TWI746512B - Physical machine fault classification processing method and device, and virtual machine recovery method and system - Google Patents

Abstract

The embodiments of the present invention provide a clustered physical machine fault classification processing method and device, and a virtual machine recovery method and system. The physical machine fault classification processing method includes: obtaining a physical machine fault information list from a physical machine fault information storage center; if the physical machine fault information is detected in the physical machine fault information list due to a network attack, triggering the physical machine Processing by the security attack protection center outside the cluster; if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine fault information list, send an instruction to shut down the failed physical machine to the failed physical machine; and by The virtualization interface migrates the virtual machine on the failed physical machine to other healthy physical machines in the cluster system. In the embodiment of the present invention, a variety of physical machine failure scenarios are refined, quickly and accurately identified, and targeted classification processing is performed, thereby realizing rapid and highly reliable physical machine damage repair processing to ensure the above Fast recovery of virtual machine services.

Description

Physical machine fault classification processing method and device, and virtual machine recovery method and system

The present invention relates to the field of communication technology, in particular to a method and device for classifying and processing physical machine faults applied to a virtualized cluster system, and a method and system for restoring virtual machines.

With the rapid development of computer technology, people began to pay more and more attention to how to reduce energy consumption and improve resource utilization, and cloud computing models came into being. Cloud computing abstracts all computers into specific computing resources, and then provides these computing resources to users instead of directly providing one or more computers as in the traditional way. The biggest advantage of the cloud computing model is that users can invent resources according to their needs, avoid unnecessary waste of resources, and improve resource utilization.

In the cloud computing environment, virtualization cluster technology is one of the key technologies. The virtualization cluster composes multiple virtualized servers into an organic whole, thereby obtaining high computing speed and improving the overall computing power of the virtualized system. The virtualization cluster manages multiple servers in a unified manner, and uses virtualization technology to abstract physical resources into storage, computing, network and other resources A large resource pool is formed, and virtual machines are provided to users by inventing resources on demand.

With the gradual expansion of the scale of virtualized clusters, the probability of physical machine failures due to the hardware and software problems of the physical machines in the cluster has gradually increased. A physical machine failure will directly affect the virtual machine services running on it. In order to ensure the normal operation of the virtual machine business, it is necessary to discover the physical machine failure in time and deal with it promptly to restore the virtual machine business; otherwise, the virtual machine users will be affected by the physical machine failure, and business continuity cannot be guaranteed. The existing technology can regularly monitor the state of the physical machine. When a physical machine fails, the virtual machine on it will be shut down and then turned on again; or the failed physical machine will be shut down and the virtual machine on it will be migrated to other clusters. On the physical machine.

However, physical machine failures are usually caused by different reasons, and there are many types of physical machine failures. However, the prior art does not finely classify physical machine failures, and does not specifically classify them. Therefore, In actual commercial use, there will be more misjudgments and missed judgments, so that high availability (HA) of virtual machines on physical machines cannot be realized after physical machine failures.

Therefore, how to classify and repair physical machine faults more accurately, efficiently, and in a targeted manner has become a technical problem that needs to be solved by those skilled in the art.

In view of the above problems, the embodiments of the present invention are proposed in order to provide a A method and device for classifying and processing physical machine faults and a virtual machine recovery method and system applied to a virtualized cluster system that solve the above-mentioned problems or at least partially solve the above-mentioned problems.

The invention discloses a clustered physical machine fault classification processing method, including: obtaining a physical machine fault information list from a physical machine fault information storage center; if the physical machine fault information is detected in the physical machine fault information list due to a network attack, the physical machine fault , The security attack protection center outside the cluster is triggered to process; if a hardware and software failure that cannot be repaired by the physical machine itself is detected in the physical machine fault information list, an instruction to shut down the failed physical machine is sent to the failed physical machine ; And migrate the virtual machine on the failed physical machine to other healthy physical machines in the cluster system through the virtualization interface.

The present invention also discloses a clustered physical machine fault classification processing device, which includes: an acquisition module for acquiring a physical machine fault information list from a physical machine fault information storage center; a first processing module for receiving fault information on the physical machine If it is detected in the fault information list that a physical machine fails due to a network attack, it will trigger processing by the security attack protection center outside the cluster; the second processing module further includes: a shutdown processing unit for Machine fault information list If a hardware and software failure that cannot be repaired by the physical machine itself is detected, an instruction to shut down the failed physical machine is sent to the failed physical machine; the migration processing unit is used to migrate the virtual machine on the failed physical machine to the On other healthy physical machines in the cluster system.

The invention also discloses a virtual machine recovery method, which is applied to a virtualized cluster system, and the method includes: the physical machine in the virtualized cluster system autonomously detects its own failure dynamics; if the physical machine itself is automatically detected, it can be fault-tolerant and repaired. Hardware failures can be repaired by fault tolerance; if software and hardware failures that can be restarted and repaired by the physical machine are detected autonomously, they can be repaired by restarting the physical machine; a list of physical machine failure information can be obtained from the physical machine failure information storage center; If it is detected in the physical machine fault information list that the physical machine fails due to a network attack, the security attack protection center outside the cluster is triggered to process; if the physical machine itself is detected in the physical machine fault information list For hardware and software failures that cannot be repaired, an instruction to shut down the failed physical machine is sent to the failed physical machine; and virtual machines on the failed physical machine are migrated to other healthy physical machines in the cluster system through a virtualization interface.

Correspondingly, the present invention discloses a virtual machine recovery system, including Including: physical machine damage repair device, which is applied to the physical machine in the virtualized cluster system to autonomously detect the fault dynamics of the physical machine itself. If the software and hardware faults that the physical machine itself can be fault-tolerantly repaired are automatically detected, the fault-tolerant method is used to repair it; If the hardware and software failures that the physical machine itself can be restarted and repaired are detected autonomously, the physical machine can be repaired by restarting the physical machine; the physical machine fault information storage center is used to integrate all reported physical fault information into the physical machine fault information list; The machine fault classification processing device is used to obtain the physical machine fault information list from the physical machine fault information storage center. If the physical machine fault information is detected in the physical machine fault information list due to a network attack, it will trigger all the physical machine faults. Processing by the security attack protection center outside the cluster; if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine fault information list, an instruction to shut down the failed physical machine is sent to the failed physical machine, and by The virtualization interface migrates the virtual machine on the failed physical machine to other healthy physical machines in the cluster system.

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: The embodiments of the present invention can be used in large-scale cloud computing clusters to perform refined fault fast and accurate identification of various physical machine failure scenarios. And targeted classification processing, so as to achieve rapid and highly reliable physical machine damage repair processing, to ensure the rapid recovery of virtual machine services on it.

Further, in the embodiment of the present invention, the physical machine autonomously detects its own failure dynamics, and specifically classifies and repairs the physical machine fault conditions that the physical machine itself can repair; the physical machine fault conditions that cannot be repaired by the physical machine itself , By using the physical machine fault classification processing module outside the cluster to perform targeted classification and repair processing, thereby effectively reducing the occurrence of false judgments and missed judgments of physical machine faults, and automatically recovering virtual machines more safely, stably and quickly.

In addition, the embodiments of the present invention aim at physical machine failures that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, it can also be classified by physical machine failures outside the cluster. The processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet commercial standards, and at the same time ensure the effectiveness of automated physical machine isolation.

In addition, the embodiments of the present invention also take into account the possibility of physical machine-scale failures in large-scale cloud computing clusters, by judging whether the number of failed physical machines constitutes the computer room level, and adopting different repair processing methods in a targeted manner . Especially in the case of large-scale physical machine failures, manual processing is used to repair, so as to effectively avoid the occurrence of system performance impacts due to frequent migration of virtual machines on the failed physical machine.

Of course, any product implementing the present invention does not necessarily need to achieve all the advantages described above at the same time.

110-130‧‧‧step

210-260‧‧‧Step

310-390‧‧‧Step

500‧‧‧Physical machine damage repair device

510‧‧‧Autonomous detection module

511‧‧‧Detection unit

512‧‧‧Reporting unit

520‧‧‧Autonomous Processing Module

521‧‧‧Fault Tolerant Unit

522‧‧‧Restart unit

523‧‧‧Shutdown unit

530‧‧‧Out-of-band management module

600‧‧‧Physical machine fault classification processing device

610‧‧‧Get Module

620‧‧‧First processing module

630‧‧‧Second processing module

640‧‧‧Third processing module

650‧‧‧Fourth processing module

660‧‧‧Fifth Processing Module

670‧‧‧Determine the module

700‧‧‧Virtualized Cluster System

710‧‧‧Physical machine damage repair device

711‧‧‧Autonomous detection module

712‧‧‧Autonomous Processing Module

713‧‧‧Out-of-band management module

720‧‧‧Physical machine fault classification processing device

721‧‧‧Get Module

722‧‧‧First Processing Module

723‧‧‧Second Processing Module

724‧‧‧Third processing module

725‧‧‧Fourth processing module

726‧‧‧Fifth Processing Module

727‧‧‧Determining Module

730‧‧‧Physical machine fault information storage center

740‧‧‧Security Attack Protection Center

750‧‧‧Physical machine fault detection module

760‧‧‧Physical machine fault collection module

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following will briefly introduce the drawings that need to be used in the embodiments. Obviously, the drawings in the following description are only some implementations of the present invention. For example, for those of ordinary skill in the art, without creative work, other schemas can be obtained from these schemas.

Figure 1 is a step flow diagram of an embodiment of a cluster entity machine fault classification processing method of the present invention; Figure 2 is a step flow diagram of another embodiment of a cluster entity machine fault classification processing method of the present invention; Figure 3 is a kind of embodiment of the present invention Step flowchart of an embodiment of a virtual machine recovery method; FIG. 4 is a step flowchart of another embodiment of a virtual machine recovery method of the present invention; FIG. 5 is a structural block diagram of an embodiment of a physical machine damage repair device of the present invention; 6 is a structural block diagram of an embodiment of a cluster physical machine fault classification processing device of the present invention; FIG. 7 is a structural block diagram of an embodiment of a virtual machine recovery system of the present invention.

The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described implementations The embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art fall within the protection scope of the present invention.

In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention will be further described in detail below with reference to the drawings and specific embodiments.

In order to facilitate the understanding of the embodiments of the present invention, firstly, several elements involved in the description of the embodiments of the present invention are introduced here:

A. Cloud computing

Cloud computing is a mode of increasing, using, and delivering related services based on Internet technology. It is a server cluster that implements distributed computing on all servers used. In other words, cloud computing provides a virtualized and on-demand flexible resource platform that dynamically supplies hardware, software, and data sets.

B. Virtual cluster

Cluster management on the cloud computing platform constitutes a virtual cluster. The so-called virtual cluster is the use of virtualization technology to virtualize multiple computing nodes, thereby constructing a cluster system that is similar to a physical cluster and has a huge scale. In other words, a virtual cluster is a system that connects multiple homogeneous or heterogeneous computers that coordinate to complete specific tasks.

C, physical machine

Multiple computers in a virtual cluster system that cooperate to complete a specific task are cluster physical computers, referred to as cluster physical machines. Among them, one or more virtual computers can be simulated on a physical machine.

D, virtual machine

With virtual machine software, one or more virtual computers can be simulated on a physical machine, and these virtual machines work like real computers. The virtual machines can be installed with operating systems and applications. Can access network resources. For applications running in a virtual machine, the virtual machine is like working in a real computer.

The embodiments of the present invention can be applied to a large-scale cloud computing virtualized cluster system, and the physical machine in the cluster system can autonomously detect its own failure dynamics, and then it can be targeted to the physical machine failures that the physical machine itself can repair. Carry out classification repair processing; and for the physical machine failures that cannot be repaired by the physical machine itself, the physical machine fault classification processing module outside the cluster is used for targeted classification and repair processing, thereby effectively reducing the misjudgment and missed judgment of the physical machine failure When the situation occurs, the automatic recovery of the virtual machine is safer, more stable, and faster.

The physical machine failure phenomena that affect the operation and management of virtual machines can be summarized as follows:

1. The physical machine's network is unavailable

The main reasons include: physical machine downtime, network card abnormality, uplink switch failure, hardware abnormality, kernel module abnormality, physical machine restart, network Distributed Denial of Service (DDoS), etc.

2. Physical machine loss

The main reasons for this include: high load on physical machines, switching of Internet access equipment, and network DDoS attacks.

3. Physical machine hardware system failure

For example, physical machine disks, memory, central processing unit (Central Processing Unit, CPU) failures, etc.

4. The physical machine software is abnormal

For example, software abnormalities at the operating system level such as the file system of the physical machine, virtualization-related modules, and operating system kernel modules.

5. The remote access channel of the physical machine is blocked

The main reasons include: network packet loss, system service abnormality, file system abnormality, etc.

6. Abnormal physical machine performance

For example, it may appear as physical machine input/output (Input/Output, I/O) stuck, high load, etc. The main reasons include: physical machine hardware failure, physical machine kernel module abnormality, physical machine user mode process abnormality, etc.

It can be seen that the above phenomenon of physical machine failure is not static, but can be transformed into each other within a certain period of time, and even related and intertwined. In addition, the reasons behind the same physical machine phenomenon may be different. Therefore, the repair method of the faulty physical machine needs to be specifically distinguished. For example, the network failure of a physical machine caused by a network DDoS attack and the physical machine are The physical machine network failure caused by the machine needs to be treated differently. If the virtual machine on the physical machine is being attacked by the network DDoS attack, the virtual machine on it is migrated to other physical machines. Physical machines are successively attacked and become unavailable, which may eventually cause flooding of the entire cluster network equipment, leading to the risk of failure of the entire cluster of physical machines.

Based on the in-depth cause analysis of the above-mentioned physical machine failure phenomena and abnormalities, in the embodiment of the present invention, the physical machine failures can be classified into the following categories:

A. The type of hardware and software faults that the physical machine itself can fault-tolerantly repair

For example, the disk that stores the data is faulty, the kernel module related to virtualization is abnormal, and the file system that stores the data is abnormal.

B. The type of hardware and software failures that the physical machine itself can restart and repair

For example, root file system read-only exceptions, network card driver restarting repairable exceptions, operating system kernel module exceptions, etc.

C. Types of hardware and software failures that cannot be repaired by the physical machine itself

For example, physical machine crashes, physical machine CPU abnormalities, physical machine memory abnormalities, physical machine power modules, and other hardware problems.

In addition, it also includes the types of failures with unknown causes, such as system load, system network, and hardware failure. Although this type of failure is essentially It is difficult to find, but the symptoms of this type of failure are very clear, mainly: physical machine network packet loss, physical machine management channel access abnormality, physical machine performance usage abnormality.

D. Types of physical machine failures caused by cyber attacks on physical machines

For example, network DDoS type security attacks have caused a large number of network packet loss or even network disconnection. Such failure phenomena mainly include: physical machine network failure, network packet loss, management channel failure, etc.

Therefore, the embodiment of the present invention performs rapid and accurate identification of refined faults for various physical machine failure scenarios, and targeted classification processing, thereby realizing rapid and highly reliable physical machine damage repair processing to ensure its Fast recovery of virtual machine services on the Internet. For example, the embodiment of the present invention can process and complete the recovery of a virtual machine on a failed physical machine within ten minutes, and the function of the virtual machine has a commercial availability standard of more than 99.95%.

Example one

1, there is shown a step flow chart of an embodiment of a clustered physical machine fault classification processing method of the present invention. The physical machine fault classification processing method can be applied to a virtualized cluster system, and specifically may include the following steps: Step 110, Obtain the physical machine fault information list from the physical machine fault information storage center; it should be noted that the physical machine fault information list includes: the physical machine fault detection module outside the cluster obtains the physical machine fault information from the faulty physical machine The physical machine fault information detected and reported to the physical machine fault information storage center, and the physical machine fault collection module outside the cluster is collected from the faulty physical machine and reported to the physical machine fault information Physical machine failure information in the storage center.

Step 120: If it is detected in the physical machine fault information list that the physical machine fails due to a network attack, the security attack protection center outside the cluster is triggered to process; it is understandable that in practical applications, all After the security attack protection center outside the cluster is triggered, the security cleaning program will be activated, such as traffic cleaning, etc., so as to restore the health of the failed physical machine. It should be noted that the network failure of a physical machine caused by a network DDoS attack and the physical machine network failure caused by a physical machine crash need to be treated differently. If the physical machine is suffering from network DDoS Migrating the virtual machines on it to other physical machines during an attack will produce a domino effect, leading to an enlarged risk of failure, that is, other physical machines are successively attacked and become unavailable, which may eventually cause flooding of the entire cluster of network equipment. Lead to the risk of failure of the entire cluster of physical machines.

Step 130: If a hardware and software failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, send an instruction to shut down the failed physical machine to the failed physical machine; and migrate the failure through the virtualization interface The virtual machine on the physical machine is connected to other healthy physical machines in the cluster system; preferably, if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, the physical machine is sent to the failed entity machine Send an instruction to shut down the failed physical machine to instruct the failed physical machine to shut down the failed physical machine autonomously or through an out-of-band management module on the physical machine.

It should be noted that the types of hardware and software failures that cannot be repaired by the physical machine itself can include: physical machine crashes, physical machine CPU abnormalities, physical machine memory abnormalities, physical machine power modules, and other hardware problems abnormal . This type of failure will directly cause the physical machine to be unavailable and require the replacement of the hardware module to be repaired. Therefore, the embodiment of the present invention isolates the failed physical machine from the cluster before performing hardware replacement or maintenance on the failed physical machine. .

In addition, in the case of hardware and software failures that cannot be repaired by the physical machine itself, the out-of-band management and control system on the traditional physical machine usually has an availability of about 90% or even lower due to hardware failure rate and cost issues. In the cloud computing service itself With a commercial availability requirement of at least 99.95%, the total unavailability for the whole year is 262.8 minutes. If a failed physical machine cannot be repaired in time, the failure of a physical machine will directly cause tens of minutes of manual processing time. Therefore, the availability index of the out-of-band management and control system in the prior art cannot match the failure recovery service-level agreement (SLA) of commercial cloud computing services. The technical solution provided by the embodiment of the present invention improves the traditional out-of-band management and control system. When the availability of the out-of-band management module does not reach commercial standards, the instructions of the physical machine fault classification processing module outside the cluster can be used. Instruct the faulty physical machine to shut down autonomously, and then the physical machine fault classification processing module outside the cluster migrates the faulty physical machine through the virtualization interface The virtual machine is added to other healthy physical machines in the cluster system; thus, the repair time of the failed physical machine is greatly shortened, and the commercial availability of the system is improved.

The embodiments of the present invention can perform rapid and accurate fault identification in a large-scale cloud computing cluster through various physical machine failure scenarios, and perform targeted classification processing, thereby realizing fast and highly reliable entities. The machine is damaged and repaired to ensure the rapid recovery of the virtual machine service on it.

In addition, the embodiments of the present invention aim at physical machine failures that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, it can also be classified by physical machine failures outside the cluster. The processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet commercial standards, and at the same time ensure the effectiveness of automated physical machine isolation.

Example two

2, there is shown a step flow chart of another embodiment of a clustered physical machine fault classification and processing method of the present invention, which may specifically include the following steps: Step 210: Obtain a physical machine fault information list from a physical machine fault information storage center; It is noted that the physical machine fault information list includes: the physical machine fault detection module outside the cluster obtains information from the faulty physical machine The physical machine fault information detected and reported to the physical machine fault information storage center, and the physical machine fault collection module outside the cluster is collected from the faulty physical machine and reported to the physical machine fault information Physical machine failure information in the storage center.

Step 220: If it is detected in the physical machine fault information list that the physical machine fails due to a network attack, the security attack protection center outside the cluster is triggered to process; it is understandable that in practical applications, all After the security attack protection center outside the cluster is triggered, the security cleaning program will be activated, such as traffic cleaning, etc., so as to restore the health of the failed physical machine. It should be noted that the network failure of a physical machine caused by a network DDoS attack and the physical machine network failure caused by a physical machine crash need to be treated differently. If the physical machine is suffering from network DDoS Migrating the virtual machines on it to other physical machines during an attack will produce a domino effect, leading to an enlarged risk of failure, that is, other physical machines are successively attacked and become unavailable, which may eventually cause flooding of the entire cluster of network equipment. Lead to the risk of failure of the entire cluster of physical machines.

Step 230: If a hardware and software failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, send an instruction to shut down the failed physical machine to the failed physical machine; and migrate the failure through the virtualization interface The virtual machine on the physical machine is connected to other healthy physical machines in the cluster system; preferably, if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, the physical machine is sent to the failed entity machine Send an instruction to shut down the failed physical machine to instruct the failed physical machine to shut down the failed physical machine autonomously or through an out-of-band management module on the physical machine.

It should be noted that the types of hardware and software failures that cannot be repaired by the physical machine itself can include: physical machine crashes, physical machine CPU abnormalities, physical machine memory abnormalities, physical machine power modules, and other hardware problems abnormal . This type of failure will directly cause the physical machine to be unavailable and require the replacement of the hardware module to be repaired. Therefore, the embodiment of the present invention isolates the failed physical machine from the cluster before performing hardware replacement or maintenance on the failed physical machine. .

In addition, in the case of hardware and software failures that cannot be repaired by the physical machine itself, the out-of-band management and control system on the traditional physical machine usually has an availability of about 90% or even lower due to hardware failure rate and cost issues. In the cloud computing service itself With a commercial availability requirement of at least 99.95%, the total unavailability for the whole year is 262.8 minutes. If a failed physical machine cannot be repaired in time, the failure of a physical machine will directly cause tens of minutes of manual processing time. Therefore, the availability index of the out-of-band management and control system in the prior art cannot match the failure recovery service-level agreement (SLA) of commercial cloud computing services. The technical solution provided by the embodiment of the present invention improves the traditional out-of-band management and control system. When the availability of the out-of-band management module does not reach commercial standards, the instructions of the physical machine fault classification processing module outside the cluster can be used. Instruct the faulty physical machine to shut down autonomously, and then the physical machine fault classification processing module outside the cluster migrates the faulty physical machine through the virtualization interface The virtual machine is added to other healthy physical machines in the cluster system; thus, the repair time of the failed physical machine is greatly shortened, and the commercial availability of the system is improved.

Step 240: If it is detected in the physical machine fault information list that the physical machine network is completely disconnected and the duration of the network disconnection reaches the preset time; determine whether the number of physical machines with the network disconnected exceeds the preset number, and if so, notify Operation and maintenance personnel manually repair; otherwise, the virtual machine on the failed physical machine is migrated to other healthy physical machines in the cluster system through the virtualization interface; wherein, the preset time can be set to 3 minutes according to the actual situation , 5 minutes and other suitable time periods.

It should be noted that when it is detected that the network of the physical machine is completely disconnected and the duration of the network disconnection reaches the preset time, the embodiment of the present invention needs to further check whether the number of faulty physical machines with the network disconnected exceeds the entity of one cabinet. If the number of machines or the number of physical machines connected to a switch exceeds, it is considered to be a cluster-scale network failure, and it needs to be repaired manually by the operation and maintenance personnel through telephone calls instead of automatic processing. This is because for large-scale physical machine failures, when isolating physical machines and migrating virtual machines, a large number of physical machines will be shut down. When the equipment in the computer room (network equipment or electrical equipment, etc.) is restored, the physical machines need to be restarted again. Then restore the virtual machine. This series of operations will directly cause the manual processing time to double or even more, thereby greatly increasing the unavailable time of the virtual machine. Therefore, the method provided by the embodiment of the present invention differentiates and processes this type of physical machine failure, which can greatly shorten the failure of the physical machine. The repair time of the server, thereby greatly shortening the time when the virtual machine on it is unavailable, thereby improving the commercial availability of the system.

Preferably, the method according to the embodiment of the present invention may further include: step 250, if it is detected in the physical machine fault information list that the physical machine has a network failure but the network failure duration has not reached the preset time, the network It returns to normal, and it is determined that the physical machine's network failure is caused by the restart of the physical machine, then determine whether the current physical machine is healthy, if it is healthy, restart the virtual machine on the physical machine through the virtualization interface, if it is unhealthy, borrow Migrate the virtual machine on the faulty physical machine to other healthy physical machines in the cluster through the virtualization interface; step 260, if it is detected in the physical machine fault information list that the physical machine network is unstable and the network is not When the stable duration reaches the preset time, an instruction is sent to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or to shut down the faulty physical machine through the out-of-band management module on the physical machine; and by virtual Migrate the virtual machine on the failed physical machine to other healthy physical machines in the cluster system; it should be noted that the network of the physical machine is unstable and the duration of network instability reaches the preset time It is mainly caused by some unknown reasons, such as system load, system network, hardware failure, etc. Although the essential causes of this type of failure are difficult to investigate, the symptoms of such failures are clear, mainly: physical machine network packet loss, physical machine management channel access abnormalities, and physical machine performance usage abnormalities. For this type of physical machine failure, the same processing method can be adopted, that is, to the failed physical machine The device sends instructions to instruct the faulty physical machine to shut down the faulty physical machine autonomously or to shut down the faulty physical machine through the out-of-band management module on the physical machine; and to migrate the virtual machine on the faulty physical machine through the virtualization interface To other healthy physical machines in the cluster system.

Preferably, in an embodiment of the present invention, the healthy physical machine is determined by the following method: matching all physical machines in the cluster in the physical machine fault information list; there will be no physical machines that are successfully matched Determined as a healthy physical machine.

The embodiments of the present invention can perform rapid and accurate fault identification in a large-scale cloud computing cluster through various physical machine failure scenarios, and perform targeted classification processing, thereby realizing fast and highly reliable entities. The machine is damaged and repaired to ensure the rapid recovery of the virtual machine service on it.

Further, in the embodiment of the present invention, the physical machine autonomously detects its own failure dynamics, and specifically classifies and repairs the physical machine fault conditions that the physical machine itself can repair; the physical machine fault conditions that cannot be repaired by the physical machine itself , By using the physical machine fault classification processing module outside the cluster to perform targeted classification and repair processing, thereby effectively reducing the occurrence of false judgments and missed judgments of physical machine faults, and automatically recovering virtual machines more safely, stably and quickly.

In addition, the embodiment of the present invention is aimed at the physical machine failure that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, the physical machine can also be shut down by an external cluster. The physical machine fault classification processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet the commercial standard, while also ensuring the effectiveness of automated physical machine isolation.

In addition, the embodiments of the present invention also take into account the possibility of physical machine-scale failures in large-scale cloud computing clusters, by judging whether the number of failed physical machines constitutes the computer room level, and adopting different repair processing methods in a targeted manner . Especially in the case of large-scale physical machine failures, manual processing is used to repair, so as to effectively avoid the occurrence of system performance impacts due to frequent migration of virtual machines on the failed physical machine.

Example three

3, there is shown a schematic diagram of an embodiment of a virtual machine recovery method of the present invention, which may specifically include the following steps: Step 310, the physical machine in the virtualized cluster system autonomously detects its own failure dynamics; preferably, each machine The physical machine can autonomously detect its own fault dynamics at regular intervals, for example, once every 30 seconds.

In step 320, if the hardware and software faults that the physical machine itself can be fault-tolerantly repaired are detected autonomously, the fault-tolerant method is used to repair; Including: Disk failure for storing data, abnormal kernel module related to virtualization, and abnormal file system for storing data Wait. For example, for the failure of the disk that stores data, the fault-tolerant repair method is specifically to isolate the disk first, and then use the mechanism of clustering to store multiple copies of the data, so that the data on the disk is automatically copied to other healthy disks, which can effectively ensure After the faulty disk is isolated, it will not affect the stable operation of the system. Similarly, for the file system that stores data is damaged, the purpose of fault-tolerant repair can also be achieved by isolating the disk mounted on the file system.

In step 330, if the hardware and software failures that the physical machine itself can be restarted and repaired are autonomously detected, repaired by restarting the physical machine; it can be understood that the hardware and software failures that can be repaired by the physical machine itself in the embodiment of the present invention are: It can include: root file system read-only exceptions, network card driver restarting repairable exceptions, operating system kernel module exceptions, etc. Such hardware and software failures can be repaired by restarting the physical machine.

Step 340: Obtain the physical machine fault information list from the physical machine fault information storage center; it should be noted that the physical machine fault classification processing module obtains the physical machine fault information list from the physical machine fault information storage center. The physical machine fault information list includes: physical machine fault information detected by a physical machine fault detection module outside the cluster from the faulty physical machine and reported to the physical machine fault information storage center, and The physical machine fault collection module outside the cluster collects the physical machine fault information from the faulted physical machine and reports it to the physical machine fault information storage center.

Step 350, if it is detected in the physical machine fault information list If a physical machine fails due to a network attack, the security attack protection center outside the cluster is triggered to process; it is understandable that in practical applications, the security attack protection center outside the cluster will be triggered. Cleaning procedures, such as flow cleaning, etc., so as to restore the health of the failed physical machine. It should be noted that the network failure of a physical machine caused by a network DDoS attack and the physical machine network failure caused by a physical machine crash need to be treated differently. If the physical machine is suffering from network DDoS Migrating the virtual machines on it to other physical machines during an attack will produce a domino effect, leading to an enlarged risk of failure, that is, other physical machines are successively attacked and become unavailable, which may eventually cause flooding of the entire cluster of network equipment. Lead to the risk of failure of the entire cluster of physical machines.

Step 360: If a hardware and software failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, an instruction is sent to the failed physical machine to instruct the failed physical machine to shut down the failed physical machine autonomously or by all means. The out-of-band management module on the physical machine shuts down the faulty physical machine; and the virtual machine on the faulty physical machine is migrated to other healthy physical machines in the cluster system through a virtualization interface.

It should be noted that the types of hardware and software failures that cannot be repaired by the physical machine itself can include: physical machine crashes, physical machine CPU abnormalities, physical machine memory abnormalities, physical machine power modules, and other hardware problems abnormal . This type of failure will directly cause the physical machine to be unavailable and require the replacement of the hardware module to be repaired. Therefore, the embodiment of the present invention isolates the faulty physical machine from the cluster and then replaces the hardware of the faulty physical machine. Or maintenance.

In addition, in the case of hardware and software failures that cannot be repaired by the physical machine itself, the out-of-band management and control system on the traditional physical machine usually has an availability of about 90% or even lower due to hardware failure rate and cost issues. In the cloud computing service itself With a commercial availability requirement of at least 99.95%, the total unavailability for the whole year is 262.8 minutes. If a failed physical machine cannot be repaired in time, the failure of a physical machine will directly cause tens of minutes of manual processing time. Therefore, the availability index of the out-of-band management and control system in the prior art cannot match the failure recovery service-level agreement (SLA) of commercial cloud computing services. The technical solution provided by the embodiment of the present invention improves the traditional out-of-band management and control system. When the availability of the out-of-band management module does not reach commercial standards, the instructions of the physical machine fault classification processing module outside the cluster can be used. Instruct the faulty physical machine to shut down autonomously, and then the physical machine fault classification processing module outside the cluster migrates the virtual machine on the faulty physical machine to other healthy physical machines in the cluster system through the virtualization interface; thereby a large number of Shorten the repair time of failed physical machines, thereby improving the commercial availability of the system.

Preferably, the method according to the embodiment of the present invention may further include: step 370, if it is detected in the physical machine fault information list that the physical machine’s network is completely disconnected and the duration of the network disconnection reaches a preset time; Whether the number of unreachable physical machines exceeds the preset number, if so, notify the operation and maintenance personnel to repair manually; otherwise, use the virtualization interface to migrate the virtual machines on the failed physical machine to the cluster system On other healthy physical machines.

Wherein, the preset time can be set to a suitable time period such as 3 minutes, 5 minutes, etc. according to actual conditions.

It should be noted that when it is detected that the network of the physical machine is completely disconnected and the duration of the network disconnection reaches the preset time, the embodiment of the present invention needs to further check whether the number of faulty physical machines with the network disconnected exceeds the entity of one cabinet. If the number of machines or the number of physical machines connected to a switch exceeds, it is considered to be a cluster-scale network failure, and it needs to be repaired manually by the operation and maintenance personnel through telephone calls instead of automatic processing. This is because for large-scale physical machine failures, when isolating physical machines and migrating virtual machines, a large number of physical machines will be shut down. When the equipment in the computer room (network equipment or electrical equipment, etc.) is restored, the physical machines need to be restarted again. Then restore the virtual machine. This series of operations will directly cause the manual processing time to double or even more, thereby greatly increasing the unavailable time of the virtual machine. Therefore, the method provided by the embodiment of the present invention differentiates and handles this type of physical machine failure, which can greatly shorten the repair time of the failed physical machine, thereby greatly reducing the time that the virtual machine on it is unavailable, thereby improving the commercial use of the system. Availability.

Preferably, the method according to the embodiment of the present invention may further include: step 380, if it is detected in the physical machine fault information list that the physical machine has a network disconnection but the network disconnection duration has not reached the preset time, the network It returns to normal again, and it is determined that the physical machine’s network disconnection is caused by the restart of the physical machine, then it is determined whether the current physical machine is healthy, and if it is healthy, the virtual machine on the physical machine is restarted through the virtualization interface, if If it is unhealthy, the virtual machine on the failed physical machine is migrated to other healthy physical machines in the cluster through the virtualization interface.

Preferably, the method according to the embodiment of the present invention may further include: step 390, if it is detected in the physical machine fault information list that the physical machine network is unstable and the network instability duration reaches a preset time, then Send an instruction to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through the out-of-band management module on the physical machine; and migrate the faulty physical machine to the faulty physical machine through the virtualization interface To other healthy physical machines in the cluster system.

It should be noted that the situation that the physical machine network is unstable and the network instability lasts for the preset time is mainly due to some unknown reasons causing physical machine failures, such as system load, system network, and hardware failure. Class etc. Although the essential causes of this type of failure are difficult to investigate, the symptoms of such failures are clear, mainly: physical machine network packet loss, physical machine management channel access abnormalities, and physical machine performance usage abnormalities. For this type of physical machine failure, the same processing method can be used, that is, sending an instruction to the malfunctioning physical machine to instruct the malfunctioning physical machine to autonomously shut down the malfunctioning physical machine or shutting down the malfunctioning entity through the out-of-band management module on the physical machine Machine; and migrate the virtual machine on the failed physical machine to other healthy physical machines in the cluster system through the virtualization interface.

Preferably, in an embodiment of the present invention, the healthy physical machine is determined by the following method: matching all the physical machines in the cluster in the physical machine fault information list There is a physical machine; the physical machine that has not been matched successfully is determined as a healthy physical machine.

The embodiments of the present invention can perform rapid and accurate fault identification in a large-scale cloud computing cluster through various physical machine failure scenarios, and perform targeted classification processing, thereby realizing fast and highly reliable entities. The machine is damaged and repaired to ensure the rapid recovery of the virtual machine service on it.

Further, in the embodiment of the present invention, the physical machine autonomously detects its own failure dynamics, and specifically classifies and repairs the physical machine fault conditions that the physical machine itself can repair; the physical machine fault conditions that cannot be repaired by the physical machine itself , By using the physical machine fault classification processing module outside the cluster to perform targeted classification and repair processing, thereby effectively reducing the occurrence of false judgments and missed judgments of physical machine faults, and automatically recovering virtual machines more safely, stably and quickly.

In addition, the embodiments of the present invention aim at physical machine failures that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, it can also be classified by physical machine failures outside the cluster. The processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet commercial standards, and at the same time ensure the effectiveness of automated physical machine isolation.

In addition, the embodiments of the present invention also take into account the possibility of physical machine-scale failures in large-scale cloud computing clusters. By judging whether the number of failed physical machines constitutes the level of the computer room, and taking targeted measures The same repair method. Especially in the case of large-scale physical machine failures, manual processing is used to repair, so as to effectively avoid the occurrence of system performance impacts due to frequent migration of virtual machines on the failed physical machine.

Embodiment four

4, there is shown a schematic diagram of an embodiment of another virtual machine recovery method of the present invention, which may specifically include the following steps: the physical machine fault detection module checks the network status of each physical machine in the cluster every 30 seconds, and Update to the physical machine fault information storage center; each physical machine in the cluster system autonomously detects its own fault conditions, and updates to the physical machine fault information storage center through the physical machine fault collection module.

For the scenario of hardware and software failures that can be repaired by the physical machine itself, the physical machine itself will be repaired by fault-tolerant methods; for the software and hardware failures that can be restarted and repaired by the physical machine itself, the physical machine itself will be restarted by the physical machine. Repair it in a way; if it is a hardware and software failure that cannot be repaired by the physical machine itself, it will be shut down.

The physical machine fault classification processing module obtains the physical machine fault information list from the physical machine fault information storage center every 1 minute; judges whether the physical machine fault information list is empty, if so, returns to the loop; otherwise, continues to judge the physical machine In the fault information list, whether there is a physical machine failure due to a network attack, if so, the security attack protection center outside the cluster will be triggered to process; otherwise, continue to judge the entity Whether there are any software and hardware failures in the machine failure information list that cannot be repaired by the physical machine itself, if so, send an instruction to the failed physical machine to instruct the failed physical machine to shut down the failed physical machine autonomously or by means of the physical machine The out-of-band management module on the above shuts down the failed physical machine; and then uses the virtualization interface to migrate the virtual machine on the failed physical machine to other healthy physical machines in the cluster system.

If it is determined that the physical machine failure information list does not cause a physical machine failure due to a network attack, continue to determine whether there is a physical machine failure information list in the physical machine failure information list that is completely disconnected and the network is disconnected. Duration Reach the preset time, such as 3 minutes; if there is, then determine whether the number of physical machines that are not connected to the network exceeds the preset number, for example, whether the number of failed physical machines exceeds the number of physical machines in a cabinet or the number of physical machines connected to a switch. If it exceeds, it is considered to be a cluster-scale network failure, and it is necessary to take a telephone call to the operation and maintenance personnel to repair it manually instead of automatically processing it. Otherwise, the virtual machine on the failed physical machine is migrated to the cluster system to other healthy physical machines through the virtualization interface.

Determine whether the physical machine's network failure is detected in the physical machine failure information list, but the network failure duration has not reached the preset time and the network returns to normal, and it is determined that the physical machine's network failure is caused by the restart of the physical machine Yes, it is judged whether the current physical machine is healthy, if it is healthy, the virtual machine on the physical machine is restarted through the virtualization interface, and if it is unhealthy, the virtual machine on the failed physical machine is migrated to the On other healthy physical machines in the cluster.

If it is detected in the physical machine fault information list that the network of the physical machine is unstable and the network instability lasts for a preset time, send an instruction to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or Shut down the faulty physical machine through the out-of-band management module on the physical machine; and migrate the virtual machine on the faulty physical machine to other healthy physical machines in the cluster system through a virtualization interface.

It should be noted that the situation that the physical machine network is unstable and the network instability lasts for the preset time is mainly due to some unknown reasons causing physical machine failures, such as system load, system network, and hardware failure. Class etc. Although the essential causes of this type of failure are difficult to investigate, the symptoms of such failures are clear, mainly: physical machine network packet loss, physical machine management channel access abnormalities, and physical machine performance usage abnormalities. For this type of physical machine failure, the same approach can be used.

Preferably, in an embodiment of the present invention, the healthy physical machine is determined by the following method: matching all physical machines in the cluster in the physical machine fault information list; there will be no physical machines that are successfully matched Determined as a healthy physical machine.

The embodiments of the present invention can perform rapid and accurate fault identification in a large-scale cloud computing cluster through various physical machine failure scenarios, and perform targeted classification processing, thereby realizing fast and highly reliable entities. The machine is damaged and repaired to ensure the rapid recovery of the virtual machine service on it.

Further, in the embodiment of the present invention, the physical machine autonomously detects itself The failure dynamics of the physical machine itself can be repaired by the physical machine itself, and the physical machine fault conditions that can be repaired by the physical machine itself are classified and repaired; for the physical machine fault conditions that the physical machine itself cannot be repaired, the physical machine fault classification processing module outside the cluster has Targeted classification and repair processing, thereby effectively reducing the occurrence of false judgments and missed judgments of physical machine faults, and more secure, stable, and rapid automatic recovery of virtual machines.

In addition, the embodiments of the present invention aim at physical machine failures that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, it can also be classified by physical machine failures outside the cluster. The processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet commercial standards, and at the same time ensure the effectiveness of automated physical machine isolation.

In addition, the embodiments of the present invention also take into account the possibility of physical machine-scale failures in large-scale cloud computing clusters, by judging whether the number of failed physical machines constitutes the computer room level, and adopting different repair processing methods in a targeted manner . Especially in the case of large-scale physical machine failures, manual processing is used to repair, so as to effectively avoid the occurrence of system performance impacts due to frequent migration of virtual machines on the failed physical machine.

It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the embodiments of the present invention are not limited by the described sequence of actions, because According to the embodiments of the present invention, some steps may be in other order or Simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

Embodiment five

5, there is shown a structural block diagram of an embodiment of an apparatus for repairing damage to a physical machine of the present invention. The apparatus 500 for repairing damage to a physical machine is applied to a physical machine in a virtualized cluster system, and may specifically include: an autonomous detection module Group (selfChecker) 510, autonomous processing module (selfHnadler) 520; wherein: the autonomous detection module 510 specifically includes: a detection unit 511 for autonomously detecting the failure dynamics of the physical machine itself; preferably, the detection unit 511 can be fixed The time interval of the self-detection of the failure dynamics of the physical machine itself, for example, once every 30 seconds.

The autonomous processing module 520 specifically includes: a fault-tolerant unit 521, which is used to repair the hardware and software faults in a fault-tolerant manner if the detection unit 511 detects the physical machine's own fault-tolerant repair; it is understandable that the implementation of the present invention Examples of the hardware and software faults that the physical machine itself can fault-tolerate and repair may include: disk failures that store data, virtualization-related kernel module abnormalities, and file system abnormalities that store data. For example, for the failure of a disk that stores data, the fault-tolerant repair method is specifically to isolate the disk first, and then use the mechanism of clustering to store multiple copies of data to realize the data on the disk is automatically copied to other healthy disks. This can effectively ensure that the faulty disk will not affect the stable operation of the system after isolation. Similarly, for the file system that stores data is damaged, the purpose of fault-tolerant repair can also be achieved by isolating the disk mounted on the file system.

The restart unit 522 is configured to, if the detection unit 511 detects a hardware and software failure that can be restarted and repaired by the physical machine itself, it is repaired by restarting the physical machine.

It is understandable that the hardware and software faults that the physical machine itself can restart and repair in the embodiment of the present invention may include: root file system read-only exceptions, network card driver restart repairable exceptions, operating system kernel module exceptions, etc. . Such hardware and software failures can be repaired by restarting the physical machine.

Preferably, the autonomous processing module 520 may further include: a shutdown unit 523, configured to: if the detection unit 511 detects a hardware and software failure that cannot be repaired by the physical machine itself, follow the physical machine outside the cluster The instruction of the fault classification processing module or the out-of-band management module 530 on the physical machine shuts down the faulty physical machine, and the physical machine fault classification processing module outside the cluster migrates the faulty entity through the virtualization interface Virtual machines on the machine to other healthy physical machines in the cluster system.

It should be noted that in the case of hardware and software failures that cannot be repaired by the physical machine itself, the out-of-band control system on the traditional physical machine usually has an availability of around 90% or even lower due to hardware failure rate and cost issues. Under the commercial availability requirement of at least 99.95% of the computing service itself, the total unavailability of the whole year is 262.8 minutes. If a physical machine fails If the device cannot be repaired in time, the failure of a physical machine will directly cause tens of minutes of manual processing time. Therefore, the availability indicators of the out-of-band control system in the prior art cannot match the failure recovery service level of commercial cloud computing services. Agreement (Service-Level Agreement, SLA). The technical solution provided by the embodiment of the present invention improves the traditional out-of-band management and control system. When the availability of the out-of-band management module 530 does not meet the commercial standard, the fault classification processing module of the physical machine outside the cluster can be used. The instruction instructs the faulty physical machine to shut down autonomously, and the physical machine fault classification processing module outside the cluster migrates the virtual machine on the faulty physical machine to other healthy physical machines in the cluster system through the virtualization interface; Significantly shorten the repair time of failed physical machines, thereby improving the commercial availability of the system.

Preferably, the autonomous detection module 510 may further include: a reporting unit 512, configured to use the physical machine fault collection module when the detection unit 511 autonomously detects a physical machine failure due to a network attack The physical machine fault information is reported to the physical machine fault information storage center, and the physical machine fault classification processing module outside the cluster triggers the processing by the security attack protection center outside the cluster.

Wherein, after the security attack protection center outside the cluster is triggered, a security cleaning program is started, such as traffic cleaning, etc., so as to restore the health of the failed physical machine. It should be noted that the network failure of a physical machine caused by a network DDoS attack and the physical machine network failure caused by a physical machine crash need to be treated differently. If the physical machine is suffering from network DDoS Migrate the virtual machine on it to it during an attack Other physical machines will produce a domino effect, leading to an enlarged risk of failure, that is, other physical machines are successively attacked and become unavailable, which may eventually cause flooding of the entire cluster of network equipment, leading to the risk of failure of the entire cluster of physical machines.

The embodiments of the present invention can perform rapid and accurate fault identification in a large-scale cloud computing cluster through various physical machine failure scenarios, and perform targeted classification processing, thereby realizing fast and highly reliable entities. The machine is damaged and repaired to ensure the rapid recovery of the virtual machine service on it.

Further, in the embodiment of the present invention, the physical machine autonomously detects its own failure dynamics, and specifically classifies and repairs the physical machine fault conditions that the physical machine itself can repair; the physical machine fault conditions that cannot be repaired by the physical machine itself , By using the physical machine fault classification processing module outside the cluster to perform targeted classification and repair processing, thereby effectively reducing the occurrence of false judgments and missed judgments of physical machine faults, and automatically recovering virtual machines more safely, stably and quickly.

In addition, the embodiments of the present invention aim at physical machine failures that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, it can also be classified by physical machine failures outside the cluster. The processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet commercial standards, and at the same time ensure the effectiveness of automated physical machine isolation.

Example Six

Referring to FIG. 6, there is shown a structural block diagram of an embodiment of a cluster entity machine fault classification and processing apparatus of the present invention. The entity machine fault classification and processing apparatus 600 may specifically include the following modules: an acquisition module 610 for obtaining data from entities The machine fault information storage center obtains a physical machine fault information list; it should be noted that the physical machine fault information list includes: the physical machine fault detection module outside the cluster detects the faulty physical machine from the faulty physical machine and reports it to the company The physical machine fault information of the physical machine fault information storage center, and the physical machine fault information collected by the physical machine fault collection module outside the cluster from the faulted physical machine and reported to the physical machine fault information storage center .

The first processing module 620 is configured to trigger processing by the security attack protection center outside the cluster if it is detected in the physical machine fault information list that the physical machine fails due to a network attack; it is understandable that, In actual applications, after the security attack protection center outside the cluster is triggered, a security cleaning program, such as traffic cleaning, is activated, so that the failed physical machine can be restored to health. It should be noted that the network failure of a physical machine caused by a network DDoS attack and the physical machine network failure caused by a physical machine crash need to be treated differently. If the physical machine is suffering from network DDoS Migrating the virtual machines on it to other physical machines during an attack will produce a domino effect, leading to an enlarged risk of failure, that is, other physical machines are successively attacked and become unavailable, which may eventually cause flooding of the entire cluster of network equipment. Lead to the risk of failure of the entire cluster of physical machines.

The second processing module 630 further includes: a shutdown processing unit, configured to send a shutdown of the failed physical machine to the failed physical machine if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list Preferably, the instruction can instruct the faulty physical machine to autonomously shut down the faulty physical machine or shut down the faulty physical machine through the out-of-band management module on the physical machine; the migration processing unit is used for virtual Migrate virtual machines on the failed physical machine to other healthy physical machines in the cluster system through a standardized interface.

It should be noted that the types of hardware and software failures that cannot be repaired by the physical machine itself can include: physical machine crashes, physical machine CPU abnormalities, physical machine memory abnormalities, physical machine power modules, and other hardware problems abnormal . This type of failure will directly cause the physical machine to be unavailable and require the replacement of the hardware module to be repaired. Therefore, the embodiment of the present invention isolates the failed physical machine from the cluster before performing hardware replacement or maintenance on the failed physical machine. .

Preferably, the physical machine fault classification processing device 600 may further include a third processing module 640, and the third processing module 640 specifically includes: a notification processing unit configured to: It is detected that the physical machine’s network is completely unreachable and the duration of the unreachable network reaches the preset time, and the number of physical machines whose network is unreachable exceeds one, and the operation and maintenance personnel are notified to repair it manually; the migration processing unit is used if the entity is in Machine fault information list If it is detected that the physical machine is completely disconnected from the network and the duration of the network disconnection reaches the preset time, and the number of physical machines that are disconnected from the network does not exceed the preset number, the virtual machine on the failed physical machine is migrated through the virtualization interface To other healthy physical machines in the cluster system.

Wherein, the preset time can be set to a suitable time period such as 3 minutes, 5 minutes, etc. according to actual conditions.

It should be noted that when it is detected that the network of the physical machine is completely disconnected and the duration of the network disconnection reaches the preset time, the embodiment of the present invention needs to further check whether the number of faulty physical machines with the network disconnected exceeds the entity of one cabinet. If the number of machines or the number of physical machines connected to a switch exceeds, it is considered to be a cluster-scale network failure, and it needs to be repaired manually by the operation and maintenance personnel through telephone calls instead of automatic processing. This is because for large-scale physical machine failures, when isolating physical machines and migrating virtual machines, a large number of physical machines will be shut down. When the equipment in the computer room (network equipment or electrical equipment, etc.) is restored, the physical machines need to be restarted again. Then restore the virtual machine. This series of operations will directly cause the manual processing time to double or even more, thereby greatly increasing the unavailable time of the virtual machine. Therefore, the method provided by the embodiment of the present invention differentiates and handles this type of physical machine failure, which can greatly shorten the repair time of the failed physical machine, thereby greatly reducing the time that the virtual machine on it is unavailable, thereby improving the commercial use of the system. Availability.

Preferably, the physical machine fault classification processing device 600 may further include a fourth processing module 650, and the fourth processing module 650 specifically includes: The restart processing unit is used for detecting that the physical machine's network failure is detected in the physical machine failure information list, but the network failure duration has not reached the preset time and the network returns to normal, and determining that the physical machine's network failure is a physical entity If the machine restarts, if it is determined that the current physical machine is healthy, restart the virtual machine on the physical machine through the virtualization interface; the migration processing unit is used for if in the physical machine fault information list It is detected that the network of the physical machine is unreachable but the duration of the unreachable network has not reached the preset time and the network returns to normal, and it is determined that the unreachable network of the physical machine is caused by the restart of the physical machine, then it is determined that the current physical machine is unhealthy In this case, the virtual machine on the failed physical machine is migrated to other healthy physical machines in the cluster through the virtualization interface.

Preferably, the physical machine fault classification processing device 600 may further include a fifth processing module 660, and the fifth processing module 660 specifically includes: a shutdown processing unit, configured to: If it detects that the network of the physical machine is unstable and the duration of network instability reaches a preset time, it sends an instruction to the faulty physical machine to instruct the faulty physical machine to autonomously shut down the faulty physical machine or use the belt on the physical machine The external management module shuts down the faulty physical machine; the migration processing unit is used to migrate the virtual machine on the faulty physical machine to other healthy physical machines in the cluster system through a virtualization interface.

It should be noted that the network of the physical machine is unstable and the network is unstable The situation where the predetermined duration reaches the preset time is mainly due to some unknown reasons causing physical machine failures, such as system load, system network, and hardware failure. Although the essential causes of this type of failure are difficult to investigate, the symptoms of such failures are clear, mainly: physical machine network packet loss, physical machine management channel access abnormalities, and physical machine performance usage abnormalities. For this type of physical machine failure, the same processing method can be used, that is, sending an instruction to the malfunctioning physical machine to instruct the malfunctioning physical machine to autonomously shut down the malfunctioning physical machine or shutting down the malfunctioning entity through the out-of-band management module on the physical machine Machine; and migrate the virtual machine on the failed physical machine to other healthy physical machines in the cluster system through the virtualization interface.

Preferably, the physical machine fault classification processing device 600 may further include: a determining module 670, configured to match all physical machines in the cluster in the physical machine fault information list, and there will be no entities that are successfully matched. The machine is determined to be a healthy physical machine.

The embodiments of the present invention can perform rapid and accurate fault identification in a large-scale cloud computing cluster through various physical machine failure scenarios, and perform targeted classification processing, thereby realizing fast and highly reliable entities. The machine is damaged and repaired to ensure the rapid recovery of the virtual machine service on it.

Further, in the embodiment of the present invention, the physical machine autonomously detects its own failure dynamics, and specifically classifies and repairs the physical machine fault conditions that the physical machine itself can repair; the physical machine fault conditions that cannot be repaired by the physical machine itself , By means of physical machine failures outside the cluster The class processing module performs targeted classification and repair processing, thereby effectively reducing the occurrence of misjudgments and missed judgments of physical machine failures, and automatically recovering virtual machines in a safer, stable, and faster manner.

In addition, the embodiments of the present invention aim at physical machine failures that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, it can also be classified by physical machine failures outside the cluster. The processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet commercial standards, and at the same time ensure the effectiveness of automated physical machine isolation.

In addition, the embodiments of the present invention also take into account the possibility of physical machine-scale failures in large-scale cloud computing clusters, by judging whether the number of failed physical machines constitutes the computer room level, and adopting different repair processing methods in a targeted manner . Especially in the case of large-scale physical machine failures, manual processing is used to repair, so as to effectively avoid the occurrence of system performance impacts due to frequent migration of virtual machines on the failed physical machine.

Example Seven

Referring to FIG. 7, there is shown a structural diagram of an embodiment of a virtual machine recovery system of the present invention. The virtual machine recovery system includes: a physical machine damage repair device 710, which is applied to each physical machine in the virtualized cluster system 700 ; Physical machine fault classification processing device 720 and physical machine fault information storage center 730; among them: The physical machine damage repairing device 710 may specifically include: an autonomous detection module 711 and an autonomous processing module 712; wherein: the autonomous detection module 711 is used for autonomously detecting the failure dynamics of the physical machine itself; the autonomous processing module 712 is used for The autonomous detection module 711 detects a hardware and software fault that the physical machine itself can be fault-tolerantly repaired, and then uses a fault-tolerant method to repair it; it is also used to if the autonomous detection module 711 detects a software and hardware fault that the physical machine itself can restart and repair. , Repair by restarting the physical machine.

Preferably, the autonomous processing module 712 can also be used to classify and process the module according to the physical machine faults outside the cluster if the autonomous detection module 711 detects a hardware and software fault that cannot be repaired by the physical machine itself 720 or through the out-of-band management module 713 on the physical machine to shut down the faulty physical machine, and the physical machine fault classification processing module 720 outside the cluster migrates the faulty physical machine through the virtualization interface The virtual machine is connected to other healthy physical machines in the cluster system.

Preferably, the autonomous detection module 712 can also be used to report the physical machine fault through the physical machine fault collection module 760 when the autonomous detection module 711 autonomously detects that the physical machine fails due to a network attack. The information is sent to the physical machine fault information storage center 730, and the physical machine fault classification processing module 720 outside the cluster triggers the security attack protection center 740 outside the cluster to process.

It should be noted that, in another embodiment of the present invention, the autonomous detection module 711 and the autonomous processing module 712 may be software modules deployed on each physical machine in the cluster, and are automatically activated when the physical machine is turned on. The operation of the autonomous detection module 711 and the autonomous processing module 712 does not depend on the file system The system only relies on CPU and memory.

The physical machine fault information storage center 730 is used to integrate all reported physical fault information into a physical machine fault information list; wherein, the physical machine fault information list includes: a physical machine fault detection module external to the cluster The group 750 detects the physical machine fault information from the faulty physical machine and reports it to the physical machine fault information storage center 730, and the physical machine fault collection module 760 outside the cluster collects and combines the physical machine fault information from the faulty physical machine. The physical machine fault information reported to the physical machine fault information storage center 730.

The physical machine fault classification processing device 720 is used to obtain a physical machine fault information list from the physical machine fault information storage center 730 through the acquisition module 721. If a physical machine fails due to a path attack, the first processing module 722 triggers the security attack protection center 740 outside the cluster to process; if software that cannot be repaired by the physical machine is detected in the physical machine fault information list If the hardware is faulty, the second processing module 723 sends an instruction to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or to shut down the faulty physical machine through the out-of-band management module 713 on the physical machine , And migrate the virtual machine on the failed physical machine to other healthy physical machines in the cluster system through the virtualization interface.

Preferably, the physical machine fault classification processing device 720 may further include a third processing module 724 for detecting that the physical machine’s network is completely unavailable and the network is unavailable in the physical machine fault information list. The renewal time reaches the preset time; judge whether the number of physical machines with unreachable network exceeds the preset number, and if so, notify the operation and maintenance personnel to repair it manually; otherwise, use the virtualization interface to migrate the virtual machine on the failed physical machine to the In the cluster system to other healthy physical machines.

Preferably, the physical machine fault classification and processing device 720 may further include a fourth processing module 725, which is used to detect in the physical machine fault information list that the physical machine’s network is unavailable but the duration of the network disconnection has not been reached. After the preset time is reached, the network returns to normal, and it is determined that the physical machine's network failure is caused by the restart of the physical machine, then determine whether the current physical machine is healthy, and if it is healthy, restart the physical machine through the virtualization interface If the virtual machine is unhealthy, the virtual machine on the failed physical machine is migrated to other healthy physical machines in the cluster through a virtualization interface.

Preferably, the physical machine fault classification processing device 720 may further include a fifth processing module 726 for detecting in the physical machine fault information list that the physical machine network is unstable and the network instability continues When the time reaches the preset time, send an instruction to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through the out-of-band management module on the physical machine; and through the virtualization interface Migrating the virtual machine on the failed physical machine to other healthy physical machines in the cluster system.

Preferably, the physical machine fault classification processing device 720 may further include a determining module 727 for matching all physical machines in the cluster in the physical machine fault information list, and the matching will not succeed. The physical machine of is determined to be a healthy physical machine.

It should be noted that, for the specific structures of the physical machine damage repairing device 710 and the physical machine fault classification processing device 720, please refer to the detailed description of the foregoing embodiment, which will not be repeated here.

It should be noted that, in another embodiment of the present invention, the physical machine fault classification processing device 720, the physical machine fault detection module 750, and the physical machine fault collection module 760 in the virtual machine recovery system are all deployed in a virtualized cluster. The software modules on physical machines other than the system 700 can be independently deployed on different physical machines, or can be combined and deployed on the same physical machine. In addition, the physical machine failure information storage center 730 is a database system deployed outside the virtualized cluster system 700. The security attack protection center 740 can directly adopt the existing security attack protection system. The embodiment of the present invention does not limit this.

The embodiments of the present invention have the following advantages: The embodiments of the present invention can be used in a large-scale cloud computing cluster to perform refined, fast and accurate fault identification through multiple physical machine failure scenarios, and perform targeted classification processing. , So as to achieve fast and highly reliable physical machine damage repair processing to ensure the rapid recovery of virtual machine services on it.

Further, in the embodiment of the present invention, the physical machine autonomously detects its own failure dynamics, and specifically classifies and repairs the physical machine fault conditions that the physical machine itself can repair; the physical machine fault conditions that cannot be repaired by the physical machine itself , Through the physical machine fault classification processing module outside the cluster to perform targeted classification and repair processing, thereby effectively reducing The occurrence of false judgments and missed judgments of physical machine failures makes it safer, more stable, and faster to automatically recover virtual machines.

In addition, the embodiments of the present invention aim at physical machine failures that cannot be repaired by the physical machine itself. In addition to shutting down the failed physical machine through the out-of-band management module on the failed physical machine, it can also be classified by physical machine failures outside the cluster. The processing module instructs the faulty physical machine to shut down autonomously, so as to make up for the problem that the availability of the out-of-band management module to call the shutdown operation cannot meet commercial standards, and at the same time ensure the effectiveness of automated physical machine isolation.

In addition, the embodiments of the present invention also take into account the possibility of physical machine-scale failures in large-scale cloud computing clusters, by judging whether the number of failed physical machines constitutes the computer room level, and adopting different repair processing methods in a targeted manner . Especially in the case of large-scale physical machine failures, manual processing is used to repair, so as to effectively avoid the occurrence of system performance impacts due to frequent migration of virtual machines on the failed physical machine.

As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiment.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other.

Those skilled in the art should understand that the embodiments of the embodiments of the present invention can be provided as methods, devices, or computer program products. Therefore, the present invention The embodiment may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may adopt computer program products implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program codes. form.

In a typical configuration, the computer equipment includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory. Memory may include non-permanent memory, random access memory (RAM) and/or non-volatile memory in computer-readable media, such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media. Computer-readable media include permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), and other types of random access memory (RAM) , Read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only CD-ROM (CD-ROM), digital multi-function Optical discs (DVD) or other optical storage, magnetic cassettes, magnetic tape storage or other magnetic storage devices, or any other non-transmission media, can be used to store information that can be accessed by computing devices. According to the definition in this article, computer-readable media does not include non-persistent computer-readable media (transitory media), such as modulated data signals and carrier waves.

The embodiments of the present invention are described with reference to the flowcharts and/or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and/or block in the flowchart and/or block diagram, and the combination of the process and/or block in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, dedicated computers, embedded processors or other programmable data processing terminal equipment to generate a machine, so that the processor of the computer or other programmable data processing terminal equipment The executed instruction generates a device for realizing the function specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal equipment to work in a specific manner, so that the instructions stored in the computer-readable memory generate instructions including the manufacturing of the instruction device The instruction device realizes the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

These computer program instructions can also be loaded on a computer or other programmable data processing terminal equipment, so that a series of operation steps are executed on the computer or other programmable terminal equipment to generate computer-implemented processing, so that the computer or other programmable terminal equipment The instructions executed on the modified terminal device provide steps for implementing the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

Although the preferred embodiments of the embodiments of the present invention have been described, once those skilled in the art know the basic creative concepts, they can implement these Additional changes and modifications are made to the embodiment. Therefore, the scope of the attached patent application is intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

Finally, it should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply these entities. Or there is any such actual relationship or sequence between operations. Moreover, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements not only includes those elements, but also includes those elements that are not explicitly listed. Other elements listed, or also include elements inherent to this process, method, article, or terminal device. Without more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other same elements in the process, method, article, or terminal device that includes the element.

The above provides a detailed introduction to the method and device for repairing physical machine damage applied to the virtualized cluster system, and the cluster physical machine fault classification processing method, device, and virtual machine recovery method and system provided by the present invention. Examples illustrate the principles and implementation of the present invention. The descriptions of the above examples are only used to help understand the methods and core ideas of the present invention; at the same time, for those of ordinary skill in the art, in accordance with the ideas of the present invention, specific implementation There may be changes in the method and the scope of application. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims (32)

A clustered physical machine fault classification processing method, which is characterized in that it comprises: obtaining a physical machine fault information list from a physical machine fault information storage center; if the physical machine fault information is detected in the physical machine fault information list due to a network attack , The security attack protection center outside the cluster is triggered to process; if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine fault information list, an instruction to shut down the faulty physical machine is sent to the failed physical machine; and Migrate the virtual machine on the failed physical machine to other healthy physical machines in the cluster system through the virtualization interface. The method according to claim 1, wherein the method further includes: if it is detected in the physical machine fault information list that the physical machine has a complete network disconnection and the network disconnection duration reaches a preset time; determining that the network is disconnected Whether the number of physical machines exceeds the preset number, if so, notify the operation and maintenance personnel to manually repair; otherwise, use the virtualization interface to migrate the virtual machines on the failed physical machine to the cluster system to other healthy physical machines. The method according to claim 1, wherein the method further includes: if it is detected in the physical machine fault information list that the physical machine has a network failure but the network failure duration has not reached the preset time, the network returns to normal , And it is determined that the physical machine’s network failure is caused by the physical machine’s restart, then determine whether the current physical machine is healthy, and if it is healthy, use The virtualization interface restarts the virtual machine on the physical machine, and if it is unhealthy, the virtual machine on the failed physical machine is migrated to other healthy physical machines in the cluster through the virtualization interface. The method according to claim 1, wherein the method further includes: if it is detected in the physical machine fault information list that the physical machine has a network instability and the network instability lasts for a preset time, reporting to the fault The physical machine sends instructions to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through the out-of-band management module on the physical machine; and migrate the virtual machine on the faulty physical machine through the virtualization interface To other healthy physical machines in the cluster system. The method according to claim 1, wherein, if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, a command to shut down the failed physical machine is sent to the failed physical machine The steps include: sending an instruction to shut down the faulty physical machine to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through an out-of-band management module on the physical machine. The method according to claim 1, wherein the healthy physical machine is determined by the following method: matching all physical machines in the cluster in the physical machine fault information list; determining the physical machine that has not been successfully matched as the healthy entity machine. The method according to claim 1, wherein the physical machine fault information list includes: a physical machine fault detection module outside the cluster The physical machine fault information detected by the faulty physical machine and reported to the physical machine fault information storage center, and the physical machine fault collection module outside the cluster is collected from the faulty physical machine and reported to the entity Physical machine fault information in the machine fault information storage center. A clustered physical machine fault classification and processing device, which is characterized by comprising: an acquisition module for acquiring a physical machine fault information list from a physical machine fault information storage center; a first processing module for detecting fault information on the physical machine If a physical machine failure is detected due to a network attack in the list, the security attack protection center outside the cluster will be triggered to process; the second processing module further includes: a shutdown processing unit, which is used if the physical machine failure information list is If a hardware and software failure that cannot be repaired by the physical machine itself is detected, an instruction to shut down the failed physical machine is sent to the failed physical machine; the migration processing unit is used to migrate the virtual machine on the failed physical machine to the failed physical machine through a virtualization interface On other healthy physical machines in the cluster system. The device according to claim 8, wherein the device further includes a third processing module, and the third processing module includes: a notification processing unit configured to detect the physical machine network in the physical machine fault information list If the path is completely blocked and the network failure duration reaches the preset time, and the number of the physical machine that is blocked by the network exceeds one, the operation and maintenance personnel will be notified to repair it manually; the migration processing unit is used if the physical machine is in the fault information list It is detected that the network of the physical machine is completely disconnected and the duration of the network disconnection reaches the preset time, and the number of the physical machine that is disconnected from the network does not exceed the preset number, then the virtual machine on the faulty physical machine is migrated through the virtualization interface Go to other healthy physical machines in the cluster system. The device according to claim 8, wherein the device further includes a fourth processing module, and the fourth processing module includes: a restart processing unit, configured to detect the physical machine network in the physical machine fault information list When the road is blocked but the network is blocked for less than the preset time, the network returns to normal, and it is determined that the network failure of the physical machine is caused by the restart of the physical machine. If it is determined that the current physical machine is healthy, Restart the virtual machine on the physical machine through the virtualization interface; the migration processing unit is used to detect that the physical machine has a network failure but the network failure duration has not reached the preset time in the physical machine fault information list. The path returns to normal, and it is determined that the physical machine’s network failure is caused by the physical machine’s restart. When it is determined that the current physical machine is unhealthy, the virtual machine on the failed physical machine is migrated to it through the virtualization interface. On other healthy physical machines in the cluster. The device according to claim 8, wherein the device further includes a fifth processing module, and the fifth processing module includes: a shutdown processing unit, configured to detect the physical machine network in the physical machine fault information list If the path is unstable and the network is unstable for a preset period of time, then an instruction is sent to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or through the belt on the physical machine. The external management module shuts down the faulty physical machine; the migration processing unit is used to migrate the virtual machine on the faulty physical machine to other healthy physical machines in the cluster system through a virtualization interface. The device according to any one of claims 8-11, wherein the shutdown processing unit is configured to: if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, send The faulty physical machine sends an instruction to shut down the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through the out-of-band management module on the physical machine. The device according to any one of Claims 8-11, wherein the device further includes: a determining module for matching all physical machines in the cluster in the physical machine fault information list, and there will be no successful matching The physical machine is determined to be the healthy physical machine. The device according to claim 8, wherein the physical machine fault information list includes: a physical machine fault detection module outside the cluster detects from the faulty physical machine and reports it to the physical machine fault information storage center Physical machine fault information, and physical machine fault information collected from the faulty physical machine by a physical machine fault collection module outside the cluster and reported to the physical machine fault information storage center. A method for recovering a virtual machine, which is characterized in that it is applied to a virtualized cluster system, and the method includes: the physical machine in the virtualized cluster system autonomously detects its own failure actions. State; if the hardware and software failures that the physical machine can be repaired by itself are automatically detected, repaired by fault-tolerant methods; if the software and hardware failures that the physical machine can be restarted and repaired are automatically detected, the physical machine can be repaired by restarting the physical machine. ; Obtain a physical machine fault information list from the physical machine fault information storage center; if it is detected in the physical machine fault information list that the physical machine fails due to a network attack, it will trigger the processing of the security attack protection center outside the cluster; if If a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine fault information list, the faulty physical machine is sent an instruction to shut down the faulty physical machine; and the virtualized interface is used to migrate the faulty physical machine. Virtual machine to other healthy physical machines in the cluster system. The method according to claim 15, wherein the method further includes: if it is detected in the physical machine fault information list that the physical machine has a complete network disconnection and the network disconnection duration reaches a preset time; determining that the network is disconnected Whether the number of physical machines exceeds the preset number, if so, notify the operation and maintenance personnel to manually repair; otherwise, use the virtualization interface to migrate the virtual machines on the failed physical machine to the cluster system to other healthy physical machines. The method according to claim 15, wherein the method also includes Including: If it is detected in the physical machine failure information list that the physical machine has a network failure but the network failure duration has not reached the preset time, the network returns to normal, and it is determined that the physical machine has a network failure is the physical machine restart If caused, determine whether the current physical machine is healthy. If it is healthy, restart the virtual machine on the physical machine through the virtualization interface. If it is unhealthy, use the virtualization interface to migrate the virtual machine on the failed physical machine to the cluster. On the other healthy physical machines within. The method according to claim 15, wherein the method further includes: if it is detected in the physical machine fault information list that the physical machine has a network instability and the network instability lasts for a preset time, reporting to the fault The physical machine sends instructions to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through the out-of-band management module on the physical machine; and migrate the virtual machine on the faulty physical machine through the virtualization interface To other healthy physical machines in the cluster system. The method according to claim 15, wherein the healthy physical machine is determined by the following method: matching all physical machines in the cluster in the physical machine fault information list; determining the physical machine that does not match successfully as the healthy entity machine. The method according to claim 15, wherein the step of obtaining the physical machine fault information list from the physical machine fault information storage center includes: The physical machine fault classification processing module should obtain the physical machine fault information list from the physical machine fault information storage center. The method according to claim 15, wherein, if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine failure information list, a command to shut down the failed physical machine is sent to the failed physical machine The steps include: sending an instruction to shut down the faulty physical machine to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through an out-of-band management module on the physical machine. The method according to claim 15, wherein the physical machine fault information list includes: a physical machine fault detection module outside the cluster detects from the faulty physical machine and reports it to the physical machine fault information storage center Physical machine fault information, and physical machine fault information collected from the faulty physical machine by a physical machine fault collection module outside the cluster and reported to the physical machine fault information storage center. A virtual machine recovery system, characterized in that the system includes: a physical machine damage repair device, applied to a physical machine in a virtualized cluster system to autonomously detect the failure dynamics of the physical machine itself, and if the physical machine itself can be detected autonomously The hardware and software faults of fault-tolerant repair are repaired by fault-tolerant methods; if the software and hardware faults that the physical machine itself can be restarted and repaired are detected autonomously, it is repaired by restarting the physical machine; the physical machine fault information storage center is used to restore All reported physical machine fault information is integrated into a physical machine fault information list; The physical machine fault classification processing device is used to obtain the physical machine fault information list from the physical machine fault information storage center, and trigger if the physical machine fault information is detected in the physical machine fault information list due to a network attack The security attack protection center outside the cluster handles; if a software and hardware failure that cannot be repaired by the physical machine itself is detected in the physical machine fault information list, the faulty physical machine is sent an instruction to shut down the faulty physical machine, and by The virtualization interface migrates the virtual machine on the failed physical machine to other healthy physical machines in the cluster system. The system according to claim 23, wherein the physical machine fault classification processing device is further used for: if it is detected in the physical machine fault information list that the physical machine has a complete network disconnection and the duration of the network disconnection reaches a preset time ; Determine whether the number of the physical machine that is not connected to the network exceeds the preset number, if so, notify the operation and maintenance personnel to repair it manually; otherwise, use the virtualization interface to migrate the virtual machine on the failed physical machine to the cluster system to other healthy entities On the machine. The system according to claim 23, wherein the physical machine fault classification processing device is further used for: if it is detected in the physical machine fault information list that the physical machine has a network disconnection but the network disconnection duration has not reached the preset time After the network returns to normal, and it is determined that the physical machine's network failure is caused by the restart of the physical machine, it is determined whether the current physical machine is healthy, and if it is healthy, the virtual machine on the physical machine is restarted through the virtualization interface. If it is unhealthy, use the virtualization interface to migrate the virtual machine on the failed physical machine to the cluster. On other healthy physical machines in the set. The system according to claim 23, wherein the physical machine fault classification processing device is further used for: if it is detected in the physical machine fault information list that the physical machine has an unstable network and the duration of network instability reaches a preset time At time, send an instruction to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or shut down the faulty physical machine through the out-of-band management module on the physical machine; and migrate the fault through the virtualization interface The virtual machine on the physical machine is connected to other healthy physical machines in the cluster system. The system according to claim 23, wherein the physical machine fault classification processing device is further used to: match all physical machines in the cluster in the physical machine fault information list, and determine the physical machine that has not been successfully matched as a healthy entity machine. The system according to claim 23, wherein the physical machine fault classification processing device is further configured to: send an instruction to shut down the faulty physical machine to the faulty physical machine to instruct the faulty physical machine to shut down the faulty physical machine autonomously or by The out-of-band management module on the physical machine shuts down the failed physical machine. The system according to claim 23, wherein the physical machine fault information list includes: a physical machine fault detection module outside the cluster detects from the faulty physical machine and reports it to the physical machine fault information storage center Physical machine fault information, and the physical machine fault collection module outside the cluster collects from the faulty physical machine and reports it to the physical machine Physical machine fault information in the physical machine fault information storage center. The system according to claim 23, wherein the physical machine damage repair device includes: an autonomous detection module, including: a detection unit, for autonomously detecting the failure dynamics of the physical machine itself; an autonomous processing module, including: a fault-tolerant unit , Used for if the detection unit detects the hardware and software faults that the physical machine itself can be fault-tolerantly repaired, it is repaired by fault-tolerant methods; the restarting unit is used if the detection unit detects that the physical machine itself can be repaired by the hardware and software faults that can be restarted If the body fails, it can be repaired by restarting the physical machine. The system according to claim 30, wherein the autonomous processing module further includes: a shutdown unit, configured to: if the detection unit detects a software and hardware failure that cannot be repaired by the physical machine itself, follow the physical machine outside the cluster The instructions of the fault classification processing module or the out-of-band management module on the physical machine shut down the faulty physical machine, and the physical machine fault classification processing module outside the cluster migrates the faulty physical machine through the virtualization interface Virtual machine to other healthy physical machines in the cluster system. The system according to claim 30, wherein the autonomous detection module further includes: a reporting unit, configured to use the physical machine when the autonomous detection module autonomously detects that the physical machine fails due to a network attack Fault collection The module reports the physical machine fault information to the physical machine fault information storage center, and the physical machine fault classification processing module outside the cluster triggers the security attack protection center outside the cluster to process.

Images