WO2022151776A1 - Cloud platform virtual machine recycling method and computer device - Google Patents

Abstract

The present invention provides a cloud platform virtual machine recycling method and a computer device. The method comprises: a front end receives a deletion instruction for a virtual machine, places the virtual machine in a front-end recycle bin deployed at the front end, and then sends a notification to a rear end; the rear end receives the notification sent by the front end and performs permission check on the deletion instruction, and for the deletion instruction that passes permission check, separately and correspondingly places, according to the current state of a virtual machine to be deleted, the virtual machine in a recycling thread of a rear-end recycle bin deployed at the rear end; and the virtual machine to be deleted that is placed in the rear-end recycle bin does not unbind a binding relationship between the virtual machine to be deleted and a virtual resource before executing a corresponding recycling thread. The present invention improves the security and stability when executing the deletion operation of the virtual machine, ensures the data security of a user, reduces the resource consumption of the virtual machine placed in the recycle bin, reduces the response time of a concurrent operation, and greatly improves the user experience.

Description

A kind of cloud platform virtual machine recycling method and computer equipment

This application claims the priority of the Chinese patent application whose filing date is January 12, 2021, the application number is 202110034978X, and the invention title is "A method for recycling a virtual machine on a cloud platform and computer equipment", the entire contents of which are incorporated by reference in in this application.

technical field

The invention relates to the technical field of cloud computing, and in particular, to a cloud platform virtual machine recycling method and computer equipment.

Background technique

Performing the recovery operation on the virtual machine in the cloud platform is a function of recovering the virtual machine as needed after the virtual machine deletion operation, so as to effectively recover the virtual machine deletion caused by the user's misoperation during the operation process, and improve the security and stability of the virtual machine. It effectively protects the user's data security and ensures a good user experience.

In the prior art, the recycle operation of virtual machines is usually bound to the orchestration engine of the cloud platform, and the virtual machines put into the recycle bin still reside in the memory for a period of time. Physical memory and/or virtual memory cause huge resource consumption. At the same time, when different types of cloud platforms perform resource orchestration, if they are bound to each other with the orchestration engine, the function of recycling virtual machines cannot be realized.

After searching, the applicant found that the Chinese invention patent with publication number CN109117251A discloses the prior art of "a method, device and readable storage medium for realizing a virtual machine recycle bin", which is based on the operating environment of the cloud platform of OpenStack. In the prior art, in order to recycle a virtual machine, the virtual machine goes through a pause state and a shutdown state in sequence during the whole recycle operation, and completely deletes the information used by the virtual machine in the recycle bin after staying in the shutdown state for a period of time. When the suspended state enters the shutdown state, the system disk and data disk of the virtual machine are converted into images and stored in the image storage. At the same time, after staying in the shutdown state for a period of time, the system disk data and data disk data information in the recycle bin are deleted.

The applicant pointed out that the prior art needs to store the data information of the virtual machine in the memory regardless of the state, so that the resource overhead of the computer forming the cloud platform is too large; The disk and the data disk are saved separately, in fact, not only will the data remain, but also the operation will be cumbersome, and more seriously, the virtual machine will not be able to encapsulate the corresponding system disk and data disk into the original virtual machine when it is restored; In addition, when performing virtual machine recovery, the prior art points out that a virtual machine is created by starting an image, a volume is created from a data disk image, and then the data disk is mounted on the virtual machine, so that virtual recovery can be completed. The applicant points out that the technical solution completely deviates from the basic common sense of virtual machines.

Finally, the above-mentioned prior art defines the implementation device of the virtual machine recycle bin in a functionally limited description manner. However, the prior art does not explain how to implement it in the cloud platform of OpenStack from the perspective of cloud computing and the infrastructure of the virtual machine. The so-called virtual machine recycling mechanism and the specific implementation process. In addition, the applicant points out that other related existing technologies cannot solve the recovery operation of virtual machines created and running in different types of cloud platforms in the scenario of hosting multiple different types of cloud platforms.

In view of this, it is necessary to improve the cloud platform virtual machine recycling method in the prior art to solve the above problems.

SUMMARY OF THE INVENTION

The purpose of the present invention is to disclose a cloud platform virtual machine recycling method and computer equipment, which are used to solve the problem of performing efficient and safe recycling operations on virtual machines in different current states in the cloud platform, especially to avoid performing recycling operations on virtual machines. In the process, resources such as computing resources and memory resources are occupied and wasted.

In order to achieve the above-mentioned first purpose of the invention, the present invention provides a cloud platform virtual machine recycling method, including:

The front-end receives the deletion instruction for the virtual machine, and after placing the virtual machine into the front-end recycle bin deployed on the front-end, the front-end sends a notification to the back-end;

The back-end receives the notification from the front-end and performs permission verification on the deletion instruction. For the deletion instruction that has passed the permission verification, according to the current state of the virtual machine to be deleted, the corresponding virtual machine is put into the back-end deployed on the back-end for recycling. In the recycling thread of the back-end recycling bin, the virtual machine to be deleted placed in the back-end recycling bin does not release the binding relationship between the virtual machine to be deleted and the virtual resource before executing the corresponding recycling thread.

As a further improvement of the present invention, after the recycle thread that puts the virtual machine into the back-end recycle bin of the deployment back-end is executed, it also includes:

The back-end recycle bin periodically checks the virtual machine that fails to release the binding relationship between the virtual machine to be deleted and the virtual resource, and performs forced recycling of the virtual machine.

As a further improvement of the present invention, the front end judges the current state of the virtual machine to be deleted, and only executes the recycling thread put into the back-end recycle bin for the virtual machine to be deleted whose current state is the shutdown state, the power-on state, the abnormal state and the unknown state. , returns an alarm message to the deletion command initiated by the virtual machine to be deleted that is not in the shutdown state, power-on state, abnormal state or unknown state, and refuses to put the virtual machine into the recycling thread of the back-end recycle bin deployed in the back-end.

As a further improvement of the present invention, the receiving instruction to delete the virtual machine is received in the front-end visual interface;

The permission verification performed by the backend for the deletion instruction includes at least one of the following verification contents: user id, virtual machine id, virtual machine name, virtual machine state, operating system running the virtual machine, hardware information forming the virtual machine, deploying the virtual machine The cloud platform information and the cluster attribute information of the cluster where the virtual machine is located;

When the back-end fails to perform permission verification on the deletion instruction based on the verification content, the back-end sends an instruction to abort the deletion operation to the front-end recycle bin, and the back-end receives the notification from the front-end and performs permission verification on the deletion instruction And after the permission verification of the deletion instruction is successful based on the verification content, the binding relationship between the virtual machine to be deleted and the virtual resource is released for the virtual machine to be deleted placed in the back-end recycle bin.

As a further improvement of the present invention, the binding relationship between the virtual machine and the virtual resource includes the binding established between the virtual machine and one or more of the virtual resources in the virtual network, security group, key pair or virtual disk. established relationship.

As a further improvement of the present invention, the front end includes at least one cloud platform, the front end is managed by the back end, the front end and the back end are logically independent of each other, and the back end is configured as a hybrid cloud management platform.

As a further improvement of the present invention, it also includes:

Create multiple availability zones at the front end, where the availability zone is a virtual digital center, and establish an association relationship between virtual machines and different permissions configured for users in the availability zone, the availability zone is created by the hybrid cloud management platform, and the The availability zone includes at least one cloud platform, and the availability zone is managed by the backend.

As a further improvement of the present invention, the front-end is configured with a first database, and the back-end is configured with a second database, the first database is used to record the status data of the virtual machines in the availability zone, and the second database is used to record The verification content associated with the permission verification performed by the terminal on the deletion instruction, when the virtual machine to be deleted is recycled, the first database is updated through the second database; the availability zone stores at least one cloud operating system or virtual machine machine management program.

As a further improvement of the present invention, when the current state of the virtual machine to be deleted is an unknown state, the virtual machine to be deleted is added to the list of virtual machines to be deleted, and the back-end creates an asynchronous virtual machine deletion thread to delete the virtual machine to be deleted The delete operation is performed immediately after the machine is placed in the back-end recycle bin.

As a further improvement of the present invention, when the current state of the virtual machine to be deleted is the shutdown state, the field information retained in the second database of the virtual machine to be deleted is modified, and the modified field information is compared with the virtual machine to be deleted. The mounted disk information and network configuration information are recorded in a log file, and the log file is saved to the second database for visual display to the front end.

As a further improvement of the present invention, when the current state of the virtual machine to be deleted is a power-on state or an abnormal state, the back-end performs the operation of adding the virtual machine in the power-on state and the deleted virtual machine to the list of virtual machines to be shut down, and the back-end Creates an asynchronous shutdown virtual machine thread.

As a further improvement of the present invention, it also includes: using a comparison and exchange algorithm to update the virtual machine state data recorded in the first database with the underlying state data stored in the availability zone of at least one cloud operating system or a virtual machine management program; the The hybrid cloud management platform invokes the updated virtual machine state data in the first database, and updates the attributes of the same field in the second database.

As a further improvement of the present invention, it also includes: using a comparison and exchange algorithm to update the virtual machine state data recorded in the first database with the underlying state data stored in the availability zone of at least one cloud operating system or a virtual machine management program; the The second database directly calls the updated virtual machine state data in the first database, and updates the attributes of the same field in the second database.

As a further improvement of the present invention, the front end is configured as a distributed system architecture.

Based on the same inventive idea, the present application also discloses a computer device, including:

processor, memory, and

a computer program stored in the memory and configured to be executed by the processor;

When executing the computer program, the processor executes the steps in the cloud platform virtual machine recycling method according to any one of the above inventions.

Compared with the prior art, the beneficial effects of the present invention are:

First of all, the cloud platform virtual machine recycling method disclosed in the present application solves the problem that the cloud platform, especially the virtual machines with different current states including different types of cloud platforms, performs efficient and safe recycling operations, and significantly reduces the need for virtual machines. It reduces the occupation and waste of resources such as computing resources and memory resources during the process of recycling operations by the computer, which reduces the response time of concurrent operations;

At the same time, in the cloud platform virtual machine recycling method disclosed in the present application, since the front-end recycling bin located in the front end and the back-end recycling bin located in the back end are separated from each other, the virtual machine recycling operation is effectively avoided. The defect of unbinding the various resources previously configured by the virtual machine makes it possible to establish resource binding with the various resources and policies bound to the virtual machine before the virtual machine is executed and deleted after the virtual machine is restored due to the erroneous deletion operation. The fixed relationship shortens the recovery speed of the virtual machines that have been put into the back-end recycle bin and the resources consumed by the recovery process, thereby indirectly improving the user experience.

Description of drawings

Fig. 1 is the main flow chart of the cloud platform virtual machine recycling method;

Figure 2 is an example diagram in which the hybrid cloud management platform creates multiple availability zones and each availability zone is composed of multiple cloud platforms of the same type;

Figure 3 is an example diagram in which the hybrid cloud management platform creates multiple availability zones and each availability zone is composed of multiple cloud platforms of different types;

Fig. 4 is the architecture diagram of realizing the virtual machine recycle bin in the cloud platform;

FIG. 5 is a detailed flowchart of a method for reclaiming a virtual machine on a cloud platform according to the present invention, which includes the main thread of reclaiming the virtual machine, and, according to whether the state of the virtual machine is a power-on state or an unknown state, respectively starting asynchronously closing the virtual machine thread and asynchronously delete virtual machine threads;

6 is an example diagram of a virtual machine recycle bin implementation device running a cloud platform virtual machine recycling method of the present invention, performing resource deletion and resource recovery on resources formed by each node in the cloud platform during the virtual machine recycling process;

FIG. 7 shows that the availability zone including at least one cloud platform is managed by the hybrid cloud management platform, and the updated virtual machine state data in the first database is called during the recovery operation for the virtual machine, and the same field in the second database is called. Schematic diagram of the properties being updated in one instance;

FIG. 8 shows that the availability zone including at least one cloud platform is managed by the hybrid cloud management platform, and the updated virtual machine state data in the first database is called during the recovery operation of the virtual machine, and the same field in the second database is called. Schematic diagram of the properties being updated in another instance;

Fig. 9 is the detailed flow chart of back-end putting virtual machines into the recycle bin in batch operation mode;

Figure 10 is a detailed flowchart of the back-end putting virtual machines into the recycle bin in batch operation mode and performing verification by the back-end;

FIG. 11 is a topology diagram of a computer device running a cloud platform virtual machine recycling method of the present invention.

Detailed ways

The present invention will be described in detail below with reference to the various embodiments shown in the accompanying drawings, but it should be noted that these embodiments do not limit the present invention. Equivalent transformations or substitutions all fall within the protection scope of the present invention.

The term "logic" includes any physical and tangible function for performing a task. For example, each operation shown in the flowchart corresponds to a logical component for performing that operation. Operations may be performed using, for example, software running on a computer device, hardware (eg, chip-implemented logic functions), etc., and/or any combination thereof. When implemented by a computing device, logical components represent electrical components that are physical parts of a computer system, however implemented.

The phrase "configured to" or the phrase "configured to" includes any manner in which any kind of physical and tangible function can be constructed to perform the identified operation. Functions may be configured to perform operations using, for example, software running on a computer device, hardware (eg, chip-implemented logic functions), etc., and/or any combination thereof.

Example 1:

Referring to FIG. 1 to FIG. 10, the detailed implementation process of a cloud platform virtual machine recycling method (hereinafter referred to as "recycling method") of the present invention is disclosed. The recycling method disclosed in this embodiment is initiated by the user in the visual operation interface in the first cloud platform 11 and/or the second cloud platform 21 in the availability zone 10 and/or the availability zone 20 . The visual operation interface is embedded to form an operation window for administrators or users to perform recovery operations on virtual machines. Administrators or users can initiate operations on one or more virtual machines or several types of virtual machines of a cloud platform in the visual operation window. One or more virtual machines of the cloud platform perform the recycling operation.

The recycling method includes the following steps S1 to S2.

Step S1: the front end receives the deletion instruction for the virtual machine, and after placing the virtual machine in the front-end recycle bin deployed in the front-end, the front-end sends a notification to the back-end. The front end includes at least one cloud platform, the front end is managed by the back end, the front end and the back end are logically independent from each other, and the back end is configured as the hybrid cloud management platform 100 . The front end is configured as a distributed system architecture, and the front end can deploy multiple cloud platforms. The use of a distributed system architecture is conducive to concurrent responses with multiple users, preprocessing and allocation of virtual resources, recording and calling of log files, authority verification, and data monitoring. The virtual machine performs specific operation events such as recycling operations.

In this embodiment, the front end can be understood as a visual operation interface capable of initiating a virtual machine recycling operation instruction. The command to delete the virtual machine is received in the front-end visual interface. In the front-end, pages such as the display of virtual machines, the deletion process, and the dialog box formed by deletion events are formed visually.

The front end judges the current state of the virtual machine to be deleted (that is, the virtual machine to be reclaimed), and only executes the execution of the virtual machine to be deleted whose current state is the shutdown state, the power-on state, the abnormal state and the unknown state, and puts it into the back-end recycle bin The recycle thread of the server returns an alarm message to the deletion command initiated by the virtual machine to be deleted that is not in the shutdown state, startup state, abnormal state or unknown state, and refuses to put the virtual machine into the back-end recycle bin deployed on the back-end for recycling. in the thread. Therefore, in this embodiment, the term "to be deleted" and the term "to be recovered" have the same meaning.

The user can first select the virtual machine that needs to be put into the front-end recycle bin on the visual operation interface formed by the hybrid cloud management platform 100, and click the "Delete" button. "Shutdown", "Power on", "Unknown" and "Abnormal" status of virtual machines to perform recycling operations), a prompt box will pop up on the front-end page to prompt the user, and after the user confirms the name of the virtual machine that needs to be placed in the front-end recycling bin for the second time , after clicking the "Confirm" button, the request data is sent to the backend, which is deployed and runs in the hybrid cloud management platform 100 .

If the "Cancel" button is clicked, the operation request will be terminated. When the user clicks the "Confirm" button, the request url "placeVmInRecycleBin" will be used to send various data required for the operation based on the recycling virtual machine to the backend, including the user id. , virtual machine id, etc. to point to the determined virtual machine that needs to be reclaimed.

Step S2: the back end receives the notification sent by the front end and performs permission verification on the deletion instruction, and executes the corresponding deletion instruction according to the current state of the virtual machine to be deleted according to the current state of the virtual machine to be deleted. In the recycling thread of the back-end recycling bin, the virtual machine to be deleted placed in the back-end recycling bin does not release the binding relationship between the virtual machine to be deleted and the virtual resource before executing the corresponding recycling thread. Preferably, in this embodiment, after the recycle thread of placing the virtual machine into the back-end recycle bin of the deployment backend is executed, the method further includes: the back-end recycle bin regularly checks the binding relationship between the virtual machine to be deleted and the virtual resource Decommission the failed virtual machine and perform a forced reclamation of the virtual machine.

In this embodiment, the binding relationship between the virtual machine and the virtual resource includes the binding relationship established between the virtual machine and one or more virtual resources of a virtual network, a security group, a key pair, or a virtual disk. Taking the key pair as an example, if the key pair has been bound to the virtual machine and the virtual machine is placed in the front-end recycle bin, when the user deletes the key pair from the key pair list, the front-end recycle bin will It will first check the record that the corresponding key pair has been saved in the first database 15. When it is checked that the key pair has been bound to the virtual machine, an error will be thrown and an error message will be returned to the front-end, resulting in the Termination of the virtual machine's reclamation operation. After the front end gets the returned error message, it will prompt the user with the error message in a visual form such as a warning box. Therefore, through the above-mentioned binding relationship established for the key pair, it is possible to effectively avoid the erroneous recovery operation of the virtual machine caused by human error operation or malicious recovery operation during the process of recovering the virtual machine. It has important technical value in scenarios such as payment systems and settlement systems of online shopping platforms; and because the recovery method disclosed in this embodiment does not require the recovered virtual machine to reside in memory during the entire process of performing the recovery operation on the virtual machine , thereby effectively avoiding the disorderly occupation of physical resources and virtual resources of various types of cloud platforms, and improving the overall stability of the cloud platform.

The binding relationship between the virtual machine and other types of virtual resources is referred to above, and is not repeated here.

Through the above technical solutions, the security and stability of the virtual machine deletion operation are improved, and the user data security is ensured; at the same time, it is also possible to effectively avoid the previously configured virtual machine caused by the recovery operation of the virtual machine. The defect of unbinding various resources makes it possible to establish resource binding relationships with various resources and policies bound to the virtual machine before the virtual machine performs and delete operations after the virtual machine is restored due to the wrong deletion operation, which shortens the period of time. The recovery speed of virtual machines placed in the back-end recycle bin and the resources consumed by the recovery process. In this embodiment, the term "resource" may be understood as a physical resource and/or a virtual resource forming a virtual machine.

Referring to FIG. 6 , the virtual resources include resources 2a corresponding to computing nodes in the cloud platform, resources 2b corresponding to network nodes, resources 2c corresponding to storage nodes or resources 2d corresponding to control nodes. The virtual machine recycle bin implementation device 400 running the cloud platform virtual machine recycling method of the present invention performs resource deletion and resource recovery on the resources formed by each node in the cloud platform during the virtual machine recycling process. After the virtual machine is reclaimed, the occupation of the above-mentioned one or more virtual resources by the virtual machine performing the reclamation operation is released. The virtual machine recycle bin implementation apparatus 400 configures the cloud orchestration engine 1001, the virtualization engine 1002, and one or more hardware devices 20i. The hardware device 20i may be a CPU, memory, firewall, memory, dedicated chip, physical network card, disk, and the like. At the same time, as shown in FIG. 4 , the cloud orchestration engine 1001 can be implemented by using a container (Docker) technology or Kubernetes. The virtualization engine 1002 may employ a VMware virtualization engine.

In the recycling method disclosed in this embodiment, the virtual machine and the function of performing recycling on the virtual machine are not bound to the underlying specific cloud orchestration engine 1001 or hardware device 20i on which a specific type of cloud platform depends, so it can be compatible with almost All mainstream cloud environments and mainstream hardware platforms can be adapted to operating systems such as Kirin, Kingdee, and Great Wall, or X86 hardware platforms or ARM architecture hardware platforms or MIPS architecture hardware platforms of different manufacturers.

Meanwhile, in this embodiment, the recycle bin back-end service 17 in the first cloud platform in FIG. 7 or FIG. 8 , the back-end service 102 in the hybrid cloud management platform, the front-end recycle bin and the back-end recycle bin, etc. The Docker method is packaged and deployed to form independent microservices (and developed based on the Java language), and all run in any Linux distribution server. At the same time, the development based on the Java language is inherently cross-platform, and the stored data is based on relational open source databases (for example, the first database 15 and the second database 25 ), which are mounted by the back-end service 102 in the hybrid cloud management platform in FIG. 7 . The field "is_in_recycle_bin" contained in virtual_machine stored in the second database 25 records (ie, determines) whether the virtual machine is in the front-end recycle bin. Meanwhile, the first database 15 and the second database 25 can be deployed in various environments. The front-end recycle bin is positioned as a sub-function on the side of the first cloud platform 11 or the second cloud platform 21 , while the back-end recycle bin is on the side of the hybrid cloud management platform 100 . The first cloud platform 11 or the second cloud platform 21 can call the adapted gateway service in different cloud environments to complete the callback operation of the virtual machine state data and other data during the virtual machine recycling operation.

In this embodiment, the binding relationship between the virtual machine entering the front-end recycle bin and the virtual network, security group, key pair, virtual disk, etc. still exists. If the deleted resource has been bound to the virtual machine in the front-end recycle bin, then Operation will be restricted. There is no limit to the storage time of virtual machines in the front-end recycle bin. Only virtual machines in the "off", "on" and "unknown" states can be placed in the front-end recycle bin, and virtual machines in the "unknown" state are deleted immediately after being placed in the front-end recycle bin.

The permission verification performed by the backend for the deletion instruction includes at least one of the following verification contents: user id, virtual machine id, virtual machine name, virtual machine state, operating system running the virtual machine, hardware information forming the virtual machine, deploying the virtual machine The cloud platform information and the cluster attribute information of the cluster where the virtual machine is located. When the back-end fails to perform permission verification on the deletion instruction based on the verification content, the back-end sends an instruction to abort the deletion operation to the front-end recycle bin, and the back-end receives the notification from the front-end and performs permission verification on the deletion instruction And after the permission verification of the deletion instruction is successful based on the verification content, the virtual machine to be deleted placed in the back-end recycle bin is released from the binding relationship between the virtual machine to be deleted and the virtual resource. In this embodiment, the back-end recycle bin is logically located in the recycle bin interface 101 of the hybrid cloud management platform 100, the front-end recycle bin is logically located in the recycle bin interface 16 of the first cloud platform, and the front-end recycle bin is connected to the back-end recycle bin. Recycle bins are independent of each other. Meanwhile, the front-end recycle bin and the back-end recycle bin in this embodiment are processes running in the hybrid cloud management platform 100 .

In the permission verification stage, the affiliation formed by the virtual resources can be checked at the same time. If it is found that the project to which the virtual machine belongs is different from the project actually owned by the user, it will be regarded as an unauthorized operation, and any related operations to the virtual machine will be terminated. operation, and return the corresponding error information to the front end, and the front end will pop up a corresponding warning box to remind the user that the user does not have the permission to operate. The affiliation formed by the virtual resources here includes not only the type of virtual resources associated with the virtual machine, but also the quota of the virtual resources associated with the virtual machine (for example, VM1 configures a 40G disk, VM2 configures a 50G disk), and even includes a virtual machine. The configured virtual intranet IP address and virtual gateway, etc. As long as the virtual machine to be reclaimed cooperates with one or more of the above virtual resource types and/or virtual resources in the permission verification phase, or the configuration is inconsistent with the configuration previously issued by the user, it is determined that the user has no right to the virtual machine for recycling.

After the permission verification is successful, the hybrid cloud management platform 100 will continue to query the information corresponding to the virtual machine id in the first database 15, including but not limited to the name, status, cloud environment information, running status, task status, and cluster of the virtual machine. information and more. If the virtual machine information in the first database 15 is queried and returned as Null, the hybrid cloud management platform 100 will also interrupt the recycling operation and return an error code and prompt information to the front end, indicating that the information sent by the user is wrong and not Find the corresponding virtual machine information.

In this embodiment, only the current status information and cluster information of the virtual machine are concerned, and only the virtual machines whose status is "shutdown", "startup", "abnormal", and "unknown" can be put into the front-end recycle bin. It will compare the queried virtual machine status with the above four statuses one by one. If it does not meet the status requirements, any recycling operation of the virtual machine will be prohibited, and an error code and prompt information will be returned to the front end to remind the user of the current status. The virtual machine state of the operation does not meet the requirements to be placed in the front-end recycle bin.

In this embodiment, the reason why the virtual machines in the above four states can only be put into the front-end recycle bin is mainly because when the virtual machine is performing some important operations, suddenly putting it into the front-end recycle bin will affect the operation of the virtual machine. For example, when the virtual machine is creating, restarting, modifying the configuration, migrating, restoring, mirroring, and backing up, it is not supported to put it in the front-end recycle bin, so as to avoid responding to some important operations that the user's virtual machine is performing. Or when sensitive data is modified, accidental loss or damage caused by events placed in the front-end recycle bin.

Referring to FIG. 9 and FIG. 10 , in this embodiment, the virtual machine recycle bin supports batch virtual machine operations and single virtual machine operations, and the single virtual machine operations and batch operations are substantially the same. Whether it is a batch or non-batch operation, the backend is required to query the virtual machine information from the second database 25, such as the data configured before the virtual machine is reclaimed, such as fields such as "environment", "zone_id", etc. data to determine the corresponding cloud environment and cluster information, and determine the corresponding request cloud environment according to the above field data. The applicant will describe the detailed implementation process of the non-batch and batch operations of recycling virtual machines respectively below.

During the batch operation, three queues are first established to store virtual machines that need to be processed asynchronously and virtual machines that need to be processed synchronously. Recycle Bin List", which is used to cache the corresponding virtual machine.

For the batch operation of asynchronous operations, the polled virtual machine information will be put into different queues according to different states. If it is a virtual machine in the "on" state, it will be added to the "standby virtual machine list" , if the current state of the virtual machine is "unknown", it will be added to the "list of virtual machines to be deleted". When the polling is over, it will check whether the "list of virtual machines to be shut down" and "list of virtual machines to be deleted" are empty. The steps of the recycle operation of the virtual machine. If the "list of virtual machines to be deleted" is not empty, the steps of reclaiming the virtual machines whose current state is in the "unknown" state will be performed one by one, as described below.

For the batch operation of the synchronization operation, the polled virtual machine information will be put into different queues according to different states. ", if it is a virtual machine in the "abnormal" state, it will also be added to the "directly placed in the recycle bin list". When the polling is over, it will check whether "Put it directly into the recycle bin list" is empty (Null). If "Put it directly into the recycle bin list" is not empty, it will be executed one by one. For the steps of recycling a virtual machine in the "shutdown" state, please refer to the following.

Specifically, as shown in FIG. 9 , step 401 : the id of the virtual machine to be put into the recycle bin. The recycle bin here is the front-end recycle bin.

Step 402: Query the corresponding virtual machine information in the first database 15 according to the virtual machine id.

Step 403: Determine whether the authority of the virtual machine meets the requirements, that is, the authority verification operation of the subject who initiates the operation on the virtual machine.

Step 405: Determine whether the state of the virtual machine is a power-on, power-off or abnormal state, if yes, execute step 406 in sequence; if not, jump to execute step 404: end and return an error prompt.

Step 406: Obtain information such as cloud environment and cluster corresponding to the virtual machine.

Step 407: Invoke the gateway service corresponding to the cloud environment for asynchronous processing.

Step 408 : Determine whether the batch processing ends, if so, directly end and exit, if not, jump to step 401 .

Referring to FIG. 10 , step 501 is first performed: invoking the gateway service of the first cloud platform.

Step 502 : Send virtual machine information such as virtual machine id, user id, etc. to the first cloud platform management system, and the first cloud platform management system runs in the first cloud platform 11 .

Step 503: The first cloud platform management system executes virtual machine recycling logic.

Step 504: Determine whether the execution is completed.

Step 505: The first cloud platform management system notifies the cloud gateway of the first cloud platform.

Step 506 : The first cloud platform re-queries the corresponding virtual machine to assemble the latest data and calls back to the hybrid cloud management platform 100 .

Non-batch operation refers to performing the recycling operation on the virtual machines whose current status is "shutdown", "startup", "abnormal", and "unknown" one by one. For details, see below.

In this embodiment, the recycling method further includes: creating a plurality of availability zones at the front end, where the availability zones are virtual digital centers, and establishing an association relationship between virtual machines and different permissions configured for users in the availability zones 10 and 20, The availability zones 10 and 20 are created by the hybrid cloud management platform 100 , the availability zones 10 and 20 include at least one cloud platform, and the availability zones 10 and 20 are managed by the backend. The recycle bin interface 101 in the hybrid cloud management platform 100 in FIG. 7 or FIG. 8 and the backend service 102 in the hybrid cloud management platform jointly define the backend, which can be directly responded by the user. The recycle bin interface 101 in the hybrid cloud management platform 100 sends a data request to the back-end service 102 in the hybrid cloud management platform, and the back-end service 102 in the hybrid cloud management platform sends the data request to the recycle bin interface 101 in the hybrid cloud management platform 100 Return user information. The back-end service 102 in the hybrid cloud management platform and the first cloud platform 11 (or the second cloud platform 21 ) perform a call function/synchronize state. As for whether the availability zone 10 (or availability zone 20 ) managed by the hybrid cloud management platform 100 includes one or several types of cloud platforms, whether the cloud platform is based on Archer OS (a subordinate concept of the first cloud platform 11 ) or WMware ESXi (The subordinate concept of the second cloud platform 21 ) is irrelevant. Meanwhile, the same Availability Zone 10 (or Availability Zone 20 ) may include the same cloud platform (as shown in FIG. 2 ), and the same Availability Zone 10 (or Availability Zone 20 ) may also include different cloud platforms (see FIG. 3 ). Show).

7 and 8, preferably, a first database 15 can be configured for the front end, and a second database 25 is configured for the back end, and the first database 15 is used to record the status of the virtual machines in the availability zones 10 and 20. Data, the second database 25 is used to record the verification content associated with the permission verification of the deletion instruction performed by the backend. When the virtual machine to be deleted is recycled, the second database 25 is used to update the first database 15 and Read the updated virtual machine state data. Availability zones 10 and 20 store at least one cloud operating system or hypervisor. The recycle bin interface 16 of the first cloud platform in FIG. 7 and FIG. 8 and the recycle bin backend service 17 in the first cloud platform jointly define the front end, which can be directly responded by the user.

In this embodiment, the recycling method further includes: using a compare and swap algorithm (Compare And Swap, CAS mechanism) to store the underlying state of at least one cloud operating system or hypervisor in the availability zone 10 (or the availability zone 20 ). The data updates the virtual machine state data recorded in the first database 15 . The hybrid cloud management platform 100 calls the updated virtual machine state data in the first database 15 and updates the attributes of the same fields in the second database 25 .

For example, the "state" field contained in the virtual machine state data will be updated to "Stopping", the "update_time" field will be updated to the current time, and the corresponding virtual machine information, mounted disk information, network information, etc. will be recorded in the log. in the file.

As shown in FIG. 7 and FIG. 8 , preferably, the recovery method further includes: using a comparison and exchange algorithm to store the underlying state data of at least one cloud operating system or a hypervisor for the first database 15 in the available zones 10 and 20 . Update the virtual machine state data recorded in the second database 25; the second database 25 directly calls the updated virtual machine state data in the first database 15, and updates the attributes of the same fields in the second database 25.

Before updating the state of the virtual machine, first copy the original information stored in the first database 15 into the memory, and modify the latest information of the virtual machine in the memory, and compare the first database before saving it to the virtual machine record corresponding to the first database 15 Whether the latest information saved in 15 is the same as the original copy information in the memory; if it is the same, it means that no one else has performed an operation update during this period of time, so the virtual machine record information corresponding to the first database 15 can be updated, avoiding the need for Dirty data; if it is different, it means that someone else has done an operation update during this period of time. At this time, if you continue to update the virtual machine record information corresponding to the database, dirty data will be generated.

If it is found that the latest information of the virtual machine saved in the first database 15 is different from the original copy information in the memory, it means that other users have changed the information of the virtual machine in the first database 15 during this period. Copy information and data modification are invalidated, re-copy the latest information of the database into memory, and repeat the previous process until the update is successful, or the conditions are not met and terminated. By comparing and exchanging algorithms, it is ensured that only one user's operation can be fully realized at any time, avoiding the generation of dirty data.

In this embodiment, the state machine (State Machine) can be used to limit the recycling operation performed by the user on the virtual machine. For example, when multiple users such as user A, user B, and user C operate the virtual machine at the same time, the CAS mechanism ensures that only one user can operate at a certain time. Assuming that user A successfully obtains the operation permission, when user A After the virtual machine is shut down, the state of the virtual machine changes to "shutting down", and the state machine mechanism ensures that all other operations cannot be performed. operation.

After the update of the virtual machine state data recorded in the first database 15 is completed, the cloud gateway of the first cloud platform 11 will be notified. At this time, the state of the virtual machine changes from "shutdown" state to "shutdown". After the cloud gateway is notified, it will re-query the latest information of the corresponding virtual machine in the first cloud platform 11 , and package the queried information and call back to the hybrid cloud management platform 100 . After the hybrid cloud management platform 100 obtains the callback information, it will update the "is_in_recycle_bin" field in the corresponding virtual machine record in the database on the hybrid cloud management platform 100 side (ie, the second database 25 ), change it to "true", and change "update_time" Modify the field to "Current Time" and update the "state" field to "Stop". At the same time, the corresponding virtual machine information, mounted disk information, network information, etc. are recorded in the log file, and the execution result is returned to the front end. The execution result returned to the front end includes the name of the virtual machine and the completion of the corresponding operation, etc. Then the front end will visually display the returned information to the user in the form of a prompt box or dialog box. Specifically, in this embodiment, the log file is stored in the hybrid cloud management platform 100 .

The log file displays all information related to virtual machine operations, including but not limited to modification time, virtual machine name, modification content, operator, etc., which also brings a more intuitive method for later troubleshooting, the operation of virtual machine deletion. It will also perform detailed logging of the whole process. At the same time, the back-end log file recording interface provides powerful search and analysis functions in the back-end. Users can sort, search and paginate by name, modification time, virtual machine status, etc. Operation, the user can quickly locate the corresponding event in the massive log records, and learn the specific modification information or influence scope. If the customer virtual machine or data is lost due to human factors, the log file records can quickly locate the The specific person who initiates the virtual machine recycling operation, such as an administrator or user who does not have the operation authority.

When the current state of the virtual machine to be deleted is unknown, the virtual machine to be deleted is added to the list of virtual machines to be deleted, and the back-end creates an asynchronous virtual machine deletion thread to put the to-be-deleted virtual machine in the back-end recycle bin delete immediately after.

If the current state of the virtual machine to be reclaimed is "unknown", the hybrid cloud management platform 100 will repackage the queried virtual machine information, the specific cloud environment where it is located, and the original request parameters (including the virtual machine ref_id , user id, etc.) to the cloud gateway (eg, ArcherOS gateway) of the first cloud platform, and the cloud gateway of the first cloud platform calls the corresponding first cloud platform 11 . Among them, "unknown" refers to the state in which the state of the virtual machine changes and the exact state cannot be sensed due to the misoperation or error of the underlying layer.

The detailed execution steps in the first cloud platform 11 are shown in FIG. 5 , which will not be repeated here. When the state of the virtual machine of the first cloud platform 11 in the front end changes (the current state of the virtual machine changes from "unknown" state to "deleting" state), the cloud gateway of the first cloud platform will be notified, and the first The cloud gateway of a cloud platform will re-initiate a query request to the first database 15 to query the latest information of the virtual machine corresponding to the first cloud platform 11, and package the queried information and call it back to the hybrid cloud management platform 100. The hybrid cloud After the management platform 100 obtains the message, it will update the data recorded in the second database 25 corresponding to the virtual machine. For example, at this time, the "state" field will be updated to "Deleting", and the "update_time" field will be updated to "current time". The machine information, mounted disk information and network information are recorded in the log file.

After the operation of the first cloud platform 11 is completed, it will notify the cloud gateway of the first cloud platform (Note: the corresponding virtual machine has been deleted in the first cloud platform at this time), and the cloud gateway of the first cloud platform will call back the hybrid cloud management after being notified. After obtaining the callback information, the platform 100 and the hybrid cloud management platform 100 will first query the resources associated with the virtual machine in the second database 25, such as disk information, network card information, key pair information, etc., and then delete all associated virtual machines. Resource data information, and release virtual resources to the resource pool at the same time. When all the associated resource data are deleted successfully, the records in the first database 15 corresponding to the virtual machine will be deleted, and the corresponding virtual machine detailed information, as well as the mounted disk information, network information, etc. will be recorded in the log. in the file. Push the final execution result of the virtual machine recycling operation in the unknown state to the front end and display it to the user.

If the virtual machine to be reclaimed is in the "on" state, the hybrid cloud management platform 100 will repackage the queried virtual machine information, the specific cloud environment where it is located, and the original request parameters (including the virtual machine ref_id, user id etc.) are forwarded to the cloud gateway of the first cloud platform, and the cloud gateway of the first cloud platform invokes the corresponding cloud environment of the first cloud platform 11 . The detailed execution steps in the first cloud platform 11 are shown in FIG. 5 , which will not be repeated here.

When the state of the virtual machine in the first cloud platform 11 changes (for example, the state of the virtual machine changes from the "on" state to the "shutdown" state), the cloud gateway of the first cloud platform will be notified, and the cloud gateway of the first cloud platform will be notified. The gateway will re-initiate a query request to the first database 15 to query the latest information of the corresponding virtual machine in the first cloud platform 11, and wrap the queried information and call back the hybrid cloud management platform 100. After the hybrid cloud management platform 100 obtains the message The data recorded in the second database 25 corresponding to the virtual machine will be updated. For example, the "state" field will be updated to "Stopping" and the "update_time" field will be updated to "current time". The disk information, network information, etc. are recorded in the log file.

After the operation is completed, the cloud gateway of the first cloud platform will be notified (the current state of the virtual machine changes from "shutdown" state to "shutdown" state), and the cloud gateway of the first cloud platform will re-report to the first database after being notified. 15 initiates a query request to query the latest information of the corresponding virtual machine in the first cloud platform 11, and wraps the queried information and calls it back to the hybrid cloud management platform 100, and the hybrid cloud management platform 100 will update the second database after obtaining the callback information 25 corresponds to the "is_in_recycle_bin" field in the virtual machine record, change it to "true", change the "update_time" field to "current time", and update the "state" field to "Stop". At the same time, the corresponding virtual machine information, mounted disk information and network information are recorded in the log file, and the execution result is returned to the front end, including the virtual machine name and the completion of the corresponding operation, etc., and then the front end will return the information through the Visual forms such as prompt boxes are fed back to users.

If the virtual machine to be reclaimed is in the "shutdown" state, the hybrid cloud management platform 100 will repackage the queried virtual machine information, the specific cloud environment where it is located, and the original request parameters (including the virtual machine ref_id, user id, etc. ) is forwarded to the cloud gateway of the first cloud platform, and the cloud gateway of the first cloud platform invokes the corresponding cloud environment of the first cloud platform 11 . The detailed flow of execution in the first cloud platform 11 is shown in FIG. 5 , and details are not repeated here.

When the operation of the first cloud platform 11 is completed, it will notify the cloud gateway of the first cloud platform 11. After the cloud gateway of the first cloud platform 11 is notified, it will re-query the latest information of the corresponding virtual machine in the first cloud platform 11, and will query The received information is packaged and called back to the hybrid cloud management platform 100. After the hybrid cloud management platform 100 obtains the callback information, it will modify the "is_in_recycle_bin" field in the corresponding virtual machine record in the virtual machine in the second database 25, and change it to "true" ", modify the "update_time" field to "current time". At the same time, the corresponding virtual machine information, mounted disk information, network information, etc. are recorded in the log, and the execution result is returned to the front end. The aforementioned execution result includes the name of the virtual machine and the completion of the corresponding operation, etc., and then the front end will The returned information is fed back to the user through visual means such as a prompt box. The virtual machine in the "abnormal" state and the virtual machine in the "shutdown" state perform the same recycling operation manner and specific implementation process, so the description is omitted.

When the current state of the virtual machine to be deleted is the "shutdown" state, modify the field information retained in the second database 25 of the virtual machine to be deleted, and associate the modified field information with the disk mounted on the virtual machine to be deleted. The information and network configuration information are recorded in a log file, and the log file is saved to the second database 25 and displayed to the front end visually.

When the current state of the virtual machine to be deleted is the power-on state or abnormal state, the back-end executes the operation of adding the virtual machine in the power-on state and executing the deletion to the list of virtual machines to be shut down, and the back-end creates a thread to shut down the virtual machine asynchronously.

Referring to FIG. 5 , for a virtual machine in a powered-on state, the virtual machine reclaims the main thread 100 to jump to execute the asynchronous shutdown of the virtual machine thread 200, and for a virtual machine in an unknown state, the virtual machine reclaims the main thread 100 to jump to execute asynchronous deletion of the virtual machine thread. 300. The applicant elaborates the above process in detail.

start.

Step 101: Determine the identifier of the virtual machine to be put into the recycle bin, so as to determine the status data of the virtual machine to be reclaimed.

Step 102: Determine whether the virtual machine is in a state to be recycled, if not, end; if yes, jump to step 103.

Step 103: Determine whether the virtual machine is in a powered-on state, if so, go to step 21: add the virtual machine to the list of virtual machines to be shut down; if not, go to step 104: determine whether the virtual machine is in an unknown state, if yes, go to step 22 : add the virtual machine to the list of virtual machines to be deleted; if not, go to step 105 .

Step 105: Update the virtual machine recycle bin status to true, that is, the hybrid cloud management platform 100 modifies the obtained callback information in the "is_in_recycle_bin" field in the corresponding virtual machine record in the virtual machine in the second database 25, and changes it to " true", then jump to step 106.

Step 106: Determine whether the batch processing ends, if yes, skip to step 107; if not, return to step 101 to traverse and execute the above steps 101 to 105.

Step 107 : determine whether the list of virtual machines to be shut down is empty, if not, jump to step 108 : create a new asynchronous shutdown virtual machine thread; if so, jump to step 109 . At the same time, step 109 is also jumped to after the execution of the newly created asynchronous shutdown virtual machine thread is completed.

Step 109 : determine whether the list of virtual machines to be deleted is empty, if not, skip to step 110 : create a new asynchronous deletion virtual machine thread, if so, end directly.

When it is detected that the virtual machine is in the "on" state, the asynchronous shutdown virtual machine thread 200 shown in FIG. 5 will be newly started, and it will immediately return to the front end that the corresponding virtual machine is in operation, and the front end will display the returned information to the user.

In the asynchronous shutdown of the virtual machine thread 200, the corresponding virtual machine database record (ie, the first database 15) in the first cloud platform 11 is first modified, and the field data indicating the virtual machine state is changed to "shutdown", and then the The virtual machine performs any other operations, thus ensuring the idempotency of interface operations by means of a state machine. After that, the asynchronous shutdown virtual machine thread 200 will read the previously queried virtual machine information, such as the ref_id, state, name and other fields corresponding to the virtual machine in the cloud orchestration engine 1001 and repackage them, and at the same time obtain the corresponding information through oAuth authentication the authorization code X-Auth-Token, and then call the shutdown service in the first cloud platform 11 and pass in the packaged request parameters.

When the request parameters passed in by the first cloud platform 11 are successfully called, the state of the virtual machine corresponding to the bottom layer will also become "shutdown". At this time, the shutdown thread of the first cloud platform 11 that asynchronously shuts down the virtual machine thread will be timed. Polling for changes in the state of the bottom layer. After the cloud orchestration engine 1001 completes the virtual machine shutdown operation, the first cloud platform 11 polls the cloud orchestration engine 1001 for a change in the state of the virtual machine, and the first cloud platform 11 will update the latest bottom layer state. Data update the database information of the first cloud platform 11, for example, change the "is_in_recycle_bin" field to "true", change the "update_time" field to "current time", and change the "state" field to "Stop" Shut down state. At the same time, the corresponding virtual machine details, mounted disk information, network information, etc. are recorded in the log file, and the final execution result is pushed to the front end.

The details of the asynchronous shutdown of the virtual machine thread 200 are as follows.

The asynchronous shutdown virtual machine thread 200 invokes the list of virtual machines to be shut down by jumping to step 23 .

Step 201: Obtain a list of virtual machines to be shut down.

Step 202: Read the virtual machine information in the list of virtual machines to be shut down one by one, such as the aforementioned fields such as ref_id, status, and name.

Step 203: Create a new shutdown virtual machine thread, and perform polling check. If the virtual machine thread is not shut down at this time, create a new thread to execute step 206: call the shutdown interface; after the polling check, jump to step 204: determine to be shutdown Whether the virtual machine list is empty, if yes, go to step 205: end; if not, go back to step 202. Wherein, invoking the shutdown interface in step 206 is specifically calling a shutdown instruction adapted to the first cloud platform 11 and/or the second cloud platform 21 from the hybrid cloud management platform 100, and the shutdown instruction can be regarded as a process of performing a shutdown operation on the virtual machine between method calls.

After step 206, step 207 is executed: the state of the virtual machine is updated to shutdown, and finally step 208 is executed: the state of the recycle bin of the virtual machine is updated to true, that is, the callback information obtained by the hybrid cloud management platform 100 is modified in the second database 25. Change the "is_in_recycle_bin" field in the corresponding virtual machine record in the virtual machine to "true".

The asynchronous deletion of the virtual machine thread 300 is as follows.

The asynchronous deletion virtual machine thread 300 calls the list of virtual machines to be deleted by jumping to step 24 .

Step 301: Obtain a list of virtual machines to be deleted.

Step 302: Take out the virtual machine information in the list of virtual machines to be deleted one by one.

Step 303: Create a new virtual machine deletion thread, and perform polling check. If the virtual machine thread is not deleted at this time, create a new thread to execute Step 306: Call the delete virtual machine interface; after the polling check, jump to step 304: Judgment Whether the list of virtual machines to be deleted is empty, if yes, go to step 305 and end; if not, return to step 302 .

After the end of step 306, step 307 is executed: polling the deletion result, and finally jump to execute step 308: delete the corresponding virtual machine information from the second database 25.

To sum up, the method for recycling a virtual machine of a cloud platform disclosed in this embodiment solves the problem that the cloud platform, especially the virtual machines with different current states in different types of cloud platforms, can perform efficient and safe recycling operations. Significantly reduces the occupation and waste of resources such as computing resources and memory resources during the recovery operation of the virtual machine, and reduces the response time of concurrent operations;

At the same time, in the cloud platform virtual machine recycling method, since the front-end recycling bin located in the front-end and the back-end recycling bin located in the back-end are separated from each other, the virtual machine caused by performing the recycling operation on the virtual machine is effectively avoided. The defect of unbinding of various resources previously configured makes it possible to establish a resource binding relationship with various resources and policies bound to the virtual machine before performing and deleting operations after the virtual machine is restored due to the wrong deletion operation , which shortens the recovery speed of the virtual machine that has been put into the back-end recycle bin and the resources consumed by the recovery process, thereby indirectly improving the user experience.

Embodiment 2:

Referring to FIG. 11 , this embodiment discloses a computer device 800 . The computer device 800 executes the recovery method disclosed in the first embodiment.

The computer device 800 includes: a processor 51, a memory 52, and a computer program stored in the memory 52 and configured to be executed by the processor 51, the processor 51 executing the computer program described in the first embodiment when executing the computer program The steps in the cloud platform virtual machine recycling method.

Specifically, the memory 52 is composed of several storage units, namely, the storage unit 521 to the storage unit 52i, wherein the parameter i takes a positive integer greater than or equal to two. Both the processor 51 and the memory 52 are connected to the system bus 53 . The form of the system bus 53 does not need to be specifically limited. I2C bus, SPI bus, SCI bus, PCI bus, PCI-e bus, ISA bus, etc., can be reasonably changed according to the specific type of the electronic device 500 and the requirements of application scenarios. The system bus 53 is not the invention of the present application, so it will not be described in the present application.

It should be noted that the storage unit 52 in this embodiment may be a storage unit in a physical state, so that the computer device 800 can be understood as a computer or a computer cluster or a cluster server in a physical state; at the same time, the storage unit 52 may also be a virtual state. The storage unit, for example, is based on the virtual storage space formed by the physical storage device through the underlying virtualization technology, so that the computer device 800 is configured as a virtual device such as a cloud platform, a virtual server, and a virtual cluster, or the computer device 800 is understood as a PC, Tablets, smartphones, smart wearable electronics, physical clusters or data centers, etc.

For the technical solution of the same part of the computer device 800 shown in this embodiment and the first embodiment, please refer to that shown in the first embodiment, and details are not repeated here.

The various illustrative logic blocks, or units described in the embodiments of the present invention can be implemented by general-purpose processors, digital signal processors, application specific integrated circuits (ASICs), field programmable gate arrays or other programmable logic devices, discrete Gate or transistor logic, discrete hardware components, or any combination of the above are designed to implement or operate the functions described. A general-purpose processor may be a microprocessor, or alternatively, the general-purpose processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, multiple microprocessors, one or more microprocessors in combination with a digital signal processor core, or any other similar configuration. accomplish.

The series of detailed descriptions listed above are only specific descriptions for the feasible embodiments of the present invention, and they are not used to limit the protection scope of the present invention. Changes should all be included within the protection scope of the present invention.

It will be apparent to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments, but that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments are to be regarded in all respects as illustrative and not restrictive, and the scope of the invention is to be defined by the appended claims rather than the foregoing description, which are therefore intended to fall within the scope of the claims. All changes within the meaning and scope of the equivalents of , are included in the present invention. Any reference signs in the claims shall not be construed as limiting the involved claim.

In addition, it should be understood that although this specification is described in terms of embodiments, not each embodiment only includes an independent technical solution, and this description in the specification is only for the sake of clarity, and those skilled in the art should take the specification as a whole , the technical solutions in each embodiment can also be appropriately combined to form other implementations that can be understood by those skilled in the art.

Claims (15)

1. A method for reclaiming a cloud platform virtual machine, comprising:

   The front-end receives the deletion instruction for the virtual machine, and after placing the virtual machine into the front-end recycle bin deployed on the front-end, the front-end sends a notification to the back-end;

   The back-end receives the notification from the front-end and performs permission verification on the deletion instruction. For the deletion instruction that has passed the permission verification, according to the current state of the virtual machine to be deleted, the corresponding virtual machine is put into the back-end deployed on the back-end for recycling. In the recycling thread of the back-end recycling bin, the virtual machine to be deleted placed in the back-end recycling bin does not release the binding relationship between the virtual machine to be deleted and the virtual resource before executing the corresponding recycling thread.
2. The cloud platform virtual machine recycling method according to claim 1, wherein after the recycling thread of placing the virtual machine into the back-end recycling bin of the deployment back-end is executed, the method further comprises:

   The back-end recycle bin periodically checks the virtual machine that fails to release the binding relationship between the virtual machine to be deleted and the virtual resource, and performs forced recycling of the virtual machine.
3. The cloud platform virtual machine recycling method according to claim 1, wherein the front-end determines the current state of the virtual machine to be deleted, and only the virtual machine to be deleted whose current state is a shutdown state, a power-on state, an abnormal state and an unknown state is to be deleted. The machine executes the recycling thread placed in the back-end recycle bin, returns alarm information to the deletion command initiated by the virtual machine to be deleted that is not in the shutdown state, power-on state, abnormal state or unknown state, and refuses to put the virtual machine in the In the recycling thread of the backend recycle bin of the backend.
4. The method for reclaiming a virtual machine on a cloud platform according to claim 3, wherein the receiving the deletion instruction for the virtual machine is received in a visual interface of the front end;

   The permission verification performed by the backend for the deletion instruction includes at least one of the following verification contents: user id, virtual machine id, virtual machine name, virtual machine state, operating system running the virtual machine, hardware information forming the virtual machine, deploying the virtual machine The cloud platform information and the cluster attribute information of the cluster where the virtual machine is located;

   When the back-end fails to perform permission verification on the deletion instruction based on the verification content, the back-end sends an instruction to abort the deletion operation to the front-end recycle bin, and the back-end receives the notification from the front-end and performs permission verification on the deletion instruction And after the permission verification of the deletion instruction is successful based on the verification content, the binding relationship between the virtual machine to be deleted and the virtual resource is released for the virtual machine to be deleted placed in the back-end recycle bin.
5. The method for reclaiming a virtual machine on a cloud platform according to claim 1, wherein the binding relationship between the virtual machine and the virtual resource comprises one of a virtual machine and a virtual network, a security group, a key pair or a virtual disk or The binding relationship established between several virtual resources.
6. The cloud platform virtual machine recycling method according to any one of claims 1 to 5, wherein the front end includes at least one cloud platform, the front end is managed by the back end, and the front end and the back end logic independent of each other, the backend is configured as a hybrid cloud management platform.
7. The cloud platform virtual machine recycling method according to claim 6, further comprising:

   Create multiple availability zones at the front end, where the availability zone is a virtual digital center, and establish an association relationship between virtual machines and different permissions configured for users in the availability zone, the availability zone is created by the hybrid cloud management platform, and the The availability zone includes at least one cloud platform, and the availability zone is managed by the backend.
8. The cloud platform virtual machine recycling method according to claim 6, wherein the front end is configured with a first database, the back end is configured with a second database, and the first database is used to record the state of the virtual machine in the availability zone data, the second database is used to record the verification content associated with the permission verification performed by the back end on the deletion instruction, and after the virtual machine to be deleted is recycled, the first database is updated through the second database; The availability zone stores at least one cloud operating system or a hypervisor.
9. The cloud platform virtual machine recycling method according to claim 8, wherein when the current state of the virtual machine to be deleted is an unknown state, the virtual machine to be deleted is added to the list of virtual machines to be deleted, and the back-end creates an asynchronous Delete virtual machine threads to delete the virtual machine to be deleted as soon as it is placed in the back-end recycle bin.
10. The cloud platform virtual machine recycling method according to claim 8, wherein when the current state of the virtual machine to be deleted is a shutdown state, the field information retained in the second database of the virtual machine to be deleted is modified, and the modified virtual machine is modified. The latter field information and the disk information and network configuration information mounted on the virtual machine to be deleted are recorded in a log file, and the log file is saved to the second database for visual display to the front end.
11. The cloud platform virtual machine recycling method according to claim 8, wherein when the current state of the virtual machine to be deleted is a powered-on state or an abnormal state, the back-end executes adding the virtual machine in the powered-on state and executing the deletion to the to-be-deleted virtual machine. Shutdown the operation of the virtual machine list, and the asynchronous shutdown virtual machine thread created by the backend.
12. The cloud platform virtual machine recycling method according to claim 8, further comprising: using a comparison and exchange algorithm to store the underlying state data of at least one cloud operating system or a hypervisor in the first database in the availability zone. The recorded virtual machine state data is updated; the hybrid cloud management platform calls the updated virtual machine state data in the first database, and updates the attributes of the same field in the second database.
13. The cloud platform virtual machine recycling method according to claim 8, further comprising: using a comparison and exchange algorithm to store the underlying state data of at least one cloud operating system or a hypervisor in the first database in the availability zone. The recorded virtual machine state data is updated; the second database directly calls the updated virtual machine state data in the first database, and updates the attributes of the same field in the second database.
14. The method for recycling a virtual machine on a cloud platform according to claim 1, wherein the front end is configured as a distributed system architecture.
15. A computer equipment, characterized in that, comprising:

    processor, memory, and

    a computer program stored in the memory and configured to be executed by the processor;

    The processor executes the steps in the cloud platform virtual machine recycling method according to any one of claims 1 to 14 when executing the computer program.

Images