Dell BigIron RX Series, Руководство по настройке

Страница: 104 / 1458

28 BigIron RX Series Configuration Guide
53-1002253-01
How management module redundancy works
3
Syslog and SNMP traps
When a switchover occurs, the BigIron RX system sends a Syslog message to the local Syslog buffer
and also to the Syslog server, if you have configured the system to use one. In addition, if you have
configured an SNMP trap receiver, the system sends an SNMP trap to the receiver.
When the system is powered on or otherwise reset normally, the system sends a cold start
message and trap. However, if the system is reset as the result of switchover to the standby
management module, the system instead sends a warm start message and trap.
MAC address changes
The MAC addresses in theBigIron RX Series system are based on the MAC address of the BigIron
RX Series devices. During switchover, the system's MAC addresses change and the system sends
out gratuitous ARP requests to flush the old MAC addresses from the ARP caches on attached IP
devices, and update the caches with the system’s new MAC addresses.
Layer 2 Hitless Failover
The Layer 2 Hitless Failover feature provides automatic failover from the active management
module to the standby management module without interrupting operation of any interface
modules in the device. Configuration changes made from the CLI to the active management
module are also written to the standby management module even if they are not written to flash
memory.
NOTE
Since both the standby and active management modules run the same code, a command that
brings down the active management module will most likely bring down the standby management
module. Because all configuration commands are synchronized from active to standby
management module in real time, both management modules will crash at almost the same time.
This in turn causes the system to reset all interface modules (similar to the behavior when the
'reboot' command is executed) and causes packet loss associated with a system reboot.
Once booted, the redundant management module keeps up-to-date copies of the active module's
running configuration. Layer 2 protocols such as STP, RSTP, MRP, and VSRP are run concurrently on
both the active and standby management modules. Upon the failover of the active management
module, the standby module takes over as the active management module and picks up where the
active module left off, without interrupting any Layer 2 traffic.
The interface modules are not reset, as they are with the previous cold-restart redundancy feature.
The interface modules continue to forward traffic while the standby management module takes
over operation of the system. The new now-active management module receives updates from the
interface modules and sends verification information to the interface modules to ensure that they
are synchronized.
If the new active management module becomes out-of-sync with an interface module, information
on the interface module can be overwritten in some cases which can cause an interruption of
traffic forwarding. Layer 3 hitless failover is not supported in this release. Consequently, a failover
will result in a re-synchronization of Layer 3 data structures