摘要 |
A computer is connected to redundant network switches by primary and secondary connections, respectively. Test messages are sent across each connection to the attached switches. A break in a connection, or a faulty connection, is detected upon a failed response to one of the test messages. In response to this failure, traffic is routed across the remaining good connection. To facilitate fast protocol rerouting, a test message is sent across the now active connection bound for the switch connected to the failed connection. This message therefore traverses both switches causing each to learn the new routing. Rerouting is therefore accomplished quickly. |