Cisco Systems 6503 manual SRM and IP Multicast, Supervisor and MSFC Failover Tests

Page 16
SRM and IP Multicast

SRM and IP Multicast

With SRM, the VLAN interfaces on the nondesignated router are in a down state. Even after a failover, these interfaces will not move into an up state until the supervisor engine verifies that the VLAN has at least one connected physical interface in the forwarding state. This interruption causes the supervisor engine to delete all multicast entries in the PFCx and this disrupts multicast forwarding. As an enhancement to the original SRM implementation, Cisco Catalyst OS release 7.1 provides support for IP Multicast stateful redundancy. When SRM is enabled in Cisco Catalyst OS version 7.1, the multicast flows are preserved during a failover.

Supervisor and MSFC Failover Tests

A test environment was created to demonstrate sample high-availability switchover scenarios and record the corresponding failover times. The test setup included one Cisco Catalyst 6509 chassis with dual supervisor engine 2 line cards running Cisco Catalyst OS version 7.2.2 or MSFC2 hardware running Cisco IOS Software Release 12.1(11b)E4. These tests were intended to be basic and easily demonstrated. The testing mechanism was pings between two end devices connected directly to the switch. Spanning-Tree was enabled on the ports in use for all scenarios. Tests were initiated by resetting the switch supervisor engine. Each scenario was tested eight times and the results were averaged.

Layer 2 Failover

For Layer 2 traffic between two end stations within the same VLAN , the failover resulted in a one or two ping timeout (approximately one- or two-second failover time).

Note: This test can be performed on a supervisor engine without an MSFC.

Layer 3 Failover

For Layer 3 traffic, a common test environment was created for measuring the supervisor or MSFC failover of a single Cisco Catalyst 6500 Series configured first with DRM and then with SRM. The two pinging devices were placed in separate VLANs. The basic software configuration included enabling the High Availability feature on the Cisco Catalyst OS and then either DRM or SRM redundancy in the Cisco IOS Software on the MSFC2. The complete configurations are shown below.

SRM

hostname SRM

!

redundancy high-availability single-router-mode

!

interface Vlan20

ip address 10.20.1.3 255.255.255.0 no ip redirects

!

interface Vlan30

ip address 10.30.1.3 255.255.255.0 no ip redirects

!

end

Cisco Systems, Inc.

All contents are Copyright © 1992–2002 Cisco Systems, Inc. All rights reserved. Important Notices and Privacy Statement.

Page 16 of 19

Image 16
Contents High Availability for the Cisco Catalyst 6500 Series Switches White PaperSwitching Fabric Failover Switch Fabric Module RedundancySwitching Fabric Operation Example 1 Fabric-Enabled System CLI output description for show fabric channel switchmodeExample 2 Classic and Fabric-Enabled System Supervisor Fast Switchover Supervisor Engine SwitchoverRedundant Supervisor Engines Supervisor Stateful Protocol Redundancy Supervisor High Availability FeatureSup-A enable set system highavailability enable and incompatible Supervisor Engine Software Image Upgrades Supervisor Engine Image SynchronizationProtocol Database Standby SupervisorSupervisor Engine Versioning Feature Sup-A enable show system highavailabilityCisco Catalyst OS Image Upgrade Procedure 6. Sup-A enable set boot system flash bootflashcat6000-sup2k8.7-2-2.binMSFC High Availability Features Dual Router Mode MSFC Configuration SynchronizationDRM Operation WAN Interfaces in DRM DRM ChallengesSingle Router Mode SRM OperationSRM Configuration and Conversion Procedure WAN Interfaces in SRMSRM and IP Multicast Supervisor and MSFC Failover TestsRedundant Power Supplies Conclusion Corporate Headquarters European HeadquartersAmericas Headquarters Asia Pacific Headquarters