-id are configured then they should be notified by a remote-lag id command.





When configuring  mc-lag one side should be active while the other side should be in passive mode.

On the CE all the links should be active however the links going to the standby PE should be in an operational down state.

MC-LAG connection is made at the ip level so it is not required for both the nodes to be directly connected.

A keep-alive mechanism is present between the active and the passive nodes which can detect the failure pretty easily.

Authentication can also be configured between MC lag peers but the configuration is not mandatory.

Lag id configured in both the boxes are generally same if different lag-id are configured then they should be notified by a remote-lag id command.








Hold on neighbor failure indicates that if 3 KA messages are missed the neighbor is considered as down.

*A:SRC_R2>config>redundancy# info
----------------------------------------------
        multi-chassis
            peer 10.10.10.1 create
                mc-lag
                    lag 1 lacp-key 1 system-id 00:00:00:00:00:01 system-priority
 100
                    no shutdown
                exit
                no shutdown
            exit
        exit
----------------------------------------------
*A:SRC_R2>config>redundancy#



*A:SRC_R1>config>lag# show lag 1 detail

===============================================================================
LAG Details
===============================================================================
Description        : N/A
-------------------------------------------------------------------------------
Details
-------------------------------------------------------------------------------
Lag-id              : 1                     Mode                 : access
Adm                 : up                    Opr                  : down
Thres. Exceeded Cnt : 3                     Port Threshold       : 0
Thres. Last Cleared : 01/22/2011 14:45:20   Threshold Action     : down
Dynamic Cost        : false                 Encap Type           : null
Configured Address  : 78:64:ff:00:01:41     Lag-IfIndex          : 1342177281
Hardware Address    : 78:64:ff:00:01:41     Adapt Qos (access)   : distribute
Hold-time Down      : 0.0 sec               Port Type            : standard
Per FP Ing Queuing  : disabled
LACP                : enabled               Mode                 : active
LACP Transmit Intvl : fast                  LACP xmit stdby      : enabled
Selection Criteria  : highest-weight        Slave-to-partner     : disabled
Number of sub-groups: 1                     Forced               : -
System Id           : 78:64:ff:00:00:00     System Priority      : 32768
Admin Key           : 32768                 Oper Key             : 1
Prtr System Id      : 78:68:ff:00:00:00     Prtr System Priority : 32768
Prtr Oper Key       : 32768

MC Peer Address     : 10.10.10.2            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    : standby
MC Lacp ID in use   : true                  MC extended timeout  : false
MC Selection Logic  : local master decided
MC Config Mismatch  : no mismatch

-------------------------------------------------------------------------------
Port-id        Adm     Act/Stdby Opr     Primary   Sub-group     Forced  Prio
-------------------------------------------------------------------------------
1/1/4          up      standby   down    yes       1             -       32768

-------------------------------------------------------------------------------
Port-id        Role      Exp   Def   Dist  Col   Syn   Aggr  Timeout  Activity
-------------------------------------------------------------------------------
1/1/4          actor     No    No    No    No    No    Yes   Yes      Yes
1/1/4          partner   No    No    No    No    Yes   Yes   Yes      No
===============================================================================
*A:SRC_R1>config>lag#


MAC flush occurs during mc-lag switchovers.




No comments:

Post a Comment