Service request processing method and device

Abstract

The embodiment of the invention provides a service request processing method and device which are applied to a service node. The method comprises the following steps: receiving a service request sent by a service proxy; setting the state of the service node as the offline state when detecting that the preset offline condition is triggered, thereby enabling the service proxy to stop sending service request to the service node when detecting that the service node is in the offline state; processing the received service request, and closing the service node after all service requests are completely processed. By using the processing method provided by the embodiment of the invention, the normal processing on the service request is maintained to an extreme, the jittering produced on the service calling is avoided, the redundancy resource is unnecessary to configure, the service proxy is unnecessary to update and restart, the service node completely and actively controls the online and offline by itself, and the smooth processing on the service request is realized.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a service request processing method and a service request processing device. Background technique [0002] In the topology of large-scale services, it is generally deployed in the form of service agent + multiple service nodes to ensure high availability of service nodes and improve service reliability. However, when the service node releases the service or fails, it usually needs to update the code, which will cause the service node to restart and interrupt the processing of service requests. [0003] In one solution, the service agent can identify the health status of the service node through the health check to ensure that the service request is sent to the service node in a normal state for processing, but due to the interval of the health check, if the service node is in the interval If the version is upgraded or fails during the period, the service agent will not be able to ...

AI Technical Summary

Problems solved by technology

However, when the service node releases the service or fails, it usually needs to update the code, which will cause the service node to restart and interrupt the processing of service requests

[0003] In one solution, the service agent can identify the health status of the service node through the health check to ensure that the service request is sent to the service node in a normal state for processing. However, due to the health check interval, if the service node is in the interval If the version is upgraded or fails during the period, the service agent will not be able to detect it in time, so it will continue to forward the service request to the service node, resulting in the loss of the service request, and when the service request is in a high concurrency situation, the short interval will cause A large number of service requests are lost

[0004] In another solution, a batch of new service nodes can be deployed, and service nodes that need to be released or have failed service nodes on the service agent can be removed to ensure that service requests are sent to service nodes in normal state. processing, but it needs to use redundant service nodes, usually using 1:1 redundancy, which increases the cost, and the service agent must be updated after the service release or failure occurs, which may also cause short-term service processing interruption

Images