Cisco Unified MeetingPlace, Release 7.0 -- Application Server Failover

From DocWiki

Revision as of 23:35, 17 December 2009 by Docwikibot (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Main page: Cisco Unified MeetingPlace, Release 7.0

Up one level: Configuration

You can deploy two Application Servers as part of one logical site. Database replication enables the Application Servers to synchronize user profiles, user groups, and meeting data. Only one Application Server is active at any time. If the active server fails, then you can activate the standby server and place the previously active server in standby mode. Each Application Server in a site is called a node.

If the site includes two Media Servers, one Media Server will be associated with each Application Server. If the site includes only one Media Server, it is always associated with the active Application Server.

In a failover deployment, each Application Server is configured with two IP addresses-one for each of the following interfaces:

  • eth0 -- Physical network interface
    • Assign the same shared hostname and IP address to eth0 on both Application Servers.
    • Anyone who tries to access the shared hostname or IP address will reach the active Application Server.
    • The eth0 interface is disabled on the standby Application Server.
  • eth0:0 -- Virtual network interface
    • Assign a unique hostname and IP address to eth0:0 on each Application Server.
    • Use the eth0:0 hostname or IP address to access an Application Server that is in standby mode.
    • The system uses the eth0:0 virtual network interface for database replication between the two nodes.

Note: The interface eth0 mentioned throughout this module actually refers to the physical Ethernet port 1 on the Application Server.

When a node is rebooted, interface eth0 is initially disabled. However, during startup of the Cisco Unified MeetingPlace application, the node checks for an active Application Server on the network by attempting to ping the eth0 IP address, which is identical on both nodes. If the ping is unsuccessful, the rebooted node enables the eth0 interface and becomes the active server.

Related Topics

Rating: 0.0/5 (0 votes cast)

Personal tools