TMS zl Management and Configuration Guide ST.1.1.100430

8-2
High Availability
Overview
Overview
High availability (HA) is a strategy for minimizing network downtime so that
users can access the network with minimal interruption in the event that a
network device fails. The best approach for providing HA for the Threat
Management Services (TMS) zl Module is to implement an HA cluster—a
group of modules that can take over the workload of another module if it fails.
Two TMS zl Modules can be clustered for HA. Most connection-state informa-
tion is synchronized between the members of the cluster for stateful failover;
that is, most of the active connections that were being processed by one
TMS zl Module will fail over to the other member without significant interrup-
tion. To see what is synchronized between cluster members, refer to
Table 8-1 on page 8-6.
The two modules can be in the same switch chassis (intra-chassis cluster) or
in two different switch chassis (inter-chassis cluster). Only one HA cluster per
switch chassis is supported.
When HA is enabled, the module’s internal port 2 becomes an untagged
member of the HA VLAN and is dedicated to HA traffic. All messages such as
HA control protocol, synchronization, and HA management messages are
communicated through this interface.
Note HA is supported in routing mode only, so both of your modules must be in
routing mode. If you require redundancy for TMS zl Modules in monitor mode,
you can mirror the same traffic to two modules.
In a cluster, the TMS zl Modules operate in active-standby mode.
Active-Standby Mode
In active-standby mode, data processing is done by the master, and the other
module acts as a passive backup to the master. The other member (partici-
pant) becomes active only when the master fails and it is elected the new
master.