
White Paper
All contents are Copyright © 1992–2007 Cisco Systems, Inc. All rights reserved. This document is Cisco Public Information. Page 45 of 51
place in one to three seconds and requires no network reconvergence for Layer 2, 3, and 4
protocols. From a router perspective, the MSFC engines can be configured for redundancy as well
with hybrid software. For more information on hybrid High Availability, refer to the white paper at
http://www.cisco.com/warp/public/cc/pd/si/casi/ca6000/tech/hafc6_wp.pdf.
Cisco IOS Software on the Cisco Catalyst 6500 supports Route Processor Redundancy (RPR,
also known as Enhanced High System Availability [EHSA]), Route Processor Redundancy Plus
(RPR+), and Nonstop Forwarding with Stateful Switchover (NSF/SSO). In this operational model,
one supervisor/MSFC pair is fully operational and the other pair is in standby mode. The show
module command lists the current and standby supervisors. There are heartbeat messages
between two pairs to help ensure rapid failure detection. The RPR and RPR+ implementations do
not synchronize protocol state between the two supervisor engines; however, the SSO
implementation synchronizes protocol state. The following section provides an overview of
supervisor redundancy characteristics that are equivalent between RPR, RPR+, and SSO.
In Cisco IOS Software, the supervisor and MSFC are each responsible for different functions and
protocols (Layer 2 vs. Layer 3). However, the system is dependent on both engines being available
for proper operation. Failure of either the supervisor or the MSFC in RPR/RPR+/SSO mode will
cause a switchover from the active supervisor to the standby supervisor/MSFC. Note that in the
Cisco Catalyst OS, the supervisor can remain fully operational if one MSFC fails; an MSFC failure
does not necessarily cause a supervisor switchover, but may cause an MSFC failover only,
allowing a cross model where the active PFC and the switch processor running the Cisco Catalyst
OS are fully functional in slot 1 whereas the route processor/MSFC are fully functional in slot 2.
There is no stateful protocol redundancy between supervisor engines with RPR or RPR+. The
SSO redundancy mode provides the stateful protocol redundancy between supervisor engines in
Cisco IOS Software and is equivalent in terms of functionality to the Cisco Catalyst OS High
Availability redundancy mode.
The following section provides an overview of supervisor redundancy characteristics with RPR,
RPR+, and NSF/SSO.
Route Processor Redundancy (RPR)
With RPR enabled, the active supervisor and MSFC are operational and responsible for all packet
forwarding and features. The standby supervisor and MSFC are out of reset, but not all
subsystems are booted. The standby supervisor is booted to the point where the gigabit uplink
ports are operational, but no protocols are running on the supervisor or MSFC.
Upon an active supervisor failure, RPR detects the loss of the active supervisor and causes a
switchover. The line cards are power-cycled, the supervisor and MSFC finish booting, and all
Layer 2 and Layer 3 protocols are initialized. The failover time for the system to start forwarding
traffic in EHSA is approximately 90 seconds. The actual failover time is dependent on the size and
complexity of the configuration.
With RPR, the startup configuration and boot variables are synchronized between the active and
standby supervisor.
Route Processor Redundancy Plus (RPR+)
With RPR+ enabled, the active supervisor and MSFC are operational and responsible for all
packet forwarding and features. The standby supervisor and MSFC are fully booted and running
on standby. RPR+ is an enhancement to RPR. Since the standby supervisor is further along in the
boot process, RPR+ provides a faster supervisor failover than RPR. In addition, the line card state
Komentáře k této Příručce