| 摘要 |
The disclosed embodiments include a method, apparatus, and computer program product for improving network monitoring. For example, in one embodiment, current inaccuracies due to the imperfect counter frame injection error between ingress counters and egress counters is eliminated, by monitoring, using an ingress counter, for a time period that is slightly greater than the time it takes for a packet/frame to pass from one end to the other, for “no packets/frames”. In one embodiment, once this no packet time occurs, a signaling event is triggered. A notification is sent from the start of the path counter to the end of the path counter to start counting. A “stop counting/restart counting” signal is then sent from the start of the path counter to the end of the path counter the next time that no packets/frames are observed for the same period. |
| 主权项 |
1. A method for improving network monitoring comprising:
monitoring, using a processor, protocol data units that are being communicated from a first network device to a second network device; determining, with the processor, a path delay time for the protocol data units to be communicated from the first network device to the second network device; detecting, with the processor, that no protocol data units are being communicated from the first network device to the second network device for a first time period greater than the path delay time; responsive to identifying that no protocol data units have been communicated for the first time period, transmitting a first signal from the first network device to the second network device to initiate a second count by initiating counting of the protocol data units received at the second network device that are sent from the first network device; responsive to identifying that no protocol data units have been communicated for the first time period, initiating, at the first network device, a first count by initiating counting of the protocol data units that are transmitted to the second network device; detecting, with the processor, that no protocol data units are being communicated from the first network device to the second network device for a second time period greater than the path delay time, wherein the second time period occurs after protocol data units are communicated subsequent to the first time period; responsive to identifying that no protocol data units have been communicated for the second time period, transmitting a second signal from the first network device to the second network device to terminate the second count by terminating counting of the protocol data units received at the second network device; responsive to identifying that no protocol data units have been communicated for the second time period, terminating, at the first network device, the first count by terminating counting of the protocol data units that are transmitted to the second network device; and determining, with at least one of a network management system, the first network device, or the second network device, frame loss by comparing the first count of the protocol data units that are transmitted by the first network device and the second count of the protocol data units that are received by the second network device, during the period between the first time period and the second time period. |
