Service fault noticing method and server using same

Abstract

The invention provides a service fault noticing method and further provides a corresponding server. The service fault noticing method comprises the steps that when a service fault situation reaches a preset condition, the server is marked to be in a fault mode, and a fault noticing page and a corresponding link address are generated; a service request sent by a client is responded, and whether the server is in the fault mode or not is recognized; when the recognizing result indicates that the server is in the fault mode and the connection of the client and the server is a long connection, the link address of the fault noticing page is returned to the long-connection client through an OOB command; when the recognizing result indicates that the server is in the fault mode and the connection of the client and the server is a short connection, the link address of the fault noticing page is added to an Http Header and is returned to the short-connection client. According to the service fault noticing method and the corresponding server, when a service request is sent to a service which normally works by the client, fault noticing information of service abnormity in the server can be acquired so as to reduce the possibility that service requests are sent to fault services by the client when large-scaled service faults exist in the server, and the risk that service avalanches of the server exist is effectively reduced.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method for announcing service faults when a client communicates with a server and a server using the method. Background technique [0002] In mobile Internet applications, the high availability of cloud services is a key indicator. However, due to the instability of the mobile network, the complexity of the business, and the high concurrency pressure brought about by the effect of the beginning of the month, cloud services will inevitably experience failures of varying degrees. Perhaps a small loophole may cause a large area of ​​users to be unavailable. Greatly affect the user experience. [0003] Cloud services include three links: cloud (server), end (client) and pipeline (network). Cloud faults manifest as abnormal business logic, excessive service load, denial of service, etc.; pipeline faults manifest as network abnormalities, such as IDC (Internet Data Center, Intern...

AI Technical Summary

Problems solved by technology

[0004] When a large-scale failure occurs on the server, for example, many services fail, and the business requests sent by the client to these failed services fail during this period, although the traditional method also gives certain failure announcements, if a large number of Sending business requests to these faulty services during this period can easily lead to a service avalanche, and traditional fault announcements cannot effectively reduce the risk of service avalanche

Images