Server management method and server
By dividing the server into computing components and storage components and utilizing the automated configuration of the BMC and processor systems, the problem of low server maintenance efficiency is solved, and plug-and-play and rapid recovery of computing components are achieved.
Patent Information
- Application Number
- CN202011583673.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2016-12-30
- Publication Date
- 2025-09-16
- Estimated Expiration
- 2036-12-30
AI Technical Summary
The integrated configuration of various components in existing servers leads to low maintenance efficiency, especially the inconsistent upgrade cycles of different components, which makes maintenance difficult and inefficient.
The server is divided into two parts: computing components and storage components. They are designed with detachable connections. The computing components include components with short maintenance cycles, while the storage components include components with long maintenance cycles. The BMC and processor system are used to achieve automatic configuration and driver package copying, making the computing components plug-and-play.
It improves the flexibility and efficiency of server maintenance, reduces maintenance time and difficulty, and enables rapid recovery and automatic configuration of computing components.
Smart Images

Figure CN112732289B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computers, and in particular to a server management method and a server. Background Art
[0002] As business grows, the server's processing power may not meet the needs of business operations, necessitating regular server upgrades and replacement of components with more powerful processing capabilities. A server includes numerous components, such as fans, memory, network cards, disks, and processors, all integrated onto a single motherboard. Each component has a different upgrade cycle. For example, the motherboard, processor, and memory have shorter upgrade cycles, typically 2 to 3 years, while the disk upgrade cycle is longer, typically 5 to 7 years. Current server architectures show that all server components are integrated onto a single motherboard. Maintenance on multiple components requires technicians to remove and insert each component individually, resulting in low maintenance efficiency. Summary of the Invention
[0003] The present application discloses a server management method and a server, which can flexibly replace computing components or storage components of the server to improve maintenance efficiency.
[0004] In a first aspect, the present application discloses a server comprising a computing component and a storage component. The computing component comprises a BMC (Baseboard Management Controller), a processor system, a computing motherboard computing component, and a first connector. The computing motherboard serves as a carrier for the computing components in the computing component. The BMC, processor system, computing components, and first connector may be located on the computing motherboard. The processor system is a processor network consisting of one or more processors, which may be single-core or multi-core processors. The computing components are components that perform specific business functions and may include network cards, sound cards, MEZZ cards, or graphics processing cards. The processor system is connected to the BMC and the first connector, respectively. The storage component comprises a disk, a storage motherboard, a secondary memory, and a second connector. The storage motherboard serves as a carrier for the components in the storage component. The disk, secondary memory, and second connector are located on the storage motherboard. The number of disks may be one or more, and the disk type may be HDD or SDD. If there are multiple disks, the multiple disks may form a RAID array. The secondary memory stores backup configuration information for the computing components, which is used to restore the configuration parameters of the computing components. The secondary memory is non-volatile memory, characterized by data protection during power outages. The computing motherboard and the storage motherboard are two independent motherboards, and the computing motherboard and the storage motherboard are connected in a detachable manner through a first connector and a second connector, and the first connector and the second connector match each other, for example: the first connector is a USB (Universal Serial Bus, USB for short) plug, and the second connector is a USB socket; or the first connector is a USB socket, and the second connector is a USB plug; or the first connector is a PCIE (Peripheral Component Interconnect Express, a special peripheral component interconnection standard, PCIE for short) male head, and the second connector is a PCIE female head; or the first connector is a PCIE female head, and the second connector is a PCIE (Serial Advanced Technology Attachment, SATA for short) male head; or the first connector is a SATA plug, and the second connector is a SATA socket. It can be understood that the first connector and the second connector can be the above-mentioned male-female connection method, or a cable connection method, or other detachable connection methods, and this application does not limit it. When the first connector and the second connector are docked, the BMC is connected to the second memory through the first connector and the second connector, and the processor system is connected to the disk through the first connector and the second connector.
[0005] In the above embodiment, the server is divided into two parts, a computing component and a storage component, according to the maintenance cycle of the components. The computing component and the storage component are connected in a detachable manner. The computing component includes components with a short maintenance cycle, and the storage component includes components with a long maintenance cycle. In this way, the computing component or the storage component can be flexibly replaced when maintaining the server, and the maintenance efficiency is high.
[0006] In a possible implementation of this aspect, the functions of the components in the server are as follows:
[0007] The BMC is configured to restore configuration parameters of the computing component according to the computing component backup configuration information stored in the second memory when the computing component is replaced;
[0008] The processor system is used to copy the driver package of the computing component stored in the preset first memory to the disk, and start from the disk after the copying is completed.
[0009] By implementing the above embodiment, the processor system configures the computing components according to its own service channels, thereby reducing resource usage and enabling the computing components to quickly return to a normal state.
[0010] In a possible implementation of this aspect, the processor system is used to restore the configuration parameters of the computing component according to the backup configuration information of the computing component stored in the second memory when the computing component is replaced, copy the driver package of the computing component stored in the preset first memory to the disk, and boot from the disk after the copying is completed.
[0011] In the above embodiment, when the computing component is updated, the computing component is driven according to the driver package of the first memory, and the computing component is configured according to the backup configuration information pre-stored in the second memory, so that the server automatically returns to normal state without reinstalling the operating system, and the configuration information is quickly restored, thereby realizing plug-and-play of the computing component, reducing the maintenance time and difficulty of server upgrades.
[0012] In one possible implementation of this aspect, the BMC is configured to determine whether the computing component has been replaced, and if so, set a preset first memory as a first boot device and restart the computer, wherein the first memory stores a driver package of the computing component, and the driver package of the computing component is used to drive the computing component;
[0013] The processor system is used to restart from the first memory, read the driver package of the computing component from the first memory after restarting, and copy the driver package of the computing component to the disk;
[0014] The BMC is also used to set the disk as the first boot device and restart after the copy is completed;
[0015] The processor system is further configured to restart from the disk, and after restarting, drive the computing component according to the driver package of the computing component stored in the disk.
[0016] In a possible implementation of this aspect, the first memory is located in the computing component of the server, that is, the computing component includes the first memory, the BMC is connected to the first memory, and the processor system is connected to the first memory; or the first memory is located in the management server, the processor system and the BMC are both connected to the first memory, the management server manages multiple servers, and the server of this embodiment is any one of the multiple servers managed by the management server. In the above embodiment, the deployment method of the first memory located in the computing component is suitable for a single-machine scenario and has the characteristic of a short recovery time. The deployment method of the first memory located in the management server is suitable for a cluster scenario, which can effectively reduce the storage space occupied by the driver package of the computing component and facilitate centralized management.
[0017] In one possible implementation of this aspect, the second memory further stores computing motherboard backup configuration information, and the BMC is further configured to read the computing motherboard backup configuration information from the second memory and configure the computing motherboard according to the computing motherboard backup configuration information. For example, the computing motherboard backup configuration information includes BIOS (Basic Input Output System) configuration information.
[0018] In a possible implementation of this aspect, the second memory further stores computing component backup identity information and computing mainboard backup identity information.
[0019] The BMC performing the step of determining whether the computing component is updated includes:
[0020] The BMC obtains the identity information of the computing component and obtains the backup identity information of the computing component from the second memory; the identity information includes but is not limited to one or more of a model, a serial number, and a MAC address.
[0021] comparing the identity information of the computing component and the backup identity information of the computing component to determine whether they are the same;
[0022] If not, it is determined that the computing component is updated.
[0023] The above embodiment can automatically identify whether the computing mainboard and the network card have been updated after the computing component and the storage component are docked, and has the characteristics of fast detection speed and low resource occupation.
[0024] In a possible implementation manner of this aspect, the BMC is further configured to update the computing mainboard identity information based on the identity information of the computing mainboard when the computing mainboard is replaced.
[0025] In the case where the computing component is replaced, the computing component backup identity information stored in the second memory is updated according to the identity information of the computing component.
[0026] In a possible implementation of this aspect, the disk also stores RAID configuration information, the storage component has multiple disks, and the processor system is further configured to read the RAID configuration information from the disk and configure the disk according to the RAID configuration information.
[0027] In one possible implementation of this aspect, the storage assembly further includes a RAID card, which is disposed on the storage mainboard and connected to the disks and the second connector. When the first connector and the second connector are mated, the processor system is connected to the RAID card via the first connector and the second connector. The processor system is further configured to read RAID configuration information from the RAID card and configure the disks based on the RAID configuration information.
[0028] In a second aspect, the present application discloses a server management method, and the upgrade method is applied to a server. The server of the present application includes but is not limited to a video server, a web server, a file server, or a data center server. The computing component includes a BMC, a computing motherboard, a processor system, a computing component, and a first connector. The storage component includes a disk, a storage motherboard, a second memory, and a second connector. The first connector and the second connector match each other, so that the computing component and the storage component are connected in a detachable manner. When either the computing component or the storage component needs to be upgraded, the old computing component or the storage component can be directly unplugged and replaced with a new computing component or storage component. In the computing component, the processor system is respectively connected to the BMC, the computing component, and the first connector. The computing motherboard serves as a carrier for each component in the computing component. The BMC, the computing component, the processor system, and the first connector are arranged on the computing motherboard. In the storage component, the storage motherboard serves as a carrier for each component in the storage component. The second connector, the disk, and the second memory are arranged on the storage motherboard. The second connector is respectively connected to the disk and the second memory. When the first connector and the second connector are docked, the BMC is connected to the second memory via the first connector and the second connector, and the processor system is connected to the disk via the first connector and the second connector. The computing component mainly performs computing functions, including but not limited to data processing, sound processing, text processing and control functions; accordingly, the computing component can be a network card, a graphics card, an image processing card, a MEZZ card or other types of components. The storage component is mainly used to store large-capacity data. The type of disk in the storage component is not limited, and the number of disks can be one or more. In the case of multiple disks, the disk types of the multiple disks can all be HDD (Hard Disk Drive, HDD for short), all SSD (Solid State Drives, solid-state drives) or a mixture of HDD and SSD. The first memory and the second memory are non-volatile memories (NVM). NVM has the characteristic that data is not lost after power failure. NVM includes but is not limited to ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory) and Flash Memory. The first memory stores a driver package of the computing component, which is used to drive the computing component. The first memory can be set in the computing component of the server or outside the server. The second memory stores computing component backup configuration information, which is used to restore the configuration parameters of the computing component.The computing mainboard and the storage mainboard are two independent mainboards, which are connected in a detachable manner through a first connector and a second connector. The server management method includes: when a computing component is replaced, restoring the configuration parameters of the computing component according to the computing backup configuration information stored in the second memory, and driving the computing component according to the driver package of the computing component stored in the first memory.
[0029] In a possible implementation of this aspect, when a computing component is replaced, the BMC restores configuration parameters of the computing component according to the computing backup configuration information stored in the second memory;
[0030] The processor system drives the computing component according to the driver package of the computing component stored in the first memory.
[0031] In a possible implementation of this aspect, when a computing component is replaced, the processor system restores the configuration parameters of the computing component based on the computing component backup configuration information stored in the second memory, copies the driver package of the computing component stored in the preset first memory to the disk, and boots from the disk after the copying is completed.
[0032] In a possible implementation of this aspect, copying the driver package of the computing component stored in the preset first memory to the disk includes: starting from the first memory, reading the driver package of the computing component from the first memory after starting, and copying the driver package of the computing component to the disk.
[0033] In a possible implementation of this aspect, the BMC determines whether the computing component has been updated. If so, it continues to determine whether the computing component supports out-of-band configuration. If so, the BMC can directly configure the computing component. The BMC reads the computing component backup configuration information from the second memory and performs relevant configurations on the computing component according to the computing component backup configuration information. For example, if the computing component is a network card, the MAC (Media Access Control, Media Access Control, abbreviated as MAC) address, IP address, gateway address or other addresses of the network card are configured according to the network card backup configuration information; the BMC sets the first memory as the first boot device (First Boot Device) and restarts. The processor system restarts from the first memory, reads the driver package of the computing component in the first memory after restarting, and copies the driver package to the disk. After the above-mentioned copy operation is completed, the BMC sets the disk as the first boot device and restarts; the processor system restarts from the disk, searches for the driver package of the computing component in the disk, and drives the computing component according to the driver package.
[0034] In the above embodiment, when the computing component of the server is replaced, the computing component is driven according to the driver package in the first memory, and the computing component is configured according to the backup configuration information pre-stored in the second memory, so that the server automatically returns to normal state without reinstalling the operating system, and the configuration information is quickly restored, thereby realizing plug-and-play of the computing components, reducing the maintenance time and difficulty of server upgrades.
[0035] In one possible implementation of this aspect, the BMC determines whether a computing component has been updated. If so, it reads the computing component backup configuration information from the second memory, stores the read computing component backup configuration information in the first memory, sets the first memory as the first boot device, and restarts. The processor system restarts from the first memory, which stores a streamlined operating system and management software. The streamlined operating system retains only basic functions and has the characteristics of being small in size and occupying few resources. The streamlined operating system can be a Linux operating system, a Unix operating system, a Windows PE operating system, or other operating systems, and this application does not limit this. After restarting, the management software in the operating system runs. The management software reads the computing component driver package from the first memory, copies the computing component driver package to the disk, reads the computing component backup configuration information from the first memory, and configures the computing component according to the computing component backup configuration information. After completing the copying and configuration, the BMC sets the disk as the first boot device and restarts. The processor system restarts from the disk, and after restarting, drives the computing component according to the computing component driver package stored on the disk.
[0036] In one possible implementation of this aspect, computing motherboard backup configuration information is stored in the first memory, and the server management method further includes: the BMC reading the computing motherboard backup configuration information from the second memory, and configuring the computing motherboard based on the computing motherboard backup configuration information. In the above embodiment, after the computing motherboard configuration information is changed, the computing motherboard backup configuration information originally stored in the second storage area is updated based on the changed configuration information, thereby achieving timely backup of the computing motherboard configuration information.
[0037] In a possible implementation of this aspect, the second memory further stores computing component backup identity information and computing mainboard backup identity information. The BMC obtains the computing mainboard identity information and reads the computing mainboard backup identity information from the second memory, and temporarily stores the read computing mainboard backup identity information in a designated location in the computing component. For example, the BMC compares the computing mainboard identity information and the read computing mainboard backup identity information to see if they are the same. If they are not the same, it is determined that the computing mainboard has been replaced, that is, the current computing mainboard is the replaced computing mainboard, and the corresponding entire computing component has also been replaced. The BCM obtains the computing component identity information and reads the computing component backup identity information from the second memory, and temporarily stores the read computing component backup identity information in a designated location in the computing component. The BMC compares the two to see if they are the same. If they are not the same, it is determined that the current computing component has been updated.
[0038] In a possible implementation of this aspect, it also includes: when the computing mainboard is replaced, updating the computing mainboard backup identity information stored in the second memory according to the identity information of the computing mainboard; when the computing component is updated, updating the computing component backup identity information in the second memory according to the identity information of the computing component.
[0039] In a possible implementation of this aspect, the method further includes: when the configuration information of the computing mainboard is changed, updating the computing mainboard backup configuration information stored in the second memory according to the changed configuration information;
[0040] When the configuration information of the computing component is changed, the computing component backup configuration information stored in the second memory is updated according to the changed configuration information.
[0041] In one possible implementation of the present aspect, the storage component has multiple disks, and the multiple disks are combined into a RAID (Redundant Arrays of Independent Disks, an array of independent disks with redundancy, abbreviated as RAID) through software. The disks store RAID configuration information, and the processor system reads the RAID configuration information from the disks and configures the disks according to the RAID configuration information. The RAID configuration information includes RAID levels, such as RAID0 to RAID50.
[0042] In a possible implementation of this aspect, the storage component has multiple disks, and the multiple disks form a RAID through hardware. The storage component also includes a RAID card. The processor system reads RAID configuration information from the RAID card and configures the disks according to the RAID configuration information.
[0043] In a third aspect, this embodiment provides a business system including a management server and at least one business server. The management server includes a first memory, which stores a driver package of a computing component included in each business server. Each business server includes a storage component and a computing component. The computing component includes a BMC, a processor system, a computing motherboard, a computing component, and a first connector. The BMC, the processor system, the computing component, and the first connector are arranged on the computing motherboard. The processor system is respectively connected to the BMC, the first memory, the computing component, and the first connector. The storage component includes a second memory, a storage motherboard, a disk, and a second connector. The second memory, the disk, and the second connector are arranged on the storage motherboard. The second connector is respectively connected to the disk and the second connector. The first connector and the second connector match each other. When the first connector and the second connector are docked, the BMC is connected to the second memory through the first connector and the second connector, and the processor system is connected to the disk through the first connector and the second connector. For each business server, the functions of each component in the business server are as follows:
[0044] The BMC of the business server is configured to determine whether the computing component has been updated, and if so, identify whether the computing component supports out-of-band configuration. If so, obtain computing component backup configuration information from the second memory, configure the computing component according to the computing component backup configuration information, set the first memory as the first boot device, and restart the computing component.
[0045] a processor system of the business server, configured to restart from the first memory, read the driver package of the computing component from the first memory after restarting, and copy the driver package of the computing component to the disk of the business server;
[0046] The BMC of the business server is further used to set the disk as the first boot device and restart the server after the copy is completed.
[0047] The processor system of the service server is further configured to restart from the disk, and after restarting, drive the network card according to the driver package of the network card stored in the disk. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0049] Figure 1 This is a schematic diagram of the structure of a server disclosed in an embodiment of the present invention;
[0050] Figure 2 A schematic diagram of the structure of another server disclosed in an embodiment of the present invention;
[0051] Figure 3a A schematic diagram of the structure and flow of a business system disclosed in an embodiment of the present invention;
[0052] Figure 3b A schematic diagram of the structure of another server disclosed in an embodiment of the present invention;
[0053] Figure 3c A schematic diagram of the structure of a server disclosed in an embodiment of the present invention;
[0054] Figure 4 This is a flow chart of a server management method disclosed in an embodiment of the present invention;
[0055] Figure 5 A flow chart of another server management method disclosed in an embodiment of the present invention;
[0056] Figure 6 This is a flow chart of another server management method disclosed in an embodiment of the present invention. DETAILED DESCRIPTION
[0057] The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.
[0058] An embodiment of the present invention discloses a server management method, a server and a business system. The server includes a computing component and a storage component. The computing component and the storage component are connected in a detachable manner. The computing component includes components with a short maintenance cycle, and the storage component includes components with a long maintenance cycle. In this way, the computing component or the storage component can be flexibly replaced when maintaining the server, and the maintenance efficiency is high.
[0059] See Figure 1 , Figure 1It is a structural diagram of a server provided by an embodiment of the present invention. In an embodiment of the present invention, the server includes: a computing component and a storage component. The computing component includes a computing motherboard 10, a first memory 11, a processor system 12, a computing component 13, a baseboard controller 14 and a first connector 15. The first memory 11 is a non-volatile memory. The first memory 11 can be ROM, EPROM, EEPROM or Flash Memory. The first memory 11 stores a driver package of the computing component 13, and the driver package of the computing component 13 is used to drive the computing component 13; the processor system 12 is a processor network composed of one or more processors, each processor can be a single-core processor or a multi-core processor. In the case where the processor system 12 is a plurality of processors, the processors communicate with each other through an interconnection channel, and the interconnection channel can be QPI (QuickPath Interconnect, abbreviated as QPI) or Hyper Transport (HyperTransport); the computing component 13 is a component that performs specified computing tasks, including but not limited to packet processing, image processing, audio processing, or video processing. The computing component 13 includes but is not limited to a network card, a sound card, a MEZZ card, or a graphics processing card. The first memory 11, the processor system 12, the computing component 13, the baseboard management controller 14, and the first connector 15 are arranged on the computing motherboard 10, wherein the first connector 15 is arranged on the computing motherboard 10. The processor system 12 is connected to the first memory 11, the baseboard management controller 14, the computing component 13, and the first connector 15 respectively, and the baseboard controller 14 is connected to the first connector 15 and the first memory 11 at the same time.
[0060] The storage component includes a second connector 16, a storage motherboard 17, a disk 18, and a second memory 19. The second connector 16, disk 18, and second memory 19 are arranged on the storage motherboard 17. The second connector 16 and the first connector 15 match each other. The second memory 19 stores computing component backup configuration information, which is used to restore the configuration parameters of the computing component 13. The specific structures of the first connector 15 and the second connector 16 can be: the first connector 15 is a USB male connector and the second connector 16 is a USB female connector; or the first connector 15 is a USB male connector and the second connector 16 is a USB female connector; or the first connector 15 is a PCIE male connector and the second connector 16 is a PCIE female connector; or the first connector 15 is a PCIE female connector and the second connector 16 is a PCIE male connector. The number of disks 18 can be one or more. In the case of multiple disks 18, the disk types of disks 18 can be all HDDs, all SSDs, or a mixture of HDDs and SSDs. This application does not impose any restrictions. Multiple disks can be combined into a RAID array through software. The second memory 19 is a non-volatile memory. The second memory 19 can be a ROM, EPROM, EEPROM, or Flash Memory. The second memory 19 stores the computing component backup configuration information. The computing motherboard 10 and the storage motherboard 17 are two independent motherboards. When the first connector 15 and the second connector 16 are docked, the processor system 12 is connected to the disks 18 through the first connector 15 and the second connector 16, and the baseboard controller 14 is connected to the second memory 19 through the first connector 15 and the second connector 16. See Figure 1 In the illustrated embodiment, the processor system 12 is connected to the P1 pin of the first connector 15, the baseboard controller 14 is connected to the P2 pin of the first connector 15, the disk 18 is connected to the P1 pin of the second connector 16, and the second memory 19 is connected to the P2 pin of the second connector 16. When the first connector 15 and the second connector 16 are docked, the P1 pin of the first connector 15 is connected to the P1 pin of the second connector 16, and the P2 pin of the first connector 15 is connected to the P2 pin of the second connector 16.
[0061] In one possible implementation, the functions of the components in the server are as follows:
[0062] The baseboard management controller 14 is used to restore the configuration parameters of the computing component 13 according to the computing component backup configuration information stored in the second memory 19 when the computing component 13 is replaced;
[0063] The processor system 12 is used to copy the driver package of the computing component 13 stored in the first memory 11 to the disk 18, and start from the disk 18 after the copying is completed.
[0064] In one possible implementation, the functions of the various components in the server are:
[0065] The processor system 12 is used to restore the configuration parameters of the computing component 13 according to the computing component backup configuration information stored in the second memory 19 when the computing component 13 is replaced, copy the driver package of the computing component 13 stored in the first memory 11 to the disk 18, and start from the disk 18 after the copying is completed.
[0066] In a possible implementation, the processor system 12 executes the copying of the driver package of the computing component 13 stored in the preset first memory 11 to the disk 18, including:
[0067] The system starts up from the first memory 11 , reads the driver package of the computing component 13 from the first memory 11 after starting up, and copies the driver package of the computing component 13 to the disk 18 .
[0068] In one possible implementation, the functions of the various components in the server are:
[0069] The baseboard management controller 14 is used to determine whether the computing component 13 has been updated. If so, it identifies whether the computing component 13 supports out-of-band configuration. If so, the baseboard management controller 14 reads the computing component backup configuration information from the second memory 19, configures the computing component 13 according to the computing component backup configuration information, sets the first memory 11 as the first boot device, and restarts. Among them, out-of-band configuration refers to a control method in which the management channel and the business channel are separated. The management channel is used to transmit control data, and the business channel is used to transmit business data. The baseboard management controller 14 can obtain the identity information of the computing component 13 based on the mapping relationship between the pre-stored identity information and the status flag bit indicating whether out-of-band configuration is supported, and query the associated status flag bit based on the mapping relationship to determine whether the computing component 13 supports out-of-band configuration.
[0070] In a possible implementation, the baseboard management controller 14 may first determine whether the computing motherboard 10 is replaced before determining whether the computing component 13 is replaced. If the computing motherboard 10 is updated, it may then continue to determine whether the computing component 13 is replaced.
[0071] Processor system 12 is configured to restart from first memory 11, read the driver package of computing component 13 stored in first memory 11 after restarting, copy the driver package of computing component 13 to disk 18, and notify baseboard management controller 14 after the copying is complete. A streamlined operating system is stored in the first memory, retaining only basic functions. The streamlined operating system is characterized by its small size and low resource consumption. After restarting from first memory 11, processor system 12 runs the operating system, and the copy of the driver package of computing component 13 is executed in the operating system environment.
[0072] The baseboard management controller 14 is further configured to set the disk 18 as the first boot device and restart the computer 13 when the driver package copying and configuration of the computer 13 are completed.
[0073] Processor system 12 is further configured to boot from disk 18, search disk 18 for a driver package for computing component 13, and drive computing component 13 based on the driver package. In this embodiment, the storage component has not been replaced, and disk 18 stores the server's operating system. After rebooting from disk 18, processor 12 runs the server's operating system. While the operating system is running, processor 12 searches disk 18 for a driver package for computing component 13 and drives computing component 13 based on the driver package.
[0074] In a possible implementation of this embodiment, the baseboard management controller 14 determines whether the computing component 13 has been replaced. If so, it reads the computing component backup configuration information from the second memory 19, configures the computing component 13 according to the computing component backup configuration information, sets the first memory 11 as the first startup device, and restarts.
[0075] The processor system 12 is used to start from the first memory 11, read the driver package of the computing component 13 from the first memory 11 after startup, copy the driver package of the computing component 13 to the disk 18, and read the computing component backup configuration information from the first memory 11, and configure the computing component 13 according to the computing component backup configuration information.
[0076] The baseboard management controller 14 is further configured to set the disk 18 as the first boot device and restart the computer after the copying and configuration are completed.
[0077] The processor system 12 is further configured to boot from the disk 18 and, after booting, drive the computing component 13 according to the driver package of the computing component 13 stored on the disk. In some embodiments of this embodiment, after determining that the computing component 13 has been replaced, the baseboard management controller 14 may further identify whether the computing component 13 supports out-of-band configuration. If out-of-band configuration is not supported, subsequent steps are performed.
[0078] In a possible implementation of this embodiment, the second memory 19 also stores computing component backup identity information and computing mainboard backup identity information. The baseboard management controller 14 is used to obtain the identity information of the computing mainboard 10, and read the computing mainboard backup identity information from the second memory 19, compare the identity information of the computing mainboard 10 with the read computing mainboard backup identity information to see if they are the same, and if they are not the same, determine that the computing mainboard 10 has been replaced and the computing mainboard 10 is the replaced computing mainboard. The baseboard management controller 14 obtains the identity information of the computing component 13, and obtains the computing component backup identity information from the second memory 19, compares the identity information of the computing component 13 with the read computing component backup identity information to see if they are the same, the identity information includes but is not limited to one or more of model information, serial number, and MAC address, and if they are not the same, determine that the computing component 13 has been updated and the computing component 13 is the updated computing component.
[0079] In a possible implementation of this embodiment, the baseboard management controller 14 is further configured to update the computing motherboard backup identity information stored in the second memory 19 according to the identity information of the computing motherboard 10 when the computing motherboard 10 is replaced;
[0080] When the computing component is updated, the computing component backup identity information stored in the second memory 19 is updated according to the identity information of the computing component 13 .
[0081] In a possible implementation, the baseboard management controller 14 is further configured to update the computing component backup configuration information stored in the second memory 19 according to the changed configuration information when the configuration information of the computing component 13 is changed.
[0082] In one possible implementation, disk 18 also stores RAID configuration information. Processor system 12 is further configured to read the RAID configuration information stored on disk 18, where the RAID configuration information indicates the RAID mode of multiple disks in disk 18, and configure disk 18 according to the RAID configuration information. If the RAID configuration information changes, the RAID configuration information stored on disk 18 is updated according to the changed RAID configuration information.
[0083] In one possible implementation, the second memory 19 also stores computing motherboard backup configuration information. If the computing motherboard 10 supports out-of-band configuration, the baseboard management controller 14 reads the computing motherboard backup configuration information from the second memory 19 and configures the computing motherboard 10 according to the computing motherboard backup configuration information. If the configuration information of the computing motherboard 10 changes, the baseboard management controller 14 updates the computing motherboard backup configuration information stored in the second memory 19 according to the changed configuration information.
[0084] In one possible implementation, the baseboard management controller 14 determines that the computing motherboard 10 has been replaced, but the computing component 13 has not been updated, and the identity information of the computing component 13 remains unchanged. In this case, it is only necessary to restore the configuration environment of the computing component 13, and there is no need to update the driver package of the computing component 13. In the case where the computing component 13 supports out-of-band configuration, the baseboard management controller 14 obtains the computing component backup configuration information from the second memory 19, and the processor system 12 starts from the disk 18 by default. After startup, since the storage component has not been replaced, the disk 18 stores the operating system and the driver package of the computing component before replacement. The identity information of the computing component before replacement and the computing component 13 are the same, and the processor system can directly use the driver package of the computing component before replacement to drive the computing component 13.
[0085] When the computing component 13 does not support out-of-band configuration, the baseboard management controller 14 reads the computing component backup configuration information from the second memory 19, and stores the read computing component backup configuration information in the first memory 11. The processor system 12 obtains the computing component backup configuration information from the first memory 11, and configures the computing component 13 according to the computing component backup configuration information. The processor system 12 starts directly from the disk 18. After startup, the processor system 12 reads the driver package of the computing component 13 from the disk 18 to drive the computing component 13.
[0086] In the above embodiment, when the computing component of the server is replaced, the computing component is driven according to the driver package in the first memory, and the server is configured according to the backup configuration information pre-stored in the second memory, so that the server automatically returns to normal state without the need to reinstall the operating system, and the configuration information is quickly restored, thereby realizing plug-and-play of the computing component, reducing the maintenance time and difficulty of server upgrades.
[0087] See also Figure 2 , is another structural diagram of a server provided by an embodiment of the invention, the server includes a computing component and a storage component, Figure 2 Servers and Figure 1The server structure differs only in that the storage component includes a RAID card 20, which is mounted on the storage motherboard 17. Multiple disks 18 are arranged into a RAID array via the RAID card 20. The processor system 12 is connected to the RAID card 20 via a first connector 15 and a second connector 16. If the computing motherboard 10 is replaced, the processor system 12 reads the RAID configuration information from the RAID card 20 and configures the disks 18 accordingly.
[0088] See also Figure 3a , is a schematic diagram of the structure of a business system provided by an embodiment of the present invention. In this embodiment of the present invention, the business system includes a management server 300 and multiple business servers 301~30n. The management server 300 is responsible for managing the multiple business servers. The management server 300 includes a first memory, which stores the driver package of the computing component included in each of the business servers 300~30n. For duplicate driver packages, only one driver package can be retained to reduce storage space usage. The structure of each business server is as follows: Figure 3b and Figure 3c It should be noted that the structure of multiple business servers can all be Figure 3b , or all of Figure 3c or a combination of the two, which is not limited in this embodiment.
[0089] See also Figure 3b , is a schematic diagram of the structure of a business server in a business system. In an embodiment of the present invention, the business server includes a computing component and a storage component. The computing component includes a computing motherboard 30, a processor system 31, a computing component 32, a baseboard management controller 33, and a first connector 34. The processor system 31, computing component 32, baseboard management controller 33, and first connector 34 are disposed on the computing motherboard 30. The processor system 31 comprises a processor network consisting of one or more processors. Each processor can be a single-core processor or a multi-core processor. When the processor system 31 comprises multiple processors, the processors communicate with each other via interconnection channels, which can be QPI or HT. The computing component 32 is a component that performs designated computing tasks and includes, but is not limited to, a network card, a sound card, and a graphics processing card. The computing motherboard 30 is the carrier of various computing components. The processor system 31, computing component 32, baseboard management controller 33, and first connector 34 are disposed on the computing motherboard 30, wherein the first connector 34 is disposed on the computing motherboard 30. The processor system is connected to the computing component 32, the first memory in the management server, the baseboard management controller 33, and the first connector, and the baseboard management controller 33 is connected to the first memory.
[0090] The storage assembly includes a second connector 35, a storage motherboard 36, disks 37, and a second memory 38. The storage motherboard 36 serves as a carrier for the various components in the storage assembly. The second connector 35, disks 37, and second memory 38 are located on the storage motherboard 36. The storage motherboard 36 and the computing motherboard 30 are two independent motherboards, and the second connector 35 and the first connector 34 mate with each other. There can be one or more disks 37. If there are multiple disks 37, they can all be HDDs, all SSDs, or a mix of HDDs and SSDs. This is not a limitation in this embodiment, and multiple disks can be configured into a RAID array using software. The second memory 38 is a non-volatile memory, including but not limited to ROM, EPROM, EEPROM, or Flash Memory. It stores backup configuration information for the computing components. The computing motherboard 30 and the storage motherboard 36 are two independent motherboards. When the first connector 34 and the first connector 35 are mated, the processor system 31 connects to the disks 37 via the first connector 34 and the second connector 35, and the baseboard management controller connects to the first memory 38 via the first connector 34 and the second connector 35. For specific connection methods, see Figure 1 The description is not repeated here.
[0091] In one possible implementation, the functions of the components in the server are:
[0092] The baseboard management controller 33 is used to restore the configuration parameters of the computing component 32 according to the computing component backup configuration information stored in the second memory 38 when the computing component 30 is replaced;
[0093] The processor system 31 is used to copy the driver package of the computing component 32 stored in the first memory of the management server 300 to the disk 37, and start it from the disk 37 after the copying is completed.
[0094] In one possible implementation, the functions of the various components in the server are:
[0095] The processor system 31 is used to restore the configuration parameters of the computing component 32 according to the computing component backup configuration information stored in the second memory 38 when the computing component 32 is replaced, copy the driver package of the computing component 32 stored in the first memory of the management server 300 to the disk 37, and start from the disk 37 after the copy is completed.
[0096] In a possible implementation, the processor system 31 executes the copying of the driver package of the computing component 32 stored in the preset first memory to the disk 37, including:
[0097] The system is started from the first memory 300 of the management server, and after starting, the driver package of the computing component 32 is read from the first memory, and the driver package of the computing component 32 is copied to the disk 37 .
[0098] In one possible embodiment, the functions of the various components in the business server are as follows: Baseboard management controller 33 is used to determine whether computing component 32 has been updated. If so, it identifies whether computing component 13 supports out-of-band configuration. If so, baseboard management controller 33 obtains computing component backup configuration information from the second memory and restores the configuration environment of computing component 32 based on the computing component backup configuration information. Baseboard management controller 32 sets the first memory in the associated management server 300 as the first boot device and restarts. In one possible embodiment, before determining whether computing component 32 has been replaced, baseboard management controller 33 may also determine whether computing motherboard 30 has been replaced. If so, it then determines whether computing component 32 has been replaced.
[0099] The processor system 31 is configured to restart from the first memory of the management server, read the driver package of the computing component 32 stored in the first memory after restarting, copy the driver package of the computing component 32 to the disk 18, and notify the baseboard management controller 33 after the copying is completed;
[0100] The baseboard management controller 33 is further configured to set the disk 37 as the first boot device and restart the computer 32 after the driver package copying is completed.
[0101] The processor system 31 is further configured to restart from the disk 37 , search for a driver package of the computing component 32 stored in the disk 37 after restarting, and drive the computing component 32 according to the driver package.
[0102] In a possible implementation of this embodiment, the baseboard management controller 33 determines whether the computing component 32 has been replaced, reads the computing component backup configuration information from the second memory 38, configures the computing component 32 according to the computing component backup configuration information, sets the first memory as the first boot device, and restarts.
[0103] The processor system 31 is used to restart from the first memory, read the driver package of the computing component 13 from the first memory after restarting, copy the driver package of the computing component 32 to the disk 37, and read the computing component backup configuration information from the first memory, and configure the computing component 32 according to the computing component backup configuration information.
[0104] The baseboard management controller 33 is further configured to set the disk 37 as the first boot device and restart the computer after the copying and configuration are completed.
[0105] The processor system 31 is further configured to restart from the disk 37 and, after restarting, drive the computing component 32 according to the driver package of the computing component 32 stored in the disk 37. In some embodiments of this embodiment, after determining that the computing component 32 has been sent for replacement, the baseboard management controller 33 may further identify whether the computing component 32 supports out-of-band configuration, and if not, execute subsequent steps.
[0106] In one possible implementation of this embodiment, the second memory 38 further stores computing component backup identity information and computing mainboard backup identity information. The baseboard management controller 33 is configured to obtain the identity information of the computing mainboard 30, read the computing mainboard backup identity information from the second memory 38, compare the identity information of the computing mainboard 30 with the read computing mainboard backup identity information to see if they are the same, and if not, determine that the computing mainboard 30 has been replaced and that the computing mainboard 30 is the replaced computing mainboard. The baseboard management controller 33 obtains the identity information of the computing component 32, obtains the computing component backup identity information from the second memory 38, compares the identity information of the computing component 32 with the read computing component backup identity information to see if they are the same, and if not, determine that the computing component 32 has been updated and that the computing component 32 is the updated computing component.
[0107] In a possible implementation of this embodiment, the baseboard management controller 33 is further configured to update the computing motherboard backup identity information stored in the second memory 38 according to the identity information of the computing motherboard 30 when the computing motherboard 30 is replaced;
[0108] When the computing component is updated, the computing component backup identity information stored in the second memory 38 is updated according to the identity information of the computing component 32 .
[0109] In a possible implementation, the baseboard management controller 33 is further configured to update the computing component backup configuration information stored in the second memory 32 according to the changed configuration information when the configuration information of the computing component 32 is changed.
[0110] In one possible implementation, disk 37 further stores RAID configuration information. Processor system 31 is further configured to read the RAID configuration information stored in disk 37, where the RAID configuration information indicates the RAID mode of multiple disks in disk 37, and configure disk 37 according to the RAID configuration information. If the RAID configuration information changes, the RAID configuration information stored in second memory 38 is updated according to the changed RAID configuration information.
[0111] In one possible implementation, the second memory 38 also stores computing motherboard backup configuration information. If the computing motherboard 30 supports out-of-band configuration, the baseboard management controller 33 reads the computing motherboard backup configuration information from the second memory 38 and configures the computing motherboard 30 based on the computing motherboard backup configuration information. If the configuration information of the computing motherboard 30 changes, the baseboard management controller 33 or the processor system 31 updates the computing motherboard backup configuration information stored in the second memory 38 based on the changed configuration information.
[0112] See also Figure 3c , is another structural diagram of a business server in a business system. In the structure and Figure 3b The only difference is that the storage component also includes a RAID card 39, which is mounted on the storage motherboard 36 and connected to the disks 37. The processor system is connected to the RAID card 39 via a first connector 34 and a second connector 35. There are multiple disks 37, which are combined into a hardware RAID array using the RAID card 39. If the computing motherboard 30 is replaced, the processor system reads the RAID configuration information stored in the RAID card and configures the disks 37 accordingly.
[0113] In the above embodiment, when a computing component of a server is replaced, the computing component is driven according to the driver package in the first memory, and the server is configured according to the backup configuration information pre-stored in the second memory, so that the server automatically returns to normal status without reinstalling the operating system and quickly restoring the configuration information, achieving plug-and-play of the computing component, reducing the maintenance time and difficulty of server upgrades. At the same time, the driver packages of the computing components in each server are centrally stored in the first memory, which can effectively reduce the storage space occupied by the driver packages.
[0114] See also Figure 4 , is a flow chart of a server management method provided in an embodiment of the present invention. In an embodiment of the present invention, the method includes:
[0115] S401. When a computing component is replaced, restore the configuration parameters of the computing component according to the computing backup configuration information stored in the second memory.
[0116] Specifically, the server includes a computing component and a storage component. The computing component is mainly used to perform computing functions, such as data packet processing, image processing, sound processing, etc. The storage component is mainly used to store large-capacity data. The storage component includes a disk. The number of disks can be one or more. Multiple disks can form a RAID. RAID can be implemented by hardware or software. If implemented by hardware, the storage component also includes a RAID card. The computing component and the storage component are connected in a detachable manner to facilitate plugging and unplugging between the computing component and the storage component. When the computing component or storage component needs to be upgraded or fails, the entire computing component or storage component can be replaced. Backup configuration information is pre-stored in the second memory. The backup configuration information includes but is not limited to backup network card configuration information, backup graphics card configuration information, and backup RAID level information. The server reads the backup configuration information pre-stored in the second memory and configures the server according to the backup configuration information.
[0117] S402: Drive the computing component according to the driver package of the computing component stored in the preset first memory.
[0118] Specifically, if it is determined that the current computing component has been replaced, the server obtains a driver package for the computing component. If the computing component includes a network card, the driver package for the computing component includes the driver package for the network card. The driver package for the computing component may be stored in a first memory, which may be located in the computing component or in a management server external to the server. The management server is used to centrally manage each server.
[0119] When the above embodiment is implemented, when the computing component of the server is replaced, the computing component is driven according to the driver package in the first memory, and the server is configured according to the backup configuration information pre-stored in the second memory, so that the server automatically returns to normal state without the need to reinstall the operating system, and the configuration information is quickly restored, thereby realizing plug-and-play of the computing component, reducing the maintenance time and difficulty of server upgrades.
[0120] See also Figure 5 , is another flow chart of a server management method provided by an embodiment of the present invention. In the embodiment of the present invention, the method includes:
[0121] S501: The BMC determines that a computing component is updated and the computing component supports out-of-band configuration.
[0122] Specifically, the server of this embodiment includes a computing component and a storage component. The computing component includes a BMC, a processor system, a computing component, a computing motherboard and a first connector. The BMC is a controller that monitors and manages various hardware on the server. The BMC can use IPMI (Intelligent Platform Management Interface), UART (Universal Asynchronous Receiver / Transmitter) interface or other types of interfaces to communicate with various hardware in an out-of-band manner. The processor system is a processor network composed of one or more processors. Each processor can be a single-core processor or a multi-core processor. Any two processors can communicate using QPI channels and HT channels. The computing component is hardware that performs a certain specified task, such as performing data packet processing, sound processing or graphics processing tasks. The computing component can be a network card, sound card, MEZZ card, graphics card or other types of components. The first connector is a connector with plug-in and unplugging functions. The first connector and the second connector match each other so that the computing component and the storage component can be connected in a detachable manner. In the computing component, the processor system is respectively connected to the BMC, the computing component and the first connector, wherein the connection bus includes but is not limited to a PCIE bus, a USB bus, a SATA bus or a SAS (Serial Attached Smallcomputer System Interface, SAS for short) bus, and the processor system, the BMC, the computing component and the first connector can all be set on the computing motherboard, and the first connector is set on the computing motherboard.
[0123] The storage component includes a disk, a second memory, and a second connector. The number of disks can be one or more, and the types of disks include, but are not limited to, SSDs or HDDs. Multiple disks can be configured into a RAID array using software. The second memory is a non-volatile memory that maintains data after a power outage. The second memory includes, but is not limited to, ROM, EPROM, EEPROM, or Flash Memory. The second memory stores backup configuration information for the computing component. It should be noted that the second memory also stores a streamlined operating system that retains only basic functions, is small in size, and consumes few resources. The operating system includes, but is not limited to, the Linux operating system, the Unix operating system, or the Windows PE operating system. The second connector is a connector with pluggable functionality, and the first connector and the second connector cooperate with each other. In one possible embodiment, the first connector and the second connector can be USB connectors, SATA connectors, PCIE connectors, or other types of connectors, which are not limited in this embodiment. When the first connector and the second connector are docked, the corresponding pins of the first connector and the second connector will touch and connect. The BMC is connected to the second memory via the first connector and the second connector, and the processor system is connected to the disk via the first connector and the second connector.
[0124] The BMC determines whether the computing motherboard has been replaced. If so, it continues to determine whether the computing component has been updated. If so, it continues to determine whether the computing component supports out-of-band configuration. Out-of-band configuration means configuration is performed through channels other than the service channel without starting the operating system. If so, execution proceeds to S502.
[0125] In one possible implementation, the BMC determines whether the computing component has been replaced. If so, it reads the computing component backup configuration information stored in the second memory, stores the computing component backup configuration information in the first memory, sets the first memory as the first boot device, and restarts. The processor system restarts from the first memory, which stores the driver package, streamlined operating system, and management software of the computing component. After restarting, the management software on the streamlined operating system runs. The management software reads the driver package of the computing component from the first memory, copies the driver package of the computing component to the disk, and reads the computing component backup configuration information from the first memory, and configures the computing component according to the computing component backup configuration information. When the copying and configuration are completed, the BMC sets the disk as the first boot device and restarts. The processor system restarts from the disk, and after restarting, drives the computing component according to the driver package of the computing component stored in the disk.
[0126] S502: The BMC reads computing component backup configuration information from the second memory, and configures the computing component according to the computing component backup configuration information.
[0127] Specifically, the computing component backup configuration information represents the configuration information of the computing component in the server before the computing motherboard is replaced. For example, if the computing component is a network card, the computing component backup configuration information includes, but is not limited to, one or more of a MAC address, an IP address, a WWN, a WWPN, and a UUID. The BMC reads the computing component backup configuration information from the second memory and restores the original configuration on the computing component based on the computing component backup configuration information.
[0128] S503: The BMC sets the first memory as the first boot device and restarts.
[0129] Specifically, the first memory is a non-volatile memory, including but not limited to ROM, EPROM, EEPROM or Flash Memory. In one possible implementation, the first memory may be located in a storage component, and the processor system is connected to the first memory. In another possible implementation, the first memory is located outside the server, inside a management server managed by the server, and the processor system is connected to the first memory. The BMC sets the first memory as the first boot device (First Boot Device), and the first boot device indicates the storage device that the server first accesses when it starts.
[0130] S504: The processor system restarts from the first memory, reads the driver package of the computing component from the first memory after restarting, and copies the driver package of the computing component to the disk.
[0131] Specifically, the processor system restarts from the first memory, which stores a streamlined operating system and management software. The processor system runs the management software in the streamlined operating system. In the operating system environment, the management software reads the driver package of the computing component from the first memory, copies the driver package of the computing component to the disk, and after completing the copy operation of the driver package of the computing component, the processor system sends a copy completion message to the BMC.
[0132] S505. When the copy is completed, the BMC sets the disk as the first boot device and restarts.
[0133] Specifically, when the BMC determines that the driver package of the computing component is copied, it sets the disk as the first boot device. If there are multiple disks, the disk of the active partition is set as the first boot device, and restarts after the setting is completed.
[0134] S506: The processor system restarts from the disk, and after restarting, drives the computing component according to the driver package of the computing component stored in the disk.
[0135] Specifically, in an embodiment of the present invention, the storage component has not been replaced, so the operating system of the server is stored in the disk. After the processor restarts from the disk, it runs the operating system in the disk. At this time, the operating system is an operating system with complete functions installed on the server. The processor system searches for the driver package of the computing component from the disk and drives the computing component according to the driver package of the computing component.
[0136] In the above embodiment, when the computing component of the server is replaced, the computing component is driven according to the driver package in the first memory, and the server is configured according to the backup configuration information pre-stored in the second memory, so that the server automatically returns to normal state without the need to reinstall the operating system, and the configuration information is quickly restored, thereby realizing plug-and-play of the computing component, reducing the maintenance time and difficulty of server upgrades.
[0137] See also Figure 6 , is another flow chart of a server management method provided by an embodiment of the present invention. In the embodiment of the present invention, the method includes:
[0138] S601, BMC is powered on and started.
[0139] Specifically, the BMC is used to monitor and manage various hardware on the server. The BMC can communicate with the various hardware in an out-of-band manner using the IPMI interface, without using an operating system. After the BMC is powered on and started, S602 is executed.
[0140] S602: The BMC determines whether the computing mainboard has been replaced.
[0141] Specifically, the server of this embodiment includes a computing component and a storage component, and the computing component includes a BMC, a processor system, a computing part, a computing motherboard and a first connector. The processor system is a processor network composed of one or more processors, each processor can be a single-core processor or multiple processors, and any two processors can communicate through a QPI channel or an HT channel. The computing part is hardware that performs a certain specified task, and the computing part includes but is not limited to a network card, a MEZZ card, a sound card, a graphics card or other types of components. The first connector is a connector with a plug-in function, and the first connector and the second connector match each other so that the computing component and the storage component can be connected in a detachable manner. In the computing component, the processor system includes the BMC and the computing part respectively, and the processor system is connected to the BMC, the computing part and the first connector respectively. The connection bus includes but is not limited to a PCIE bus, a USB bus, a SATA bus or a SAS bus. The computing motherboard is the carrier of each component in the computing component. The processor system, the BMC, the computing part and the first connector can all be set on the computing motherboard, and the first connector is set on the computing motherboard.
[0142] The storage component includes a disk, a second memory, and a second connector. The number of disks can be one or more, and the types of disks include, but are not limited to, SSDs or HDDs. Multiple disks can be configured into a RAID array using software. The second memory is a non-volatile memory, including, but not limited to, ROM, PROM, EPROM, EEPROM, or Flash Memory. The second memory stores computing motherboard backup identity information, computing component backup identity information, and computing component backup configuration information. It should be noted that the second memory also stores a streamlined operating system that retains only basic functions, is compact, and consumes few resources. Such operating systems include, but are not limited to, Linux, Unix, Windows PE, or other operating systems. The second connector is a plug-and-unplug connector that mates with the first connector. For example, the first and second connectors can be USB, SATA, PCIE, or other types of connectors, although this embodiment does not limit this. When the first and second connectors are docked, corresponding pins of the first and second connectors touch and connect. The BMC is connected to the second memory via the first and second connectors, and the processor system is connected to the disk via the first and second connectors.
[0143] The method for the BMC to determine whether the computing mainboard has been replaced may be: the BMC obtains the identity information of the computing mainboard and obtains the backup identity information of the computing mainboard from the second memory, and compares whether the identity information of the computing mainboard and the backup identity information of the computing mainboard are the same; if they are the same, it is determined that the computing mainboard has not been replaced, and S603 is executed; if they are not the same, it is determined that the computing mainboard has been replaced, and S604 is executed.
[0144] It should be noted that when the BMC determines that the computing motherboard has been replaced, it updates the computing motherboard backup identity information stored in the second memory according to the computing motherboard identity information, and overwrites the computing motherboard backup identity information stored in the second memory with the computing motherboard identity information.
[0145] S603: The processor system starts from the disk.
[0146] Specifically, the computing motherboard has not been replaced. The computing motherboard is the carrier of each component in the computing assembly. The server's default first boot device is the disk. The processor system boots from the disk. The disk stores the operating system and configuration information of each component. The server runs normally.
[0147] S604: The BMC determines whether the computing component is updated.
[0148] Specifically, the BMC may determine whether a computing component has been updated by: obtaining the identity information of the computing component and obtaining backup identity information of the computing component from the second memory, comparing the identity information of the computing component with the identity information of the computing component; if they are different, determining that the computing component has been updated and executing S608; if they are the same, determining that the computing component has not been updated and executing S605.
[0149] It should be noted that when the BMC determines that a computing component has been updated, it obtains the identity information of the computing component, updates the computing component backup identity information stored in the second memory according to the identity information of the computing component, and overwrites the computing component backup identity information stored in the second memory with the computing component identity information.
[0150] S605: The BMC identifies whether the computing component supports out-of-band configuration.
[0151] Specifically, the computing component pre-stores a mapping relationship between identity information and status flags. The identity information represents the identity information of the computing component, and the status flag represents whether out-of-band configuration is supported. For example, 0 indicates that out-of-band configuration is not supported, and 1 indicates that out-of-band configuration is supported. The computing component obtains the identity information of the computing component and queries the corresponding status flag based on the mapping relationship to determine whether out-of-band configuration is supported. If yes, execute S606; otherwise, execute S605.
[0152] S606: The BMC reads the computing component backup configuration information from the second memory, and configures the computing component according to the computing component backup configuration information.
[0153] Specifically, the BMC reads the computing component backup configuration information from the second memory and configures the computing component according to the computing component backup configuration information. The BMC can detect whether the first boot device is a disk. If not, the first boot device is set to a disk. If yes, the process proceeds to S607.
[0154] S607: The processor system boots from the disk.
[0155] Specifically, the storage component has not been updated, the operating system is stored in the disk, and the processor system runs the operating system after starting from the disk. Since the identity information of the computing component has not changed, the driver package of the computing component before replacement stored in the disk is still valid, and the processor system drives the computing component according to the original driver package.
[0156] S608: The BMC identifies whether the computing component supports out-of-band configuration.
[0157] The specific process is described in S605 and will not be repeated here. If yes, execute S609.
[0158] S609: The BMC reads the computing component backup configuration information from the second memory and configures the computing component according to the computing component backup configuration information. For example, if the computing component is a network card, the computing component backup configuration information includes but is not limited to one or more of a MAC address, an IP address, a UUID, a WWN, and a WWPN.
[0159] In a possible implementation, when the configuration information of the computing mainboard is changed, the BMC updates the computing mainboard backup configuration information stored in the second memory according to the changed configuration information;
[0160] When the configuration information of the computing component is changed, the BMC updates the computing component backup configuration information stored in the second memory according to the changed configuration information.
[0161] S610: The BMC sets the first storage device as the first boot device and restarts.
[0162] Specifically, the first memory may be located in a computing component of the server, or in a management server outside the server.
[0163] S611. The processor system reads the driver package of the computing component from the first memory, and copies the driver package of the computing component to the disk.
[0164] Specifically, the processor system restarts from the first memory, which stores a streamlined operating system. The processor system runs the streamlined operating system, reads the driver package of the computing component from the first memory in the operating system environment, copies the driver package of the computing component to the disk, and then completes the copy operation of the driver package of the computing component. The processor system sends a notification message to the BMC.
[0165] S612: The processor system notifies the BMC to complete the copy and configuration operations.
[0166] Specifically, after the processor system determines that the copy operation of the driver package of the computing component and the configuration operation of the BMC on the computing component are completed, the processor system sends a notification message to the BMC.
[0167] S613. The BMC sets the disk as the first boot device and restarts the computer.
[0168] Specifically, the BMC receives a notification message sent by the processor system to determine that the driver package of the computing component has been copied and the configuration of the computing component has been completed, and the disk is set as the first boot device. If there are multiple disks, the disk of the active partition is set as the first boot device. After the setting is completed, the system restarts.
[0169] S614: The processor system starts from the disk and drives the computing component according to the driver package of the computing component.
[0170] Specifically, in an embodiment of the present invention, the storage component has not been replaced, so the operating system of the server is stored in the disk. After the processor starts from the disk, it runs the operating system in the disk. At this time, the operating system is an operating system with complete functions installed on the server. The processor system searches for the driver package of the computing component from the disk and drives the computing component according to the driver package of the computing component.
[0171] S615. The BMC reads the computing component backup configuration information from the second memory, and stores the computing component backup configuration information in the first memory.
[0172] Specifically, when the judgment result of S5608 is no, the BMC reads the computing component backup configuration information from the second memory and stores the computing component backup configuration information in the first memory.
[0173] S616: The BMC sets the first storage device as the first boot device and restarts.
[0174] S617: The processor system reads the driver package and the computing component backup information of the computing component from the first memory, copies the driver package of the computing component to the disk, and configures the computing component according to the computing component backup configuration information.
[0175] Specifically, the processor system restarts from the first memory, which stores the driver package of the computing component, the streamlined operating system and the management software. After restarting, the management software on the streamlined operating system runs, and the management software reads the driver package of the computing component from the first memory, copies the driver package of the computing component to the disk, and reads the computing component backup configuration information from the first memory, and configures the computing component according to the computing component backup configuration information.
[0176] S618: The processor system notifies the BMC to complete the copy and configuration operations.
[0177] S619, BMC sets the disk as the first boot device and restarts.
[0178] S620: The processor system starts from the disk and drives the computing component according to the driver package of the computing component.
[0179] In a possible implementation, when the judgment result of S604 is yes, it is not necessary to identify whether the computing component supports out-of-band configuration, and S615 to S620 are directly executed.
[0180] In a possible implementation manner, the disk further stores RAID configuration information; and the method further includes:
[0181] The processor system reads the RAID configuration information from the disk and configures the disk according to the RAID configuration information.
[0182] Specifically, the number of disks in the storage component is multiple, and the multiple disks form a RAID through software. When the processor system determines that the computing motherboard has been replaced, it reads the RAID configuration information stored in the second memory and configures the disks according to the RAID configuration information.
[0183] In a possible implementation, the storage component further includes a RAID card, the RAID card is connected to the disk, and the RAID card is connected to the processor system via the second connector and the first connector;
[0184] The server management method further includes:
[0185] The processor system reads RAID configuration information from the RAID card and configures the disk according to the RAID configuration information.
[0186] Specifically, there are multiple disks in the storage component, and the multiple disks form a RAID through hardware. When the processor system determines that the computing motherboard has been replaced, it reads the RAID configuration information stored in the RAID card and configures the disks according to the RAID configuration information.
[0187] To sum up, when the computing components of the server are replaced, the computing components are driven according to the driver package in the first memory, and the server is configured according to the backup configuration information pre-stored in the second memory, so that the server automatically returns to normal state without the need to reinstall the operating system, and the configuration information is quickly restored, thereby realizing plug-and-play of the computing components, reducing the maintenance time and difficulty of server upgrades.
[0188] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.
[0189] The above embodiments merely disclose optional embodiments of the present invention and are not intended to limit the scope of the present invention. A person skilled in the art will appreciate that all or part of the processes of the above embodiments and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.
Claims
1. A server, characterized in that: The system comprises a computing component and a storage component, wherein the computing component and the storage component are detachably connected via a connector, and a maintenance period of the computing component is shorter than a maintenance period of the storage component; The storage component is used to back up configuration information of computing components in the computing component; The computing component is configured to restore configuration parameters of the computing component according to the configuration information of the computing component backed up by the storage component when any computing component in the computing components is updated; The computing component includes a baseboard management controller BMC, a processor system, a first memory and a first connector; the storage component includes a disk, a second memory and a second connector; The processor system is connected to the disk via a first connector and a second connector; The processor system is configured to copy the driver package of the computing component stored in the preset first memory to the disk when the computing component is replaced, and boot from the disk after the copying is completed; The baseboard management controller BMC is further configured to: update the computing mainboard backup identity information stored in the second memory according to the identity information of the computing mainboard when the computing mainboard is replaced; When any computing component in the computing assembly is replaced, the computing component backup identity information stored in the second memory is updated according to the identity information of the computing component.
2. The server according to claim 1, wherein: The computing component and the storage component are connected via the first connector and the second connector respectively, and the first connector and the second connector match each other.
3. The server according to any one of claims 1 to 2, characterized in that: An update of any computing component in the computing assembly includes component replacement; The BMC is used to determine whether any computing component in the computing assembly has been replaced; and when any computing component in the computing assembly has been replaced, restore the configuration parameters of the computing component according to the computing component backup configuration information stored in the second memory.
4. The server according to claim 3, wherein: The BMC is specifically used to set the first memory as the first startup device and perform a restart when it is determined that any computing component in the computing assembly has been replaced, wherein the first memory stores a driver package of the computing component, and the driver package of the computing component is used to start the computing component.
5. The server according to claim 3, wherein: The BMC is specifically used to obtain the identity information of the computing component and obtain the backup identity information of the computing component from the second memory; wherein the identity information includes but is not limited to one or more of model, serial number, and MAC address.
6. The server according to claim 5, wherein: The BMC is further configured to compare the identity information of the computing component and the backup identity information of the computing component to determine whether they are the same; and determine, based on the judgment result, that any computing component in the computing assembly has been replaced.
7. The server according to any one of claims 1 or 2, characterized in that: The storage component also includes a RAID card; The processor system is further configured to read RAID configuration information from the RAID card and configure the disk according to the RAID configuration information, wherein the RAID configuration information is used to identify RAID modes of multiple disks in the storage component.
Citation Information
Patent Citations
1U storage hot plugging achieving method
CN104199514A
Information processing apparatus, control method for the information processing apparatus
CN105868051A