Back-end server health detection method and server based on dpvs

Abstract

The invention discloses a DPVS-based back-end server health detection method and server, belonging to the technical field of cloud computing. The method includes: the high availability process constructs the status detection message of the back-end server corresponding to the target client through the kernel protocol stack, and sends the status detection message to the DPVS process; the DPVS process is based on the pre-received target client Communication configuration information, through SNAT and DNAT operations, forward the state detection message and its corresponding state response message between the high availability process and the backend server; the high availability process obtains the After receiving the status response message fed back by the back-end server, determine the health status of the back-end server according to the analysis result of the status response message. The invention can reduce the resource consumption of the system.

Description

technical field [0001] The invention relates to the technical field of cloud computing, in particular to a DPVS-based back-end server health detection method and server. Background technique [0002] In the cloud computing system, the shared data center regards each customer providing services as a tenant. Through multi-tenant technology, it can provide the same or customizable services for most clients with a single system architecture while ensuring the isolation of customer data. . In order to achieve customer data isolation, the shared data center sets up a namespace for each customer. Customer data is stored in the corresponding namespace, and various services are provided to customers through this namespace. [0003] In order to provide customers with stable cloud computing services, it is necessary to determine the connection status of the back-end servers providing services in a timely manner. Therefore, the node servers in the cloud computing system can periodicall...

AI Technical Summary

Problems solved by technology

[0005] The high-availability process used for health detection in the node server usually uses the kernel protocol stack to construct the status detection message. Therefore, the high-availability process itself does not have the concept of multi-tenancy, and cannot use the same high-availability process for the back-end servers of different customers. Simultaneous health checks can only start one high-availability process in the corresponding namespace for each customer. However, the number of customers sharing the data center is large, and running multiple high-availability processes on the node server at the same time will consume a lot of resources. System resources increase the difficulty of operation and maintenance

Images