Foundry NetIron M2404C and M2404F Metro Access Switches
Configuring Interfaces (Rev. 03)
Network Wide Resilience
© 2008 Foundry Networks, Inc.
Page 46 of 57
Network Wide Resilience
The goal of the Resilient Link is to enable carrier-grade reliability across an Ethernet network
(Carrier-grade is a term that refers to public network telecommunications products that require
99.999% to 99.9999% reliability, which translates into between 30 seconds and 5 minutes of
downtime per year). This requires the application Resilient Link software in order to deliver
recovery from network faults without interrupting network applications or users.
The Resilient Link feature enables the user to protect critical links and prevent network downtime
if those links fail. A resilient link consists of a main link and a standby link, which together form a
resilient-link pair. If the main link fails, the standby link immediately and automatically takes over
the task of the main link.
The Resilient Link can operate between a Foundry Networks switch and other devices in the
network, such as router switches and servers, providing a fully redundant network and ensuring that
there will be no link failures.
During normal operation, the main port is enabled and the standby port is disabled. If the main link
fails, the main port is disabled and the standby port is enabled. If the main link becomes operational,
the user can then re-enable the main port and disable the standby port again.
Switchover time to the backup link is less than 1 second, which ensures that no session timeouts
take place and that system timeouts are prevented.
NOTE
Resilient Link feature cannot be enabled on the M2404 enhanced uplink ports.
Overview
The Resilient Link can be deployed on the enterprise backbone, enterprise premises edge and/or
service provider edge. This provides resilience across all segments of a network, delivering fault
recovery so fast that a fault is actually transparent to network applications or users because of the
near-zero packet loss in the network.
Preferred Port
A port is said to be
preferred
if it is always the main port as long as it has a link. Traffic is switched
back to the main port once its connection is recovered.
There are several ways to establish the preferred port:
•
The port with the higher bandwidth (the port operational speed) gets the preference
automatically, although it does not show in the Resilient Link display commands. The user can
change these settings.
NOTE
The choice of the preferred port is determined by comparing the ports’ operational
speed. Therefore, both ports must be connected before the preferred port can be
chosen. If, for example a 100M port is connected to a 100M port and a Giga port is
connected to a 100M port and both are in autonegotiate mode, the Giga port cannot
be set as preferred.