Service unit offline method based on micro-service unit architecture and related device

Abstract

The embodiment of the invention provides a service unit offline method based on a micro-service unit architecture and a related device. The method comprises the following steps: firstly, acquiring a service unit offline instruction comprising the name of a to-be-offline service unit; then, according to the name of the to-be-offline service unit, determining whether the to-be-offline service unit is in a running state or not, and when it is determined that the to-be-offline service unit is in the running state according to the name of the to-be-offline service unit, logging out the name of the to-be-offline service unit in a registration service unit list of the micro-service unit architecture, and starting timing at the same time to obtain offline waiting duration; and when the offline waiting duration of the to-be-offline service unit meets a waiting duration threshold value of the to-be-offline service unit, making the to-be-offline service unit be offline. According to the service unit offline method based on the micro-service unit architecture and the related device provided by the embodiment of the invention, elegant offline of the service unit can be ensured, and the problem of request cutoff is avoided.

Description

technical field [0001] The embodiments of the present invention relate to the field of computers, and in particular to a method for offline service units based on a micro-service unit architecture and related devices. Background technique [0002] In the distributed architecture of microservice units, it is very important to manage the online and offline of microservice units. [0003] In the native scenario of Kubernetes, although the smallest object of the system is a Pod (microservice unit group), the creation and destruction of Pods (or online and offline) are judged by the Kubernetes system itself, and the operator cannot directly target each Each Pod performs online and offline operations, but uses registration centers such as Eureka for indirect management. [0004] In the existing technology, when Pods are indirectly managed by a registration center such as Eureka, Pods in the Terminating state will be directly removed from the registration list (EndPoint, etc.) of ...

AI Technical Summary

Problems solved by technology

[0003] In the native Kubernetes scenario, although the smallest object of the system is a Pod (microservice unit group), the creation and destruction of Pods (or online and offline) are judged by the Kubernetes system itself, and the operator cannot directly target each Each Pod performs online and offline operations, but uses registration centers such as Eureka for indirect management

[0004] In the existing technology, when using a registration center such as Eureka to indirectly manage Pods, Pods in the Terminating state will be directly removed from the registration list (EndPoint, etc.) of the registration center, resulting in inaccessible requests. The specific Pod will eventually cause the request to be cut off

Images