Network fault management, a key part of the today Network Management architecture, covers functions such as detect, isolate, determine the cause and correct malfunctions in a network. The objectives of doing fault management are to increase network availability, reduce network downtime and restore network failure quickly.

The basic requirements for a fault management system are:

• Monitoring and collect of statistics on network devices, traffic conditions and usage in real-time to avoid and forecast potential faults
• Setting thresholds and alarms that may cause network failure to warn the network admin
• Setting alarms that warns of performance degradation on network devices and links
• Setting alarms of network resource (such as hard disk space) usage and limitation problems
• Remotely control network devices for rebooting, shutting down etc.
• Have a centralized consol to perform all of the above functions

A typical fault management system follows these steps:

When an error occurs, a report is generated and is sent to the fault analyzer. The fault analyzer diagnoses and records the problem. Finally, a system or a person uses the information from the fault analyzer to take appropriate actions such as isolating the error, black-listing failing or failed components, automatically restarting/restoring services, and alerting the system administrator.

Related Terms: Network management, Performance management, Configuration management, Security management