A split-brain-preventing OpenStack virtual machine high-availability system, comprising management-end devices, a management network, computing node devices, and a shared storage device. At least two management-end devices communicate by means of the management network to form a management cluster. The management-end devices are communicatively connected with the computing node devices by means of the management network. The computing node devices are connected to the shared storage device. Each management-end device comprises: a Nova control module; a cluster management module; and a high-availability module, used to perform high-availability management on the computing node devices. The computing node devices, besides having a cloud computing virtual machine VM program installed thereon, are further provided with: a nova-computer computer module; a libvirt management module, used to provide a standard hypervisor API interface management process on a KVM; a lock management module, cooperating with the libvirt management module, and used to update and monitor a lock heartbeat of the shared storage device; and a high-availability computing node module, used at least to report the lock heartbeat to a management-end device.
展开▼