Lenovo Flex System Fabric CN4093, Application Manual

Page: 516 / 634

516
CN4093 Application Guide for N/OS 8.4
Auto Monitoring LAG Links
Layer 2 Failover can be enabled on any LAG in the CN4093, including LACP
LAGs. LAGs can be added to failover trigger groups. Then, if some specified
number of trigger links fail, the switch disables all the internal ports in the switch
(unless VLAN Monitor is turned on). When the internal ports are disabled, it
causes the NIC team on the affected server blades to failover from the primary to
the backup NIC. This process is called a failover event.
When the appropriate number of links in a trigger group return to service, the
switch enables the internal ports. This causes the NIC team on the affected server
blades to fail back to the primary switch (unless Auto ‐ Fallback is disabled on the
NIC team). The backup switch processes traffic until the primary switch’s internal
links come up, which can take up to five seconds.
VLAN Monitor
The VLAN Monitor allows Layer 2 Failover to discern different VLANs. With
VLAN Monitor turned on:
 If enough links in a trigger fail (see “Setting the Failover Limit” on page 518 ), the
switch disables all internal ports that reside in the same VLAN membership as
the LAG(s) in the trigger.
 When enough links in the trigger return to service, the switch enables the
internal ports that reside in the same VLAN membership as the LAG(s) in the
trigger.
If you turn off the VLAN Monitor (
CN 4093# no failover vlan
), only one
failover trigger is allowed. When a link failure occurs on the trigger, the switch
disables all internal server ‐ blade ports.
Auto Monitor Configurations
Figure 57 is an example of Layer 2 Failover. One CN4093 is the primary and the
other is used as a backup. In this example, all external ports on the primary switch
belong to a single LAG, with Layer 2 Failover enabled and Failover Limit set to 2. If
two or fewer links in trigger 1 remain active, the switch temporarily disables all
internal server ‐ blade ports that reside in VLAN 1. This action causes a failover
event on Server 1 and Server 2.