A network fault detection method and device

Abstract

The embodiment of the present application provides a method and device for network fault detection, which is applied to the first device in the first service node in the service cluster, including: calculating, for each link to be tested, the network failure detection method and device through the link to be tested within the statistical cycle The message delay of sending the message is taken as the link delay, the link to be tested is the link between each second device and the first device in the service node to be tested different from the first service node, and the message delay is the link to the second device The delay between sending the message and receiving the feedback response from the second device; for each second device, according to the link delay of the link under test whose end is the second device, calculate the time delay sent to the second device in the characterization statistical period The message delay of the message is used as the integrated message delay; according to the number of second devices whose integrated message delay exceeds the preset delay in the service node to be tested, it is detected whether there is a network fault in the service node to be tested. Applying the solutions provided by the embodiments of the present application can improve the accuracy of network fault detection for service nodes.

Description

Technical field [0001] The present application relates to the field of network technology, and in particular, to a network fault detection method and device. Background technique [0002] As the number of users increases, user demand is growing, and it is usually provided to provide users based on business clusters containing multiple business nodes. For example, the above-described service cluster can be a storage cluster, or the like for providing a storage service. Among them, a plurality of devices for providing services can be included in the service node. [0003] However, the above-mentioned business clusters may work in a sub-health network environment, resulting in a network failure of various business nodes in the business cluster, and the operational clusters can work in a network environment in the business cluster. Stable, for example, there is a business oscillate, business node, zero, business node, and equipment frequently in the business node. In severe cases, it...

AI Technical Summary

Problems solved by technology

However, in this case, OSD1 only considers that a single OSD, OSD2, has a network fault, and does not consider that the entire service node has a network fault. When subsequent OSD isolation is performed, only OSD2 is isolated.

Therefore, from the perspective of detecting whether there is a network fault in the storage node, the accuracy of network fault detection using the above method is low

[0005] In addition, in this case, there may be other OSDs with network failures in the storage node to which OSD2 belongs, but the above-mentioned storage nodes still provide storage services normally, which may cause business fluctuations in the entire storage cluster, and it is difficult to ensure that the storage cluster provides continuous storage business

Images