Tuesday, May 29, 2012

Failover Server

A Secondary installation of the CONTROL-M/Server resides on a separate computer. Typically the database of the secondary CONTROL-M/Server, created as part of the CONTROL-M/Server installation, acts as a mirror database.

The secondary CONTROL-M/Server, including its database, acts as a backup to the primary CONTROL-M/Server. This secondary CONTROL-M/Server computer is sometimes called the Mirror Environment.

The secondary CONTROL-M/Server duplicates all of the primary CONTROL-M/Server’s links to CONTROL-M/Agents, and CONTROL-M/Server Configuration Agent. The secondary CONTROL-M/Server is defined in each agent’s list of authorized server hosts.

During normal operation, the secondary database is constantly updated to remain identical with the primary database. The secondary CONTROL-M/Server remains inactive.

If the primary CONTROL-M/Server fails due to a hardware/software problem on the primary computer, the administrator activates failover on the secondary CONTROL-M/Server. The secondary CONTROL-M/Server performs all normal CONTROL-M/Server functions until the primary CONTROL-M/Server is repaired.

Jobs submitted to agent computers before failover continue executing. The secondary CONTROL-M/Server polls the agent computers to determine the status of jobs listed in the Active Jobs file.

When the primary CONTROL-M/Server is stable, normal operation is restored. The failover server is shut down, the primary database is restored, and the primary CONTROL-M/Server is restarted. The secondary server returns to passive readiness.

Implementing a Failover Server

A failover server can be implemented any time after CONTROL-M/Server has been installed.

The secondary CONTROL-M/Server installation should be built on a separate computer and database server, independent of the computers that host the primary CONTROL-M/Server and its database.

Several administrative functions are used to initialize and use the failover server, and restore the primary computer. These functions are accessed from the Database Mirroring menu of the CONTROL-M/Server Main Menu.

Preparing a Failover server

Verify the following conditions for the secondary installation:

The Agent-to-Server Port Number and Server-to-Agent Port Number communication parameters must be the same on both CONTROL-M/Server computers. These ports must also be dedicated to the use of CONTROL-M/Server. You should also ensure that both computers are using the same date.

  1. The failover server can be configured to use a dedicated or a non-dedicated SQL database server. However, it should not use the same instance of the database server used by the primary environment.

  2. The Database (Data Portion) Size parameter should be assigned the same value as the current size of the primary database.

  3. All agent computers that are defined on the primary CONTROL-M/Server should also be defined on the failover server. These definitions must be identical. If you installed CONTROL-M/Agent on the primary CONTROL-M/Server computer, it should also be installed on the failover server. The parameters should be the same on both CONTROL-M/Server installations.

The host name of the failover server must be added to the list of authorized hosts of each agent.


Unknown said...

what happens to job running on Control-M Server during failover?
if that was submitted right before failover.

Mahi said...

Job will run normally if its submitted already as it will be run by the Agent... once the failover is done the job status will be updated.

Post a Comment

Twitter Delicious Facebook Digg Stumbleupon Favorites More

Design by Awesome Deals | Bloggerized by Mahendra - 24x7FunOnline