Legacy platform

Cold site recovery

You can perform a cold site recovery, though it might be time consuming.

A cold site recovery can be daunting, especially for a large complex system like Sterling™ Order Management System Software. At a minimum, you have to procure, install, and configure all the hardware equipment needed by the application ranging from network equipment, load balancers, mid-tier and database nodes, SAN, cabling for the SAN, and so forth.

Next, you have to install and configure all the system software ranging from the operating system, database management system, application server, Sterling Order Management System Software, and so forth. It is critically important that the software version and release, and even the same fix packs be installed the same as the primary data center. Installing different software versions may result in unexpected behavior.

Next, you have to configure the environment. At a minimum, this includes:

  • Defining all the service, host, and server names to DNS
  • Defining the message queues
  • Setting all the configuration and performance parameters (for example, the operating system kernel parameters, the database parameters). Again, it is important that these parameters be set to the same values as the corresponding parameters in the primary site.
  • Installing and preparing the SAN including defining the storage and file systems
  • Loading the application database schema

After the infrastructure and environment is available:

  • Restore the database from the backup tapes
  • Roll forward all the transaction logs to bring the database up to the latest consistent state
  • Configure the application servers (for example, connection pool)
  • Restore the messages from the integration queues – if you do not have a backup of the messages, all the of the unprocessed messages in the integration queues are lost
  • Install Sterling Order Management System Software and reapply all the custom code, extensions, custom XMLs, and the property files
  • Reconfigure the load balancer or proxy to the application server cluster
  • Define the service names to the IP address at the recovery site
  • Establish connections to all the external systems (for example, credit card companies for credit authorization)

A cold site recovery could easily take days.