Cabinet fault diagnosis system, method and device and storage medium

Abstract

The invention discloses a cabinet fault diagnosis system, method and device and a storage medium, relates to the field of servers, and aims to solve the problem of low efficiency of troubleshooting cable backboard faults one by one in a traditional electrical test. The first manager is a manager in a node included in the cabinet under different networking configurations, the first manager is configured to generate a routing table corresponding to each node of the cabinet and burn the routing table into the cable backboard, and in response to a received fault monitoring instruction, the corresponding second manager is accessed through the switching component according to the routing table, and the second manager is configured to perform fault monitoring according to the received fault monitoring instruction. Obtaining a fault diagnosis result; and the second manager is configured to obtain the current state of the communication link of the node corresponding to the second manager and update the link state column corresponding to the second manager in the routing table according to the current state of the communication link. According to the invention, local fault diagnosis can be carried out on the cable backboard under the condition that the remote management software is not deployed, and the fault diagnosis efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of servers, and in particular to a cabinet fault diagnosis system, method, device and storage medium. Background Art

[0002] With the rapid development of artificial intelligence technology, the demand for computing resources has shown an exponential growth. To meet this demand, AI (Artificial Intelligence) Rack cabinets came into being, which are characterized by higher integration, density and speed, as well as more complex interconnection networking structures. In the AI ​​Rack cabinet, GPU (Graphics Processing Unit) nodes and switching nodes are connected to the RackCable Tray (cable backplane) through high-density connectors to achieve high-speed interconnection between nodes. However, in the early stages of system construction, fault detection for Cable Tray can only rely on traditional electrical testing methods to check one by one, which makes it difficult to detect potential problems in a timely manner and the troubleshooting efficiency is low.

[0003] Therefore, how to provide a solution to the above technical problems is a problem that those skilled in the art need to solve at present. Summary of the invention

[0004] The purpose of the present invention is to provide a cabinet fault diagnosis system, method, device and storage medium, which can perform local fault diagnosis on the cable backplane when the remote management software has not yet been deployed, thereby improving the fault diagnosis efficiency.

[0005] In order to solve the above technical problems, the present invention provides a cabinet fault diagnosis system, comprising:

[0006] A switching component disposed on a cable backplane of the cabinet, wherein a plurality of ports of the switching component are correspondingly connected to preset communication ports of a plurality of nodes in the cabinet;

[0007] a first manager, the first manager being a manager in a node included in the cabinet under different networking configurations, the first manager being configured to generate a routing table corresponding to each of the nodes in the cabinet and burn it into the cable backplane, and in response to receiving a fault monitoring instruction, accessing the corresponding second manager through the switching component according to the routing table to obtain a fault diagnosis result; the routing table includes a link status column for recording the current status of the communication link of each of the nodes, and the second manager being a manager in the node other than the resident node in the cabinet;

[0008] The second manager is configured to obtain the current state of the communication link of the node corresponding to itself, and update the link state column corresponding to itself in the routing table according to the current state of the communication link.

[0009] Optionally, the second manager is specifically configured to read the value of the link status register of the node corresponding to itself according to the first preset period, determine the current status of the communication link of the node corresponding to itself according to the value of the link status register, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

[0010] Optionally, the first manager is specifically configured to generate a routing table corresponding to each of the nodes of the cabinet and burn it into the cable backplane. In response to receiving a fault monitoring instruction, the first manager is configured to determine the node whose current state in the link status column of the routing table is an abnormal state as the target node, communicate with the second manager corresponding to the target node through the switching component, obtain the abnormal information of the target node, and obtain the fault diagnosis result based on the abnormal information.

[0011] Optionally, the first manager is further configured to obtain the node type and topological relationship of each of the nodes in the cabinet, so as to generate the routing table based on the node type and topological relationship of each of the nodes.

[0012] Optionally, the cabinet fault diagnosis system further includes multiple alarm devices, and the multiple alarm devices correspond to different alarm priorities;

[0013] The first manager is further configured to determine a fault alarm priority corresponding to the target node based on the abnormal information of the target node, and control an alarm device corresponding to the fault alarm priority to output the alarm information of the target node.

[0014] Optionally, the cabinet fault diagnosis system further includes a user management component configured to send a fault monitoring instruction to the first manager at a second preset period and prompt an acquired fault diagnosis result.

[0015] Optionally, it includes multiple second managers, including an auxiliary manager, and the auxiliary manager is configured to obtain the current status of the communication link of the node corresponding to itself when the first manager is in normal working state, and update the link status column corresponding to itself in the routing table according to the current status of the communication link; when the first manager is in an abnormal state, output a replacement prompt instruction, and in response to receiving a fault monitoring instruction output by the user management component based on the replacement prompt signal, access the corresponding second manager through the switching component according to the routing table to obtain the fault diagnosis result.

[0016] In order to solve the above technical problems, the present invention further provides a cabinet fault diagnosis method, which is applied to a resident node in a cabinet, wherein the cabinet includes multiple nodes, and the resident node is the node included in the cabinet under different networking configurations. The cabinet fault diagnosis method includes:

[0017] Generate a routing table corresponding to each of the nodes in the cabinet and burn it into the cable backplane of the cabinet; the routing table includes a link status column for recording the current status of the communication link of each of the nodes; a switching component is provided on the cable backplane of the cabinet, and multiple ports of the switching component are correspondingly connected to the preset communication ports of multiple nodes in the cabinet;

[0018] In response to receiving a fault monitoring instruction, the corresponding second manager is accessed through the switching component according to the routing table to obtain a fault diagnosis result; the second manager is a manager in the nodes in the cabinet except the resident node, and the second manager is configured to obtain the current status of the communication link of the node corresponding to itself, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

[0019] In order to solve the above technical problem, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the cabinet fault diagnosis method described above are implemented.

[0020] In order to solve the above technical problems, the present invention further provides a cabinet fault diagnosis device, which is applied to a resident node in a cabinet, wherein the cabinet includes multiple nodes, and the resident node is the node included in the cabinet under different networking configurations. The cabinet fault diagnosis device includes:

[0021] A processing module, used for generating a routing table corresponding to each of the nodes of the cabinet and burning it into the cable backplane of the cabinet; the routing table includes a link status column for recording the current status of the communication link of each of the nodes; a switching component is provided on the cable backplane of the cabinet, and a plurality of ports of the switching component are correspondingly connected to the preset communication ports of the plurality of nodes in the cabinet;

[0022] A diagnosis and determination module is used to respond to receiving a fault monitoring instruction and access the corresponding second manager through the switching component according to the routing table to obtain a fault diagnosis result; the second manager is a manager in the node in the cabinet except the resident node, and the second manager is configured to obtain the current status of the communication link of the node corresponding to itself, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

[0023] The present invention provides a cabinet fault diagnosis system. For a cabinet equipped with a cable backplane, a switching component is arranged on the cable backplane, and then the preset communication port corresponding to the manager of each node is connected to the switching component, so as to ensure data communication between each node. The manager in a resident node included in the networking configuration in the cabinet is used as the first manager, and then a routing table is generated according to each node in the cabinet. The link status column in the routing table is automatically updated by each second manager according to the current status of its own communication link. After receiving the fault monitoring instruction, the first manager accesses the corresponding second manager according to the routing table to obtain the fault diagnosis result. When the remote management software has not been deployed, the cable backplane is locally diagnosed, so as to improve the fault diagnosis efficiency, not rely on the deployment of the remote management software, and improve the self-diagnosis ability of the cabinet. Through the automated fault diagnosis process, the operation of the maintenance personnel is simplified and the manual intervention is reduced.

[0024] The present invention also provides a cabinet fault diagnosis method, device and computer-readable storage medium, which have the same beneficial effects as the above cabinet fault diagnosis system. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 A schematic diagram of the structure of a cabinet fault diagnosis system provided by the present invention;

[0027] Figure 2 A flowchart of the steps of a cabinet fault diagnosis method provided by the present invention;

[0028] Figure 3 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention;

[0029] Figure 4 This is a schematic structural diagram of a cabinet fault diagnosis device provided by the present invention. DETAILED DESCRIPTION

[0030] The core of the present invention is to provide a cabinet fault diagnosis system, method, device and storage medium, which can perform local fault diagnosis on the cable backplane when the remote management software has not been deployed, thereby improving the fault diagnosis efficiency.

[0031] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] First, please refer to Figure 1 The present invention provides a cabinet fault diagnosis system, comprising:

[0033] A switching component 1 is arranged on the cable backplane of the cabinet, and a plurality of ports of the switching component 1 are correspondingly connected to preset communication ports of a plurality of nodes in the cabinet;

[0034] A first manager 2, the first manager 2 is a manager in a node included in the cabinet under different networking configurations, the first manager 2 is configured to generate a routing table corresponding to each node of the cabinet and burn it into the cable backplane, and in response to receiving a fault monitoring instruction, access the corresponding second manager 3 through the switching component 1 according to the routing table to obtain a fault diagnosis result; the routing table includes a link status column for recording the current status of the communication link of each node, and the second manager 3 is a manager in a node other than a resident node in the cabinet;

[0035] The second manager 3 is configured to obtain the current state of the communication link of the node corresponding to itself, and update the link state column corresponding to itself in the routing table according to the current state of the communication link.

[0036] In this embodiment, considering the characteristics of distributed management, there is no direct communication management between the managers (BMC (Baseboard Management Controller)) in each node in the cabinet. First, the architecture and application scenarios of the entire cabinet are analyzed to determine the resident nodes that exist under different networking configurations. In this embodiment, the resident nodes are selected as the interface for the management of the entire cabinet. The management of the entire cabinet is achieved through the BMC of the resident nodes, which facilitates the local BMC to diagnose the faults of the cable backplane, and to achieve real-time, fast and convenient data aggregation, analysis and diagnosis of potential problems and faults of the cable backplane without relying on management software. Among them, each node in the cabinet includes but is not limited to a switching node, a GPU Box node (i.e., a node encapsulated with a GPU), etc. A switching component 1 is provided on the cable backplane, and the switching component 1 can be constructed by an I2C Switch chip board. The connector reserved pin of each node (i.e., the communication port in this embodiment) is connected to the port on the I2C Switch chip board. It can be understood that in this embodiment, the resident node is configured to obtain data through the I2CSwitch chip board, and other nodes except the resident node are configured to output data to the I2CSwitch chip board.

[0037] The first manager 2 in the resident node first generates a routing table corresponding to each node in the cabinet and burns it into the FRU (Field Replaceable Unit) of the cable backplane. The routing table includes but is not limited to a link status column, which is used to record the current status of the communication link of each node. Each node updates the link status column in the routing table according to the current status of its corresponding communication link. After receiving the fault monitoring instruction, the first manager 2 accesses the second manager 3 in the corresponding node connected to the switching component 1 according to the information in the routing table, thereby obtaining the fault diagnosis result. Specifically, the first manager 2 obtains the node type and topological relationship of each node in the cabinet, so as to generate a routing table based on the node type and topological relationship of each node. Specifically, the first manager 2 generates a routing table for the 112G link of the Ethernet switching chip according to the advance hardware link, and burns it into the FRU of the cable backplane.

[0038] It can be seen that in this embodiment, for a cabinet equipped with a cable backplane, a switching component 1 is arranged on the cable backplane, and then the preset communication port corresponding to the manager of each node is connected to the switching component 1, ensuring data communication between each node, and the manager in a resident node included in the networking configuration in the cabinet is used as the first manager 2, and then a routing table is generated according to each node in the cabinet. The link status column in the routing table is automatically updated by each second manager 3 according to the current status of its own communication link. After receiving the fault monitoring instruction, the first manager 2 accesses the corresponding second manager 3 according to the routing table to obtain the fault diagnosis result. When the remote management software has not been deployed, the cable backplane is locally diagnosed, the fault diagnosis efficiency is improved, and it does not rely on the deployment of the remote management software, which improves the self-diagnosis capability of the cabinet. Through the automated fault diagnosis process, the operation of maintenance personnel is simplified and manual intervention is reduced.

[0039] Based on the above embodiments:

[0040] In an exemplary embodiment, the second manager 3 is specifically configured to read the value of the link status register of the node corresponding to itself according to the first preset period, determine the current status of the communication link of the node corresponding to itself according to the value of the link status register, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

[0041] In this embodiment, the second manager 3 of each node uses the management data input / output (MDIO) interface to periodically communicate with the Retimer (Ethernet switching chip). MDIO is a standard protocol for accessing the status and control registers of network physical layer devices. Through the MDIO interface, the BMC can read the registers about the link status in the Retimer to monitor the interconnection signal status in both directions. This means that the BMC can monitor the data link status sent and received by the node to ensure the integrity of the link. By analyzing the value of the link status register, the second manager 3 can determine whether the communication link of the node where it is located is currently working properly. Once the link status is determined, the second manager 3 will update the link status column corresponding to the node in the routing table according to the current status of the link. In this way, the first manager 2 can obtain the link status of the entire cabinet by checking the routing table. The link status register usually provides information such as link on / off, signal quality, and error count. Communication links include but are not limited to high-speed communication links of the node, such as PCIe links, Ethernet links, etc.

[0042] In an exemplary embodiment, the first manager 2 is specifically configured to generate a routing table corresponding to each node of the cabinet and burn it into the cable backplane. In response to receiving a fault monitoring instruction, the node whose current state in the link status column of the routing table is an abnormal state is determined as the target node, and the second manager 3 corresponding to the target node is communicated with through the switching component 1 to obtain the abnormal information of the target node, and obtain the fault diagnosis result based on the abnormal information.

[0043] In this embodiment, the first manager 2 is responsible for generating a routing table corresponding to each node in the cabinet. This routing table contains the link status information of each node. After generating the routing table, the first manager 2 burns the routing table into the cable backplane to ensure that the routing information can be accessed by the components in the cabinet. When the first manager 2 receives a fault monitoring instruction, it checks the link status column in the routing table to determine which nodes have abnormal current status, and the nodes whose link status is listed as abnormal are determined as target nodes, which may be the source of the fault or affected by the fault. The first manager 2 communicates with the second manager 3 (i.e., the BMC of the node) corresponding to the target node through the switching component 1 to obtain the abnormal information of the target node. The second manager 3 provides detailed abnormal information of the target node, which may include the value of the link status register, error log, diagnostic code, etc. The first manager 2 performs fault diagnosis analysis based on the abnormal information obtained from the second manager 3 to obtain the fault diagnosis result.

[0044] In this embodiment, the first manager 2 generates and manages the routing table, thereby realizing centralized monitoring and management of the entire cabinet. Through the pre-stored link status information, the first manager 2 can quickly respond to the fault monitoring instruction and locate the abnormal node in time. The communication between the first manager 2 and the second manager 3 ensures the rapid transmission of fault information, which helps to quickly diagnose the fault. The system has a high degree of automation and can automatically complete fault diagnosis based on preset logic and algorithms, reducing manual intervention. This configuration improves the automation level and efficiency of the cabinet fault diagnosis system.

[0045] In an exemplary embodiment, the cabinet fault diagnosis system further includes multiple alarm devices, and the multiple alarm devices correspond to different alarm priorities;

[0046] The first manager 2 is further configured to determine the fault alarm priority corresponding to the target node based on the abnormal information of the target node, and control the alarm device corresponding to the fault alarm priority to output the alarm information of the target node.

[0047] In this embodiment, the system also includes a variety of alarm devices, which may include sound and light alarms, email notification systems, short message services, mobile application push notifications, etc. Each alarm device may be suitable for different alarm scenarios, for example, sound and light alarms are suitable for on-site instant alarms, while emails or short messages are suitable for remote notifications.

[0048] This embodiment defines different alarm priorities for different fault types or severity levels. The priority can be high, medium, low, or a more detailed level. The alarm priority helps maintenance personnel quickly identify and handle the most serious or urgent problems. The first manager 2 is configured to analyze the abnormal information of the target node and determine the corresponding fault alarm priority, including but not limited to the identification of the fault type (such as hardware failure, software error, network interruption, etc.), and the assessment of the impact of the fault on the system. The first manager 2 controls the corresponding alarm device to output the alarm information according to the determined fault alarm priority. For example, if a high-priority fault is determined, the system may simultaneously activate the sound and light alarm and send an email or SMS notification to ensure that maintenance personnel can respond quickly to troubleshoot and repair the fault.

[0049] In this way, the cabinet fault diagnosis system can provide a more intelligent and efficient alarm mechanism to help maintenance personnel quickly identify and handle potential system problems.

[0050] In an exemplary embodiment, the cabinet fault diagnosis system further includes a user management component configured to send a fault monitoring instruction to the first manager 2 according to a second preset period and prompt the acquired fault diagnosis result.

[0051] In this embodiment, the debugging personnel can implement cable tray fault monitoring of the entire cabinet through the user management component and the first manager 2, and obtain link abnormality information in the entire cabinet conveniently and quickly by reading the link status, thereby realizing convenient fault diagnosis.

[0052] In an exemplary embodiment, a plurality of second managers 3 are included, wherein the plurality of second managers 3 include an auxiliary manager, and the auxiliary manager is configured to obtain the current state of the communication link of the node corresponding to itself when the first manager 2 is in a normal working state, and update the link state column corresponding to itself in the routing table according to the current state of the communication link, and output a replacement prompt instruction when the first manager 2 is in an abnormal state, and in response to receiving the fault monitoring instruction output by the user management component based on the replacement prompt signal, access the corresponding second manager 3 through the switching component 1 according to the routing table to obtain the fault diagnosis result. In this embodiment, when the first manager 2 is in a normal working state, the auxiliary manager is responsible for obtaining the current state of the communication link of its corresponding node. The auxiliary manager will update the link state column corresponding to itself in the routing table according to the acquired communication link state. This ensures that the routing table always reflects the latest network state, which facilitates fast and accurate fault diagnosis. When the first manager 2 is in an abnormal state, the auxiliary manager will output a replacement prompt instruction to notify the user management component that the first manager 2 may have a problem, and the user management component will output a fault monitoring instruction to the auxiliary manager after receiving the replacement prompt signal. The auxiliary manager responds to the fault monitoring instruction and accesses the corresponding second manager 3 through the switching component 1 according to the routing table to obtain the fault diagnosis result. In this way, even if the first manager 2 fails, the system can still perform fault diagnosis and response. When the first manager 2 fails, the auxiliary manager may assume more management responsibilities, such as taking over part or all of the functions of the first manager 2. Provide redundant paths and backup communication links to ensure that the system can still communicate and diagnose when the main link fails. Furthermore, the auxiliary manager may also monitor its own status and issue an alarm when it fails.

[0053] With these configurations, the cabinet fault diagnostic system has higher reliability and self-healing capabilities, and can continue to operate and diagnose faults when problems occur in key components.

[0054] In summary, the present invention can provide a method for local fault diagnosis of Cable Tray based on the AI ​​Rack cabinet equipped with Cable Tray when the remote management software has not yet been deployed. This solution can identify the fault of the Cable Tray without disassembling the Cable Tray, and the granularity of the fault diagnosis can be as fine as a single lane.

[0055] Second, please refer to Figure 2 The present invention also provides a cabinet fault diagnosis method, which is applied to a resident node in a cabinet, wherein the cabinet includes multiple nodes, and the resident node is a node included in the cabinet under different networking configurations. The cabinet fault diagnosis method includes:

[0056] S101: Generate a routing table corresponding to each node of the cabinet and burn it into the cable backplane of the cabinet; the routing table includes a link status column for recording the current status of the communication link of each node; a switching component is provided on the cable backplane of the cabinet, and multiple ports of the switching component are correspondingly connected to preset communication ports of multiple nodes in the cabinet;

[0057] S102: In response to receiving a fault monitoring instruction, access the corresponding second manager through the switching component according to the routing table to obtain a fault diagnosis result; the second manager is a manager in a node other than a resident node in the cabinet, and the second manager is configured to obtain the current status of the communication link of the node corresponding to itself, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

[0058] Third, please refer to Figure 3 The present invention further provides a computer-readable storage medium 10, on which a computer program 11 is stored. When the computer program 11 is executed by a processor, the steps of the above cabinet fault diagnosis method are implemented.

[0059] The computer-readable storage medium 10 may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0060] Fourth, please refer to Figure 4 The present invention also provides a cabinet fault diagnosis device, which is applied to a resident node in a cabinet. The cabinet includes multiple nodes. The resident node is a node included in the cabinet under different networking configurations. The cabinet fault diagnosis device includes:

[0061] The processing module 21 is used to generate a routing table corresponding to each node of the cabinet and burn it into the cable backplane of the cabinet; the routing table includes a link status column for recording the current status of the communication link of each node; the cable backplane of the cabinet is provided with a switching component, and multiple ports of the switching component are correspondingly connected to the preset communication ports of multiple nodes in the cabinet;

[0062] The diagnosis and judgment module 22 is used to access the corresponding second manager through the switching component according to the routing table in response to receiving a fault monitoring instruction to obtain the fault diagnosis result; the second manager is a manager in the node other than the resident node in the cabinet, and the second manager is configured to obtain the current status of the communication link of the node corresponding to itself, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

[0063] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0064] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A cabinet fault diagnosis system, characterized in that: include: A switching component disposed on a cable backplane of the cabinet, wherein a plurality of ports of the switching component are correspondingly connected to preset communication ports of a plurality of nodes in the cabinet; a first manager, the first manager being a manager in a node included in the cabinet under different networking configurations, the first manager being configured to generate a routing table corresponding to each of the nodes in the cabinet and burn it into the cable backplane, and in response to receiving a fault monitoring instruction, accessing the corresponding second manager through the switching component according to the routing table to obtain a fault diagnosis result; the routing table includes a link status column for recording the current status of the communication link of each of the nodes, and the second manager being a manager in the node other than the resident node in the cabinet; The second manager is configured to obtain the current state of the communication link of the node corresponding to itself, and update the link state column corresponding to itself in the routing table according to the current state of the communication link.

2. The cabinet fault diagnosis system according to claim 1, characterized in that: The second manager is specifically configured to read the value of the link status register of the node corresponding to itself according to the first preset period, determine the current status of the communication link of the node corresponding to itself according to the value of the link status register, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

3. The cabinet fault diagnosis system according to claim 1, characterized in that: The first manager is specifically configured to generate a routing table corresponding to each of the nodes of the cabinet and burn it into the cable backplane. In response to receiving a fault monitoring instruction, the first manager determines the node whose current state in the link status column of the routing table is an abnormal state as the target node, communicates with the second manager corresponding to the target node through the switching component, obtains the abnormal information of the target node, and obtains the fault diagnosis result based on the abnormal information.

4. The cabinet fault diagnosis system according to claim 3, characterized in that: The first manager is further configured to obtain the node type and topological relationship of each of the nodes in the cabinet, so as to generate the routing table based on the node type and topological relationship of each of the nodes.

5. The cabinet fault diagnosis system according to claim 3, characterized in that: The cabinet fault diagnosis system also includes multiple alarm devices, and the multiple alarm devices correspond to different alarm priorities; The first manager is further configured to determine a fault alarm priority corresponding to the target node based on the abnormal information of the target node, and control an alarm device corresponding to the fault alarm priority to output the alarm information of the target node.

6. The cabinet fault diagnosis system according to claim 1, characterized in that: The cabinet fault diagnosis system also includes a user management component configured to send a fault monitoring instruction to the first manager according to a second preset period and prompt the acquired fault diagnosis result.

7. The cabinet fault diagnosis system according to claim 6, characterized in that: It includes multiple second managers, including an auxiliary manager. The auxiliary manager is configured to obtain the current status of the communication link of the node corresponding to itself when the first manager is in normal working state, and update the link status column corresponding to itself in the routing table according to the current status of the communication link; when the first manager is in an abnormal state, output a replacement prompt instruction, and in response to receiving the fault monitoring instruction output by the user management component based on the replacement prompt signal, access the corresponding second manager through the switching component according to the routing table to obtain the fault diagnosis result.

8. A cabinet fault diagnosis method, characterized in that: Applied to a resident node in a cabinet, the cabinet includes a plurality of nodes, the resident node is the node included in the cabinet under different networking configurations, and the cabinet fault diagnosis method includes: Generate a routing table corresponding to each of the nodes in the cabinet and burn it into the cable backplane of the cabinet; the routing table includes a link status column for recording the current status of the communication link of each of the nodes; a switching component is provided on the cable backplane of the cabinet, and multiple ports of the switching component are correspondingly connected to the preset communication ports of multiple nodes in the cabinet; In response to receiving a fault monitoring instruction, the corresponding second manager is accessed through the switching component according to the routing table to obtain a fault diagnosis result; the second manager is a manager in the nodes in the cabinet except the resident node, and the second manager is configured to obtain the current status of the communication link of the node corresponding to itself, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the cabinet fault diagnosis method according to claim 8 are implemented.

10. A cabinet fault diagnosis device, characterized in that: Applied to a resident node in a cabinet, the cabinet includes multiple nodes, the resident node is the node included in the cabinet under different networking configurations, and the cabinet fault diagnosis device includes: A processing module, used for generating a routing table corresponding to each of the nodes of the cabinet and burning it into the cable backplane of the cabinet; the routing table includes a link status column for recording the current status of the communication link of each of the nodes; a switching component is provided on the cable backplane of the cabinet, and a plurality of ports of the switching component are correspondingly connected to the preset communication ports of the plurality of nodes in the cabinet; A diagnosis and determination module is used to respond to receiving a fault monitoring instruction and access the corresponding second manager through the switching component according to the routing table to obtain a fault diagnosis result; the second manager is a manager in the node in the cabinet except the resident node, and the second manager is configured to obtain the current status of the communication link of the node corresponding to itself, and update the link status column corresponding to itself in the routing table according to the current status of the communication link.

Images