Content loop detection processing method, device and computer-readable storage medium

Abstract

Embodiments of the present invention provide a content loop detection processing method, device, and computer-readable storage medium. The method includes: sending a back-to-source request to the content center; the back-to-source request carries the edge node ID and the Content distribution network CDN operator identifier; generate a first back-to-source request log according to the content center address, and send it to a preset back-to-source request analysis device; the first back-to-source request log is used for the back-to-source request The analysis device judges the looping of the content; receives the request termination instruction, and terminates the back-to-source request operation; the request termination command is sent when the back-to-source request analysis device determines that the content is looped.

Description

technical field [0001] The present invention relates to the technical field of mobile communication, in particular to a content loop detection processing method, device and computer-readable storage medium. Background technique [0002] Both the content center and the edge service nodes of the Content Delivery Network (CDN) support static back-to-origin and dynamic back-to-origin methods, and support the configuration of back-to-origin policies, including back-to-origin methods and back-to-origin addresses. The free configuration of back-to-origin policies can easily lead to loops in CDN content access. For example, domain name A is configured to back-to-origin to domain name B, and domain name B is back-to-origin to domain name A. According to the differences between domain names A and B on CDN service nodes, there are the following three scenarios: There are two kinds of ring formation situations: Case 1: A loop is formed within a CDN node; Case 2: A loop is formed between...

AI Technical Summary

Problems solved by technology

[0004] Problems in the pre-prevention method: the current pre-prevention method can only prevent content loops in a single CDN system, including self-forming loops within CDN nodes and inter-CDN node loops, and cannot prevent content loops between CDNs; for static back-to-source mode, the CDN operation manager can only obtain the IP addresses of all nodes in the CDN, and can only prohibit the content center from returning to the node of the CDN, and cannot prohibit the content center from returning to other CDN nodes; for the dynamic back-to-source method, it is forbidden to configure The source domain name is the CNAME domain name of this CDN, the domain name served by this CDN, and the back-to-source domain name cannot be prohibited from being the domain name served by other CDNs; the CDN operation manager can only prohibit the content center from initiating domain name resolution requests to the CDN scheduling node to avoid secondary scheduling To this CDN, the domain name resolution result cannot be forbidden to be the IP of the edge service node of other CDN

[0005] Problems with the post-processing method: Based on the method of waiting for response time detection, it can only handle suspected content loops, and there is a certain probability that normal requests will be disrupted. Form a loop; at present, the back-to-origin log of the content center and the user access log of the edge node are generated based on completed requests. For unfinished requests waiting for a response message, no back-to-origin log and user access log are generated, and there is no valid information for post-event analysis

Images