Implementation method of openstack full-distributed dhcp service

Abstract

The invention discloses an implementation method of an openstack full-distributed type dhcp service, relating to the field of cloud computing. The method comprises steps of deploying dhcp services onall the computing nodes; creating a namespace on each computing node; creating a trunk port on the virtual machine network bridge and associating with the namespaces; setting a rule of the flow table,only allowing network flow of a related port of a virtual machine to flow through the trunk port, establishing a dhcp service process after the port of the virtual machine is bound with a computing node, and only allowing a dhcp service of the computing node to run in the namespace of the computing node. According to the method provided by the invention, the service of the dhcp is only provided by the process of the node; the network does not pass through the network card and directly corresponds to the operating system, so that the performance of the system can be greatly improved, the configuration file of the dnsmasq is simplified, only the information related to the single network card needs to be configured, and the complexity of the configuration file is reduced.

Description

technical field [0001] The invention discloses a method for realizing an openstack fully distributed dhcp service, and relates to the field of cloud computing. Background technique [0002] The dhcp service of openstack is implemented by starting the dnsmasq process on the user-defined dhcp node. For each network, there is a dedicated dnsmasq process responsible for providing network IP and routing information. There are some problems in this mode: the centralized distribution of dhcp limits the total number of networks; the centralized dhcp service needs to occupy multiple IP addresses to ensure high service availability; the virtual machine and the dhcp server are not on the same host, which affects performance At the same time, if the physical network fails, the normal service cannot be guaranteed; DHCP broadcast packets are flooded on the second layer or even the third layer network, which has the risk of affecting the stability of the basic network. [0003] The invent...

AI Technical Summary

Problems solved by technology

There are some problems in this mode: the centralized distribution of dhcp limits the total number of networks; the centralized dhcp service needs to occupy multiple IP addresses to ensure high service availability; the virtual machine and the dhcp server are not on the same host, which affects performance At the same time, if the physical network fails, the normal service cannot be guaranteed; DHCP broadcast packets are flooded on the second layer or even the third layer network, which has the risk of affecting the stability of the basic network.

Images