CN121029475B - Server storage controller recovery methods, electronic devices - Google Patents

Abstract

The application discloses a recovery method of a storage controller of a server and electronic equipment, which are characterized in that after operating system state information of the storage controller is monitored in real time and an abnormal storage controller and a normal storage controller are identified, the abnormal storage controller is controlled to generate a starting control instruction to start the abnormal storage controller, after a basic input and output system of the abnormal storage controller is successfully started, the state of the system of the abnormal storage controller is further checked, a target normal storage controller is determined according to the operating system resource utilization rate of the normal storage controller, and a target service mode of diskless remote starting operating system service of the target normal storage controller is determined according to the state of the system of the abnormal storage controller, so that the abnormal storage controller can be quickly recovered. Therefore, the abnormal storage controller can be automatically repaired when the system is in operation, the system downtime caused by the failure of the storage controller is reduced, and the reliability and the overall stability of the storage controller are obviously improved.

Description

Recovery method of storage controller of server and electronic equipment

Technical Field

The present application relates to the field of server technologies, and in particular, to a method for recovering a storage controller of a server, and an electronic device.

Background

The importance of storage is continuously improved with the increase of data volume, the progress of technology and the change of business demands, and storage devices are generally used for storing important information, so that the reliable operation of the storage devices becomes a key for guaranteeing the data security and the reliability of an information system. When a system disk of the storage controller fails, the system disk is generally required to be manually processed on site, and the problems of untimely processing time, low processing efficiency and the like exist, so that a large amount of data is lost, leaked and even maliciously tampered, and huge losses are brought to individuals, enterprises and institutions. In the related art, a separate external device such as a usb disk is generally required to repair a damaged storage controller of a system disk, but cannot repair the damaged storage controller in the operation process of the storage system.

Disclosure of Invention

The application provides a recovery method of a storage controller of a server and electronic equipment, which at least solve the problem that the storage controller cannot be automatically repaired when a storage system runs in the related technology.

The application provides a recovery method of a storage controller of a server, which comprises the steps of obtaining operating system state information of a plurality of storage controllers, determining at least one abnormal storage controller and at least one normal storage controller according to the operating system state information, generating a starting control instruction according to restarting guiding information of the abnormal storage controller to start the abnormal storage controller, obtaining resource utilization rate of the at least one normal storage controller under the condition that a basic input and output system of the abnormal storage controller is successfully started, determining a target normal storage controller according to the resource utilization rate of the at least one normal storage controller, and determining a target service mode of the target normal storage controller according to system disk state information of the abnormal storage controller so that the normal storage controller reloads an operating system of the abnormal storage controller according to the target service mode.

The application also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the recovery method of the storage controller of the server is realized when the processor executes the computer program.

According to the application, the abnormal storage controller is controlled to generate the starting control instruction after the operating system state information of the storage controller is monitored in real time and the abnormal storage controller and the normal storage controller are identified, so that the abnormal storage controller is started, after the basic input and output system of the abnormal storage controller is successfully started, the system state of the abnormal storage controller is further checked, and the service target service mode of the diskless remote starting operating system is determined according to the operating system resource utilization rate of the normal storage controller, so that the abnormal storage controller is quickly recovered. Therefore, the abnormal storage controller can be automatically repaired when the system is in operation, the system downtime caused by the failure of the storage controller is reduced, the reliability of the storage controller and the overall stability of the system are obviously improved, and the availability and the maintenance efficiency of the storage system are further enhanced.

Drawings

For a clearer description of embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described, it being apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to the drawings without inventive effort for those skilled in the art.

FIG. 1 is a schematic diagram of a plurality of controllers of a server according to some embodiments of the application;

FIG. 2 is a flow chart of a method of recovering a controller of a server according to some embodiments of the application;

FIG. 3 is a flow chart of a method of recovering a controller of a server according to further embodiments of the present application;

FIG. 4 is a flow chart of a method of recovering a controller of a server according to further embodiments of the present application;

fig. 5 is a block schematic diagram of an electronic device according to some embodiments of the application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without making any inventive effort are within the scope of the present application.

It should be noted that in the description of the present application, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. The terms "first," "second," and the like in this disclosure, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order.

The present application will be further described in detail below with reference to the drawings and detailed description for the purpose of enabling those skilled in the art to better understand the aspects of the present application.

The specific application environment architecture or specific hardware architecture upon which execution of the recovery method of the storage controller of the server depends is described herein.

In some embodiments, referring to fig. 1, the server includes a plurality of memory controllers, each including a BMC (Baseboard Management Controller, baseboard management memory controller), a BIOS (Basic Input/Output System), and a motherboard. In the embodiment of the present application, the server includes three storage controllers, specifically, the storage controller a, the storage controller B, and the storage controller C, which are described as examples, but are not limiting to the present application. The storage controller A comprises a substrate management storage controller A, a basic input/output system A and a main board A, the storage controller B comprises a substrate management storage controller B, a basic input/output system B and a main board B, and the storage controller C comprises a substrate management storage controller C, a basic input/output system C and a main board C.

The embodiment of the application provides a recovery method of a storage controller of a server, and the method is described in detail by combining an execution flow of the recovery method of the storage controller of the server.

FIG. 2 is a flow chart of a method of recovering a storage controller of a server according to some embodiments of the application. Referring to fig. 2, a method for restoring a storage controller of a server according to an embodiment of the present application may include the steps of:

S110, acquiring operating system state information of a plurality of storage controllers, and determining at least one abnormal storage controller and at least one normal storage controller according to the operating system state information.

Specifically, when the storage controller is in operation, the baseboard management storage controller monitors the operating system state information of the storage controller in real time, specifically, the baseboard management storage controller a monitors the operating system state information of the storage controller a in real time, the baseboard management storage controller B monitors the operating system state information of the storage controller B in real time, the baseboard management storage controller C monitors the operating system state information of the storage controller C in real time, for example, the operating system of the storage controller a sends its state information to the baseboard management storage controller a through an LPC (a computer bus) interface in real time, the operating system of the storage controller B sends its state information to the baseboard management storage controller B through an LPC interface in real time, and the operating system of the storage controller C sends its state information to the baseboard management storage controller C through an LPC interface in real time. The abnormal Operating System state information includes that a certain component of the Operating System is abnormal and the BMC cannot receive the Operating System state information, such as RAID (Redundant Array of INDEPENDENT DISKS ) card faults, system disk faults, trojan horse viruses in a System OS (Operating System) and the like.

In the following description, an abnormal storage controller in which the storage controller a is the operating system state information and is in an abnormal state, and a normal storage controller in which the storage controller B and the storage controller C are in a normal state are described as an example, but the present application is not limited thereto. It should be noted that the storage controller may be a storage controller.

S120, generating a starting control instruction according to the restarting guiding information of the abnormal storage controller so as to start the abnormal storage controller.

Specifically, when the abnormal storage controller is judged, the BMC of the abnormal storage controller generates a starting control instruction of the abnormal storage controller according to restarting guiding information (such as firmware version, starting parameter, configuration identifier and the like) of the abnormal storage controller, specifically, the BMC of the abnormal storage controller generates a starting control instruction of a main board of the abnormal storage controller and a starting control instruction of a basic input output system so as to control the basic input output system and the main board of the abnormal storage controller to start.

For example, when the BMC a determines that the storage controller a is an abnormal storage controller, the BMC a issues a start control instruction of the basic input/output system to the BIOS a through the IIC interface, and then the BMC a issues a start control instruction of the motherboard to the CPLD (Complex Programmable Logic Device ) a through the IIC (Inter-INTEGRATED CIRCUIT, integrated circuit bus) interface, and when the CPLD a receives the start control instruction of the motherboard, the BIOS a restarts to control the storage controller a when receiving the start control instruction of the basic input/output system.

S130, under the condition that the basic input and output system of the abnormal storage controller is started successfully, acquiring the resource utilization rate of at least one normal storage controller, and determining the target normal storage controller according to the resource utilization rate of the at least one normal storage controller.

Specifically, under the condition that the BIOS of the abnormal storage controller is successfully started, BMCs of the normal storage controllers acquire the operating system resource utilization rate of the normal storage controller, compare the operating system resource utilization rates of the normal storage controllers, determine the normal storage controller corresponding to the minimum value in the operating system resource utilization rates of the normal storage controllers as a target normal storage controller, and the diskless remote start operating system service of the corresponding target normal storage controller is the target diskless remote start operating system service.

Illustratively, when the BIOS A is successfully started, the BMC B acquires the operating system resource utilization rate of the storage controller B, and the BMC C acquires the operating system resource utilization rate of the storage controller C, and determines the target normal storage controller by comparing the operating system resource utilization rate of the storage controller B with the operating system resource utilization rate of the storage controller C. For example, the storage controller corresponding to the smaller value of the operating system resource utilization of the storage controller B and the operating system resource utilization of the storage controller C is determined as the target normal storage controller. Further, under the condition that the storage controller B is determined to be the target normal storage controller, the diskless remote start operating system service of the storage controller B is determined to be the target diskless remote start operating system service, the BMC of the storage controller B informs the operating system of the storage controller B to open the diskless remote start operating system service, under the condition that the storage controller C is determined to be the target normal storage controller, the diskless remote start operating system service of the storage controller C is determined to be the target diskless remote start operating system service, and the BMC of the storage controller C informs the operating system of the storage controller C to open the diskless remote start operating system service.

And S140, determining a target service mode of the target normal storage controller according to the system disk state information of the abnormal storage controller so that the target normal storage controller reloads the operating system of the abnormal storage controller according to the target service mode.

Specifically, when the BIOS of the abnormal storage controller is started successfully, the BMCs of the plurality of normal storage controllers acquire the operating system resource utilization rate of the normal storage controller, and the BIOS of the abnormal storage controller acquires the system disk state information of the abnormal storage controller and transmits the system disk state information of the abnormal storage controller to the BMCs of the abnormal storage controller. For example, BIOS A will perform storage controller A enumeration to obtain system disk state information for storage controller A and transmit the system disk state information for storage controller A to BMC A.

Then, the target service mode of the target normal storage controller is determined according to the system disk state information of the storage controller a, for example, the target service mode may be determined by searching a preset relation mapping table between the system disk state information and the service modes, where the preset relation mapping table includes a plurality of system disk state information and the service mode corresponding to each system disk state information.

After determining the target service mode of the target normal storage controller, the target diskless remote start operating system service of the target normal storage controller is controlled to establish connection with the abnormal storage controller, and the target diskless remote start operating system service reloads the operating system according to the target service mode, for example, assuming that the diskless remote start operating system service of the storage controller C is determined to be the target diskless remote start operating system service, the diskless remote start operating system service of the storage controller C establishes connection with the storage controller A, and reloads the preset mirror operating system for the storage controller A according to the target service mode to recover the abnormal storage controller A.

It should be noted that, under the condition that multiple abnormal storage controllers appear at the same time, multiple target normal storage controllers can be determined, the diskless remote start operating system service of the multiple target normal storage controllers is started to reload the operating systems for the multiple abnormal storage controllers respectively, one abnormal storage controller can be randomly selected, after the diskless remote start operating system service of the target normal storage controllers is started to reload the operating systems, the next abnormal storage controller is selected first, and the target normal storage controller is redetermined to reload the operating systems of the next abnormal storage controller.

According to the method, after the operating system state information of the storage controller is monitored in real time and the abnormal storage controller and the normal storage controller are identified, the abnormal storage controller is controlled to generate a starting control instruction so as to start the abnormal storage controller, after the basic input and output system of the abnormal storage controller is successfully started, the system state of the abnormal storage controller is further checked, the target normal storage controller is determined according to the operating system resource utilization rate of the normal storage controller, and the target service mode of diskless remote starting operating system service of the target normal storage controller is determined according to the system state of the abnormal storage controller, so that the abnormal storage controller can be quickly recovered. Therefore, the abnormal storage controller can be automatically repaired when the system is in operation, the storage system downtime caused by the storage controller failure is reduced, the reliability and the overall stability of the storage controller are obviously improved, and the availability and the maintenance efficiency of the storage system are further enhanced.

In some embodiments, determining the target service mode of the target normal storage controller based on the system disk state information of the abnormal storage controller includes determining the target service mode to be a network mounted operating system mode if the system disk state information of the abnormal storage controller is in an abnormal state.

Specifically, after the system disk state information of the abnormal storage controller is determined, the target service mode of the target normal storage controller is determined according to the system disk state information of the abnormal storage controller, for example, if the system disk of the abnormal storage controller can be read, but the system disk state information is determined to be an abnormal state according to the system disk state information of the abnormal storage controller, for example, a virus in the system disk, the target service mode is determined to be a network installation operating system mode.

In some embodiments, the system state information includes virtual state information, and determining that the system state information of the abnormal storage controller is an abnormal state includes determining that the system state information of the abnormal storage controller is an abnormal state if the virtual state information is a destage identifier or a failure identifier.

Specifically, the system state information includes virtual state information, and if the virtual state information is read as a degradation identifier or a failure identifier, the system state information of the abnormal storage controller is determined to be an abnormal state.

In some embodiments, the system state information includes physical state information, and determining a target service mode of the target normal storage controller according to the system state information of the abnormal storage controller includes determining that the target service mode is a memory start operating system mode if the physical state information is a loss identifier or an offline identifier.

Specifically, the system disk state information includes physical disk state information, and if the physical disk state information is read as a loss identifier or an offline identifier, which indicates that the system disk of the abnormal storage controller cannot be read, the target service mode of the target diskless remote starting operating system service is determined to be a memory starting operating system mode.

When the system disk is readable but has faults, the fault storage controller is recovered by installing the operating system through the network, and when the system disk cannot be read, the operating system is started by using the memory to recover the fault storage controller. Therefore, the storage controller can take proper measures under different fault conditions to restore the normal operation of the storage controller, and the availability and maintenance efficiency of the storage system can be improved.

In some embodiments, determining the target normal storage controller based on the resource utilization of the at least one normal storage controller includes obtaining the resource utilization of the at least one normal storage controller, and taking the normal storage controller corresponding to the minimum value of the resource utilization as the target normal storage controller.

Specifically, in the case where the storage controller B and the storage controller C are normal storage controllers, the target normal storage controller is determined by comparing the operating system resource utilization of the storage controller B with the operating system resource utilization of the storage controller C.

The exemplary embodiment determines that the storage controller B is the target normal storage controller if the operating system resource utilization of the storage controller B is smaller than the operating system resource utilization of the storage controller C, and determines that the diskless remote start operating system service of the corresponding storage controller B is the target diskless remote start operating system service if the operating system resource utilization of the storage controller C is smaller than the operating system resource utilization of the storage controller B, and determines that the storage controller C is the target normal storage controller and the diskless remote start operating system service of the corresponding storage controller C is the target diskless remote start operating system service.

According to the application, the storage controller with the minimum resource utilization rate is selected as the target normal storage controller by comparing the operating system resource utilization rates of the normal storage controllers, so that the target diskless remote starting operating system service is determined, and the diskless remote starting service with less resource occupation can be preferentially used when the abnormal storage controller is repaired, thereby optimizing resource allocation, improving the overall operation efficiency of the system, reducing the starting failure risk caused by insufficient resources, and enhancing the stability and reliability of the system.

In some embodiments, the method further comprises the steps of obtaining numbers of the plurality of normal storage controllers when the resource utilization rates of the plurality of normal storage controllers are the same and are the minimum value, wherein the numbers are used for representing numbers of the storage controllers installed in the interconnection expansion slots of the high-speed peripheral components of the server main board, and taking the normal storage server corresponding to the minimum value or the maximum value of the numbers as the target normal storage controller.

Specifically, in the case where the storage controller B and the storage controller C are normal storage controllers, if the operating system resource utilization of the storage controller B is the same as the operating system resource utilization of the storage controller C, the numbers of the storage controller B and the storage controller C are acquired, and the numbers of the storage controller B and the storage controller C are compared to determine the target normal storage controller. Wherein the number is used to indicate the number of the memory controller installed in the server motherboard high speed peripheral component interconnect expansion slot.

For example, in the case that the normal storage server corresponding to the minimum number is used as the target normal storage controller, if the number of the storage controller B is smaller than the number of the storage controller C, the storage controller B is determined to be the target normal storage controller, and if the number of the storage controller C is smaller than the number of the storage controller B, the storage controller C is determined to be the target normal storage controller. If the number of the storage controller B is larger than the number of the storage controller C, determining that the storage controller B is the target normal storage controller, and if the number of the storage controller C is larger than the number of the storage controller B, determining that the storage controller C is the target normal storage controller.

In this way, by introducing a deterministic strong arbitration policy of the number of the memory controller installed in the interconnection expansion slot of the high-speed peripheral component of the server main board, the problem of uncertainty caused by random selection among multiple candidate normal memory controllers is effectively solved, thereby ensuring the high efficiency, no disputes and predictability of the selection process and realizing the complete automation and high reliability of the failover process.

In some embodiments, the method further comprises shutting down the target service mode of the target normal storage controller after the abnormal storage controller is successfully restored.

The example is that if the storage controller B is the target normal storage controller, if the abnormal storage controller a resumes normal operation, the diskless remote start operating system service of the storage controller B is turned off, for example, the BMC of the storage controller B notifies the operating system of the storage controller B to turn off the diskless remote start operating system service, and if the abnormal storage controller C is the target normal storage controller, the diskless remote start operating system service of the storage controller C is turned off, for example, the BMC of the storage controller C notifies the operating system of the storage controller C to turn off the diskless remote start operating system service.

Therefore, by closing the diskless remote start operating system service of the target normal storage controller after the abnormal storage controller is restored, the system resources occupied before can be effectively released, the reasonable utilization of the system resources is ensured, the resource waste is avoided, and the running efficiency is further improved.

In some embodiments, the method further comprises sending an alarm instruction to the normal storage controller to alert the normal storage controller if the basic input output system of the abnormal storage controller is not successfully started and restarted for a preset number of times. The preset number of times may be determined according to practical situations, for example, the preset number of times may be 3 times, which is not particularly limited herein.

Specifically, when the abnormal memory controller is the memory controller a, and the CPLD a receives a startup control command of the motherboard, the BIOS a restarts the memory controller a when receiving a startup control command of the BIOS. However, the storage controller a, specifically, the basic input/output system of the storage controller a may not be restarted successfully after receiving the start control instruction, so the BMC a may issue corresponding start control instructions to the BIOS a and the CPLD a multiple times to control the storage controller a to restart. If the startup times of the basic input output system of the storage controller and the main board of the storage controller reach the preset times, the basic input output system of the storage controller still fails to be started, the BMC A sends an alarm instruction to the BMC B, the BMC C or the BMC B and the BMC through the back board of the storage controller so as to control the operation systems of the storage controller B, the storage controller C or the storage controller B and the storage controller C to send out an alarm, and inform a user that the storage controller A fails and cannot recover by oneself.

If the system is unsuccessful after a plurality of attempts, an alarm instruction is sent to other normal storage controllers to trigger the normal storage controllers to send out an alarm to inform a user, so that the fault diagnosis efficiency is improved, the user is ensured to know fault information in time, so that maintenance measures can be quickly taken, and the availability and maintenance efficiency of the whole storage system are improved.

In some embodiments, reloading the operating system of the exception memory controller according to the target service mode includes downloading a network boot program, an operating system kernel, and a configuration file based on the target service mode, loading the network boot program and the operating system kernel to the exception memory controller, and loading a preset mirrored operating system to a system disk of the exception memory controller based on the network boot program and the operating system kernel, and installing the preset mirrored operating system to the system disk of the exception memory controller according to the configuration file to resume the exception memory controller.

Specifically, after the abnormal storage controller establishes a connection with the target diskless remote start operating system service, the target diskless remote start operating system service loads a preset mirror operating system according to a target service mode (for example, a network installation operating system mode) so as to recover the abnormal storage controller.

After the connection between the abnormal storage controller a and the diskless remote start operating system service of the normal storage controller B is established, a PXE (Preboot eXecution Environment, pre-boot execution environment) client of the abnormal storage controller a (a corresponding PXE client configured by a network interface card of the storage controller) downloads a network boot program (for example, a pxelinux.0), an operating system kernel and a configuration file by using the diskless remote start operating system service of the normal storage controller B, and then the PXE client of the abnormal storage controller a loads the network boot program and the operating system kernel into a memory of the abnormal storage controller a, loads a preset mirror operating system into a system disk of the abnormal storage controller a based on the network boot program and the operating system kernel, and finally installs the preset mirror operating system into the system disk of the abnormal storage controller according to the configuration file to resume the abnormal storage controller a.

In some embodiments, reloading the operating system of the abnormal storage controller according to the target service mode comprises downloading the operating system kernel and the boot file based on the target service mode, and loading the operating system kernel, the boot file and the preset mirror operating system into the abnormal storage controller to operate so as to restore the abnormal storage controller.

Specifically, after the abnormal storage controller establishes a connection with the target diskless remote start operating system service, the target diskless remote start operating system service loads a preset mirror operating system according to a target service mode (for example, a memory start operating system mode) so as to recover the abnormal storage controller.

For example, after the abnormal storage controller a establishes a connection with the diskless remote start operating system service of the normal storage controller B, the PXE client of the abnormal storage controller a (the corresponding PXE client is configured by the network interface card of the storage controller) downloads the boot item file (for example, initrd) and the operating system kernel by using the diskless remote start operating system service of the normal storage controller B, and then the PXE client of the abnormal storage controller a loads the operating system kernel, the boot item file and the preset mirror operating system into the memory of the abnormal storage controller a to directly run, so as to restore the abnormal storage controller a.

In some embodiments, the target service mode includes a network mounted operating system mode, and the method further includes re-determining the target normal storage controller if the abnormal storage controller is still in an abnormal state after reloading the operating system of the abnormal storage controller according to the network mounted operating system mode, and reloading the operating system of the abnormal storage controller according to the re-determined network mounted operating system mode of the target normal storage controller.

Specifically, after the diskless remote start operating system service of the target normal storage controller loads the preset mirror operating system according to the network installation operating system mode, the abnormal storage controller still cannot be recovered, at this time, the target normal storage controller needs to be redetermined, and a specific determining manner is described in detail in the above embodiment, which is not described herein, so that the diskless remote start operating system service of the target normal storage controller loads the preset mirror operating system according to the network installation operating system mode to recover the abnormal storage controller.

After the abnormal storage controller is failed to recover through the network installation operating system mode, the network installation operating system mode is used for retrying, so that the recovery success rate is obviously improved, and the fault tolerance of the system is enhanced.

In some embodiments, the target service mode includes a memory-enabled operating system mode, and the method further includes, after reloading the operating system of the exception memory controller according to the network-mounted operating system mode, re-determining the target normal memory controller if the exception memory controller is still in an exception state, and reloading the operating system of the exception memory controller according to the re-determined memory-enabled operating system mode of the target normal memory controller.

Specifically, after the diskless remote start operating system service of the target normal storage controller loads the preset mirror operating system according to the network installation operating system mode, the abnormal storage controller still cannot be recovered, at this time, the target normal storage controller needs to be redetermined, and a specific determining manner is described in detail in the above embodiment, which is not described herein, so that the diskless remote start operating system service of the target normal storage controller loads the preset mirror operating system according to the memory start operating system mode to recover the abnormal storage controller.

After the abnormal storage controller is failed to recover through the network installation operation system mode, the method can flexibly switch to the memory starting operation system mode for retrying, so that the recovery success rate is remarkably improved, and the fault tolerance of the system is enhanced.

FIG. 3 is a flow chart of a method of recovering a storage controller of a server according to further embodiments of the present application. Referring to fig. 3, a method for recovering a storage controller of a server according to an embodiment of the present application may include the steps of:

s210, receiving a starting control instruction of the basic input/output system.

Specifically, after the abnormal storage controller A is determined, the BMC of the abnormal storage controller A generates a starting control instruction of the basic input output system, and issues the starting control instruction of the basic input output system to the BIOS A through the IIC interface, and the BIOS A receives the starting control instruction of the basic input output system.

S220, under the condition that the starting control instruction of the basic input and output system is responded and the starting is successful, system disk state information of the abnormal storage controller is obtained, and the system disk state information of the abnormal storage controller is sent to the baseboard management storage controller of the abnormal storage controller, so that the baseboard management storage controller of the abnormal storage controller determines a target normal storage controller and a target service mode of the target normal storage controller.

Specifically, when the BIOS A receives a start control instruction of the basic input output system, the basic input output system is controlled to restart, and under the condition that the basic input output system is successfully started, the BIOS A enumerates the storage controller A to acquire system disk state information of the storage controller A, the system disk state information of the abnormal storage controller A comprises virtual disk state information and physical disk state information, and the system disk state information of the storage controller A is transmitted to the BMC A, so that the BMC A of the abnormal storage controller A determines a target service mode of a target normal storage controller.

For example, BMC B obtains the operating system resource utilization of storage controller B, BMC C obtains the operating system resource utilization of storage controller C, and the target normal storage controller is determined by comparing the operating system resource utilization of storage controller B with the operating system resource utilization of storage controller C. For example, the storage controller corresponding to the smaller value of the operating system resource utilization of the storage controller B and the operating system resource utilization of the storage controller C is determined as the target normal storage controller. Further, under the condition that the storage controller B is determined to be the target normal storage controller, the BMC of the storage controller B informs an operating system of the storage controller B to open the diskless remote start operating system service, and under the condition that the diskless remote start operating system service of the storage controller C is determined to be the target diskless remote start operating system service, the BMC of the storage controller C informs the operating system of the storage controller C to open the diskless remote start operating system service. After the system state information of the abnormal storage controller A is determined, determining a target service mode of the target normal storage controller according to the system state information of the abnormal storage controller A, for example, if the virtual state information of the system disk of the abnormal storage controller A is a degradation identifier or a failure identifier, determining the target service mode to be a network installation operating system mode, and if the physical state information of the system disk of the abnormal storage controller A is a loss identifier or an offline identifier, determining the target service mode to be a memory starting operating system mode.

S230, the abnormal storage controller is controlled to establish connection with the remote start operating system service of the target normal control, so that the remote start operating system service of the target normal control reloads the operating system of the abnormal storage controller according to a target service mode, wherein the target normal storage controller is determined according to the operating system resource utilization rate of the normal storage controller with the operating system state information being in a normal state, and the target service mode of the target normal storage controller is determined according to the system disk state information of the abnormal storage controller.

Specifically, after the target normal storage controller and the target service mode of the target normal storage controller are determined, for example, the storage controller B is determined to be the target normal storage controller, and the BIOS B of the target normal storage controller B controls the diskless remote start operating system service of the storage controller B to establish connection with the abnormal storage controller a, so that the diskless remote start operating system service of the storage controller B loads a preset mirror operating system for the abnormal storage controller a according to the target service mode, so as to restore the abnormal storage controller a.

In some embodiments, controlling the abnormal storage controller to establish a connection with the target diskless remote start operating system service includes controlling the abnormal storage controller to broadcast a dynamic host configuration protocol message to request the dynamic host configuration protocol service to allocate an internet protocol address, and establishing a connection between the abnormal storage controller and the diskless remote start operating system service of the target normal storage controller based on the internet protocol address.

Specifically, in the case where it is determined that the storage controller B is the target normal storage controller, the BIOS a of the abnormal storage controller a will control the network interface card of the storage controller a to broadcast a dynamic host configuration protocol message, requesting the dynamic host configuration protocol service to allocate an internet protocol address. Once the abnormal storage controller a obtains the internet protocol address through the dynamic host configuration protocol service, the abnormal storage controller a can use the internet protocol address to communicate on the network, and the abnormal storage controller a can establish a connection with the diskless remote start operating system service of the target normal storage controller B based on the internet protocol address.

In some embodiments, establishing connection between the abnormal storage controller and the diskless remote start operating system service of the target normal storage controller based on the Internet protocol address comprises initiating a transmission control protocol connection request to a preset port of the diskless remote start operating system service of the target normal storage controller according to the Internet protocol address, and receiving a connection confirmation response returned by the diskless remote start operating system service of the target normal storage controller to establish connection between the abnormal storage controller and the diskless remote start operating system service of the target normal storage controller.

Specifically, based on the internet protocol address, the abnormal storage controller initiates a TCP (Transmission Control Protocol ) connection request to a preset port of the diskless remote start operating system service of the target normal storage controller, and after receiving the request, the diskless remote start operating system service of the target normal storage controller returns a connection confirmation response, and both sides complete the establishment of the session channel of the transmission layer according to the request, thereby realizing reliable network connection between the abnormal storage controller and the diskless remote start operating system service of the target normal storage controller.

As a specific example, referring to fig. 4, the method for recovering a storage controller of a server according to an embodiment of the present application may further include the steps of:

S301, starting.

S302, judging whether an operating system of the storage controller is abnormal or not. If yes, S303 is performed, otherwise S301 is performed.

S303, restarting the abnormal storage controller.

When the abnormal storage controller is judged, the BMC of the abnormal storage controller generates a starting control instruction of a main board of the abnormal storage controller and a starting control instruction of a basic input output system so as to control the starting of the basic input output system and the main board of the abnormal storage controller.

For example, when the BMC a determines that the storage controller a is an abnormal storage controller, the BMC a issues a start control instruction of the basic input/output system to the BIOS a through the IIC interface, and then the BMC a issues a start control instruction of the motherboard to the CPLD a through the IIC interface, and when the CPLD a receives the start control instruction of the motherboard, the BIOS a restarts to control the storage controller a when receiving the start control instruction of the basic input/output system.

S304, judging whether the basic input/output system of the abnormal storage controller is started normally. If yes, S305 is performed, otherwise S306 is performed.

S305, judging system disk state information of the abnormal storage controller. The system disk state information includes virtual disk state information and physical disk state information, and if the virtual disk state information of the abnormal storage controller is a degradation flag or a failure flag, S307 is performed, and if the physical disk state information of the abnormal storage controller is a loss flag or an offline flag, S308 is performed.

The system state information includes virtual disk state information and physical disk state information, and after determining the system state information of the abnormal storage controller a, the target service mode of the target normal storage controller is determined according to the system state information of the abnormal storage controller a, for example, if the virtual disk state information of the abnormal storage controller a is a degradation identifier or a failure identifier, the target service mode of the target normal storage controller is determined to be a network installation operating system mode, and if the physical disk state information of the abnormal storage controller a is a loss identifier or an offline identifier, the target service mode of the target normal storage controller is determined to be a memory starting operating system mode.

S306, attempting to restart the abnormal storage controller, if the BIOS of the abnormal storage controller cannot be started up for the preset times, and executing S310.

Specifically, when the abnormal memory controller is the memory controller a, and the CPLD a receives a startup control command of the motherboard, the BIOS a restarts the memory controller a when receiving a startup control command of the BIOS. However, the storage controller a, specifically, the basic input/output system of the storage controller, may not be restarted successfully after receiving the start control instruction, so the BMC a may issue corresponding start control instructions to the BIOS a and the CPLD a multiple times to control the storage controller a to restart. If the startup times of the basic input output system of the storage controller and the main board of the storage controller reach the preset times, the basic input output system of the storage controller still fails to be started, the BMC A sends an alarm instruction to the BMC B, the BMC C or the BMC B and the BMC through the back board of the storage controller so as to control the operation systems of the storage controller B, the storage controller C or the storage controller B and the storage controller C to send out an alarm, and inform a user that the storage controller A fails and cannot recover by oneself.

S307, the target diskless remote start operating system service loads a preset mirror operating system according to the network installation operating system mode.

After the connection between the abnormal storage controller a and the diskless remote start operating system service of the target normal storage controller B is established, the PXE client of the abnormal storage controller a (the corresponding PXE client is configured by the network interface card of the storage controller) downloads a network boot program (for example, pxelinux.0), an operating system kernel and a configuration file by using the diskless remote start operating system service of the target normal storage controller B, and then the PXE client of the abnormal storage controller a loads the network boot program and the operating system kernel into the memory of the abnormal storage controller a, loads the preset mirror operating system into the system disk of the abnormal storage controller a based on the network boot program and the operating system kernel, and finally installs the preset mirror operating system into the system disk of the abnormal storage controller according to the configuration file to operate so as to restore the abnormal storage controller a.

S308, the target diskless remote start operating system service loads a preset mirror operating system according to the memory start operating system mode, and S310 is executed.

After the connection between the abnormal storage controller a and the diskless remote start operating system service of the target normal storage controller B is established, the PXE client of the abnormal storage controller a (the network interface card of the storage controller configures the corresponding PXE client) downloads the boot item file (for example, initrd) and the operating system kernel by using the diskless remote start operating system service of the target normal storage controller B, and then the PXE client of the abnormal storage controller a loads the operating system kernel, the boot item file and the preset mirror operating system into the memory of the abnormal storage controller a to directly run, so as to restore the abnormal storage controller a. Under the condition that the diskless remote starting operating system service of the target normal storage controller loads the preset mirror image operating system according to the memory starting operating system mode, an alarm is also given, for example, the normal storage controller B, the normal storage controller C or the operating systems of the normal storage controller B and the normal storage controller C are controlled to give an alarm, and a user is informed that the system disk of the abnormal storage controller A cannot be read.

S309, judging whether the operating system of the abnormal storage controller runs normally. If yes, S311 is performed, otherwise S308 is performed.

S310, sending out alarm information.

S311, the abnormal storage controller is restored.

As another specific example, referring to fig. 1, each storage controller includes a baseboard management storage controller, a basic input output system, a motherboard, a programmable logic device, an operating system, a diskless remote start operating system service, and a plurality of storage controllers are connected through a storage controller backplane.

When the storage controller is started up each time, the BIOS of the storage controller monitors the system disk state information of the storage controller through enumeration, if the virtual disk state information of the abnormal storage controller A is not a degradation mark or a failure mark and the physical disk state information is not a loss mark or an offline mark, the system disk of the storage controller is indicated to be in a normal state, and at the moment, the system disk of the storage controller can be booted according to a preset value of the BMC (for example, the BMC sets a first priority to boot from the system disk, the BIOS boots from the first priority, and if the first priority is set to boot from the PXE, the BIOS boots from the first priority). If the BMC has no preset value, the BIOS defaults to boot from the system disk preferentially, and under the condition that the system disk is failed to boot from PXE, the BIOS transmits the boot information to the BMC of the same storage controller through the IIC.

In the case that the virtual disk state information of the system disk of the storage controller is monitored to be a degradation identifier or a failure identifier, or the physical disk state information is monitored to be a loss identifier or an offline identifier, the storage operating system on the storage controller cannot be booted. At this time, the BIOS transmits the system state information to the BMC of the same memory controller through the IIC (Inter-INTEGRATED CIRCUIT, integrated circuit interconnection), and the BMC performs corresponding action transmission according to the system state information and other comprehensive conditions. The BMC of the storage controller mainly carries out comprehensive judgment through collecting system state information transmitted by a BIOS and collecting operating system state information transmitted by an operating system of the storage controller, and issues corresponding instruction actions, wherein the instruction actions comprise ① notifying a CPLD to issue a storage controller restarting instruction through IIC, ② notifying the BIOS to issue a boot priority starting instruction through IIC, ③ notifying BMCs of other storage controllers through TCP/IP (Internet Protocol ) networks, the BMCs of the other storage controllers notify a storage OS through LPC, and start or close DRBL (Diskless Remote Boot in Linux, diskless remote starting operating system) service, when DRBL service needs to be started, each storage controller BMC communicates with the storage OS through LPC, acquires the condition of resource utilization rate of a storage controller CPU (Central Processing Unit, a central processing unit), a memory and the like, selects the storage controller with the minimum resource utilization rate, starts DRBL service, and if the condition that the utilization rate is the same, selects the storage OS with the minimum slot number of the storage controller to start DRBL service in a normal state.

The CPLD mainly receives a restarting instruction transmitted by the BMC through the IIC and is used for restarting the storage controller, meanwhile, the IIC is communicated with the BIOS to acquire the BIOS state, if the BIOS cannot be started, the CPLD is restarted for 3 times by default, if the BIOS cannot be started for 3 times, the BMC is notified through the IIC, the BMC is notified to other storage controllers, the other storage controllers BMC are notified to the storage OS to send out alarm information, and a user is reminded of timely maintenance processing.

Each storage controller is pre-installed DRBL with service and pre-stores storage system OS image files to be installed, and the system OS pre-configures DRBL two working modes:

① The memory starts an operating system mode, namely, after the BIOS of the memory controller is started through PXE, the BIOS acquires an IP address through DHCP (Dynamic Host Configuration Protocol ), downloads a memory OS kernel and initrd to the memory through DRBL service, loads a complete system mirror image to the memory for operation, and directly starts the memory OS from the memory when the RAID card or the system disk of the memory controller fails, so that the normal operation of the memory system is ensured.

② And the network installation operating system mode is that after the BIOS of the storage controller is started through PXE, the BIOS acquires an IP address through DHCP, and downloads pxelinux.0, corresponding configuration files, a kernel and the like through a TFTP (TRIVIAL FILE TRANSFER Protocol, simple file transfer Protocol) service in DRBL services, and the complete system mirror image is installed on a system disk according to the pre-configured file requirements, so that when the file on the system disk of the storage controller is damaged or the operation of the storage OS is abnormal, the storage OS is reinstalled, and the normal operation of the storage system is ensured.

The storage controller back plane mainly provides storage OS network interconnection, BMC network interconnection and configuration information such as network configuration, VPD (Vital Product Data, key product data) information and the like of each storage controller.

In addition to running the software functions of the storage system OS itself, the storage system OS provides a service that can be turned on or off DRBL, and sends its health status to the BMC in real time through the LPC interface.

In conclusion, through combining DRBL services based on the CPLD module and the BMC module of the multi-control storage, when the file of the storage OS is damaged or abnormal and the RAID card or the system disk fails and cannot be started, the automatic processing of the failure that the storage OS cannot be started is realized, the problem that when the RAID is normal and the file of the storage OS is abnormal, the manual processing is needed is solved, the problem that when the RAID card or the system disk is abnormal and the failure that the storage OS cannot be started is solved, and the method has the advantages of low cost, high efficiency and strong stability, does not need manual intervention, and greatly improves the usability and the reliability of storage.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment.

Embodiments of the present application also provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the aforementioned method of restoring a storage controller of a server.

The embodiment of the present application further provides an electronic device, referring to fig. 5, the electronic device 400 includes a memory 410, a processor 420, and a computer program stored in the memory 410 and capable of running on the processor 420, where the processor 420 executes the computer program to implement the aforementioned method for recovering the storage controller of the server.

The description of the features in the embodiment corresponding to the electronic device may refer to the related description of the embodiment corresponding to the recovery method of the storage controller of the server, which is not described herein in detail.

In an exemplary embodiment, the computer readable storage medium may include, but is not limited to, a U disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a removable hard disk, a magnetic disk, or an optical disk, etc. various media in which a computer program may be stored.

Embodiments of the present application also provide a computer program product comprising a computer program/instruction which, when executed by a processor, implements a method of recovering a storage controller of a server as described above.

Embodiments of the present application also provide another computer program product comprising a non-volatile computer readable storage medium storing a computer program which, when executed by a processor, implements the steps of a recovery method embodiment of a storage controller of any one of the servers described above.

Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The method for recovering the storage controller of the server and the electronic equipment provided by the application are described in detail. The principles and embodiments of the present application have been described herein with reference to specific examples, the description of which is intended only to facilitate an understanding of the method of the present application and its core ideas. It should be noted that it will be apparent to those skilled in the art that various modifications and adaptations of the application can be made without departing from the principles of the application and these modifications and adaptations are intended to be within the scope of the application as defined in the following claims.

Claims (10)

1. A method of restoring a storage controller of a server, the server comprising a plurality of storage controllers, the method comprising:

Acquiring operating system state information of a plurality of storage controllers, and determining at least one abnormal storage controller and at least one normal storage controller according to the operating system state information;

Generating a starting control instruction according to the restarting guiding information of the abnormal storage controller so as to start the abnormal storage controller;

under the condition that the basic input and output system of the abnormal storage controller is successfully started, acquiring the resource utilization rate of at least one normal storage controller, and determining a target normal storage controller according to the resource utilization rate of at least one normal storage controller;

Determining a target service mode of the target normal storage controller according to the system disk state information of the abnormal storage controller so that the target normal storage controller reloads an operating system of the abnormal storage controller according to the target service mode;

the system state information comprises virtual state information, and a target service mode of the target normal storage controller is determined according to the system state information of the abnormal storage controller, wherein the target service mode is determined to be a network installation operating system mode under the condition that the virtual state information is a degradation identifier or a failure identifier;

The system state information comprises physical disk state information, and the target service mode of the target normal storage controller is determined according to the system state information of the abnormal storage controller, wherein the target service mode is determined to be a memory starting operating system mode under the condition that the physical disk state information is a loss identifier or an offline identifier.

2. The method for restoring a storage controller of a server according to claim 1, wherein determining a target normal storage controller based on a resource utilization of at least one of the normal storage controllers comprises:

acquiring the resource utilization rate of at least one normal storage controller;

and taking the normal storage controller corresponding to the minimum value of the resource utilization rate as the target normal storage controller.

3. The method for restoring a storage controller of a server according to claim 2, further comprising:

under the condition that the resource utilization rates of a plurality of normal storage controllers are the same and are the minimum value, the numbers of the plurality of normal storage controllers are obtained, wherein the numbers are used for representing the numbers of the storage controllers installed in the interconnection expansion slots of the high-speed peripheral components of the server main board;

and taking the normal storage server corresponding to the minimum number value or the maximum number value as a target normal storage controller.

4. A method of recovering a storage controller of a server according to any one of claims 1-3, further comprising:

and after the abnormal storage controller is successfully recovered, closing the target service mode of the target normal storage controller.

5. The method for restoring a storage controller of a server according to claim 1, further comprising:

And under the condition that the basic input and output system of the abnormal storage controller is not successfully started and restarted for a preset number of times, sending an alarm instruction to the normal storage controller so as to give an alarm to the normal storage controller.

6. The method of claim 1, wherein reloading the operating system of the abnormal storage controller according to the target service mode comprises:

Downloading a network bootstrap program, an operating system kernel and a configuration file based on the target service mode;

Loading the network boot program and the operating system kernel to the abnormal storage controller, loading a preset mirror image operating system to a system disk of the abnormal storage controller based on the network boot program and the operating system kernel, and installing the preset mirror image operating system to the system disk of the abnormal storage controller according to the configuration file for running so as to restore the abnormal storage controller.

7. The method of claim 1, wherein reloading the operating system of the abnormal storage controller according to the target service mode comprises:

downloading an operating system kernel and a guide item file based on the target service mode;

And loading the operating system kernel, the guide item file and the preset mirror image operating system into the abnormal storage controller to run so as to recover the abnormal storage controller.

8. The method for restoring a storage controller of a server of claim 6, wherein the target service mode comprises a network-mounted operating system mode, the method further comprising:

After reloading the operating system of the abnormal storage controller according to the network installation operating system mode, re-determining the target normal storage controller under the condition that the abnormal storage controller is still in an abnormal state;

And reloading the operating system of the abnormal storage controller according to the redetermined network installation operating system mode of the target normal storage controller.

9. The method for recovering a storage controller of a server according to claim 6, wherein the target service mode includes a memory-initiated operating system mode, the method further comprising:

After reloading the operating system of the abnormal storage controller according to the network installation operating system mode, re-determining the target normal storage controller under the condition that the abnormal storage controller is still in an abnormal state;

And restarting the operating system of the abnormal storage controller according to the memory starting operating system mode of the normal storage controller which is determined again.

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing a method of recovering a storage controller of a server according to any one of claims 1-9 when the computer program is executed.