Back-to-source processing method and device

Abstract

The invention discloses a back-to-source processing method and device, and the method comprises the steps: obtaining a query request of a user, and judging whether there is cache in an object of the query request; if NO, obtaining a set number of back-to-source IPs; taking one back-to-source IP in the set number of back-to-source IPs to carry out object return; if the returned object has an error,returning the object again through the back-to-source IP according to a preset condition or taking the next back-to-source IP in the set number of back-to-source IPs to return the object; and repeating the previous step until the returned object is correct, or the connection of the front-end server is overtime, or the number of fault switching times in the preset condition is exhausted, and ending the query request. By setting more than one switchable return source IPs, fault switching can be carried out on each random return source error in real time, so that the error rate is effectively reduced.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method and device for back-to-source processing. Background technique [0002] CDN is an edge caching system, which is an http reverse proxy and caching system. This system generates a cachekey through the uniform resource locator url of the hypertext transfer protocol http, and checks whether the requested object is in the cache through the cachekey. Hit. If it hits, it returns directly. If it does not hit the cache, the request needs to be reverse-proxyed to the origin site. The term is called back-to-origin. For the object returned by the source, it should be returned to the client, and the object should also be written into the cache. Objects in the cache follow the LRU algorithm, and the earliest objects written to the cache are deleted from the cache first. [0003] Origin station failover solves the problem of a large number of request errors caused by exces...

AI Technical Summary

Problems solved by technology

[0005] Using the health check method, it is also difficult to deal with the error requests randomly generated by the origin site. Such error requests may be discontinuous, which cannot trigger the record downtime function of the health check, resulting in the failure to reduce the error rate.

Images