Network switching method and device of virtual machine, computer device and storage medium

Abstract

The application relates to a network switching method and device of a virtual machine, a computer device and a storage medium. The method comprises the following steps: in the case that the virtual machine uses a main network for network communication, monitoring the network state of the main network; the communication link of the main network comprises a virtual switch connected with the virtual machine in communication, a network service component connected with the virtual switch in communication and a physical switch connected with the network service component in communication; in the case that the main network is in an abnormal network communication state, switching the communication network of the virtual machine to a backup network; the backup network is connected with the physical switch in communication through a user mode provided by a fast emulator; the fast emulator is used for simulating and generating the virtual machine. The network reliability of the virtual machine can be improved by adopting the method, and the high availability of the virtual machine is improved.

Description

Technical Field

[0001] This application relates to the field of cloud computing network technology, and in particular to a method, apparatus, computer equipment, storage medium and computer program product for network switching of virtual machines. Background Technology

[0002] With the rapid development of cloud computing technology, virtualization technology has demonstrated significant advantages in resource utilization, management, and allocation. OpenStack, as an open-source cloud computing platform, has become one of the preferred solutions for building private and public clouds.

[0003] Currently, most cloud terminals based on the OpenStack architecture rely on virtual networks for network traffic within virtual machines. This traffic then passes through operations such as Dynamic Host Configuration Protocol (DHCP), routing, and Network Address Translation (NAT) on network servers or network clusters before reaching the network exit.

[0004] However, in real-world applications, hardware failures of network nodes or software failures of network components such as network service components can easily lead to network link failures within virtual machines. Therefore, improving the network reliability of virtual machines and thus enhancing their high availability is an urgent problem to be solved in practical applications. Summary of the Invention

[0005] Based on this, it is necessary to provide a network switching method, apparatus, computer device, computer-readable storage medium, and computer program product for virtual machines that can improve the network reliability of virtual machines and thus enhance their high availability, in order to address the aforementioned technical problems.

[0006] Firstly, this application provides a network switching method for a virtual machine, the method comprising:

[0007] When the virtual machine uses the main network for network communication, the network status of the main network is monitored; the communication links of the main network include a virtual switch that communicates with the virtual machine, a network service component that communicates with the virtual switch, and a physical switch that communicates with the network service component.

[0008] If the main network is detected to be in an abnormal network communication state, the communication network of the virtual machine is switched to the backup network; the backup network communicates with the physical switch through the user mode provided by the fast emulator; the fast emulator is used to simulate and generate the virtual machine.

[0009] In some embodiments, the method further comprises:

[0010] in response to a type adding operation for the network capability profile, adding a primary network and a backup network in the network capability class;

[0011] in response to an interface creation operation for the primary network and the backup network, creating an Ethernet interface corresponding to the primary network and the backup network respectively in the Ethernet service;

[0012] in response to a function adding operation for the network management profile, adding network request support for the primary network and the backup network in the network management service;

[0013] in response to a scoring operation for the primary network and the backup network, configuring a first score for the primary network and a second score for the backup network in the network management service, and controlling the virtual machine to use the primary network for network communication when the first score is greater than the second score.

[0014] In some embodiments, the method further comprises:

[0015] in response to a parameter adding operation for the fast emulator, adding a user mode enabling parameter in the configuration file of the fast emulator, and determining a virtual network card corresponding to the backup network;

[0016] in response to a logic adding operation for the user mode, adding a routing configuration mode in the user mode, the routing configuration mode being used to determine a service network routing table corresponding to the virtual machine from each routing table of a host computer after the virtual machine starts the backup network, and configure the service network routing table as a default routing table of the backup network;

[0017] performing hardware abstraction processing on the virtual network card in a hardware abstraction layer of Ethernet, and starting a dynamic host configuration protocol client for the virtual network card;

[0018] based on an Ethernet interface registration file of the backup network, calling the network management service to configure the backup network.

[0019] In some embodiments, the method further comprises:

[0020] in a case where it is monitored that the primary network has network communication abnormity, counting a continuous abnormal number of the primary network having network communication abnormity;

[0021] if the continuous abnormal number is less than a preset number threshold, performing a backoff retry operation for the primary network, and returning to the step of monitoring the network state of the primary network;

[0022] In a case where the continuous abnormal number of times reaches the preset number threshold, it is determined that the main network is in a network communication abnormal state, the continuous abnormal number of times is set to an initial value, and the step of monitoring the network state of the main network is returned to be executed.

[0023] In some embodiments, the method further comprises:

[0024] In a case where the main network has a dynamic host configuration protocol abnormality or a network connectivity abnormality, it is determined that the main network has a network communication abnormality.

[0025] In some embodiments, the switching of the communication network of the virtual machine to the backup network comprises:

[0026] In a case where the first score is less than the second score, a network management service is invoked, a network evaluation of the communication network used by the virtual machine is performed, and a network switching operation is performed based on the evaluation result;

[0027] The default route of the virtual machine is set to a network route of the backup network; the network route of the backup network is configured by the route configuration mode.

[0028] In a second aspect, the application further provides a network switching device of a virtual machine, the device comprising:

[0029] A network state monitoring module is configured to monitor a network state of a main network in a case where a virtual machine uses the main network for network communication; the communication link of the main network comprises a virtual switch connected in communication with the virtual machine, a network service component connected in communication with the virtual switch, and a physical switch connected in communication with the network service component;

[0030] A network switching module is configured to switch a communication network of the virtual machine to a backup network in a case where it is monitored that the main network is in a network communication abnormal state; the backup network is connected in communication with the physical switch through a user service network mode provided by a fast emulator; the fast emulator is configured to simulate generation of the virtual machine.

[0031] In a third aspect, the application further provides a computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the above method when executing the computer program.

[0032] In a fourth aspect, the application further provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of the above method.

[0033] In a fifth aspect, the present application further provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method described above.

[0034] The network switching method, device, computer device, storage medium and computer program product of the virtual machine described above, in the case of using the main network for network communication of the virtual machine, the network state of the main network is detected, wherein the communication link of the main network comprises a virtual switch connected in communication with the virtual machine, a network service component connected in communication with the virtual switch, and a physical switch connected in communication with the network service component. In the case of monitoring that the main network is in an abnormal network communication state, in order to make the network communication of the virtual machine not be affected, the communication network of the virtual machine is switched from the main network to the backup network, wherein since the backup network is connected in communication with the physical switch through the user service network mode provided by the fast emulator, the dynamic host configuration protocol of the backup network can be directly provided by the fast emulator, and no longer depends on the virtualization network component. By configuring the backup network for the network communication of the virtual machine, the network reliability of the virtual machine can be effectively improved, so that the virtual machine can still process business when the communication link of the main network fails, thereby effectively improving the high availability of the virtual machine. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 Network topology block diagram for existing virtual machine online in some embodiments;

[0036] Figure 2 Network topology block diagram for virtual machine online in other embodiments;

[0037] Figure 3 Flowchart of the network switching method of the virtual machine in some embodiments;

[0038] Figure 4 Flowchart of the network switching method of the virtual machine in other embodiments;

[0039] Figure 5 Flowchart of the network switching method of the virtual machine in other embodiments;

[0040] Figure 6 Flowchart of the network switching method of the virtual machine in other embodiments;

[0041] Figure 7 Flowchart of the network switching method of the virtual machine in other embodiments;

[0042] Figure 8 Connection relationship between modules in the network management service of the Android system in some embodiments;

[0043] Figure 9A structural block diagram of a network switching device for a virtual machine in some embodiments;

[0044] Figure 10 An internal structural diagram of a computer device in some embodiments. DETAILED DESCRIPTION

[0045] For the purpose, technical solutions and advantages of the present application to be more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0046] In some embodiments, the network topology structure of the virtual machine currently accessing the Internet through the virtual network is as shown in Figure 1 The network topology structure as a whole includes five nodes, namely, an Android system 101, a physical switch 102, a router 103, a firewall 104 and the Internet 105.

[0047] The Android system 101 can be integrated on a terminal, connected with the router 103 through the physical switch 102, and reaches the Internet 105 through the router 103, the firewall 104, etc.

[0048] The Android system 101 includes a computing node 1011 and a network node 1012. The computing node 1011 refers to a node unit in the Android system 101 for performing various computing tasks and processing data. The computing node 1011 includes a virtual machine 10111, a virtual network card 10112 and a first virtual switch 10113. The network node 1012 is a node unit for connecting various network components in the virtual network. The network node 1012 includes a second virtual switch 10121 and a Neutron component 10122. The first virtual switch 10113 in the computing node 1011 is connected in communication with the second virtual switch 10121 in the network node 1012 through a service network in the Openstack architecture.

[0049] The virtual machine 10111 is a virtual system generated by simulating the host of the Android system based on the Quick EMUlator (QEMU). The QEMU is an open-source virtual machine monitor and emulator, which can allow users to execute virtual machines on different hardware platforms. It provides dynamic binary translation and hardware emulation functions, so that users can simulate multiple different hardware architectures and operating systems on a physical host.

[0050] In actual use, in compute node 1011, the virtual machine traffic of virtual machine 10111 can flow through virtual network card 10112 and first virtual switch 10113 to network node 1012 for DHCP configuration. The second virtual switch 10121 in network node 1012 will continue to transmit the virtual machine traffic through network service component 10122 to physical switch 102. Physical switch 102 will then transmit the virtual machine traffic to Internet 105 through router 103 connected to it and firewall 103 connected to router 103.

[0051] As can be seen from the network topology of traditional virtual machine network transmission, network traffic within a virtual machine depends on the virtual network. After passing through operations such as DHCP, routing, and NAT by a network server or network cluster, it can reach the network exit. In practical application scenarios, this network link is prone to hardware failures of network nodes or software failures of network components such as Neutron, which can lead to the failure of the network link within the virtual machine.

[0052] Based on this, such as Figure 2 As shown in the illustration, this application provides an Android system 200, which is integrated into a cloud terminal based on the OpenStack architecture. The Android system 200 also includes a compute node 1011 and a network node 1012. However, the compute node 1011 of the Android system 200 also includes a user-mode node 10114. User mode is a network mode provided by the fast emulator, also known as Slirp mode, used to implement the user-mode network stack. This allows the virtual machine 10111 to communicate with the external network without requiring a high-performance network connection. The user-mode node 10114 is a key component in the fast emulator for implementing user-mode functionality. The user-mode node 10114 communicates with the physical switch 102 through the service network in the OpenStack architecture.

[0053] In the Android software 200 of this embodiment, the virtual machine 10111 can communicate with the external network through the main network or the backup network.

[0054] The main network communication links include a virtual switch communicating with virtual machine 10111, a network service component 10122 communicating with the virtual switch, and a physical switch 102 communicating with the network service component 10122. It is understood that the virtual switch includes a first virtual switch 10113 and a second virtual switch 10121, and virtual machine 10111 needs to communicate with the first virtual switch 10113 through a virtual network interface card 10112.

[0055] The communication link of the standby network includes the user mode node 10114 connected with the virtual machine 10111, and the physical switch 102 connected with the user mode node 10114 through the service network.

[0056] In actual use, the Android system 200 can monitor the network state of the main network when the virtual machine 10111 uses the main network for network connection based on the integrated cloud terminal. When it is monitored that the main network is in an abnormal network communication state, the communication network of the virtual machine 10111 is switched to the standby network, wherein the standby network is connected with the physical switch 102 through the user mode provided by the fast emulator.

[0057] The Android system in the embodiment can effectively improve the network reliability of the virtual machine by configuring the standby network for the network communication of the virtual machine, so that the virtual machine can still process services when the communication link of the main network fails, thereby effectively improving the high availability of the virtual machine.

[0058] In some embodiments, as shown in Figure 3 , a network switching method of a virtual machine is provided. The method is applied to the Android system 200 in Figure 2 for example, and includes the following steps:

[0059] S302, when the virtual machine uses the main network for network communication, the network state of the main network is monitored.

[0060] The communication link of the main network includes a virtual switch connected with the virtual machine, a network service component connected with the virtual switch, and a physical switch connected with the network service component.

[0061] The network state monitoring of the main network is an effective monitoring operation for determining the network communication state of the main network. The Android system can call the network management service to monitor the network communication state of the main network. The network management service monitors the network communication state of the main network by receiving the signal broadcasted by the anomaly monitoring module in the Android system, so as to timely make corresponding processing when there is a communication anomaly in the main network, thereby improving the network reliability of the virtual machine.

[0062] Specifically, usually, a designer configures a default network type of an Android system, and the default network type of a virtual machine is a host network. In a case where the virtual machine uses the host network for network communication, virtual internal traffic of the virtual machine depends on a virtual network, and needs to pass through each network virtual component in a network node, and after a DHCP, routing, NAT, and the like of a network server are completed, the virtual internal traffic can reach a network outlet. In a running process, a host network link in the virtual machine is prone to failure due to a hardware fault of the network node or a software fault of each network component in the network node, and the failure of the host network link affects reliable application of the virtual machine. Therefore, in a case where the virtual machine uses the host network for network communication, the Android system needs to monitor a network state of the host network.

[0063] S304, in a case where it is monitored that the host network is in a network communication abnormal state, switching a communication network of the virtual machine to a backup network.

[0064] The backup network is in communication connection with the physical switch through a user mode provided by the fast emulator.

[0065] The network communication abnormal state of the host network is a state in which the host network cannot provide network traffic transmission services for the virtual machine, and the virtual machine cannot communicate with the outside through the host network. The network communication abnormal state of the host network can be caused by software abnormality and / or hardware abnormality.

[0066] Specifically, in a case where it is monitored that the host network is in the network communication abnormal state, it indicates that the virtual machine cannot use the host network to communicate with the outside at present, and the Android system can switch the communication network of the virtual machine to the backup network.

[0067] In the network switching method of the virtual machine, in a case where the virtual machine uses the host network for network communication, a network state of the host network is detected, and a communication link of the host network includes a virtual switch in communication connection with the virtual machine, a network service component in communication connection with the virtual switch, and a physical switch in communication connection with the network service component. In a case where it is monitored that the host network is in a network communication abnormal state, in order to make network communication of the virtual machine not be affected, the communication network of the virtual machine is switched from the host network to the backup network. Since the backup network is in communication connection with the physical switch through a user service network mode provided by the fast emulator, a dynamic host configuration protocol of the backup network can be directly provided by the fast emulator, and no longer depends on a virtualization network component. By configuring the backup network for network communication of the virtual machine, network reliability of the virtual machine can be effectively improved, and the virtual machine can still process business when the communication link of the host network fails, and high availability of the virtual machine is effectively improved.

[0068] If the Android system wants to provide two types of network communication methods, i.e., a primary network and a backup network, for a virtual machine, an interface needs to be configured in advance.

[0069] In some embodiments, as shown in Figure 4 The network switching method of the virtual machine further includes the following steps:

[0070] S402, in response to a type addition operation on a network capability configuration file, adding a primary network and a backup network in a network capability class.

[0071] The type addition operation refers to an operation of adding two types of networks in the Android system, which can be triggered by a designer based on an integrated terminal of the Android system.

[0072] The network capability configuration file is a file information for configuring the network functions that can be implemented in the Android system, and the network capability class of the Android system can be added by operating the network capability configuration file.

[0073] The network capability class (NetworkCapabilities) is an important utility class in Android development for describing network status and transmission types, which mainly provides information about network properties, such as transmission methods and network capabilities.

[0074] Specifically, the designer can trigger a type addition operation on the network capability configuration file in the Android system based on the integrated terminal of the Android system, and the Android system adds two new types in the network capability class in response to the type addition operation on the network capability configuration file, which correspond to the primary network and the backup network, respectively.

[0075] S404, in response to an interface creation operation for the primary network and the backup network, creating an Ethernet interface corresponding to the primary network and the backup network in an Ethernet service, respectively.

[0076] The interface creation operation is an operation of creating an Ethernet interface corresponding to the primary network and the backup network in the Ethernet service, which can be triggered by the designer based on the integrated terminal of the Android system.

[0077] The Ethernet service refers to a service of connecting a terminal integrated with the Android system to a local area network through an Ethernet interface, and the Ethernet interface refers to a network interface for connecting the Ethernet, which is an interface for data transmission between network devices and networks.

[0078] Specifically, the designer can trigger the interface creation operation for the primary network and the backup network based on the integrated terminal of the Android system, and the Android system creates independent NetworkAgent and NetworkFactory instances for the respective Ethernet interfaces of the primary network and the backup network in the Ethernet service in response to the interface creation for the primary network and the backup network, and implements the corresponding registration method. After the initialization of the hardware abstraction (hal) layer is completed, the instances created for the primary network and the backup network are registered in the network management service (ConnectivityService) to realize the creation of the respective Ethernet interfaces of the primary network and the backup network in the Ethernet service.

[0079] S406, in response to the function addition operation for the network management configuration file, the network request support for the primary network and the backup network is added in the network management service.

[0080] The function addition operation refers to an operation behavior of adding support for the network request of the two network capability classes of the primary network and the backup network in the Android system, and is used to support the capability of the application layer actively requesting the primary network or the backup network. The function addition operation can also be triggered by the designer based on the integrated terminal of the Android system.

[0081] Specifically, the designer can trigger the function addition operation for the network management configuration file based on the integrated terminal of the Android system, and add the network request support for the primary network and the backup network in the network management service to support the capability of the application layer actively requesting the primary network or the backup network.

[0082] S408, in response to the scoring operation for the primary network and the backup network, a first score is configured for the primary network and a second score is configured for the backup network type in the network management service. In the case where the first score is greater than the second score, the virtual machine is controlled to use the primary network for network communication.

[0083] The scoring operation is an operation behavior of setting network scores for the primary network and the backup network in the network management service. By configuring network scores for the primary network and the backup network in the network management service, the network management service can determine the default network to be used in the case where the two networks coexist. The scoring operation can also be triggered by the designer based on the integrated terminal of the Android system. It can be understood that the network score can represent the network use priority of the corresponding network in the virtual machine. For example, the higher the network score, the higher the network use priority, and the network management service will prefer the network with a higher network score as the default network of the virtual machine.

[0084] Specifically, the designer can trigger the scoring operation for the primary network and the backup network based on the integrated terminal of the Android system, and the Android system configures a first score for the primary network and a second score for the backup network in the network management service in response to the scoring operation for the primary network and the backup network. If the first score is greater than the second score, the Android system will call the network management service to control the virtual machine to use the primary network for network communication.

[0085] In some embodiments, the network score for the primary network can be configured as 70 in the network management service, and the network score for the backup network can be configured as 60, so as to ensure that the virtual machine uses the primary network as the default network in the case of coexistence of the two networks.

[0086] In the above embodiments, by configuring the Android system as described above, the Android system can support multiple Ethernet cards, and thus the Android system can support the coexistence of the primary network and the backup network at the same time, and provide a feasible use environment for the virtual machine to use the backup network.

[0087] In addition to the multiple Ethernet card support configuration on the Android system, in order to enable the virtual machine to use the backup network, in some embodiments, as shown in Figure 5 The network switching method of the virtual machine further includes:

[0088] S502, in response to the parameter addition operation for the fast emulator, a user mode enabling parameter is added to the configuration file of the fast emulator, and a virtual network card corresponding to the backup network is determined.

[0089] The parameter addition operation is to add a parameter to the simulation parameter of the fast emulator, so that the Android system can enable the user mode operation of the virtual machine. By adding a parameter to the simulation parameter of the fast emulator, a virtual network card corresponding to the backup network can be generated in the virtual machine, providing a basis for subsequent configuration of the backup network. It can be understood that the parameter addition operation for the fast emulator can be triggered by the designer based on the integrated terminal of the Android system.

[0090] The configuration file of the fast emulator is a configuration file for configuring the simulation parameter of the fast emulator, and the user mode enabling parameter is a simulation parameter for enabling the fast emulator corresponding to the user mode of the virtual machine. By adding the user mode enabling parameter to the configuration file of the fast emulator, a virtual network card corresponding to the backup network can be generated in the virtual machine.

[0091] Specifically, the designer can trigger the parameter addition operation for the fast emulator based on the integrated terminal of the Android system, and the Android system adds the user mode enabling parameter to the configuration file of the fast emulator in response to the parameter addition operation for the fast emulator.

[0092] S504, in response to the logical addition operation for the user mode, adding a routing configuration mode in the user mode.

[0093] The logical addition operation is an operation behavior for adding logic in the user mode of the fast emulator. By adding logic in the user mode of the fast emulator, the virtual machine can filter the routing table of the host through the routing configuration mode when the user service is enabled to use the backup network, determine the service network routing table corresponding to the virtual machine, and configure the service network routing table as the default routing table of the backup network. The logical addition operation can be triggered by a designer based on an integrated terminal of an Android system.

[0094] The routing configuration mode is used to determine the service network routing table corresponding to the virtual machine from each routing table of the host after the virtual machine starts the backup network, and configure the service network routing table as the default routing table of the backup network.

[0095] Specifically, the designer can trigger the logical addition operation for the user mode based on the integrated terminal of the Android system. The Android system responds to the logical addition operation for the user mode, obtains the routing configuration mode, and adds the routing configuration rule in the user mode of the fast emulator, so that the virtual machine can determine the service network routing table corresponding to the virtual machine from each routing table of the host after starting the backup network, and configure the service network routing table as the default routing table of the backup network, which can effectively avoid routing the virtual machine traffic to the management network or the storage network and other unintended networks.

[0096] S506, performing hardware abstraction processing on the virtual network card at the hardware abstraction layer of the Ethernet, and starting a dynamic host configuration protocol client for the virtual network card.

[0097] The hardware abstraction processing on the virtual network card refers to providing a unified virtual network card interface independent of specific hardware for the virtual network card in the operating system or upper layer application through the hardware abstraction layer (Hardware Abstraction Layer, HAL) technology.

[0098] Specifically, the Android system will perform hardware abstraction processing on the virtual network card at the hardware abstraction layer of the Ethernet, so that the Ethernet hardware abstraction layer initializes the virtual network card interface, and starts a dynamic host configuration protocol (DHCP) client for the virtual network card, so that the Ethernet can obtain an IP address through the DHCP service provided by the user service and connect to the public network.

[0099] S508, based on the backup network Ethernet interface registration file, calling the network management service to configure the backup network.

[0100] The Ethernet interface registration file is a registration file generated by the Android system when registering the Ethernet interface for the standby network, that is, the instance registration file corresponding to the network agent and network engineering of the standby network in the Ethernet service.

[0101] Specifically, the Ethernet framework service in the Android system can configure the standby network by calling the network management service based on the Ethernet interface file of the standby network.

[0102] In the above embodiment, by configuring the standby network in the Android system, the standby network can directly adopt the user service mode provided by the fast emulator, the DHCP service can be directly provided by the fast emulator, and the service area is limited to a single virtual machine. Compared with the IP service provided by the network node, the DHCP failure rate such as link failure, service failure and IP conflict can be significantly reduced.

[0103] In some embodiments, as Figure 6 , the network switching method of the virtual machine further includes:

[0104] S602, in the case of monitoring that the main network occurs network communication abnormality, the number of continuous abnormality of the main network occurring network communication abnormality is counted.

[0105] The number of continuous abnormality is the number of abnormality counted in the case of continuously monitoring that the main network occurs network communication abnormality. In general, when the main network occurs network communication abnormality is monitored, the Android system records the number of continuous abnormality as 1, performs a backoff retry operation on the main network, and detects the communication abnormality of the main network again. If the main network still occurs network communication abnormality is monitored, the number of continuous abnormality of the main network occurring network communication abnormality is counted as 2, and so on.

[0106] Specifically, the Android system can count the number of continuous abnormality of the main network occurring network communication abnormality in the case of monitoring that the main network occurs network communication abnormality.

[0107] S604, if the number of continuous abnormality is less than a preset number threshold, a backoff retry operation is performed on the main network, and the step of monitoring the network state of the main network is returned.

[0108] The preset number threshold is a preset judgment threshold parameter for judging whether the main network is in a network communication abnormality state. In order to avoid misjudgment of the main network occurring network communication abnormality due to signal detection error or other factors, the designer can set the preset number threshold in the Android system in advance to exclude the possibility of error and accurately judge whether the main network is in a network communication abnormality state. It can be understood that the preset number threshold can be determined by the designer according to actual needs, for example, the preset number threshold can be 5 times.

[0109] The backoff retry operation is an error handling and retry strategy commonly used in network communication and distributed systems, which means that when a retryable error occurs, the request is not immediately reinitiated, but is attempted after a period of time, and as the number of retries increases, the waiting time is gradually extended according to certain rules, for example, the Android system can use a binary exponential backoff as a backoff strategy for backoff retry.

[0110] Specifically, after the Android system counts the number of consecutive abnormal times of the main network, it can compare the number of consecutive abnormal times with the preset number threshold. If the number of consecutive abnormal times is less than the preset number threshold, it means that it cannot be determined whether the main network has a network communication abnormality at present. The Android system will perform a backoff retry operation on the main network and return to the step of monitoring the network state of the main network.

[0111] In some embodiments, if the backoff retry operation is performed on the main network, and it is monitored that the network communication of the main network returns to normal, the Android system can set the number of consecutive abnormal times to the initial value and return to the step of monitoring the network state of the main network.

[0112] S606, in the case where the number of consecutive abnormal times reaches the preset number threshold, determining that the main network is in a network communication abnormal state, setting the number of consecutive abnormal times to an initial value and returning to the step of monitoring the network state of the main network.

[0113] The initial value is a basic value preset by the designer for the number of consecutive abnormal times, for example, the initial value of the number of consecutive abnormal times can be zero, and the operation of setting the number of consecutive abnormal times to the initial value can be considered as zeroing the number of consecutive abnormal times.

[0114] Specifically, if the Android system determines that the number of consecutive abnormal times of the main network has reached the preset number threshold, it can be determined that the main network cannot currently provide network traffic transmission services for the virtual machine. The Android system can determine that the main network is in a network communication abnormal state, and can perform a network switching operation of switching the communication network of the virtual machine to the standby network. At the same time, the Android system will set the number of consecutive abnormal times to the initial value and return to the step of monitoring the network state of the main network, so as to timely switch the communication network of the virtual machine from the standby network back to the main network in the case where the main network returns to normal communication.

[0115] In the above embodiments, by performing a backoff retry operation on the main network and counting the number of consecutive abnormal times of the main network, the number of consecutive abnormal times is compared with the preset number threshold to determine whether the main network is really in a network communication abnormal state, which can effectively improve the accuracy of monitoring the network state of the main network and provide an accurate data basis for subsequent network switching.

[0116] Further, in some embodiments, the network switching method of the virtual machine further comprises: determining that the main network has a network communication exception in the case that a dynamic host configuration protocol exception or a network connectivity exception occurs in the main network.

[0117] The dynamic host configuration protocol exception refers to an exception type in which the virtual machine cannot correctly obtain an IP address and other related network configuration parameters through DHCP. For example, a DHCP process initiated by a client, such as a discover or request timeout without response or negative acknowledgment (NACK), can be counted as a dynamic host configuration protocol exception, and the number of consecutive exceptions needs to be counted. For another example, a subsequent renew or rebound timeout without response or negative acknowledgment can also be counted as a dynamic host configuration protocol exception, and the number of consecutive exceptions needs to be counted. It can be understood that the dynamic host configuration protocol exception can be determined by the Android system calling the network management service according to the notification broadcast by the DHCP client module for monitoring the main network.

[0118] The network connectivity exception refers to an exception type in which the virtual machine has a network communication connection exception. For example, the domain name server based on http and https can be continuously tested, and in the case of DNS resolution exception, or ICMP exception, or socket establishment exception, or socket read-write exception, or SSL handshake exception, it can be counted as a network connectivity exception, and the number of consecutive exceptions needs to be counted. It can be understood that the network connectivity exception can be determined by the Android system calling the network management service according to the notification broadcast by the network monitor module for monitoring the main network.

[0119] Specifically, in the network management service of the Android system, the DHCP client module for monitoring the dynamic host configuration protocol exception of the main network and the network monitor module for monitoring the network connectivity exception of the main network are provided. In the case that the DHCP client module and the monitor module monitor the dynamic host configuration protocol exception or the network connectivity exception of the main network, the statistics module of the network management service is directly broadcasted, and the statistics module determines that the main network has a network communication exception and counts the number of consecutive network communication exceptions of the main network.

[0120] In the above embodiments, by monitoring the dynamic host configuration protocol exception and the network connectivity exception of the main network, it can be determined whether the main network has a network communication exception in time, and the accuracy and efficiency of network switching are improved.

[0121] In some embodiments, the switching of the communication network of the virtual machine to the backup network in S304 comprises: in a case where the first score is less than the second score, invoking a network management service, performing network evaluation on the communication network used by the virtual machine, and performing a network switching operation based on the evaluation result. The default route of the virtual machine is set to the network route of the backup network.

[0122] The network route of the backup network is configured by a routing configuration manner.

[0123] Specifically, in a case where the Android system determines that the main network is in the network communication abnormal state, the Android system can reduce the first score of the main network, so that the first score of the main network is less than the second score of the backup network. For example, the first score of the main network is originally 70 points, and the second score of the backup network is 60 points. In a case where the Android system determines that the main network is in the network communication abnormal state, the first score can be reduced to 50 points. The Android system invokes the network management service to re-evaluate the communication network used by the virtual machine. Since the first score of the main network is less than the second score of the backup network at this time, the network management service switches the default network of the virtual machine to the backup network and notifies the network daemon (Network Daemon, netd) in the Android system to set the default routing table to the network route of the backup network, so as to realize that the upper-layer application of the Android system defaults to accessing the Internet through the backup network. It can be understood that the network route of the backup network can be determined from each routing table of the host machine by a routing configuration manner. The network route of the backup network is the business routing table corresponding to the virtual machine.

[0124] In the above embodiments, by reducing the first score of the main network and invoking the network management service to perform the network switching work, the accuracy and switching efficiency of the virtual machine network switching can be effectively improved.

[0125] In some embodiments, in a case where the Android system monitors that the main network communication is normal, the Android system can increase the first score of the main network, for example, increase the first score of the main network from 50 points to 70 points, so that the first score of the main network is greater than the second score of the backup network. The Android system invokes the network management service to re-evaluate the communication network used by the virtual machine. The network management service switches the default network of the virtual machine to the main network and notifies the network daemon (Network Daemon, netd) in the Android system to set the default routing table to the network route of the main network, so as to realize that the upper-layer application of the Android system defaults to accessing the Internet through the main network.

[0126] The business network in the Openstack architecture is a physical connection between nodes and does not depend on virtualized network components. The stability and reliability of the business network are higher, and the business network can provide a reliable backup link when the virtual network has a problem. Based on this, in some embodiments, as shown in FIG. 6, the Android system can be configured to monitor the network communication state of the main network and the backup network. In a case where the Android system determines that the main network is in the network communication abnormal state, the Android system can reduce the first score of the main network, so that the first score of the main network is less than the second score of the backup network. The Android system invokes the network management service to re-evaluate the communication network used by the virtual machine. Since the first score of the main network is less than the second score of the backup network at this time, the network management service switches the default network of the virtual machine to the backup network and notifies the network daemon (Network Daemon, netd) in the Android system to set the default routing table to the network route of the backup network, so as to realize that the upper-layer application of the Android system defaults to accessing the Internet through the backup network. Figure 7As shown, a network switching method of a virtual machine is provided, including network configuration and network switching two parts, and specifically can include the following steps:

[0127] First, the Android system is configured to enable the Android system to support multiple Ethernet cards, which can specifically include the following steps:

[0128] The designer can add two new types in the NetworkCapabilities class to correspond to the primary network and the backup network.

[0129] In the Ethernet service, independent NetworkAgent and NetworkFactory instances are created for the Ethernet interfaces eth0 and eth1 corresponding to the primary network and the backup network, and the corresponding registration method is implemented, and they are registered to ConnectivityService after the hal layer initialization is completed.

[0130] In the onnectivityService, the network request support for two NetworkCapabilities is added to support the application layer to actively request the capability of the primary network or the backup network.

[0131] In ConnectivityService, the default primary network score is 70 points, and the redundant network is 60 points. In this way, in the case of coexistence of the two networks, the primary network is used as the default network.

[0132] Then, the primary and backup networks are configured on the QEMU side, wherein the primary network configuration can specifically include the following steps:

[0133] The designer enables a tap card by configuring the QEMU parameter-netdevtap, id=tap0, ifname=tap_w1-devicevirtio-net-pci, netdev=tap0, mac=fa:00:11:22:33:44, and a corresponding network card eth0 is generated in the virtual machine to be used as the primary network. Openstack manages the traffic through the mac of the network card.

[0134] The Android virtual machine enables the Ethernet function by configuring the android system-feature, and after the Android system is started, the Ethernet hal layer is initialized to find the Ethernet card eth0 and start the DHCP client on the card. In this way, the Ethernet can perform DHCP through the virtual network provided by the Openstack network node and obtain the ip to access the public network.

[0135] The Ethernet framework service registers the corresponding type of network in the system service ConnectivityService using the NetworkAgent and NetworkFactory of the main network, and the default score is 70.

[0136] ConnectivityService registers a routing table named 200 eth0 in the Android system, and the default route points to the eth0 device and the virtual network gateway. This routing table serves as the first priority routing table of the system.

[0137] Similarly, in order for the virtual machine to use the backup network, the backup network also needs to be configured on the QEMU side, which includes the following steps:

[0138] The designer can add the QEMU parameter -netdev user, id=mynet -device virtio-net-pci, netdev=mynet to enable the Slirp network of the virtual machine. A virtual network card with the device name eth1 will be generated in the virtual machine, which will be used as the backup network.

[0139] In the Slirp implementation of QEMU, add logic to filter the host's routing table after the virtual machine starts the Slirp network, find the route corresponding to the business network card through the specified rule, and set the default route of Slirp to this route to avoid routing virtual machine traffic to non-intended networks such as management network / storage network.

[0140] Similar to the main network process, the Ethernet hal layer will initialize the eth1 interface and start the DHCP client on this card. In this way, Ethernet can obtain an IP through the DHCP service provided by Slirp and connect to the public network. QEMU will forward the virtual machine traffic to the business network card according to the logic added in the Slirp implementation.

[0141] The Ethernet framework service registers the corresponding type of network in the system service ConnectivityService using the NetworkAgent and NetworkFactory of the backup network, and the network score is set to 60. ConnectivityService registers a routing table named 201 eth1, and the default route points to eth1 and the gateway of the Slirp network. Since the score of the backup network is lower than that of the main network (60<70), the default network of the Android system is the main network, and the default routing table is 200 eth0.

[0142] After completing the configuration of the Android system for multiple Ethernet and the configuration of the main and backup networks on the QEMU side, the network switching scheme can be implemented based on the configured Android system, which can include the following steps:

[0143] The Android system can add an exception statistics module in ConnectivityService to collect and count the network exception information reported by the DHCPClient module and the NetworkMonitor module, and switch to a backup network when the threshold is reached. The relationship between the modules is as shown in Figure 8

[0144] The DHCPClient module and the NetworkMonitor module directly broadcast the network exception information to the network exception statistics module of ConnectivityService after encountering an exception, and back off and retry. The back-off strategy adopts binary exponential back-off. The network exception statistics module takes five accumulations as the threshold. When encountering a network normal notification, the exception count is cleared. When the exception count exceeds the threshold, it is checked whether the current default network is the main network type. If so, the default network is switched to the backup network.

[0145] Among them, DHCPclient counts DHCP exceptions. In the DHCP process initiated by the client, such as discover or request timeout without response or nack, it is counted as an exception count. Subsequent renew or rebound timeout or nack is counted as an exception count. NetworkMonitor counts network connectivity exceptions. Continuous timing test based on http and https domain name servers. DNS resolution exception, icmp exception, socket establishment or read-write exception, SSL handshake exception, etc. are all counted as an exception and counted.

[0146] After switching to the redundant network, the DHCPclient process and the NetworkMonitor connectivity detection module of the main network continue to work. When ConnectivityService receives the main network normal notification, the default network is switched back to the main network.

[0147] Among them, ConnectivityService can switch the default network to the backup network by reducing the score of the main network to 50 points and triggering ConnectivityService to re-evaluate the network. At this time, the score of the main network is lower than that of the backup network (50<60), ConnectivityService will switch the default network to the backup network, and notify netd to set the default routing table to 201 eth1. At this time, the upper layer application of the Android system defaults to access the Internet through the backup network.

[0148] ​The ConnectivityService switches the default network back to the primary network by increasing the primary network score back to 70 points, triggering the ConnectivityService to reevaluate the network implementation, and since the primary network score is higher, the default network switches back to the primary network with the default routing table set to 200 eth0.

[0149] The network switching method of the virtual machine in the embodiment first uses the Slirp network mode provided by QEMU for the standby network, and the DHCP service is directly provided by QEMU, and the service area is limited to a single virtual machine. Compared with the ip service provided by the network node, the DHCP failure rate such as link failure, service failure, and ip conflict can be significantly reduced. Secondly, the standby network relies on the Openstack business network, which is the physical connection between nodes, and does not rely on the virtualization network component, and has higher stability and reliability, and ensures that the service does not interrupt when the virtual network link fails.

[0150] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or steps or stages in other steps.

[0151] Based on the same inventive concept, the embodiment of the present application also provides a network switching device for virtual machine for implementing the network switching method of the virtual machine as described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more network switching device embodiments of the virtual machine provided below can refer to the limitations of the network switching method of the virtual machine in the above, which will not be repeated here.

[0152] In some embodiments, as shown in Figure 9 A network switching device 900 for virtual machine is provided, comprising a network state monitoring module 901 and a network switching module 902, wherein:

[0153] The network state monitoring module 901 is configured to monitor the network state of the primary network in a case where the virtual machine uses the primary network for network communication; the communication link of the primary network comprises a virtual switch connected with the virtual machine in communication, a network service component connected with the virtual switch in communication, and a physical switch connected with the network service component in communication;

[0154] The network switching module 902 is configured to switch the communication network of the virtual machine to a backup network in a case where it is monitored that the primary network is in an abnormal network communication state; the backup network is connected with the physical switch in communication through a user service network mode provided by a fast emulator; the fast emulator is configured to simulate and generate the virtual machine.

[0155] In some embodiments, the network switching apparatus of the virtual machine further comprises:

[0156] The type adding module is configured to add the primary network and the backup network in the network capability category in response to a type adding operation for the network capability configuration file.

[0157] The interface creating module is configured to create an Ethernet interface corresponding to the primary network and the backup network respectively in the Ethernet service in response to an interface creating operation for the primary network and the backup network.

[0158] The function adding module is configured to add network request support for the primary network and the backup network in the network management service in response to a function adding operation for the network management configuration file.

[0159] The network scoring module is configured to configure a first score for the primary network and a second score for the backup network type in the network management service in response to a scoring operation for the primary network and the backup network; and control the virtual machine to use the primary network for network communication in a case where the first score is greater than the second score.

[0160] In some embodiments, the network switching apparatus of the virtual machine further comprises:

[0161] The parameter adding module is configured to add a routing configuration mode in the user mode in response to a logical adding operation for the user mode; the routing configuration mode is configured to determine a service network routing table corresponding to the virtual machine from each routing table of the host computer after the virtual machine starts the backup network, and configure the service network routing table as a default routing table of the backup network.

[0162] The hardware abstraction processing module is configured to perform hardware abstraction processing on the virtual network card at a hardware abstraction layer of the Ethernet, and start a dynamic host configuration protocol client for the virtual network card.

[0163] The backup network configuration module is configured to register a file of the Ethernet interface of the backup network, and invoke the network management service to configure the backup network.

[0164] In some embodiments, the network switching apparatus of the virtual machine further comprises:

[0165] a continuous abnormality number counting module, configured to count a continuous abnormality number of the network communication abnormality of the main network when the network communication abnormality of the main network is monitored.

[0166] a backoff retry module, configured to perform a backoff retry operation on the main network and return to the step of monitoring the network state of the main network if the continuous abnormality number is less than a preset number threshold.

[0167] an abnormality state determining module, configured to determine that the main network is in a network communication abnormality state if the continuous abnormality number reaches the preset number threshold, set the continuous abnormality number to an initial value and return to the step of monitoring the network state of the main network.

[0168] In some embodiments, the network switching apparatus of the virtual machine further comprises:

[0169] a communication abnormality determining module, configured to determine that the main network is in a network communication abnormality state if a dynamic host configuration protocol abnormality or a network connectivity abnormality occurs in the main network.

[0170] In some embodiments, the network switching module 902 is configured to: call a network management service to perform network evaluation on the communication network used by the virtual machine and perform a network switching operation based on the evaluation result if the first score is less than the second score; set a default route of the virtual machine to a network route of the standby network; and the network route of the standby network is obtained through a routing configuration manner.

[0171] The above various modules of the network switching apparatus of the virtual machine can be all or partially realized by software, hardware and combinations thereof. The above various modules can be embedded in or independent of a processor in a computer device in a hardware form, or stored in a memory in a computer device in a software form, so as to be called and executed by a processor to perform the operations corresponding to the above various modules.

[0172] In some embodiments, a computer device is provided, which can be a cloud terminal integrated with an Android system, and an internal structure diagram of the computer device can be as shown in Figure 10As shown in the figure. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store related data for network switching of a virtual machine. The network interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement a network switching method of a virtual machine.

[0173] Those skilled in the art can understand that, Figure 10 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0174] In some embodiments, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the specific implementation steps of the network switching method of the virtual machine.

[0175] In some embodiments, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by the processor to implement the specific implementation steps of the network switching method of the virtual machine.

[0176] In some embodiments, a computer program product is provided, including a computer program, and the computer program is executed by the processor to implement the specific implementation steps of the network switching method of the virtual machine.

[0177] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties. And the acquisition, storage, processing, transmission, etc. of data comply with the relevant provisions of laws and regulations.

[0178] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0179] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0180] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A network switching method of a virtual machine, characterized by, The method comprises: monitoring a network state of a main network in a case that a virtual machine uses the main network for network communication; a communication link of the main network comprises a virtual switch in communication connection with the virtual machine, a network service component in communication connection with the virtual switch, and a physical switch in communication connection with the network service component; switching a communication network of the virtual machine to a backup network in a case that the main network is in an abnormal network communication state; the backup network is in communication connection with the physical switch through a user mode provided by a fast emulator; the fast emulator is used for simulating generation of the virtual machine.

2. The method of claim 1, wherein, The method further comprises: adding the main network and the backup network in a network capability class in response to a type adding operation for a network capability configuration file; creating an Ethernet interface of the main network and the backup network respectively in an Ethernet service in response to an interface creation operation for the main network and the backup network; adding network request support for the main network and the backup network in a network management service in response to a function adding operation for a network management configuration file; configuring a first score for the main network and a second score for the backup network in the network management service in response to a scoring operation for the main network and the backup network, and controlling the virtual machine to use the main network for network communication in a case that the first score is greater than the second score.

3. The method of claim 2, wherein, The method further comprises: adding a user mode enabling parameter in a configuration file of the fast emulator and determining a virtual network card corresponding to the backup network in response to a parameter adding operation for the fast emulator; adding a routing configuration mode in the user mode in response to a logic adding operation for the user mode, the routing configuration mode being used for determining a service network routing table corresponding to the virtual machine from each routing table of a host computer after the virtual machine starts the backup network, and configuring the service network routing table as a default routing table of the backup network; performing hardware abstraction processing on the virtual network card in a hardware abstraction layer of Ethernet, and starting a dynamic host configuration protocol client for the virtual network card; registering a file of an Ethernet interface of the backup network, and calling the network management service to configure the backup network.

4. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: counting a continuous abnormal number of network communication abnormality of the main network in a case that the main network is monitored to be in network communication abnormality; performing a back-off retry operation for the main network in a case that the continuous abnormal number is less than a preset number threshold, and returning to perform the step of monitoring the network state of the main network; determining that the main network is in the abnormal network communication state in a case that the continuous abnormal number reaches the preset number threshold, setting the continuous abnormal number as an initial value, and returning to perform the step of monitoring the network state of the main network.

5. The method of claim 4, wherein, The method further comprises: determining that the main network is in network communication abnormality in a case that the main network is in dynamic host configuration protocol abnormality or network connectivity abnormality.

6. The method of claim 3, wherein, The switching of the communication network of the virtual machine to the backup network comprises: In the case that the first score is less than the second score, calling a network management service to perform network evaluation on the communication network used by the virtual machine, and performing a network switching operation based on the evaluation result; Setting a default route of the virtual machine to a network route of the backup network; the network route of the backup network is configured by the route configuration mode.

7. A network switching apparatus of a virtual machine, characterized by comprising: The apparatus comprises: A network state monitoring module, configured to monitor a network state of a host network in the case that a virtual machine uses the host network for network communication; a communication link of the host network comprises a virtual switch connected in communication with the virtual machine, a network service component connected in communication with the virtual switch, and a physical switch connected in communication with the network service component; A network switching module, configured to switch a communication network of the virtual machine to a backup network in the case that the host network is in an abnormal network communication state; the backup network is connected in communication with the physical switch through a user service network mode provided by a fast emulator; the fast emulator is used to simulate generation of the virtual machine.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method in any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 6.

Images