| 摘要 |
<p>A method and system for monitoring nodes in a network having at least one network management station and a plurality of nodes. A queue (10) stores polling messages for transmission to the nodes where each node is indexed by its network address. The network management station sends polling messages to the nodes in sequence at a predetermined rate controlled by a rate control mechanism (12). Polling messages are sent up to four times to a particular node. The transmission of these messages are recorded in a table which is indexed by the network address of each node, and by the time of the next scheduled timeout (the time period between successive polling messages) associated with each node. The network management station determines if another polling message should be sent to each of the nodes. If the fourth polling message has been sent to a particular node, it has been unacknowledged by that node and the timeout has expired, then the node is determined to have failed. <IMAGE></p> |
