What is the contingency plan of the integration platform formed of? The integration platform can crash due to a failure in a system component – a hardware failure in a computer, or a software failure in Windows Server, SQL Server, BizTalk Server, or FRENDS Iron, for example. Based on Contingency Planning Guide for Information Technology Systems, the contingency plan results three steps. The first step documents the components of the integration platform and their connections with each other and to the internal systems (i.e. EAI) and external systems (i.e. B2Bi), if this kind of an architectural description does not exist.  The components and connections should also cover those ones planned to be deployed in the near future. The description includes quality attributes such as volume, availability, reliability, and security of the solution components and connections for receiving, processing, and sending data. The impact analysis is conducted to prioritize these solution components and connections.

The next step is to identify and design preventive controls. These controls include clustering, backups, and monitoring. Perhaps the most effective control is the high availability integration platform. For fault tolerance, clustered system components enable duplicated solution components. Depending on the risk level, there are alternatives how to cluster and duplicate components of the integration platform – to cluster only FRENDS Iron and BizTalk Server, or also SQL Server, for example. The integration platform should always be backed up. Components can be backed up through distributed drives or a centralized backup device. A full backup captures all files on the disk or within the directory selected for backup, while an incremental backup captures files that were created or changed after the last backup. Moreover, a trading partner agreement may require that messages sent to or received from an external system must be stored for a certain period. With monitoring tools, personnel can observe error patterns that anticipate a failure, and then perform corrective actions to avoid the crash of the integration platform. 

The last step prepares recovery plan that involves tasks to create backup media and record instructions for reinstalling the integration platform and restoring the data. Furthermore, the recovery plan documents detailed configurations of the components and connections as well as vendors that supply hardware, software, and services for the integration platform. Most importantly, personnel associated with the integration platform should be trained on when and how to carry out the recovery plan. No contingency plan is carved in stone. When the integration platform changes, the architectural description, preventive controls, and recovery plan should be updated to ensure their effectiveness.