Method and device for recovering virtual machine

Abstract

The invention discloses a method and a device for recovering a virtual machine. The method comprises the following steps that: a control node detects that a virtual machine needs to be subjected to remote regeneration, and a first operation state corresponding to the virtual machine which needs to be subjected to the remote regeneration is searched in a corresponding relationship among the virtual machine, a computational node and an operation state; a judgment result shows that the searched first operation state is a dummy state, the computational node corresponding to the virtual machine which needs to be subjected to the remote regeneration in the corresponding relationship is changed into a target computational node; and the control node searches a second operation state corresponding to the virtual machine which needs to be subjected to the remote regeneration in the corresponding relationship, judges the searched second operation state as the dummy state, changes the second operation state corresponding to the virtual machine which needs to be subjected to the remote regeneration in the corresponding relationship into a virtual machine under reconstruction, and sends a message of carrying out the remote regeneration to a destination computational node. By use of the method, a conflict caused in a way that two different computational nodes carries out a recovery operation on the same virtual machine is reduced.

Description

technical field [0001] The invention relates to virtual technology, in particular to a method and device for restoring a virtual machine. Background technique [0002] OpenStack is a free software and open source cloud computing management platform project authorized by the Apache license, jointly developed and initiated by NASA (National Aeronautics and Space) and Rackspace. OpenStack provides various complementary services. The infrastructure as a service (IaaS) solution is an open source project designed to provide software for the construction and management of public and private clouds. [0003] In OpenStack's implementation of virtual machines, when a computing node goes down abnormally, the virtual machine on the computing node cannot run. The machine can be restarted on an available computing node in time. The rebirth of the virtual machine in this scenario is the rebirth of the virtual machine in a different place. When the virtual machine is reborn in a different...

AI Technical Summary

Problems solved by technology

[0006] In the existing method for restoring a virtual machine, during the rebirth of the virtual machine, when the failed computing node is powered off due to reasons other than the computing node itself such as power supply, the environment may recover after a period of time or human intervention may occur. Power off and on again. During the power-on process of the computing node, in the correspondence between virtual machines, computing nodes, virtual machine status and operating status, find the virtual machine corresponding to the computing node itself and the virtual machine status to the running status. and operating state, change the operating state corresponding to the found virtual machine whose operating state is empty state to the power-on state, and start the found virtual machine whose operating state is empty state, thus causing a fault The computing node and the destination computing node configure the same virtual machine at the same time, causing a series of resource and process conflicts such as Internet Protocol (IP, Internet Protocol) addresses, and may even cause the business of the virtual machine to fail to run normally.

Images