
To improve Dispatcher availability, the Dispatcher high availability functions
as follows:
- Two Dispatchers with connectivity to the same clients, and the same cluster
of servers, as well as connectivity between the Dispatchers. Both Dispatchers
must run on the same type of operating system and platform.
- A “heartbeat” mechanism between the two Dispatchers detects a Dispatcher
failure. At least one heartbeat pair must have the NFAs of the pair as the
source and destination address. If possible, at least one of the heartbeat
pairs should be across a separate subnet than the regular cluster traffic.
Keeping the heartbeat traffic distinct will help prevent false takeovers during
very heavy network loads and also improve complete recovery times after a
failover.
- A list of reach targets, addresses that both Dispatcher machines must
be able to contact in order to load balance traffic normally. For more information,
see Detecting server failures with heartbeats and reach targets.
- Synchronization of the Dispatcher information
- Logic to elect the active Dispatcher which is in charge of a given cluster
of servers, and the standby Dispatcher which continuously gets synchronized
for that cluster of servers.
- A mechanism to perform IP takeover, when the logic or an operator decides
to switch active and standby.
Planning for high availability
When configuring
for high availability, consider that the Load Balancer machine is supported
with the following limitations or special considerations:
- If you are using IPv6 protocol on your machine and want to use high availability,
you must check to see if protocol 58 is defined to be ICMPv6 in the protocol
file.
- In the high availability configuration, both primary and backup machines
must be on the same subnet with identical configuration.
- The heartbeat pairs (which is the mechanism between the primary and standby
Dispatchers to detect Dispatcher failure) must be both IPv4 format or both
IPv6 format.
- In a high availability or a stand-alone environment, you must not alias
the cluster address against the network adaptor.
- The HighAvailChange script can be moved from the install_root/servers/samples directory to the install_root/servers/bin directory
to log high availability state changes for the Dispatcher machine, but this
script does not need to be changed.