Distributed system node fault detection method based on high-availability detection node

A distributed system and detection node technology, applied in the field of network communication, can solve problems such as heartbeat program freeze, misjudgment, network paralysis, etc., achieve the effect of fewer messages, accurate detection results, and improved completion speed

Active Publication Date: 2018-09-28
NARI TECH CO LTD +4
View PDF8 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The heartbeat solution is currently a popular node fault detection method, but it has a major flaw, that is, the heartbeat sending and receiving program runs at the user level above the operating system. Once the CPU and other resources are insufficient, the heartbeat program will freeze. will cause misjudgment
[0004] In addition, the ping command of the operating system is a common node detection method, but if each node is simply asked to ping all other nodes to obtain the status, the system will form a network detection topology, and the number of messages and the number of nodes are exponential Related, a large number of ping packets may cause network paralysis

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Distributed system node fault detection method based on high-availability detection node
  • Distributed system node fault detection method based on high-availability detection node
  • Distributed system node fault detection method based on high-availability detection node

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] In order to further describe the technical features and effects of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. In order to illustrate the fault detection process in detail, a distributed system consisting of 6 nodes is used as an example, and node1, node2, and node3 act as detection nodes (they can also be detected nodes at the same time) as an example.

[0027] Such as Figure 1-Figure 3 as shown, figure 1 It is the failover flowchart of the master detection node. In the initial state of the system, because node1 has the smallest IP address, it is elected as the main detection node, and then it receives detection requests from each node, stores the detection results in the local shared memory, and uses multicast to send the detection results to each detection node. Node1 sends heartbeat broadcasts in a certain period, and other nodes also send heartbeats at the sa...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a distributed system node fault detection method based on a high-availability detection node. The method comprises steps that 1), two or more detection nodes are set in a system for detecting faults of all other nodes; a heartbeat message is utilized among a plurality of detection nodes for election and fault switching of a main detection node, and the heartbeat message further reaches each node of the system; 2), a request is sent by a client to the main detection node, and fault detection is carried out by the main detection node for a to-be-detected node; 3), detection requests of different clients for the same to-be-detected target node are combined by the main detection node; and 4), detection on the target node is periodically carried out by the main detectionnode after the request is received, the detection result is cached in a memory of the main detection node, and the same target nodes can be merged and detected. The method is advantaged in that misjudgment of the detection result caused by a user-level program caton phenomenon is avoided, the number of messages is reduced, and detection efficiency is improved.

Description

technical field [0001] The invention relates to a distributed system node fault detection method based on a highly available detection node, which belongs to the technical field of network communication. Background technique [0002] A distributed system is composed of multiple nodes. If some nodes fail to process, such as traffic sharing, master-standby switchover and other processing actions, the functions of the entire system will be affected. Therefore, how to quickly and accurately judge node failure is very important. [0003] The heartbeat solution is currently a popular node fault detection method, but it has a major flaw, that is, the heartbeat sending and receiving program runs at the user level above the operating system. Once the CPU and other resources are insufficient, the heartbeat program will freeze. will cause misjudgment. [0004] In addition, the ping command of the operating system is a common node detection method, but if each node is simply asked to ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): H04L12/26H04L12/24
CPCH04L41/0668H04L41/0677H04L41/12H04L43/10
Inventor 高原张勇顾文杰宁剑陈鹏王毅郭骏张学勇何源
Owner NARI TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products