| 摘要 |
A system provides automatic protection switching in case of failure or degradation in a function. For example, a system node may have a working server function, a protection server function, one or more client functions, and a control function. In its baseline configuration, the working server function is active, meaning that it is operating on-line with respect to the client functions, while the protection server function is standby, meaning that it is either inactive (i.e., cold standby) or operating off-line with respect to the client functions (i.e., hot standby). The control function monitors the operations of the working server function, and, if a failure or degradation is detected, or if an externally generated switch command is received, the control function will (1) instruct the working server function to change its status from active to standby and (2) instruct the protection server function to change its status from standby to active. In addition, the control function instructs each server function about a protection switch override condition. Each server function communicates its active/standby status as well as the override condition to the client functions (e.g., using in-band signaling). In addition to monitoring the active/standby status and the override condition transmitted from each server function, each client function also performs its own independent monitoring of the status of the server functions to detect failures or other degradation in the server functions. Each client function then applies a specific set of rules to this information to determine which server function to select for its operations. The set of rules are designed to handle race conditions and other ambiguous states.
|
