Coyote Point Systems E350GX, Руководство по установке и администрированию

Страница: 37 / 344

Where Do I Go Next?
Equalizer Installation and Administration Guide 37
In the sample failover configuration above, there is no single point of failure. If a router goes down, the other router
takes over; if a link fails, requests are routed through another link. In this dual network configuration, the Equalizers
communicate over both the internal and external subnets. The connection shown between the Equalizers on the
internal interface might not be necessary in a single network configuration, or if the Equalizers can communicate
over the internal interface through an existing route.
The backup Equalizer monitors all traffic to and from the primary unit; both Equalizers periodically exchange status
messages over the local area network. The Equalizers also exchange current configuration information. When you
update the configuration on either machine, the configuration on its peer is automatically updated.
Should either Equalizer fail to respond to a status message probe, the other Equalizer begins a diagnostic cycle and
attempts to contact its peer via the other network ports. If these attempts fail, the peer is considered to be
down .
When the backup Equalizer determines that its failover peer is down, it initiates a failover process:
1. The backup Equalizer configures the virtual cluster aliases on the external port and sends out “gratuitous ARP”
packets that instruct any external-network routers to replace ARP table entries that point to the physical address
of the failed Equalizer with the physical address of the backup unit.
2. The backup Equalizer configures a failover gateway alias on the port that is local to the servers.
• With no backup configuration, the servers use the IP address of the cluster or external port as their
default gateway.
• In a hot-backup environment, the gateway address can migrate between the primary and backup unit.
This requires an additional address.
3. The Equalizer kernel changes from BACKUP mode to PRIMARY mode. The PRIMARY-mode Equalizer
performs gateway routing of packets between its cluster and external ports, address translation, and load
balancing.
When a failed unit is brought back online, it begins to exchange status messages with its failover peer. Once both
Equalizers have synchronized, the newly-started unit assumes the backup role.
For more information on setting up Failover, please see “Setting Up a Failover Configuration” on page 90.
Where Do I Go Next?
Chapter 2 through Chapter 6 tell you how to install Equalizer into your network and configure clusters and servers.
The remainder of the manual basically provides additional information to use and administer the features introduced
in these chapters:
• Chapter 2, “Installing and Configuring Equalizer Hardware”, provides comprehensive instructions for
installing Equalizer hardware and setting up Equalizer to work with your networks and servers.
• Chapter 3, “Using the Administration Interface” , discusses how to use Equalizer’s HTML-based
administration interface, including adding administrative logins with distinct permissions.
• Chapter 5, “Configuring Equalizer Operation” , tells you how to configure system and global resources
through the Equalizer Administration Interface, including setting up a failover configuration.
• Chapter 4, “Equalizer Network Configuration” , covers VLAN and switch port configuration, as well as
how to define static routes on Equalizer.
• Chapter 6, “Administering Virtual Clusters” , tells you how to add and remove virtual clusters and servers,
changing load balancing options, and shutting down servers.
If you are also deploying the optional Envoy Geographic Server Load Balancing product, refer to Chapter 9,
“Administering GeoClusters” , for Envoy configuration specifics.