Server interface detection method, system, server, device, medium and product

Abstract

The application discloses a kind of server interface detection method, system, server, equipment, medium and product, it is related to computer technical field, the detection method includes;Receive the pressure parameter and interface signal parameter at server interface, and judge whether pressure parameter and / or interface signal parameter satisfy preset fault judgment rule;If pressure parameter and / or interface signal parameter satisfy preset fault judgment rule, then according to pressure parameter and / or interface signal parameter, by searching preset fault type database, obtain the fault type of server interface;Realize fast, accurate anomaly identification;Interface identification of server interface, fault type and target maintenance suggestion corresponding to fault type are sent to man-machine interface, so that operation and maintenance personnel can quickly maintain. Solve the technical problem that existing interface troubleshooting efficiency is low, improve troubleshooting efficiency and response speed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, and in particular, to a server interface detection method and system, a server, a device, a medium and a product. BACKGROUND

[0002] Servers undertake tasks such as storage, calculation and high-speed transmission of large-scale data. The interconnection between servers and external components relies on multiple interfaces, which are responsible for not only the transmission of electrical signals, but also functions such as protocol negotiation, error checking and state management. With the continuous improvement of server integration and interface rate, server interface exceptions (such as slight contact loosening, signal integrity degradation or protocol layer negotiation exception) may cause throughput decline, link reset or even service interruption.

[0003] In related technologies, interface faults are usually located by relying on baseboard management controller logs, operating system kernel error reports or application alarms. When an error appears in the log, the log entry usually only gives an error code or a general exception identifier, which makes it difficult to distinguish between mechanical contact, signal integrity degradation or protocol layer compatibility problems. Maintenance personnel still need to use offline tools to check one by one, resulting in low fault troubleshooting efficiency. Therefore, there is an urgent need for a server interface detection method to solve the technical problem of low server interface fault troubleshooting efficiency in the related art. SUMMARY

[0004] The present application provides a server interface detection method, system, server, device, medium and product to at least solve the problem of low server interface fault troubleshooting efficiency in the related art.

[0005] The present application provides a server interface detection method, comprising:

[0006] receiving a pressure parameter and an interface signal parameter at a server interface, and determining whether the pressure parameter and / or the interface signal parameter meet a preset fault determination rule;

[0007] If the pressure parameter and / or the interface signal parameter meet the preset fault determination rule, then according to the pressure parameter and / or the interface signal parameter, a fault type of the server interface is obtained by searching a preset fault type database;

[0008] sending the interface identifier of the server interface, the fault type and the target maintenance suggestion corresponding to the fault type to a man-machine interface.

[0009] The present application provides a server interface detection system, comprising: a parameter acquisition module and a processing module;

[0010] The parameter acquisition module is configured to acquire a pressure parameter and an interface signal parameter at a server interface, and send the pressure parameter and the interface signal parameter to the processing module;

[0011] a processing module, configured to receive the pressure parameter and the interface signal parameter at the server interface, and determine whether the pressure parameter and / or the interface signal parameter meets a preset fault determination rule;

[0012] The processing module is further configured to, if the pressure parameter and / or the interface signal parameter meets the preset fault determination rule, acquire a fault type of the server interface according to the pressure parameter and / or the interface signal parameter by searching a preset fault type database, and send an interface identifier, the fault type, and a target maintenance suggestion corresponding to the fault type of the server interface to a man-machine interactive interface.

[0013] The application further provides a server comprising the detection system.

[0014] The application further provides a server interface detection device, comprising:

[0015] a data receiving unit, configured to receive the pressure parameter and the interface signal parameter at the server interface, and determine whether the pressure parameter and / or the interface signal parameter meets a preset fault determination rule;

[0016] a fault identification unit, configured to, if the pressure parameter and / or the interface signal parameter meets the preset fault determination rule, acquire a fault type of the server interface according to the pressure parameter and / or the interface signal parameter by searching a preset fault type database;

[0017] a fault alarm unit, configured to send an interface identifier, the fault type, and a target maintenance suggestion corresponding to the fault type of the server interface to a man-machine interactive interface.

[0018] The application further provides an electronic device, comprising a memory configured to store a computer program, and a processor configured to execute the computer program to implement the steps of any of the detection methods.

[0019] The application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of any of the detection methods.

[0020] The application further provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the steps of any of the detection methods.

[0021] The application realizes synchronous perception of the mechanical contact state and the signal transmission state of the interface by simultaneously acquiring the pressure parameter and the interface signal parameter at the server interface. Then, the pressure parameter and / or the interface signal parameter are determined according to the preset fault judgment rule, so that rapid and determined abnormality discovery is realized. Compared with the passive recording mode relying on logs, the time from abnormality occurrence to abnormality identification is shortened, and the dependence on manual experience and offline tools is reduced. The fault type of the server interface is retrieved and output, so that the accuracy and repeatability of fault type acquisition are improved. The interface identifier of the server interface, the fault type, and the target maintenance suggestion corresponding to the fault type are sent to the man-machine interaction interface, so that the operation and maintenance personnel can quickly perform maintenance, and the manual processing time required for interface fault troubleshooting is reduced. In summary, through synchronous acquisition, rule determination, type retrieval, and suggestion delivery, rapid identification, accurate classification, and maintenance guidance of interface abnormalities are realized, which can solve the technical problem of low interface fault troubleshooting efficiency, improve troubleshooting efficiency and response speed, and reduce server downtime and manual cost. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0023] Figure 1 The structural schematic diagram of the server interface detection system provided by the embodiments of the present application is shown in the figure.

[0024] Figure 2 The flowchart of the server interface detection method provided by the embodiments of the present application is shown in the figure.

[0025] Figure 3 The flowchart of the method for acquiring pressure characteristics provided by the embodiments of the present application is shown in the figure.

[0026] Figure 4 The flowchart of the method for marking abnormal data provided by the embodiments of the present application is shown in the figure.

[0027] Figure 5 The flowchart of the interface health analysis method provided by the embodiments of the present application is shown in the figure.

[0028] Figure 6 The structural schematic diagram of the detection device provided by the embodiments of the present application is shown in the figure.

[0029] Figure 7 The structural schematic diagram of the electronic device provided by the present application is shown in the figure.

[0030] Reference signs:

[0031] 1-detection system; 11-parameter acquisition module; 12-processing module; 111-interface pressure detection unit; 112-interface information detection unit. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0033] It should be noted that, in the description of the present application, the terms "comprise", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. The terms "first", "second" and the like in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0034] The related art mainly relies on substrate management controller logs, operating system kernel error reports or application layer alarms for interface exception positioning, and performs cyclic self-checking or loopback testing on the interface through offline diagnostic tools. Such means can only record error codes, link resets, on-off times and the like; when it is necessary to determine physical contact instability or signal integrity decline, external measurement equipment needs to be connected and manual experience needs to be relied on for summarization. This results in detection lag, scattered evidence and difficulty in comprehensive analysis, which not only affects the accuracy of judgment, but also lengthens the positioning and disposal cycle, resulting in low interface fault troubleshooting efficiency.

[0035] Based on the above technical problems and needs, the inventive concept of the present application aims to unify the mechanical contact state and signal transmission state of the server interface into the fault detection link. Specifically, the pressure parameter and the interface signal parameter are acquired for the same interface entity, and the pressure parameter and / or the interface signal parameter are determined by using a preset fault determination rule. After determining that there is an exception, the pressure parameter and / or the interface signal parameter are used as a search condition to acquire the fault type of the server interface through a preset fault type database, so that the rapid identification and accurate classification of the interface exception are realized. At the same time, the server interface identifier, the fault type and the target maintenance suggestion corresponding to the fault type are sent to the man-machine interaction interface together, so that timely feedback after fault identification and classification is realized. Through the modes of data acquisition, fault determination, type retrieval and maintenance suggestion issuance, the link from the occurrence of the exception to positioning and disposal is compressed, and the accuracy and timeliness of interface exception detection are improved.

[0036] In order for those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] Figure 1 The structural schematic diagram of the server interface detection system provided by the embodiment of the present application is shown in Figure 1 The server interface detection system 1 includes a parameter acquisition module 11 and a processing module 12. The parameter acquisition module 11 is used to acquire the pressure parameter and the interface signal parameter at the server interface, and send the pressure parameter and the interface signal parameter to the processing module 12. By collecting the pressure parameter related to mechanical contact and the interface signal parameter related to signal transmission, the synchronous perception of the physical contact state and the signal transmission state of the server interface is realized, and the data comprehensiveness used in the subsequent determination is ensured. Among them, the pressure parameter is quantifiable data reflecting the contact load and micro-vibration state of the interface; the interface signal parameter is quantifiable data reflecting the signal transmission quality, communication logic behavior and connection state of the interface. It should be noted that the parameter acquisition module 11 can be realized in various forms such as sensor acquisition circuit or firmware reading channel, and the specific form is not limited here as long as the comparable pressure parameter and interface signal parameter can be formed.

[0038] The processing module 12 is used to receive the pressure parameter and the interface signal parameter at the server interface, and determine whether the pressure parameter and / or the interface signal parameter meets the preset fault determination rule. The pressure parameter is used to reflect the stress and loose vibration state of the interface, and the interface signal parameter is used to reflect the transmission quality and communication behavior of the link. By receiving these two types of data, the omissions and deviations caused by comparison and manual information splicing are reduced, which is beneficial to the complete acquisition and utilization of abnormal clues. The processing module 12 determines the pressure parameter and / or the interface signal parameter according to the preset fault determination rule. Among them, the preset fault determination rule is a parameterized set for determining whether the parameter is in the normal working domain, which realizes the instant identification of the abnormality and reduces the waiting time for relying on log playback and manual experience comparison. It should be further noted that the fault determination rule can be a combination of static threshold range and drift tolerance, or a statistical baseline updated over time and a deviation degree measure, as long as it can give a clear judgment indication on whether the pressure parameter and / or the interface signal parameter meets the fault trigger condition.

[0039] The processing module 12 is also used to acquire the fault type of the server interface by searching the preset fault type database according to the pressure parameter and / or the interface signal parameter if the pressure parameter and / or the interface signal parameter meets the preset fault determination rule, and send the interface identifier, fault type and target maintenance suggestion corresponding to the fault type of the server interface to the man-machine interface.

[0040] In the embodiment, when the fault judgment rule is satisfied, the processing module 12 takes the current pressure parameter and / or interface signal parameter as a retrieval condition, queries a preset fault type database, and obtains the fault type of the server interface. The database maps common parameter patterns to fault types, including mechanical contact problems, signal transmission degradation, protocol compatibility exceptions, and related hardware failures. Through parameter-to-type retrieval, the type of the fault can be quickly determined, repeated trials and multiple rounds of troubleshooting are reduced, and the accuracy and timeliness of interface fault positioning are improved. After obtaining the fault type, the alarm information containing the server interface identifier, the fault type, and the target maintenance suggestion corresponding to the fault type is generated and sent to the human-computer interaction interface, so that the operation and maintenance personnel can further check and maintain according to the target maintenance suggestion, and the efficiency of server interface troubleshooting is improved. The server interface identifier is an identifier that can uniquely indicate the measured interface entity, and exemplary elements include device sequence, mainboard slot number, and port index, to ensure the accuracy and traceability of positioning. The target maintenance suggestion is a one-to-one or one-to-many associated disposal item, which is derived from the suggestion table associated with the preset fault type database or other accessible data storage, and the human-computer interaction interface is an exhibition and interaction end for operation and maintenance personnel, including but not limited to a server management interface, a remote management console, or an integrated alarm system.

[0041] In summary, the detection system 1 simultaneously covers pressure parameters and interface signal parameters to realize parallel monitoring of server interface mechanical contact and signal transmission. By using a preset fault judgment rule to trigger retrieval and confirming the fault type of the current server interface according to the fault type database, the accuracy of fault type acquisition is improved. Then, the server interface identifier, the fault type, and the target maintenance suggestion corresponding to the fault type are sent to the human-computer interaction interface, so that the operation and maintenance personnel can quickly maintain, improve the troubleshooting efficiency and response speed, and reduce the server downtime and labor cost.

[0042] In one embodiment, the parameter acquisition module 11 comprises an interface pressure detection unit 111 and an interface information detection unit 112; wherein the pressure detection end of the interface pressure detection unit 111 (such as a pressure sensor) is arranged at the server interface, and the parameter acquisition module 11 is specifically configured to collect the pressure parameter at the server interface through the interface pressure detection unit 111, and send the pressure parameter to the processing module 12. The arrangement of this structure enables the interface pressure detection unit 111 to sense the slight load change and local vibration response in the connection process in real time at the actual contact point of the interface. Since the server interface may fluctuate in contact pressure or vibrate slightly due to loosening, oxidation or structural fatigue in long-term operation, the interface pressure detection unit 111 can effectively acquire the pressure parameters reflecting the mechanical contact reliability, such as the contact pressure value, the pressure fluctuation frequency, etc., thereby providing real-time physical basis for the stability of the interface.

[0043] The interface information detection unit 112 is connected with the signal transmission line of the server interface, and at least one of the physical layer parameters, the protocol layer parameters and the interface state parameters of the server interface is collected through the interface information detection unit 112, and at least one of the physical layer parameters, the protocol layer parameters and the interface state parameters is sent to the processing module 12. Specifically, the physical layer parameters usually involve voltage and current amplitude, eye opening, jitter range, signal integrity, etc., and are used to reflect the transmission quality of the signal on the physical channel; the protocol layer parameters can include data packet format, frame check code, error retransmission times, handshake response delay, etc., and are used to reflect the processing state of the communication protocol stack; the interface state parameters may involve link connection state, online and offline event count, link reset frequency, current port working mode, etc. The collection of these signal side parameters can realize dynamic monitoring of the running state of the server interface in the information interaction process. In summary, through the interface pressure detection unit 111 and the interface information detection unit 112, the parameter acquisition module 11 can comprehensively monitor the server interface from the physical and signal levels. This multi-source data fusion sensing method not only enhances the accuracy of fault identification, but also improves the response speed of interface abnormality detection, and provides basic data support for fault judgment and maintenance suggestion generation.

[0044] In one specific embodiment, the interface pressure detection unit 111 can adopt a structure based on a piezoelectric sensor or a strain gauge sensor. Specifically, a micro piezoelectric sensor assembly is arranged at a connection area of the server interface, which is directly attached to a metal terminal or a support part of an interface slot, and can sense the contact load change and micro vibration amplitude generated by the interface during plugging or running. The sensor signal is connected to a data processing assembly through an analog acquisition circuit to realize real-time acquisition and digital conversion of the pressure value. At the same time, the high sensitivity characteristics of the piezoelectric material ensure effective response to slight mechanical disturbances, so that mechanical abnormalities of the interface caused by loose contact, metal fatigue or uneven stress can be monitored, and then the pressure parameter is generated.

[0045] The interface information detection unit 112 acquires the physical layer signal, electrical parameter and data transmission state of the interface through a signal transmission channel connected to the server interface. For example, in a Peripheral Component Interconnect Express (PCIe) or Serial Advanced Technology Attachment (SATA) interface, a high-speed differential signal sampling assembly can be configured to extract physical layer information such as signal eye diagram, clock offset, level amplitude, etc.; and a protocol analysis module is configured to analyze link negotiation, handshake process, error code, retransmission information and other protocol layer data. Through state monitoring, the on-off times, abnormal interruption, link stability and other state parameters of the interface are recorded in real time to constitute the interface operation information. It should be noted that the present embodiment is only an example, and the interface pressure detection unit 111 and the interface information detection unit 112 can adopt related technologies capable of realizing pressure acquisition and signal information acquisition, which are not limited specifically herein.

[0046] Figure 2 The flowchart of the server interface detection method provided by the embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, it includes: Figure 2

[0047] S21, receiving the pressure parameter and the interface signal parameter at the server interface, and judging whether the pressure parameter and / or the interface signal parameter meet the preset fault judgment rule.

[0048] ​In the embodiment, the pressure parameters and interface signal parameters at the server interface sent by the parameter acquisition module are received. The pressure parameters are used to reflect the physical contact state of the interface during operation, such as plugging contact load, micro-vibration trend, and contact stability fluctuation possibly caused by loosening or structural stress change; and the interface signal parameters are used to reflect the transmission state of the interface signal, including but not limited to the physical layer communication quality in the interface link, protocol layer error information and handshake behavior, and whether the interface is currently in a connection, reconnection or disconnection working state. After receiving these parameters, the pressure parameters and interface signal parameters are analyzed and compared with preset fault judgment rules. The fault judgment rules include multiple judgment conditions for identifying the occurrence of abnormal states, such as whether there is persistent or sudden contact instability, whether there is signal level anomaly, code rate drop, handshake failure number being too high, reconnection frequency being out of limit, etc. When any one or one kind of parameter meets the preset judgment condition, it is considered that the server interface has an abnormality. By receiving the interface related parameters and setting multi-dimensional judgment logic, early identification of the abnormal state of the server interface is realized, the problem interface position is quickly determined without relying on link communication failure or manual recheck, and the automation degree and real-time performance of interface abnormality diagnosis are improved.

[0049] In S22, if the pressure parameters and / or interface signal parameters meet the preset fault judgment rules, the fault type of the server interface is obtained by searching the preset fault type database according to the pressure parameters and / or interface signal parameters.

[0050] In the embodiment, when it is judged that the pressure parameters and / or interface signal parameters of the server interface meet the preset fault judgment rules, it is considered that the server interface has an abnormal state. At this time, in order to further identify the specific fault type, the current pressure parameters and / or interface signal parameters of the server interface are used as the searching basis to access the locally or remotely deployed fault type database. Different types of interface faults and corresponding characteristic parameter information are recorded in the database in advance, which are used to support fast comparison and fault type judgment. Specifically, the received parameter information is matched with the fault data stored in the fault type database to judge which known fault type the current interface abnormal state is most consistent with. For example, if the pressure parameters show unstable contact pressure and the interface signal parameters have the characteristic of frequent physical layer fluctuation, it is possible to match the contact failure type fault; if high-frequency error response appears in the protocol layer parameters and the physical layer parameters are normal, it is possible to match the protocol negotiation failure type fault, etc. Through the embodiment, automatic classification and identification of the server interface abnormality can be realized without relying on manual experience analysis or external measurement tools, the accuracy and processing efficiency of fault judgment are improved. The fault type information can be output in a structured manner, which provides a basis for subsequent maintenance decision and further reduces the probability of human interference and misjudgment in interface fault troubleshooting.

[0051] S23, send the interface identification of the server interface, the fault type, and the target maintenance suggestion corresponding to the fault type to the man-machine interactive interface.

[0052] In this embodiment, after identifying the fault type of the server interface, information related to the interface exception is sent to the man-machine interactive interface to facilitate the operation and maintenance personnel to timely grasp the interface state and take effective measures. Specifically, it includes: first, the server interface identification is used to determine the physical location and logical number of the faulty interface in the whole machine or cabinet, ensuring that the troubleshooting process is targeted; second, the fault type is used to intuitively show the user the type of the current interface exception, facilitating the analysis of the fault impact range and severity; third, the target maintenance suggestion corresponding to the fault type, for example, the operation suggestion retrieved from the preset maintenance suggestion database according to the fault type, usually including checking steps, replacement suggestions, or structure adjustment schemes. In summary, through this embodiment, real-time visual display of abnormal information can be realized. On the one hand, the information interaction method reduces the dependence on human experience and improves the standardization and executability of interface operation and maintenance; on the other hand, the target maintenance suggestion is selected and pushed in advance, which helps to shorten the interface fault handling period and further improve the overall operation stability and response efficiency.

[0053] In one embodiment, the above step S21 of determining whether the pressure parameter meets the preset fault determination rule is further described, including:

[0054] S2111, obtain at least one pressure characteristic corresponding to the pressure parameter, and if any pressure characteristic is not within the corresponding preset pressure threshold range, the pressure parameter meets the preset fault determination rule.

[0055] In the embodiment, after receiving the pressure parameters from the parameter acquisition module, at least one pressure feature reflecting the interface contact state is extracted from the pressure parameters. The pressure feature refers to a statistical quantity or dynamic index that can quantitatively represent the mechanical contact quality and stability of the interface, and generally includes the mean value, fluctuation amplitude, instantaneous gradient, micro-vibration frequency component, loading / unloading trend, etc. of the pressure value. Among them, the pressure mean value can reflect whether the interface contact load is in a reasonable range, the fluctuation amplitude indicates the stability of the interface contact, and high-frequency micro-vibration may reflect problems such as unstable plugging or resonance interference. After extracting these pressure features, they are compared with the corresponding preset pressure threshold range for judgment. Each feature has an independent upper and lower threshold range, which is determined based on a large amount of measured data and can effectively distinguish between normal and abnormal contact states. If any pressure feature exceeds its corresponding threshold range, it is considered that the current pressure parameter is abnormal, and it is further determined that the pressure parameter meets the fault judgment rule, thereby triggering the subsequent fault recognition and alarm process. In summary, through the embodiment, the sensitivity of fault detection to small abnormalities in the contact state is enhanced, and the problem of false positives or false negatives caused by a single indicator in actual operation is reduced, improving the robustness and accuracy of the judgment result. At the same time, the above method can detect possible mechanical contact degradation trend earlier, which helps to intervene and maintain in advance, thereby improving the reliability of the interface operation and the overall stability of the server.

[0056] Figure 3 The method for obtaining pressure features provided by the embodiment of the present application is shown in the flowchart. After receiving the pressure parameters and interface signal parameters at the server interface, the acquisition of pressure features is further described. Based on the above-mentioned embodiment, as shown in Figure 3 , it includes:

[0057] S31, according to the preset sampling period, time-sequencing the pressure parameters corresponding to the server interface, and obtaining a pressure sampling data set;

[0058] S32, according to the preset sliding time window, extracting features from the pressure sampling data set by the preset pressure feature extraction rule, and obtaining at least one pressure feature.

[0059] In this embodiment, in order to more comprehensively and dynamically reflect the mechanical contact state of the server interface, time sequence sampling and sliding window analysis mechanism are adopted for the collected pressure parameters. In actual operation, the pressure state at the receiving server interface continuously collects data, and according to the preset sampling period, a pressure data sequence arranged in chronological order is formed. The sampling period can be flexibly adjusted according to the server running state and the interface usage frequency, and common settings are, for example, collecting once per second or higher frequency, to ensure the continuity and timeliness of the data. The original data obtained by sampling are sorted by time stamp to form a pressure sampling data set, which is used for subsequent time sequence feature analysis. On the basis of this data set, according to the set sliding time window, the sampling data is processed in a local segment by using a moving window. Each sliding window covers a continuous pressure sampling data, and its length is usually selected according to experience, for example, set to the data within the past 5 seconds or 10 seconds, to balance the feature stability and response speed. In each window, at least one pressure feature is extracted according to the preset pressure feature extraction rule. The pressure feature extraction rule can include various statistical analysis methods, such as the mean, standard deviation, maximum-minimum difference (amplitude of fluctuation), first derivative of pressure change (indicating change trend), and frequency domain analysis features (such as frequency spectrum of micro-vibration) based on the pressure values in the window. These pressure features can show the contact stability and load uniformity of the server interface over time. Through this sliding window analysis and feature extraction rule, fine-grained dynamic monitoring of the pressure state is realized. The pressure feature extraction has real-time and continuity, and the accuracy and timeliness of fault judgment are improved.

[0060] In one specific embodiment, the interface signal parameters include at least one of a physical layer parameter, a protocol layer parameter, and an interface state parameter.

[0061] In this embodiment, the interface signal parameters are a collection of data for characterizing the electrical connection state and communication state of the server interface, and their composition covers multiple levels of communication information. Specifically, the interface signal parameters include at least one of a physical layer parameter, a protocol layer parameter, and an interface state parameter. The physical layer parameter is a reflection of the properties of the server interface at the electrical level, and usually includes link voltage, current, signal strength, eye opening, jitter amplitude, etc. These parameters can characterize the integrity of signal transmission and the stability of physical connection, for example, obtained by a high-speed sampling component or an eye diagram analysis module. The protocol layer parameter mainly reflects the state of the data link in the communication process, and usually includes retransmission times, packet loss rate, CRC error code, etc. Such parameters are generally automatically generated by the driver program or protocol stack during operation, and can reflect communication abnormalities caused by protocol incompatibility, signal interference, or link performance degradation. The interface state parameter is an abstract description of the overall connection state of the interface, such as whether the link is activated, the connection rate, the duplex state, the switching frequency of the interface online and offline, etc.

[0062] By comprehensively collecting the above three types of parameters, the running state of the interface can be comprehensively analyzed from different levels and dimensions. For example, physical layer abnormalities may indicate that the plug is loose or the contact resistance is too high, protocol layer abnormalities may indicate that the communication is disturbed or the device negotiation fails, and frequent interface state changes may reflect poor interface stability or configuration problems. Therefore, according to actual needs, one or more of the above parameters are selected as the interface signal parameter collection object and sent to the processing module for unified analysis and processing. Not only does it enhance the perception of fault modes, but it also lays the foundation for subsequent feature extraction and fault judgment, improving the precision and applicability of interface anomaly detection.

[0063] In one specific embodiment, after receiving the pressure parameters and interface signal parameters at the server interface, it further includes:

[0064] S2101, according to a preset sampling period, time-sequencing the physical layer parameters, protocol layer parameters and interface state parameters of the server interface respectively, to obtain a physical layer sampling data set, a protocol layer sampling data set and an interface state sampling data set;

[0065] S2102, according to a preset sliding time window, respectively using a preset physical layer feature extraction rule, a protocol layer feature extraction rule and an interface state feature extraction rule to extract features from the physical layer sampling data set, the protocol layer sampling data set and the interface state sampling data set, to obtain at least one physical layer feature, at least one protocol layer feature and at least one interface state feature.

[0066] In this embodiment, in order to improve the granularity and accuracy of server interface signal anomaly detection, the physical layer parameters, protocol layer parameters and interface state parameters contained in the interface signal parameters are processed respectively. Specifically, according to a preset sampling period, the above three types of parameters are received and sorted based on time tags to form a physical layer sampling data set, a protocol layer sampling data set and an interface state sampling data set. The physical layer sampling data set usually includes the original sampling value of the signal, which is used to reflect the physical transmission quality of the link. For example, it contains parameters such as signal strength (such as voltage value, current value), bit error rate, noise level, channel occupancy rate, etc. collected at a fixed sampling frequency. These parameters can reflect whether there is jitter, distortion or connection anomaly at the connection level. The protocol layer parameter sampling data set includes response delay of handshake packet, session packet loss rate, retransmission times, etc.; the interface state sampling data set may cover the number of interface online and offline switching, bandwidth, link activation state change frequency, negotiation failure times, etc.

[0067] Then, according to the window length and the window step of the sliding time window, the physical layer sampling data set, the protocol layer sampling data set and the interface state sampling data set are sequentially slid. In the time period corresponding to each sliding window, the preset feature extraction rule is used for feature extraction. Specifically, the physical layer sampling data set is processed by using the physical layer feature extraction rule to obtain at least one physical layer feature including signal quality, bit error rate, voltage fluctuation, etc.; the protocol layer sampling data set is processed by using the protocol layer feature extraction rule to obtain at least one protocol layer feature including connection state, handshake process exception, etc.; and the interface state sampling data set is processed by using the interface state feature extraction rule to obtain at least one interface state feature including device connection state, port response delay, etc. In this embodiment, the features are extracted by using the sliding time window and according to the feature extraction rules of different layers, so that the multi-dimensional features are obtained, and the accuracy and real-time performance of the interface fault identification are improved.

[0068] Specifically, the physical layer features, the protocol layer features and the interface state features are obtained, including:

[0069] S21021, according to the physical layer sampling data set in the current time window, at least one of the jitter mean square value, the bit error rate or the eye opening is extracted as the physical layer feature by using the preset physical layer feature extraction rule;

[0070] S21022, according to the protocol layer sampling data set in the current time window, at least one of the error code occurrence frequency, the protocol frame retransmission ratio or the handshake failure rate is extracted as the protocol layer feature by using the preset protocol layer feature extraction rule;

[0071] S21023, according to the interface state sampling data set in the current time window, at least one of the link width change amplitude, the power consumption fluctuation amplitude or the bandwidth utilization rate fluctuation amplitude is extracted as the interface state feature by using the preset interface state feature extraction rule.

[0072] In this embodiment, in order to more comprehensively analyze the signal quality, the protocol stability and the connection state change of the server interface during the running process, the physical layer features, the protocol layer features and the interface state features are obtained from the various types of original data collected in the current sliding time window. The feature extraction rule is an operation rule for converting the sampling data in the sliding time window into comparable and determinable features.

[0073] The physical layer sampling data set (such as a jitter time series, a bit error count series, eye diagram sampling data, level amplitude sampling data, and clock amplitude sampling data) is processed by a physical layer feature extraction rule to extract features such as signal quality fluctuation degree, continuous distortion time length, instantaneous pressure drop, and the like to determine whether there is a contact failure or signal integrity decline problem of the interface. Specifically, the jitter mean square value is calculated according to the jitter time series, the bit error rate per unit time is calculated according to the bit error count series, the eye opening is calculated for the eye diagram sampling data, the high level amplitude and the low level amplitude are calculated for the level amplitude sampling data, and the clock stability is calculated for the clock amplitude sampling data.

[0074] The protocol layer sampling data set (such as a protocol error code count, a protocol frame retransmission number, a link state machine jump number, an uncorrected error cumulative count, and an error correction code cumulative count) is processed by a protocol layer feature extraction rule to extract protocol layer features such as the number of error codes per unit time, the frame retransmission ratio per unit time, the link state machine jump number per unit time, the uncorrected error rate, and the error correction success rate, which are used to analyze whether there is a protocol compatibility problem or a device communication failure problem, and the number of handshake failures can also be extracted to calculate the handshake failure rate to measure the reliability of the connection establishment stage.

[0075] The interface state sampling data set (such as link rate, link width, power consumption state, power management level, and dynamic bandwidth utilization rate) is processed by an interface state feature extraction rule to extract features such as link rate offset per unit time, link width change amplitude, power consumption fluctuation amplitude, power management level change number, and bandwidth utilization rate fluctuation amplitude to identify a decline in interface stability. By separately extracting the above features, diagnostic information at various signal levels can be retained to achieve more accurate server interface fault judgment. It should be noted that the collected data and extracted features can be flexibly set by those skilled in the art according to requirements.

[0076] Next, it is determined whether the interface signal parameters satisfy the preset fault judgment rule in step S21, which is further described as follows based on the above embodiment, including:

[0077] S2121, at least one physical layer feature corresponding to the physical layer parameter is obtained, and if any physical layer feature is not within the corresponding physical layer threshold range, the interface signal parameters satisfy the preset fault judgment rule;

[0078] S2122, at least one protocol layer feature corresponding to the protocol layer parameter is obtained, and if any protocol layer feature is not within the corresponding protocol layer threshold range, the interface signal parameters satisfy the preset fault judgment rule;

[0079] S2123, at least one interface state feature corresponding to the interface state parameter is acquired, and if any interface state feature is not within the corresponding state threshold range, the interface signal parameter satisfies the preset fault judgment rule.

[0080] In the embodiment, at least one physical layer feature corresponding to the physical layer parameter is acquired, and if the physical layer feature is not within the corresponding physical layer threshold range, the interface signal parameter satisfies the preset fault judgment rule. For example, in a certain collection period, the extracted bit error rate is 2.5*10 -5 The corresponding preset bit error rate threshold range is 0 to 1.0*10 -5 By comparing the current value with the range, it is determined that the bit error rate exceeds the upper limit, the physical layer feature is determined to be abnormal, and it is determined that the interface signal parameter satisfies the fault judgment rule. For protocol layer features, such as retransmission rate, average handshake delay, or acknowledgement character delay time of a transmission control protocol (TCP) connection. If the retransmission rate of the transmission control protocol transmission in a certain period is 7.3%, and the preset protocol layer retransmission rate threshold upper limit is 5%, it can be concluded that the current retransmission behavior is abnormal, the protocol communication exists interruption or packet loss trend, and it is determined that the fault judgment rule is satisfied. Interface state features include link disconnection times per unit time, interface up and down frequency, etc. For example, in a 30-second sliding window, the link up and down switching times are detected to be 11 times, and the set interface state stability threshold is 3 times or less. The current data exceeds the range, and it is considered that the interface state parameter is abnormal.

[0081] In general, based on the extracted feature values and the corresponding preset thresholds, when any feature value is not within the corresponding reasonable interval, it can be determined that the interface signal parameter satisfies the preset fault judgment rule. The judgment method has strong practicability and accuracy, can realize early identification of fault problems of the server interface in multiple dimensions of physical, protocol and connection state, and provide quantitative basis for subsequent fault positioning and maintenance suggestions. The embodiment does not depend on subjective judgment, realizes automatic analysis through objective data comparison, and improves the efficiency and accuracy of interface abnormality identification.

[0082] Figure 4 The method for marking abnormal data provided in the embodiment of the application is shown in the flowchart, and after the pressure parameter and / or the interface signal parameter satisfy the preset fault judgment rule, the method comprises:

[0083] S41, marking the pressure feature not within the preset pressure threshold range as an abnormal pressure feature;

[0084] S42, the physical layer features not within the preset physical layer threshold range, the protocol layer features not within the preset protocol layer threshold range, and the interface state features not within the preset state threshold range are all marked as abnormal signal features.

[0085] In the embodiment, after obtaining the multiple types of features, threshold comparison is performed on each type of feature respectively, and whether it is an abnormal feature is marked accordingly. Specifically, whether the value of each pressure feature is within the corresponding preset pressure threshold range is judged one by one, for example, the value of a certain pressure feature is 1.2N, and the normal threshold range of the feature is 0.3N to 1.0N, so the pressure feature does not meet the threshold requirement and is marked as an abnormal pressure feature. For physical layer features, protocol layer features and interface state features, corresponding threshold ranges are also used to perform judgment, for example: if the bit error rate of a certain physical layer feature is 2.8%, and the preset upper limit of the physical layer bit error rate is 1.0%, the feature is marked as an abnormal physical layer feature. All physical layer features, protocol layer features and interface state features that are not within the corresponding preset threshold range are uniformly marked as abnormal signal features. Further, each type of abnormal feature is marked with its belonging feature category, threshold boundary exceeded or lowered and abnormal degree index when being marked. Through the embodiment, the abnormal state of the server interface related features is marked, and the abnormal feature set is constructed, which provides a structured and quantifiable basis for subsequent fault identification, type judgment and maintenance suggestions.

[0086] In one embodiment, the fault type of the server interface obtained in the above step S22 is further described. Based on the above embodiment, it includes:

[0087] S221, according to the pressure parameters and / or interface signal parameters corresponding to the server interface, abnormal pressure features and / or abnormal signal features are obtained, and according to the abnormal pressure features and / or abnormal signal features, a search feature group corresponding to the server interface is constructed;

[0088] S222, according to the search feature group, the fault type corresponding to the server interface is retrieved and obtained from the preset fault type database.

[0089] In the embodiment, the pressure parameters and interface signal parameters obtained at the server interface are processed, and the abnormal pressure features and / or abnormal signal features marked therein are extracted as the basis for subsequent fault identification. The abnormal pressure features generally include contact load fluctuations exceeding a preset threshold, abnormal frequency of micro-vibration, or pressure change trend not conforming to a preset mode; and the abnormal signal features include feature data exceeding the normal range in the physical layer, protocol layer and state layer, such as signal strength being weaker than the set lower limit, protocol response time delay being too long, or link state frequently fluctuating, etc. The abnormal pressure features and abnormal signal features are combined in a preset format to form a search feature group. The search feature group is used to describe the abnormal performance presented by the current server interface. By constructing a structured, fault-oriented feature set, an accurate input condition can be formed to ensure accurate comparison with the classified fault types in the database. Subsequently, based on the search feature group as the query condition, the preset fault type database is accessed. The database pre-stores interface fault cases and their corresponding abnormal feature patterns, and by matching the feature items in the current search feature group, the closest or completely matched historical fault types can be screened out. Finally, the determined server interface fault type is output, realizing fusion modeling and typed identification of the interface multi-source abnormal data, and improving the accuracy and response efficiency of fault judgment.

[0090] In one embodiment, the above step S23 provides an implementation manner. On the basis of the above embodiment, it comprises:

[0091] S231, according to the interface identifier and the fault type of the server interface, searching in the preset maintenance suggestion database to determine a set of maintenance suggestion entries;

[0092] S232, according to the priority of the maintenance suggestion entries, screening the maintenance suggestion entries with the highest priority from the set of maintenance suggestion entries as the target maintenance suggestion;

[0093] S233, generating alarm information carrying the interface identifier, the fault type and the target maintenance suggestion, and sending the alarm information to the man-machine interaction interface through a preset communication interface.

[0094] In the embodiment, after the type of the fault of the server interface is acquired, in order to further improve the efficiency and accuracy of the fault response, the interface identifier of the server interface (that is, the type of the interface of the server) is combined as a retrieval condition to retrieve the preset maintenance suggestion database. The interface type can include a Peripheral Component Interconnect Express (PCIe), a memory interface, and a Serial Advanced Technology Attachment (SATA), and the fault type includes, for example, “poor contact”, “protocol negotiation failure”, and “frequent link fluctuation”. Based on the two conditions, a set of maintenance suggestion entries matched with the current scenario is acquired, and each entry corresponds to a solution to a fault scenario.

[0095] To improve the actual execution efficiency, after the set of maintenance suggestion entries is acquired, the entries are further filtered according to the priority field included in the entries. The priority field is set by the maintenance suggestion database when the database is constructed, and reflects the recommended order of the entries when the entries are processed, and is usually in the form of a numerical level or a weight score. The maintenance suggestion entry with the highest priority is selected from the set as the target maintenance suggestion of the current server interface. For example, when facing the fault type of poor contact of the interface, if there are three maintenance suggestion entries of replugging the interface, replacing the interface card slot, and checking the solder joints of the mainboard, the replugging of the interface is selected as the target maintenance suggestion. Finally, the identifier information of the server interface, the corresponding fault type, and the determined target maintenance suggestion are packaged as alarm information. The alarm information includes the identifier of the interface (such as the device ID and the interface number) where the exception occurs, the fault classification (such as link fault, protocol exception, etc.), and the disposal suggestion. The alarm information is pushed to the man-machine interface in real time through a preset communication interface for viewing and execution by the operation and maintenance personnel. Through the embodiment, the automatic generation and pushing of the maintenance suggestion based on the fault situation are realized, the operation and maintenance reaction speed is improved, and the problem of low efficiency of interface exception troubleshooting is solved.

[0096] On the basis of the above-mentioned embodiments, another implementation manner is provided for acquiring the fault type, in addition to acquiring the fault type by querying the database. The implementation manner includes: if the pressure parameter and / or the interface signal parameter meet the preset fault judgment rule, then the fault type of the server interface is acquired according to the pressure parameter and the interface signal parameter by using a pre-trained fault type detection model.

[0097] In the embodiment, a pre-trained fault type detection model based on machine learning is used to judge the current data. The model learns a large amount of historical interface fault data in the training stage, and the data samples contain multi-dimensional features such as pressure change pattern, signal disturbance characteristics, protocol error statistics, and connection state fluctuation. Through supervised training, the mapping relationship between features and fault types is established, thereby having strong fault type discrimination ability. Specifically, the pressure parameters and interface signal parameters collected by the current server interface are input into the pre-trained model (or abnormal pressure features and abnormal signal features are input). The model first performs normalization processing and feature mapping conversion on the input data, and maps it to a vector expression form recognizable in the model structure. Then, the model will calculate the matching degree of each typical fault type in turn, and finally output the fault type label most consistent with the current features. For example, when the input pressure features show frequent transient load fluctuations, and the interface signal parameters show high cyclic redundancy check (CRC) error rate in the physical layer and abnormal increase in protocol layer negotiation times, etc., the model will determine this group of features as an interface fault of the contact failure type. As can be seen, inputting the collected multi-source parameters into the fault type detection model trained sufficiently realizes the automatic identification of the server interface fault type, which not only improves the accuracy of identification, but also shortens the response time of fault type identification.

[0098] Specifically, the fault type detection model adopts a lightweight deep neural network model based on multi-modal feature fusion, which is used to identify the fault type of the server interface. The model takes abnormal pressure features and abnormal signal features as input, comprehensively considers the contact state and signal transmission state of the server interface, and uses neural networks to complete fault classification analysis. Among them, the abnormal pressure features include contact load fluctuation mean, micro-vibration kurtosis, etc., and the abnormal signal features include signal amplitude change in the physical layer, error code frequency in the protocol layer, and connection stability parameters in the state layer, etc. In terms of model structure, the pressure features and signal features are respectively encoded, and each type of feature is processed by an independent fully connected encoding network to ensure that each type of feature has unified vector expression ability. Then, the two types of encoded vectors are input into the fusion layer, which uses an attention mechanism to weight different sources of abnormal features, highlight fault features, and enhance the model's ability to distinguish fault differences. The fused result enters the fault classification layer, which is composed of two fully connected networks, and outputs the corresponding fault type label.

[0099] The fault type detection model is supervised trained by using the labeled server interface exception samples, the training data includes the historical fault data collected from the actual server running environment and various typical exception samples injected through the simulation environment, so as to ensure that the model has good generalization ability. After the model is trained, it is deployed in the processing module of the server, and is used for real-time fault classification and judgment of each data exception event. The model improves the accuracy of fault identification and the degree of automation of processing, and solves the problems of poor real-time performance and high error rate caused by the traditional method of relying on log analysis and manual judgment. It should be noted that the fault type of the server interface is determined by the pre-trained fault type detection model, which is a supplementary solution for obtaining the fault type from the database. The database is established based on the actual and determined fault conditions, and has strong practical significance. Therefore, the database is preferred to determine the fault type. If the matching degree of the fault type obtained from the fault type database with the pressure parameter and / or the interface signal parameter is lower than the preset matching degree threshold, the fault type detection model can be used to obtain the fault type. For those skilled in the art, the requirements can be flexibly set, and this is only an exemplary example.

[0100] Figure 5 The interface health analysis method provided by the embodiment of the application is shown in the flowchart. Based on the above-mentioned embodiment, as shown in Figure 5 , it includes:

[0101] S51, obtaining the interface pressure parameter, physical layer parameter, protocol layer parameter and interface state parameter of the server interface in a preset time window;

[0102] S52, according to the interface pressure parameter, physical layer parameter, protocol layer parameter and interface state parameter, respectively adopting the pressure feature extraction rule, physical layer feature extraction rule, protocol layer feature extraction rule and interface state feature extraction rule, obtaining the pressure stability feature, interface signal fluctuation feature, protocol error code distribution feature and connection state change frequency feature through feature extraction;

[0103] S53, according to the pressure stability feature, interface signal fluctuation feature, protocol error code distribution feature and connection state change frequency feature, adopting the health degree calculation rule, generating the health score of each server interface;

[0104] S54, when the health score of the server interface is lower than the preset health score threshold, issuing a warning notification and sending it to the man-machine interaction interface.

[0105] In the embodiment, within a preset time window, the corresponding interface pressure parameters, physical layer parameters, protocol layer parameters and interface state parameters of the target server interface are continuously collected and recorded to form a time series data set, so as to ensure that subsequent analysis has sufficient context basis. Next, the four types of data are structured according to the feature extraction rules applicable to different types of data. For example, for the pressure parameters, the pressure mean, standard deviation and fluctuation frequency in the time window are calculated by using the pressure feature extraction rule, and finally the pressure stability feature representing the stability thereof is obtained; for the physical layer parameters, the interface signal fluctuation features are extracted by analyzing the signal amplitude change, bit error rate curve slope and the like; the protocol error code distribution features are generated by counting the distribution density of various error codes and the renegotiation frequency; and in the interface state parameters, the connection state change frequency features are extracted by calculating the frequency of connection interruption and the number of online and offline switching.

[0106] After the feature extraction is completed, according to the preset health degree calculation rule, the four features are input into a health score model, and a weighted sum is used to generate the health score of the server interface in the current time window, for example, between 0 and 100. In the scoring rule, the weight of each feature can be flexibly configured according to the server model, interface type or user demand, so as to improve the adaptability and accuracy of the health degree analysis. When it is detected that the health score of a certain interface is lower than the set health score threshold (for example, lower than 60), it is determined that the interface may be in an abnormal or degraded state. At this time, the early warning mechanism is automatically triggered, and information including the interface identifier, score details and abnormal features is constructed, which is displayed to the operation and maintenance personnel through a preset human-computer interaction interface. By setting the multi-dimensional feature fusion and scoring mechanism, the comprehensive quantitative analysis of the server interface state is realized, and the timeliness of abnormal detection and the foresight of operation and maintenance response are effectively improved.

[0107] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment.

[0108] The embodiment of the present application also provides a server comprising the above detection system.

[0109] The server comprises a parameter acquisition module and a processing module; the parameter acquisition module is configured to acquire pressure parameters and interface signal parameters at a server interface, and send the pressure parameters and the interface signal parameters to the processing module; the processing module is configured to receive the pressure parameters and the interface signal parameters at the server interface, and determine whether the pressure parameters and / or the interface signal parameters satisfy a preset fault determination rule; the processing module is further configured to, if the pressure parameters and / or the interface signal parameters satisfy the preset fault determination rule, acquire a fault type of the server interface according to the pressure parameters and / or the interface signal parameters by searching a preset fault type database, and send an interface identifier of the server interface, the fault type, and a target maintenance suggestion corresponding to the fault type to a man-machine interactive interface.

[0110] Figure 6 A structural schematic diagram of a detection device provided by an embodiment of the application is shown in FIG. 1. Figure 6 As shown in FIG. 1, the embodiment of the application further provides a server interface detection device 6, which comprises:

[0111] A data receiving unit 61 configured to receive pressure parameters and interface signal parameters at a server interface, and determine whether the pressure parameters and / or the interface signal parameters satisfy a preset fault determination rule;

[0112] A fault identification unit 62 configured to, if the pressure parameters and / or the interface signal parameters satisfy the preset fault determination rule, acquire a fault type of the server interface according to the pressure parameters and / or the interface signal parameters by searching a preset fault type database;

[0113] A fault alarm unit 63 configured to send an interface identifier of the server interface, the fault type, and a target maintenance suggestion corresponding to the fault type to a man-machine interactive interface.

[0114] The features of the embodiment corresponding to the detection device 6 can be referred to the related descriptions of the embodiment corresponding to the detection method, which will not be repeated here.

[0115] Figure 7 A structural schematic diagram of an electronic device provided by the application is shown in FIG. 2. Figure 7 As shown in FIG. 2, the electronic device 7 provided by the embodiment comprises at least one processor 71 and a memory 72. Optionally, the electronic device 7 further comprises a communication component 73. The processor 71, the memory 72, and the communication component 73 are connected through a bus 74.

[0116] In the specific implementation process, the at least one processor 71 executes computer execution instructions stored in the memory 72, so that the at least one processor 71 executes the detection method embodiments described above.

[0117] The specific implementation process of the processor 71 can refer to the method embodiments described above, which have similar implementation principles and technical effects, and thus will not be described here.

[0118] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in the application can be directly embodied as hardware processor execution, or executed by a combination of hardware and software modules in the processor.

[0119] The memory can include a random access memory (RAM), and can also include a non-volatile memory (NVM), such as at least one disk memory.

[0120] The bus can be an industry standard architecture (ISA) bus, a peripheral component (PCI) bus, an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0121] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, wherein the computer program is configured to execute the steps in any of the above detection method embodiments when running.

[0122] In an example embodiment, the above computer readable storage medium can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various computer program storage media.

[0123] The embodiment of the present application further provides a computer program product, which comprises a computer program, and the computer program realizes the steps in any one of the detection method embodiments when executed by a processor.

[0124] The embodiment of the present application further provides another computer program product, which comprises a nonvolatile computer readable storage medium, and the nonvolatile computer readable storage medium stores a computer program, and the computer program realizes the steps in any one of the detection method embodiments when executed by a processor.

[0125] Those skilled in the art can further understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in general in the above description. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0126] The above describes in detail a server interface detection method, system, server, device, medium and product provided by the present application. The principles and implementation modes of the present application are described by applying specific examples in this paper. The above description of the embodiments is only applicable to help understand the method and its core idea of the present application. It should be pointed out that for those skilled in the art, without departing from the principles of the present application, some improvements and modifications can be made to the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. A server interface detection method, characterized in that, include: Receive pressure parameters and interface signal parameters from the server interface, and determine whether the pressure parameters and interface signal parameters meet the preset fault judgment rules; The pressure parameters include the contact pressure value and the pressure fluctuation frequency; The interface signal parameters include at least one of physical layer parameters, protocol layer parameters, and interface status parameters; the physical layer parameters include jitter time, bit error count, signal eye diagram, level amplitude, and clock amplitude; the protocol layer parameters include protocol error code count, protocol frame retransmission count, link state machine transition count, uncorrected error cumulative count, and error correction code cumulative count; the interface status parameters include link rate, link width, power consumption status, power management level, and bandwidth utilization. If the pressure parameter and the interface signal parameter meet the preset fault judgment rules, then based on the pressure parameter and the interface signal parameter, the fault type of the server interface is obtained by searching the preset fault type database; Send the interface identifier of the server interface, the fault type, and the target maintenance suggestion corresponding to the fault type to the human-computer interaction interface; After the pressure parameters and interface signal parameters meet the preset fault judgment rules, the following steps are also included: Pressure features that are outside the preset pressure threshold range are marked as abnormal pressure features; Physical layer features that are not within the preset physical layer threshold range, protocol layer features that are not within the preset protocol layer threshold range, and interface state features that are not within the preset state threshold range are all marked as abnormal signal features. The step of obtaining the fault type of the server interface by searching a preset fault type database based on the pressure parameters and the interface signal parameters includes: Based on the pressure parameters and interface signal parameters corresponding to the server interface, abnormal pressure characteristics and abnormal signal characteristics are obtained, and based on the abnormal pressure characteristics and abnormal signal characteristics, a retrieval feature group corresponding to the server interface is constructed. Based on the retrieval feature group, the fault type corresponding to the server interface is retrieved from the preset fault type database.

2. The detection method according to claim 1, characterized in that, Determining whether the pressure parameter meets the preset fault judgment rules includes: Obtain at least one pressure feature corresponding to the pressure parameter. If any of the pressure features is not within its corresponding preset pressure threshold range, then the pressure parameter satisfies the preset fault judgment rule.

3. The detection method according to claim 2, characterized in that, After receiving the pressure parameters and interface signal parameters from the server interface, it also includes: According to the preset sampling period, the pressure parameters corresponding to the server interface are sorted by time to obtain the pressure sampling dataset; Based on a preset sliding time window, and using preset pressure feature extraction rules, features are extracted from the pressure sampling dataset to obtain at least one pressure feature.

4. The detection method according to claim 1, characterized in that, After receiving the pressure parameters and interface signal parameters from the server interface, it also includes: According to the preset sampling period, the physical layer parameters, protocol layer parameters and interface status parameters of the server interface are sorted by time to obtain the physical layer sampling dataset, protocol layer sampling dataset and interface status sampling dataset. Based on a preset sliding time window, preset physical layer feature extraction rules, protocol layer feature extraction rules, and interface state feature extraction rules are used respectively to extract features from the physical layer sampling dataset, the protocol layer sampling dataset, and the interface state sampling dataset, thereby obtaining at least one physical layer feature, at least one protocol layer feature, and at least one interface state feature. Then, it is determined whether the interface signal parameters meet the preset fault judgment rules, including: Obtain at least one physical layer feature corresponding to the physical layer parameter; if any physical layer feature is not within its corresponding physical layer threshold range, then the interface signal parameter satisfies a preset fault judgment rule; or, Obtain at least one protocol layer feature corresponding to the protocol layer parameter. If any of the protocol layer features is not within its corresponding protocol layer threshold range, then the interface signal parameter satisfies a preset fault judgment rule; or, Obtain at least one interface state feature corresponding to the interface state parameter. If any of the interface state features is not within its corresponding state threshold range, then the interface signal parameter satisfies the preset fault judgment rule.

5. The detection method according to claim 1, characterized in that, The step of sending the interface identifier of the server interface, the fault type, and the target maintenance suggestion corresponding to the fault type to the human-computer interaction interface includes: Based on the interface identifier and fault type of the server interface, a search is performed in the preset maintenance suggestion database to determine the set of maintenance suggestion entries; Based on the priority of the maintenance suggestion items, the highest priority maintenance suggestion item is selected from the set of maintenance suggestion items as the target maintenance suggestion; An alarm message carrying the interface identifier, the fault type, and the target maintenance suggestion is generated, and the alarm message is sent to the human-machine interface through a preset communication interface.

6. The detection method according to claim 1, characterized in that, Also includes: If the pressure parameters and the interface signal parameters meet the preset fault judgment rules, then the fault type of the server interface is obtained by using a pre-trained fault type detection model based on the pressure parameters and the interface signal parameters.

7. The detection method according to claim 4, characterized in that, The acquisition of at least one physical layer feature, at least one protocol layer feature, and at least one interface state feature includes: Based on the physical layer sampling dataset within the current time window, at least one of the following physical layer features is extracted using the preset physical layer feature extraction rules: jitter mean square value, bit error rate, and eye diagram opening degree. Based on the protocol layer sampling dataset within the current time window, at least one of the following is extracted as a protocol layer feature: the number of error codes, the protocol frame retransmission ratio, and the handshake failure rate, using the preset protocol layer feature extraction rules. Based on the interface status sampling dataset within the current time window, at least one of the following is extracted as an interface status feature: the amplitude of link width change, the amplitude of power consumption fluctuation, and the amplitude of bandwidth utilization fluctuation, using a preset interface status feature extraction rule:

8. A server interface detection system, characterized in that, include: Parameter acquisition module and processing module; The parameter acquisition module is used to acquire the pressure parameters and interface signal parameters at the server interface, and send the pressure parameters and interface signal parameters to the processing module; The processing module is used to receive pressure parameters and interface signal parameters at the server interface, and determine whether the pressure parameters and interface signal parameters meet preset fault judgment rules; wherein, the pressure parameters include contact pressure value and pressure fluctuation frequency; the interface signal parameters include at least one of physical layer parameters, protocol layer parameters, and interface status parameters; the physical layer parameters include jitter time, bit error count, signal eye diagram, level amplitude, and clock amplitude; the protocol layer parameters include protocol error code count, protocol frame retransmission count, link state machine transition count, uncorrected error cumulative count, and error correction code cumulative count; the interface status parameters include link rate, link width, power consumption status, power management level, and bandwidth utilization. The processing module is further configured to, if the pressure parameter and the interface signal parameter satisfy a preset fault judgment rule, obtain the fault type of the server interface by searching a preset fault type database based on the pressure parameter and the interface signal parameter, and send the interface identifier of the server interface, the fault type, and the target maintenance suggestion corresponding to the fault type to the human-computer interaction interface. The processing module is also used to mark pressure features that are not within the preset pressure threshold range as abnormal pressure features; and to mark physical layer features that are not within the preset physical layer threshold range, protocol layer features that are not within the preset protocol layer threshold range, and interface state features that are not within the preset state threshold range as abnormal signal features. The processing module is further configured to obtain abnormal pressure features and abnormal signal features based on the pressure parameters and interface signal parameters corresponding to the server interface, and construct a retrieval feature group corresponding to the server interface based on the abnormal pressure features and abnormal signal features; and retrieve the fault type corresponding to the server interface from the preset fault type database based on the retrieval feature group.

9. A server, characterized in that, Including the detection system as described in claim 8.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the detection method as described in any one of claims 1 to 7 when executing the computer program.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of the detection method as described in any one of claims 1 to 7.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the detection method as described in any one of claims 1 to 7.

Images