A high-availability implementation method of load balancer in openstack cloud platform

Abstract

A high-availability implementation method of a load balancer in an OpenStack cloud platform, comprising: dividing more than two availability domains in the OpenStack cloud platform, setting more than two nodes in each availability domain, and in one availability domain , create at least one type of load balancer, one for each lbaas-agent of different nodes, and the same type of load balancers between different nodes establish communication with each other; each load balancer is equipped with a QR port And the HA port, the HA port is responsible for the internal communication of the availability domain, and is used to monitor the working status of the ontology load balancer and the same type of load balancer on other LB nodes in the availability domain; the QR port is responsible for external communication outside the availability domain. The method of the present invention executes the same business service through multiple load balancers, which can reduce the impact of a single load balancer on the entire business, reduce risks, and achieve isolation by dividing available domains, which improves the security and reliability of the communication system, and can reduce The difficulty of maintaining the system.

Description

technical field [0001] The invention belongs to the technical field of communications, and in particular relates to a high-availability implementation method of a load balancer in an OpenStack cloud platform. Background technique [0002] In order to meet the growing business demand, communication service providers often set up multiple servers for the same business to reduce costs. When a large number of clients send requests to the server, the traffic can be evenly distributed to each server through the load balancer , for the client, it directly accesses the load balancer, so it only needs to obtain the VIP (virtual IP, virtual IP address) of the load balancer, and does not need to obtain the IP addresses of all servers, while the load balancer needs to obtain all servers IP address. The advantage of this method is that when the number of servers changes or the server needs to be changed, there is no need to interrupt the business. In addition, because the error probabil...

AI Technical Summary

Problems solved by technology

[0011] 1) Once there is a problem with a single load balancer, the entire business cannot continue to be used;

[0012] 2) If there is a problem on the cloud platform, and there is a running load balancer in the node that needs to be debugged, it will be difficult to find the problem because the working state of the load balancer cannot be stopped;

[0013] 3) When creating a high-performance load balancer and a low-performance load balancer, because the cloud platform cannot distinguish between different types of lbaas-agent, it will cause a serious waste of resources;

[0014] 4) For load balancers with different security levels, since it is impossible to specify which lbaas-agent the load balancer will be created in when creating, there are hidden dangers in terms of security

Images