Node equipment state detection method and node equipment state detection device

Abstract

The invention provides a node equipment state detection method and a node equipment state detection device. The method and the device are applied to an equipment cluster which is pre-divided into one or more circular test groups with the same master node. The method comprises the following steps: a master node sends a test message carrying an identity list of all nodes to first slave node equipment; if the first slave node responds to the message normally, the first slave node is self-marked as 'test passed' in the identity list and forwards the message to a second slave node; if the second slave node responds to the message normally, the second slave node is self-marked as 'test passed' in the identity list and continues to forward the message until the message is forwarded to the master node; if the second slave node cannot respond to the message normally, the first slave node marks the second slave node as 'test failed' in the identity list and reports the message to the master node; and the master node determines the state of each node in a test group based on the received test message. Through the method and the device, the processing pressure on the master node equipment can be alleviated.

Description

technical field [0001] The present invention relates to the field of communication technology, in particular to a method and device for detecting the state of node equipment. Background technique [0002] ElasticSearch is a Lucene-based search server cluster. It provides a full-text search engine with distributed multi-user capabilities and is one of the most popular enterprise search engines. [0003] The node devices in ElasticSearch's search server cluster include an elected master node device and several slave node devices. These nodes present a star topology, and each node can communicate with other nodes, so the cluster's Each node includes information about all other nodes. [0004] In the search server cluster of ElasticSearch, fault detection is one of the important basic functions in the process of node discovery and state maintenance and update. The current error detection mechanism is divided between the master node device and each slave node device. Send ping...

AI Technical Summary

Problems solved by technology

[0004] In the search server cluster of ElasticSearch, fault detection is one of the important basic functions in the process of node discovery and state maintenance and update. The current error detection mechanism is divided between the master node device and each slave node device. Send ping messages to each other for error detection to find faulty nodes in the cluster. Since the master node device needs to send ping messages to each slave node device separately, the processing pressure of the master node device is high, and in the case of a large number of nodes In the cluster, it will cause excessive consumption of bandwidth resources of the entire network

Images