Server management system using ai

The AI-powered server management system addresses the challenge of managing complex IT environments by proactively predicting and addressing server failures, reducing costs and improving operational efficiency through real-time alerts and standardized operations across multiple servers.

JP2024156646AInactive Publication Date: 2024-11-06GENIAI CO LTD

Patent Information

Application Number
JP2024070593
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-04-24
Filing Date
2024-04-24
Publication Date
2024-11-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The management of complex IT environments, particularly large-capacity servers, is hindered by frequent failures due to various factors, leading to operational inefficiencies, high costs, and potential downtime, especially in small companies lacking specialized server administrators, and existing server management systems fail to preemptively predict and address these issues effectively.

Method used

A server management system utilizing AI technology to analyze server status and failures, predict potential issues, and provide real-time alerts via text and email, while integrating with Redfish API for data collection and management across multiple servers, enabling standardized operations and proactive maintenance.

Benefits of technology

The system reduces operational costs, enhances security, and improves efficiency by preemptively predicting and addressing server failures, thereby minimizing downtime and operational disruptions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024156646000001_ABST
    Figure 2024156646000001_ABST
Patent Text Reader

Abstract

To prevent faults that are likely to occur in servers in advance to reduce damages due to the server faults.SOLUTION: A server management system managing two or more servers to be managed includes a management server which collects hardware-related data and software-related data from the servers to be managed, identifies and manages status of the servers to be managed, and provides various server management information including management service statistical data and a management service report related to the identified and managed status of the servers, to an administrator terminal used by an administrator and a customer terminal that has requested the server to be managed. The server analyzes the server to be managed by using AI technology, thereby predicting status and a fault of the server to be managed. When an occurrence of a trouble is predicted, the server sends an alarm message as a text message and e-mail to the administrator terminal and the customer terminal.SELECTED DRAWING: Figure 3
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a technique for managing multiple servers, and more particularly to a technique for managing multiple servers using AI (Artificial Intelligence) technology. [Background technology]

[0002] Recently, IT (Information Technology) environments such as servers, storage, and networks have become more complex, resulting in a lack of available work time. As computer systems become larger in capacity and faster, computer problems caused by system errors and viruses are becoming more frequent.

[0003] Especially in large-capacity servers, failures occur frequently due to various factors such as the operation of various application programs, storage, reading, and transfer of data, etc. Therefore, each company assigns separate server administrators to manage these servers and handle failures when they occur.

[0004] However, server management requires specialized skills, and hiring such specialized staff is quite costly. Therefore, particularly in small companies, the reality is that rather than hiring specialized engineers to be server administrators, they select suitable people from among the existing staff within the company and assign them as server administrators.

[0005] In such cases, server management is difficult to carry out smoothly, and it is almost impossible to deal smoothly with server failures. Even if a server administrator with specialized skills is employed for server management, if the server administrator is located far away from the servers due to a business trip or other reason, it is difficult for the administrator to be notified of the status of the servers promptly when a server failure occurs, making it difficult to deal smoothly with server failures. Even if the server administrator is notified of the occurrence of a server failure, it is difficult to deal with it immediately because he or she is located far away, and this can ultimately result in huge losses such as the server going down.

[0006] In the past, in a server integrated management system that integratedly manages multiple servers, if a failure occurs in a server, this is detected and the failure is recovered after the fact. However, in this conventional method of recovering from a failure, the operation of the server is interrupted during the recovery period, resulting in losses due to the interruption of use of the server, and there are problems with the loss due to the manpower and costs required for recovery. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Korean Patent Publication No. 10-2015-0124642 Summary of the Invention [Problem to be solved by the invention]

[0008] The present invention has been devised to solve the above-mentioned problems, and its purpose is to provide a server management system that uses AI (Artificial Intelligence) to systematize IT resources and standardize operations, thereby improving operational efficiency, reducing operational costs, and strengthening security.

[0009] The object of the present invention is not limited to the above object, and other objects not mentioned will be clearly understood by those skilled in the art from the following description. [Means for solving the problem]

[0010] To achieve this object, the present invention provides a server management system for managing two or more managed servers, which includes a database for storing data related to the managed servers, and a management server that collects hardware-related data and software-related data from the managed servers, grasps and manages the current state of each managed server, and provides various server management information including related management service statistical data and management service reports to an administrator terminal used by an administrator and a customer terminal that has requested the managed servers, and the management server uses AI technology to analyze the managed servers, thereby predicting the status and failures of the managed servers, and if a problem occurs through the prediction, transmits an alarm message to the customer terminal by text message and email.

[0011] The management server collects formatted and unformatted log data from each managed server, classifies the collected data, performs data pre-processing, and then performs learning through an AI learning data model, thereby being able to predict the status and faults of the managed servers.

[0012] The management server uses Redfish APIs to provide AI analysis capabilities that can learn what normal traffic is on each managed server, discover abnormal traffic, prioritize risks required by users, analyze problems, and provide assistance. Effect of the Invention

[0013] According to the present invention, by using AI analysis of a large number of managed servers, it is possible to preemptively predict server failures, provide warnings, and provide solutions, thereby preventing possible server failures and reducing damage caused by server failures.

[0014] Furthermore, according to the present invention, by systematizing IT resources and standardizing operations, it is possible to improve operational efficiency, reduce operational costs, and strengthen security.

[0015] Furthermore, the present invention has the effect of enabling a large number of servers to be managed more conveniently and efficiently.

[0016] Furthermore, according to the present invention, by analyzing failure patterns for a client who has requested server management, preemptively associating failures in advance, and providing server management functions, it is possible to process and transmit data suited to the client's needs. [Brief description of the drawings]

[0017] [Figure 1] 1 is a diagram conceptually illustrating an overall configuration of a server management system according to an embodiment of the present invention. [Diagram 2] 2 is a diagram conceptually illustrating an operation process in a server management system according to an embodiment of the present invention; [Diagram 3] 1 is a flowchart illustrating a method for implementing an AI (Artificial Intelligence) analysis function in a server management system according to an embodiment of the present invention. [Figure 4] 4 is a flowchart illustrating an AI learning process for shaping data among the AI ​​(Artificial Intelligence) analysis functions in a server management system according to an embodiment of the present invention. [Diagram 5] 1 is a flowchart illustrating an AI learning process for unstructured data among AI (Artificial Intelligence) analysis functions in a server management system according to an embodiment of the present invention. [Figure 6] 4 is a screen example showing functions provided by a server management system according to an embodiment of the present invention. [Figure 7] 4 is a screen example showing functions provided by a server management system according to an embodiment of the present invention. [Figure 8] 4 is a screen example showing functions provided by a server management system according to an embodiment of the present invention. [Figure 9] 4 is a screen example showing functions provided by a server management system according to an embodiment of the present invention. [Figure 10] 4 is a screen example showing functions provided by a server management system according to an embodiment of the present invention. [Figure 11] FIG. 1 is a diagram illustrating an example of a configuration of a server management system according to an embodiment of the present invention. [Figure 12] FIG. 11 is an exemplary diagram illustrating a server monitoring function using Redfish events in a server management system according to an embodiment of the present invention. [Figure 13] FIG. 11 is an exemplary diagram illustrating a server configuration job automation function using Redfish in a server management system according to an embodiment of the present invention. [Figure 14] FIG. 1 is an exemplary diagram illustrating a server configuration automation function using Redfish in a server management system according to an embodiment of the present invention. [Figure 15] 4 is a flowchart illustrating a method for managing a server by supporting multiple vendors in a server management system according to an embodiment of the present invention. [Figure 16] 4 is a flowchart illustrating a method for preventing failures by analyzing failure logs and patterns in a server management system according to an embodiment of the present invention. [Figure 17] 1 illustrates an operational model for supporting multi-vendors using Redfish API in a server management system according to an embodiment of the present invention. [Figure 18] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 19] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 20] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 21] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 22] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 23] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 24] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Diagram 25] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 26] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 27] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 28] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Figure 29] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Diagram 30] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Diagram 31] FIG. 2 is a diagram showing an example of a screen of a server management system according to an embodiment of the present invention. [Diagram 32] FIG. 2 is a diagram showing classification of system equipment according to an embodiment of the present invention. [Diagram 33] 1 is a chart illustrating hardware symptoms and their causes according to an embodiment of the present invention. [Diagram 34] 1 is a chart illustrating hardware symptoms and their causes according to an embodiment of the present invention. [Diagram 35] 4 is a flowchart illustrating a method for proactively responding to failures in a server management system according to an embodiment of the present invention. [Diagram 36] 4 is a flowchart illustrating a method for proactively responding to failures in a server management system according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] While the present invention can be modified in various ways and has various implementations, specific implementations are illustrated in the drawings and described in detail. However, it is understood that this is not intended to limit the present invention to the specific embodiments, but includes all modifications, equivalents and alternatives within the spirit and technical scope of the present invention.

[0019] The terms used in this application are merely used to describe examples in specific implementations and are not intended to limit the present invention. A singular expression includes a plural expression unless otherwise clearly meant in the context.

[0020] In this application, terms such as "comprise" or "have" are intended to specify the presence of features, numbers, steps, operations, components, parts, or combinations thereof described herein, and it is to be understood that they do not preclude the presence or additional possibility of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0021] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0022] Terms such as those commonly used and predefined should be construed to have a meaning consistent with the contextual meaning of the relevant art, and should not be construed as having an ideal or overly formal meaning unless expressly defined in this application.

[0023] In the description with reference to the accompanying drawings, the same components are denoted by the same reference numerals regardless of the reference numerals, and duplicated description thereof will be omitted. In the description of the present invention, if it is determined that a detailed description of related known technology may unnecessarily blur the gist of the present invention, the detailed description thereof will be omitted.

[0024] The present invention relates to a server management system for managing two or more managed servers, and includes an administrator terminal used by an administrator to collect hardware-related data and software-related data from the managed servers, grasp and manage the status of each managed server, and provide various server management information including related management service statistical data and management service reports to a customer terminal that has requested the managed server, and the server uses AI technology to analyze the managed servers, thereby predicting the status and failures of the managed servers, and if a problem occurs based on the prediction, transmits an alarm message via text message and email to the relevant administrator terminal and customer terminal.

[0025] The management server collects formatted and unformatted log data from each managed server, classifies the collected data, performs data pre-processing, and then performs learning through an AI learning data model, thereby being able to predict the status and faults of the managed servers.

[0026] The management server can use the Redfish API to provide AI analysis capabilities to learn what normal traffic is on each managed server, discover abnormal traffic, and provide users with the necessary risk prioritization, problem analysis, and assistance.

[0027] FIG. 1 is a diagram conceptually showing an overall configuration of a server management system according to an embodiment of the present invention, and FIG. 2 is a diagram conceptually showing an operation process in the server management system according to an embodiment of the present invention.

[0028] 1 and 2, the server management system of the present invention includes a management server 110, a database 112, an administrator terminal 120, and a customer terminal .

[0029] The administrator terminal 120 is a terminal used by an administrator who manages the server management system. The customer terminal 130 is a terminal used by each customer who has requested the managed servers 10, 20, 30, 40. In one embodiment of the present invention, the administrator terminal 120 and the customer terminal 130 can be realized in various terminal forms capable of wired and wireless communication, such as a desktop computer, a laptop computer, a tablet PC, a mobile phone, a smartphone, etc. In one embodiment of the present invention, a user terminal is a concept that includes the administrator terminal 120 and the customer terminal 130.

[0030] The database 112 stores data relating to the managed servers 10 , 20 , 30 , 40 .

[0031] The management server 110 collects data from the managed servers 10, 20, 30, 40, grasps and manages the current status of each managed server, including related management service statistical data and management service reports. The management server 110 provides various server management information to the administrator terminal 120 and the customer terminal 130.

[0032] The management server 110 can collect and store multi-vendor hardware information from multiple managed servers, and provide the stored information to the administrator terminal 120 and the customer terminal 130 so that the information can be referenced and used.

[0033] The management server 110 can collect and store multi-vendor hardware inventory information from multiple registered managed servers.

[0034] The management server 110 can facilitate firmware updates for all managed servers upon firmware update events, including emergency firmware updates.

[0035] The management server 110 can analyze logs and patterns when a failure occurs in any of the equipment of the managed servers, store the analyzed data, and when the failure is resolved, classify equipment similar to the failed equipment and perform pre-failure response processing for the classified similar equipment.

[0036] The management server 110 can use the Redfish API to collect information about the x86 servers in operation, including detailed hardware specifications, OS (Operating system) information, firmware information, and driver information of each managed server, and can perform standardized management of the x86 servers.

[0037] The management server 110 provides a preventive analysis function that analyzes failure patterns of the managed servers 10, 20, 30, and 40 and prevents similar failures from occurring. Through the preventive analysis function, when a predetermined event occurs in the managed servers 10, 20, 30, and 40, the management server 110 can preemptively send a predicted failure occurrence message to a client terminal that has requested the managed server, warning that a failure may occur due to the event.

[0038] The management server 110 can provide a history management function for managing the installation, failure, and technical support history of the managed servers 10 , 20 , 30 , and 40 .

[0039] The management server 110 can provide a delivery management function for managing the delivery history of the managed servers 10 , 20 , 30 , and 40 .

[0040] When an equipment-related event occurs in a managed server, the management server 110 classifies the dangerous equipment according to predetermined classification criteria, sends a warning message regarding the dangerous equipment to the administrator terminal 120 and the corresponding customer terminal, and can take pre-emptive troubleshooting measures for the dangerous equipment.

[0041] When an equipment-related event occurs in a managed server, the management server 110 can grasp the fault symptoms of the equipment, analyze the cause according to the fault code corresponding to the fault symptoms, send a report including a fault response plan to the administrator terminal 120 and the corresponding customer terminal, and execute fault response measures for the equipment.

[0042] In the present invention, the management server 110 can provide a data delivery service function for processing and delivering data related to management of the managed servers upon request of the client terminal 130 .

[0043] In addition, the management server 110 can analyze critical failures of the managed servers and propagate the same cases to prevent server failures in advance, and can provide failure statistics of each server to the administrator terminal 120 and the customer terminal 130 on a quarterly basis.

[0044] In the present invention, the management server can manage the history of delivered server-related equipment, provide installation / failure / technical support history management services, and manage problems by part.

[0045] The present invention relates to a server management system that manages a plurality of managed servers 10, 20, 30, and 40 at the request of a customer.

[0046] In one embodiment of the present invention, the managed server, which is the server to be managed, may be a variety of servers, such as a Dell (registered trademark) server 10, an HP (registered trademark) server 20, a Lenovo (registered trademark) server 30, or an x86 server 40.

[0047] The managed servers 10, 20, 30, 40 and the management server 110 can communicate with each other via various wired or wireless communication methods, for example, HTTP communication or a POST transfer method in JSON format.

[0048] Additionally, the managed servers 10, 20, 30, 40 can automatically execute scripts according to a defined schedule across various x86 servers in a large scale computing environment.

[0049] The administrator connects to the management server 110 via the administrator terminal 120, executes a batch program according to a schedule defined by the management server 110, and manages the change history by comparing it with existing data.

[0050] The management server 110 automatically collects hardware and software information of the managed servers 10, 20, 30, 40, and based on this, grasps the current state of each server and provides management services according to the request status of each server.

[0051] FIG. 2 is a conceptual diagram illustrating the operational process of a server management system according to one embodiment of the present invention. In FIG. 2, the managed server is a Dell (registered trademark) server 10 to which the iDRAC9 version is applied, and an example of a platform using the Redfish API (Application Programming Interface) is shown.

[0052] 2, a Get Module is executed using Flask on a user terminal, and Redfish API is used to collect iDRAC9 formatted data and unformatted data from the Dell server 10, classify the collected data, and perform data preprocessing. The preprocessed data is then stored in the database 112, and learning is performed on the data stacked in the database through an AI learning data model, data is reclassified, and data rows are generated.

[0053] Additionally, a page is called up on the user terminal using Flask, and a data analysis module searches the database 112, performs analysis, and performs data visualization, which is then transmitted to a Flask Response User Web page.

[0054] 3 is a flowchart showing a method for implementing an AI (Artificial Intelligence) analysis function in a server management system according to an embodiment of the present invention. The embodiment of FIG. 3 is an embodiment using the Redfish API.

[0055] 3, the management server 110 collects data including log data from each managed server (S4010), classifies the collected data (S4020), and pre-processes the classified data (S4030). Then, AI learning is performed on the pre-processed data (S4040). Next, the equipment status of each managed server is diagnosed through AI analysis (S4050) and a fault is predicted (S4060). When a problem occurs in the managed server (S4070), an alarm message is sent to related terminals via text message, email, etc. (S4080).

[0056] 4 is a flowchart showing an AI learning process for shaping data among the AI ​​(Artificial Intelligence) analysis functions in a server management system according to an embodiment of the present invention. The embodiment of FIG. 4 is an embodiment using Redfish API.

[0057] 4, when a log is generated in a managed server (S5010), the management server 110 judges whether or not there is a fault (S5020). If there is no fault, learning data indicating "no abnormality" is added to the log (S5080, S5090). If there is a fault, a server with similar specifications and model as the corresponding managed server is searched for (S5030) and the log of the server is extracted (S5040). Then, the log of the managed server is compared with the extracted log (S5050). If the comparison results in the same pattern, learning data indicating "abnormality" is added to the log (S5060, S5070, S5090), otherwise learning data indicating "no abnormality" is added (S5060, S5080, S5090).

[0058] 5 is a flowchart showing an AI learning process for unstructured data among the AI ​​(Artificial Intelligence) analysis functions in a server management system according to an embodiment of the present invention. The embodiment of FIG. 5 is an embodiment using Redfish API.

[0059] 5, the management server 110 collects data from the managed servers according to a preset period (S6010). For example, if the period is set to 10 seconds, data is collected from the managed servers at 10-second intervals.

[0060] Thereafter, the management server 110 performs preprocessing on the collected data (S6020) and stores it in a database (S6030). Next, the management server 110 analyzes the atypical data using the AI ​​learning data model (S6040). If the analysis reveals an anomaly in the atypical data, it adds learning data with "anomaly" to the atypical data (S6050, S6060, S6080), otherwise it adds learning data with "no anomaly" (S6050, S6070, S6080).

[0061] Figures 6 to 10 are example screens showing the functions provided by a server management system according to an embodiment of the present invention: Figure 6 is an example of a main dashboard screen.

[0062] Referring to FIG. 6, the management server 110 provides a main dashboard screen that displays important information on a single screen based on resource information collected from the managed servers 10, 20, 30, and 40 and the number of registered performance items.

[0063] The present invention supports deep analysis and continuous monitoring of specific information, and provides various information such as what equipment the user frequently uses, what task is taking up a lot of time, and whether stabilized firmware is applied to each component of the managed server through a dashboard screen, allowing the user to check important managed server information at a glance through the dashboard screen.

[0064] The example screen in Figure 6 displays server, storage, and network operation status information, providing pie charts for the total number of servers in operation and the number of servers by manufacturer. It also provides monthly performance status information, providing bar graphs for the number of tasks, changes, and failures.

[0065] In addition, when displaying information on the stabilization firmware application status, a chart is provided regarding the stabilization application rate, which is the ratio of devices with stabilization firmware applied and devices without stabilization firmware applied, such as BIOS, R / C, NIC, iDRAC, and HBA.

[0066] 7 is an example of a screen displaying the resource management function. In the present invention, the management server 110 provides a resource management function that automatically collects and organizes a list of new equipment installations and changes to servers, etc., and provides reliable data in real time.

[0067] In its resource management function, the management server 110 can collect information registered from user terminals and automatically collect resource information of servers in a data center according to a predefined period via a standardized Redfish RESTful API.

[0068] The screen example in Figure 7 displays equipment information and allows users to register or view equipment information for servers, storage, networks, SANs, backup equipment, discarded equipment, etc. Also, to provide related statistical information, it provides pie charts for equipment status such as operating, idle, pre-service, discarded, etc., and various statistical graphs such as the current status of operating equipment by year and vendor, a list of recently registered equipment, and additional customization methods.

[0069] 6 is an example of a screen displaying the performance management function. In the present invention, the management server 110 provides a performance management function for managing scheduled tasks, items changed by tasks, etc., history after a failure occurs, and improvement results. As a result, in the present invention, when the cause of a failure is clear, it is possible to record and manage the failure so that the same failure does not occur again, assign a person in charge to items that need improvement, and check the improvement results.

[0070] In addition, it can provide various performance status statistics according to the state such as year, month, data center location, pre-operation service, idle, etc.

[0071] The example screen in Figure 8 displays job history including online and offline job history management, fault history from the fault processing history manager, change history from the system change history manager, and various statistical graphs relating to backup schedule management and business performance.

[0072] 9 is an example of a screen displaying an automation management function. In the present invention, the management server 110 provides an automation management function that provides notification information on equipment requiring inspection through daily automatic inspection, and manages execution periods by group for automated collection such as setting a synchronization period (Daily / Weekly / Monthly), setting automated collection values ​​(All / Chassis / MGMT / CPU / NIC / HBA / DISK / GPU, etc.), and registering schedule information, through a standardized Redfish RESTful API.

[0073] The example screen in Fig. 9 shows a menu for daily inspections that allows you to set the collection synchronization cycle, customize the automated collection values, and register collection schedule information, as well as automatically classify equipment that requires daily inspections and check equipment with MGMT (Management Repository) connection errors.

[0074] 9, the management server 110 can display the equipment in different colors depending on the equipment status in the daily inspection menu. That is, if there is no abnormality in the equipment, it can be displayed with symbol 1 (green circle), if the equipment requires inspection by the administrator, it can be displayed with symbol 2 (red circle), if the equipment requires inspection by the naked eye, it can be displayed with symbol 3 (yellow circle), if the equipment requires inspection by the naked eye, it can be displayed with symbol 4 (gray circle), if the equipment cannot be connected to MGMT, it can be displayed with symbol 4 (gray circle).

[0075] 10 is an example of a screen displaying the automation management function. In the present invention, the management server 110 provides a configuration diagram management function, which is a configuration diagram view function required for efficiently operating and managing an IT infrastructure environment such as servers, storage, networks, SANs, etc., which are IT infrastructure components.

[0076] That is, the management server 110 also provides a configuration management function that automatically displays a view of the configuration of resources selected from a user terminal, such as servers, storage, networks, SANs, etc., thereby enabling faster decision-making when performance issues and failures occur.

[0077] Referring to FIG. 10, the configuration diagram management function provides a view function of the configuration diagram of the equipment (server, storage, network, SAN, etc.) selected from the user terminal, and provides a search and selection function based on hostname and equipment model, allowing the infrastructure configuration to be checked in real time when performance problems or failures occur.

[0078] Fig. 11 is a diagram showing a configuration example of a server management system according to an embodiment of the present invention. In the configuration example of Fig. 11, the Redfish API is used, the managed servers are connected via the MGMT network, and the administrator terminal 120 can connect to the managed servers via a Web connection method.

[0079] In one embodiment of the present invention, the server management system collects hardware system inventory information of multi-vendor x86 servers in real time using a Redfish API-based platform, and distributes BIOS settings, firmware, etc. This improves maintenance efficiency and reduces operating costs. In addition, similar equipment can be identified based on the collected logs, allowing the same failures to be prevented in advance.

[0080] 12 is an exemplary diagram illustrating a server monitoring function using a Redfish event in a server management system according to an embodiment of the present invention. Referring to FIG. 12, the management server 110 in the present invention can provide a server monitoring function using a Redfish event.

[0081] Redfish events transfer server event information to Redfish clients based on HTTPS, and when an alarm occurs in management, it is transferred to HTTP POST and can be received via HTTP GET. At this time, you can select the target servers for important notification email push, status monitoring, and daily inspections, and load the required data.

[0082] 13 is an exemplary diagram illustrating a server configuration job automation function using Redfish in a server management system according to an embodiment of the present invention. Referring to FIG. 13, in the present invention, a management server 110 can provide a server configuration job automation function using Redfish.

[0083] This function allows you to locally distribute and update BIOS setting changes, Secure Boot, iDRAC Configuration, etc. It also provides management and updates of firmware inventory for managed servers, and shortens deployment time by applying BIOS standard settings and management standard settings collectively during server deployment, and prevents input of incorrect settings with the automated management function. It also provides a function to update firmware information installed on managed servers according to a preset cycle, automatically select target equipment when emergency firmware is distributed, and push emails to administrators.

[0084] 14 is an exemplary diagram illustrating a server configuration automation function using Redfish in a server management system according to an embodiment of the present invention. In the present invention, the management server 110 can provide a server configuration automation function using Redfish.

[0085] The unique setting values ​​of a server are stored as metadata in a Server Configuration Profile (SCP), and this metadata can be configured using the Redfish API in the present invention.

[0086] The SCP can be exported, previewed, and imported, and by utilizing this, the server configuration automation function of the present invention can apply configuration information to a newly constructed server.

[0087] SCP can be shared via HTTPS, NFS, CIFS, and other methods, and is implemented in XML and JSON formats. Using the SSH protocol when configuring the server allows for stable and consistent distribution of many applications.

[0088] In the present invention, unique setting values ​​for physical server distribution are stored as metadata in a file sharing server in XML or JSON format, and the configuration information can be automatically applied to newly constructed servers connected to the management network. In this way, the configuration automation function of the present invention allows operators to quickly configure new servers without having to connect to each server separately for new server configuration.

[0089] In one embodiment of the present invention, an AI (Artificial Intelligence) analysis function using Redfish is provided, which collects formatted and unformatted log data from servers and storage devices using SRC (Server remote control) (iDRAC, iLO, IPMI), and performs data classification and pre-processing.

[0090] The learned data model is then used to predict equipment status and faults, and when important problems occur, alarm messages are sent to the user's device via text, email, etc.

[0091] The AI ​​analysis capabilities of the present invention can learn what normal traffic is, discover abnormal traffic, prioritize the required risk levels for users, and analyze and assist with problems.

[0092] It also uses AI to analyze and learn from logs collected during server operation, develops algorithms, and provides a solution to problems by sending an alert message to the client terminal 130 when log information similar to an existing problem is confirmed through the learned algorithm. In other words, the AI ​​analysis function can prevent problems in advance, detect problem occurrences, quickly share information, and perform real-time analysis.

[0093] The management server 110 checks the BBU (Backup Battery Unit) cycle of the managed server, and when a predetermined cycle is reached, it can transfer the contents of the cycle to the client terminal of the managed server.

[0094] The management server 110 checks the charging capacity of a backup battery unit (BBU) of a managed server, and when the charging efficiency of the battery drops below a predetermined value, it can notify the client terminal of the managed server of this information.

[0095] In addition, the management server 110 can check the BBU charging capacity of the managed server, and when the battery charging efficiency drops below 40%, can notify the client terminal of the managed server of this information.

[0096] For example, the management server 110 can check the BBU charging capacity of a managed server, and if the remaining battery power is below a predetermined value, can notify the customer terminal of the managed server of the content. For example, the management server 110 can check the BBU remaining power of a managed server, and if the remaining battery power is below a predetermined value, can notify the customer terminal of the managed server of the content.

[0097] In addition, the management server 110 can check the BBU write policy of the managed server, and if the write policy is changed, can notify the client terminal of the managed server of the change.

[0098] The present invention relates to a server integrated management system that integrates and manages a large number of servers, diagnoses various server functions, predicts failures in advance, issues warnings, and presents solutions.

[0099] In this invention, among various functions of a server, a BBU (Backup Battery Unit) will be exemplified for explanation. Taking a Dell (registered trademark) server as an example, in order to prevent the loss of cache data due to a battery failure in a RAID controller, it is necessary to check the status of the BBU battery and proceed with preemptive replacement.

[0100] For this purpose, the battery full charging efficiency (%) is checked by checking the Dell server log, and if any equipment has a full charging efficiency of less than 50%, a battery replacement is recommended. After 36 months, the battery charging efficiency naturally decreases to about 70%, and taking this into consideration, a battery with an additional decrease of about 20% can be determined to have poor charging efficiency.

[0101] The server integrated management system of the present invention performs periodic BBU checks, charge capacity checks, remaining capacity checks, and write policy checks, thereby preventing cache data loss and preventing risk factors of battery conditions in advance.

[0102] The server management system of the present invention diagnoses, when an event occurs, that the event may cause a failure in the server, warns the server system in advance, and transmits information regarding a solution together with the warning.

[0103] In this regard, events that occur in a server are very diverse, and a new event that has never occurred before can be generated. In the present invention, some of the events that can occur in such a server are exemplified.

[0104] 1. Fan Noise (Reading above 12,000 RPM) on Dell R720 Servers with iDRAC7 version 1.51.51 The recommended solution for this is to downgrade to iDRAC7 version 1.46.45.

[0105] 2. Rack Power usage of PDU#1 and PDU#2 is concentrated in PDU#1 Referring to FIG. 34, not only Dell® servers but also HP® servers are set to operate in Active Standby as the default for the power supply, which results in a situation where power is concentrated on one side of the Rack PDU, so in order to adjust the balance, the Primary-PSU ratio needs to be adjusted.

[0106] 3. Abnormal OS behavior after 12th to 14th generation kernel updates for Dell(R) server products In this case, if abnormal operation is detected in the OS (Operating System) after a kernel update on a Dell (registered trademark) server, the management server 110 sends a message about a predicted fault that may occur as a result to the corresponding managed server, and also transmits a solution to the predicted fault to the corresponding managed server.

[0107] 4. Service unavailable due to insufficient TCP / IP ports This is a phenomenon where a network TIME_WAIT session remains unclosed if the uptime is 497 days or more on Windows 2008. This occupies a port and becomes a problem when there are no more ports available. This applies to Windows 2008 Server and Windows 2012 Server, and the problem can be resolved by deleting the updated patch.

[0108] 5. Windows(R) 2003~2022 Event Log Generation 6. Memory production cycle diagnostics This confirms that a specific production cycle of a specific memory is faulty. The affected equipment is 13th generation equipment (R730, R930, R630), the affected OS is Windows 2012 R2 Server with the KB3064209 hotfix, and the solution is to remove the corresponding hotfix.

[0109] In the present invention, the management server 110 diagnoses the memory production cycle of the managed servers, determines that a predetermined memory production cycle is defective, and notifies the corresponding managed server of this result.

[0110] 7. When using a PCIe Type SSD, the device settings may stop responding. The solution to this is to update BIOS 1.1.4 -> 1.2.10.

[0111] After updating the BIOS on the 8.12G server, the temperature sensor did not work properly, causing the beep (Alert_) to continue. The solution to this is to diagnose the BIOS 2.5.2 version and update it with the latest firmware.

[0112] 9. After patch update, BSOD occurs and booting is not possible This event is a phenomenon caused by the Patch Tuesday update Windows error KB2982791 in August 2014. The affected system is a Windows 2008 server, and the error can be resolved through a patch update.

[0113] 10. DNS connection error occurs on a client using Windows 2012 Active Director When logging into a domain account from a server, an error "Incorrect username or password" occurs even if the account and password are correct. This is due to a product issue where AES key generation fails when updating a computer account password when the AD server is Windows Server 2012 R2 and the domain member is Windows Server 2008 R2 or Windows 7, but DES-CBC-MD5 and DES-CBC-CRC encryption is not used and only AES256-CTS-HMAC-SHA1-96, AES128-CTS-HMAC-SHA1-96, and RC4-HMAC encryption are used in Windows Server 2008 R2 and Windows 7, respectively.

[0114] 11. Vulnerabilities in GNU Bash 4.3 Shell It is known that Bash vulnerabilities allow attackers to modify web server content and code, tamper with websites, leak user data, and launch DDoS attacks. Other attack scenarios for Bash code injection vulnerabilities in various environments, such as SSH and DHCP protocols, have also been raised. The problem affects Red Hat Enterprise Linux(R) 5,6,7 servers, and the solution is a Bash update.

[0115] 12. Buffer overflow vulnerability in the GNU C library (glibc) When calling the gethostbyname() or gethostbyname2() functions, which are often used when connecting to a network, a vulnerable function is called, allowing an external attacker to remotely execute arbitrary code on the vulnerable server. The problem affects Red Hat Enterprise Linux(R) 5,6,7 servers, and the solution is a GLIBC update.

[0116] 13. Radhat V5 and V6 series OS bugs This is a bug that causes a reboot phenomenon to occur after 208.5 days on all versions of Red Hat Enterprise Linux(R) 6 or 5 that use Intel(R) CPUs. The problem affects Red Hat Enterprise Linux(R) 5 and 6 servers, and the solution is to update the kernel.

[0117] 14. Raid Controller Battery Fail I / O performance is reduced due to the inability to use the Raid Controller Cache. The problem is with the Raid Controller Battery for Dell(R) Perc 5i and 6i, and the solution to this problem is to proactively replace the Raid Controller Battery for Dell(R) Perc 5i and 6i every 4 to 5 years.

[0118] 15. System down due to CPU IERR error (SYSTEM DOWN) The problem was with a server (PE R720, PE R920) using the Intel (registered trademark) iBridge V2 CPU, and the solution was to change the BIOS settings. For example, set System Profile Settings to Custom, CPU Power Management to Maximum Performance, C1E to Disabled C States Disabled, and Monitor / Mwait to Disabled.

[0119] 16. Unable to connect to management web when using iDrac 1.50.50 F / W (Firmware) (search for applicable version) Upgrade to iDrac F / W (Firmware) 1.51.51 via F / W upgrade on the OS or via media upgrade in daily life.

[0120] This invention proposes a server management system that supports multi-vendors. For example, this invention stores information on hardware systems from three companies, such as Dell (registered trademark), HP (registered trademark), and Lenovo (registered trademark), in one inventory, and uses the information stored in the inventory to query all information related to the hardware and utilize the functions.

[0121] For the convenience of explaining the present invention, a server management system that supports multiple vendors will be described by taking manufacturers such as Dell (registered trademark), HP (registered trademark), and Lenovo (registered trademark) as examples.

[0122] 15 is a flowchart illustrating a method for managing servers by supporting multiple vendors in a server management system according to an embodiment of the present invention. In FIG. 15, the management server 110 is responsible for executing each step.

[0123] Referring to FIG. 15, the managed server is registered (S201). At this time, the target server can be registered using the management IP information of each server. For example, the target server can be registered using iDRAC for Dell (registered trademark), iLO for HP (registered trademark), or iMM for Lenovo (registered trademark).

[0124] Next, it is determined whether or not to connect to each server (S203), and multi-vendor hardware inventory information is collected (S205). In one embodiment of the present invention, the hardware inventory information of an x86 server hardware system can be collected without distinguishing between manufacturers by using the Redfish API (Application Programming Interface), which is a common hardware standard.

[0125] Then, the collected inventory information is stored (S207). If there is a firmware update event including an emergency firmware update, the firmware update is proceeded with for all managed servers (S209). Next, the changed update information is confirmed (S211). In one embodiment of the present invention, the firmware update information can be confirmed via the Redfish API.

[0126] Then, groups are set according to the security level of each server, whether or not it is subject to inspection, its importance, etc. (S215), and server information is checked in real time (S217). In this manner, in one embodiment of the present invention, various information related to the x86 servers in operation, such as detailed hardware specifications, OS (Operating system) information, firmware information, and driver information, can be collected by using the Redfish API, and standardized management of x86 servers can be performed.

[0127] 16 is a flowchart showing a method for preventing failures by analyzing failure logs and patterns in a server management system according to an embodiment of the present invention. The subject that executes each step in FIG. 16 is the management server 110.

[0128] 16, when a fault occurs in any device of the managed server (S401), the logs and patterns are analyzed (S403). The analyzed data is then stored (S405). When the fault is resolved (S407), devices similar to the fault are classified (S409), and a fault response process is performed on the classified similar devices in advance (S411). In this way, in the present invention, when a fault problem occurs, the logs and patterns are analyzed and similar devices are automatically classified, thereby making it possible to prevent faults occurring in similar devices.

[0129] 17 illustrates an operation model for supporting multi-vendor using Redfish API in a server management system according to an embodiment of the present invention. As shown in FIG. 17, the present invention uses Redfish API to collect inventory information for x86 server hardware systems regardless of manufacturer, such as Dell®, HP®, Lenovo®, etc., and to query and utilize the collected information.

[0130] For example, Dell uses iDRAC to collect data, HP uses iLO to collect data, and Lenovo uses iMM to collect data. Redfish API can also be used to distribute and install OS and firmware on multiple servers.

[0131] In this invention, the Redfish API can be used to quickly check the hardware specifications, OS information, firmware information, and so on of each server.

[0132] Additionally, the present invention can analyze patterns to predict failures, and hardware logs can be used to facilitate pattern analysis.

[0133] Since its first release in 2015, Redfish API has been continuously updated and supports multiple server manufacturers, providing the same functions as IPMI. In addition, Redfish API supports BIOS and Secure Boot configuration functions, firmware update functions, storage and server network configuration functions, and supports Open Compute Platform, Open stack, SNIA (Storage Networking Industry Association), etc., and supports network switch management and external storage management.

[0134] iDRAC, a management tool for Power Edge servers, supports Redfish RESTful API using Redfish. For example, iDRAC can perform server power (Reset, Reboot, Power Control), server hardware inventory, server monitoring and status check, system log collection, server status change check, and alarms.

[0135] PowerEdge Server can automate the initial settings of servers through Redfish, and can create templates for various configuration information such as iDRAC initial settings, BIOS, RAID controllers, and network cards to perform automated deployment of servers.

[0136] An example of using Redfish on the iDRAC of a Power Edge server for server configuration automation (Auto deployment) is as follows.

[0137] The server's unique configuration values ​​are stored as metadata in the server configuration profile (SCP) and can be configured with the Redfish API. In addition, various settings such as BIOS, iDRAC / LC, PERC RAID Controller, NIC, and HBA can be configured via the Redfish API.

[0138] SCP can be exported, previewed, and imported, allowing you to freely apply configuration information to newly created servers. SCP can be shared via methods such as HTTPS, NFS, and CIFS, and can be realized in XML and JSON file formats.

[0139] 18 to 31 are diagrams showing examples of screens of a server management system according to an embodiment of the present invention. Fig. 18 shows an example of an initial screen, which supports a dashboard to view information on inventory and logs automatically collected for a managed server at a glance.

[0140] Figure 19 shows an example of a screen that allows you to check the inventory information of the managed servers in real time. Any changes made to the information on this screen are automatically reflected in the inventory information.

[0141] In the example screen in Figure 20, each part is represented by symbol 5 (red circle) to make it easy to identify when a problem with the managed server is confirmed, and normal parts are represented by symbol 6 (green circle).

[0142] FIG. 21 is an example of a screen where real-time management information of all managed servers, including firmware (F / W) information, can be viewed.

[0143] Figure 22 shows an example of a screen where you can check the real-time CPU details and current status of all managed servers.

[0144] Figure 23 shows an example of a screen where you can check the real-time memory details and current status of all managed servers.

[0145] Figure 24 is an example of a screen where you can check the real-time Raid Controller details and current status of all managed servers.

[0146] Figure 25 is an example of a screen where you can check the real-time disk details and current status of all managed servers.

[0147] Figure 26 is an example of a screen where you can check real-time detailed information and current status of the PSU (Power supply) of all managed servers.

[0148] Figures 27 and 28 are examples of screens that allow you to check real-time detailed information about the logos collected by all managed servers. Real-time vendor HW error codes are collected and automatically classified, and problematic equipment can be checked for each error code.

[0149] FIG. 29 shows an example of a fault analysis screen, which displays fault analysis information including the cause of the fault, conclusion, and replacement time.

[0150] FIG. 30 is an example of a screen showing a fault analysis distribution chart for each server in comparison with a client company.

[0151] FIG. 31 is an example of a screen showing an example of the service report function, and shows report contents including the time of occurrence, problem details, problem resolution, and recurrence prevention measures.

[0152] FIG. 32 is a table classifying system equipment according to an embodiment of the present invention, and FIG. 33 and FIG. 34 are tables describing hardware symptoms and their causes according to an embodiment of the present invention.

[0153] 35 and 36 are a flow chart illustrating a method for proactively responding to failures in a server management system according to one embodiment of the present invention.

[0154] 35, when a hardware-related problem occurs in a managed server (S101), the management server 110 classifies similar equipment that is likely to cause a failure as risk equipment by referring to the classification table of FIG. 32 (S101, S103). Then, a warning message is sent for the classified risk equipment (S105), and countermeasures are taken before a failure occurs (S107). Referring to the classification table of FIG. 32, specific similarity judgment criteria for system equipment in one embodiment of the present invention are illustrated, and examples include classification of equipment of the same class, classification of equipment of the same CPU, classification of equipment of the same memory, classification of equipment of the same NIC, classification of equipment of the same disk, classification of equipment of the same HBA, classification of equipment of the same BIOS, classification of equipment of the same driver version, classification of equipment of the same OS, and classification of equipment of the same firmware version.

[0155] Referring to Fig. 36, when a hardware-related problem occurs in a managed server (S301), the management server 110 grasps the failure symptoms (S303). Next, referring to the diagrams of Figs. 33 and 34, the management server 110 checks the symptom code corresponding to the failure symptoms (S305). Then, the management server 110 checks the cause corresponding to the symptom code (S307) and sends a countermeasure report accordingly (S309). Next, the management server 110 executes the failure countermeasure corresponding to the failure cause (S311). If there is no symptom code corresponding to the failure symptom in step S305, a new symptom code is generated and added to the list of Figs. 33 and 34 (S313). Referring to Figs. 33 and 34, examples of failure causes corresponding to symptom codes for each failure symptom according to an embodiment of the present invention are shown.

[0156] In other words, RAC1198 is an iDrac firmware problem, pluggable memory failure is a memory problem and BIOS firmware problem, Link Failure is a NIC failure and firmware problem, numerous Link Failure Counts are NIC driver and firmware problems, NIC Link is Down is a NIC driver and firmware problem, link status and server confirmation request are NIC driver and firmware problems, HOST_DOWN occurs is a NIC driver and firmware problem, yellow light on the front of the server is an iDrac firmware problem, SWC5008: critical message output is an iDrac firmware problem, NO_PARTITION alarm is a disk failure, Reset adapte is a BIOS firmware problem, Correctable memory error is a memory problem and BIOS firmware problem, CPU performance degradation is a BIOS firmware problem, memory problems and BIOS firmware problems not displayed in Memory and Slot, Disk fault error is a disk failure, disk predicted fail is a failure due to disk BadBlock, periodic FAN 6 recognition problem is a Fan 6 failure, Fault due to light intensity below 400 is a Gbic failure, NIC Gbic failure with GBIC communication not possible, infinite system reboots are a BIOS firmware problem, specific message output on LCD Panel is an iDrac firmware problem, repeated error messages on iDRAC are an iDrac firmware problem, synchronization errors with the vCenter agent are an EXSi version and OS version problem, server reboots are a BIOS firmware problem, slow HBA write speed is an HBA firmware and driver problem, slow HBA read speed is an HBA firmware and driver problem, HBA link down is an HBA Gbic and card problem, HBA redundant node failure is an HBA Gbic and card problem, Riser1 not being recognized is a Riser Card problem, Riser2 not being recognized is a Riser Card problem, network redundancy failure is a network card problem, PSUAlert yellow LED lights up indicates PSU failure, low voltage abnormality indicates PSU failure, PXE boot is not possible due to BIOS setting and NIC firmware / driver problems, POST boot is not possible due to motherboard failure, LifeCycle connection is not possible due to motherboard failure, iDRAC Hang symptom indicates iDrac firmware problem, iDRAC Network disconnection is motherboard failure and iDrac firmware problem, iDRAC SNMP service failure is iDrac firmware problem, server suddenly turns off while in use is motherboard problem, Medium Error indicates disk failure, ERROR Event confirmation request is due to an Error Event problem, CMC connection is not possible due to a CMC firmware problem.

[0157] Also, a DSET analysis request is a failure due to analysis, a TSR Log analysis request is a failure due to analysis, failure to start NFS service is a problem due to analysis, check the NFS settings and OS settings, inability to connect to vCenter is a problem with the EXSi version and OS version, NIC Reset is a problem with the Network Card, GPU not being recognized is a GPU card failure, OS crashes are OS dump analysis, network errors / dropped packets are a problem with the network card, CRC errors are a problem with the network card, server switches are turned off is a problem with the network card, problems with smooth communication in the network (bonding) are a problem with the Network Card, the same slot event occurs after memory replacement is a memory failure or motherboard failure, disk failure that cannot be accessed in Disk Read Only state or a RAID configuration problem, the switch hangs 3 to 4 times a month is a problem with the motherboard or OS version, LACP Network Speed ​​problems are a problem with the Network Card, cluster failovers are a problem with the cluster settings or HW failure, RTSP synchronization failures are a problem with the OS settings or Network, session degradation symptoms are a problem with the Network Card or Gbic problem, unknown power cut is PSU failure, server slowness or hang phenomenon is application or HW failure, Network Ping Loss is Network Card or Gbic problem, LoadAvg increase requires CPU check, Fatal Error occurs PCI Card or Riser Card problem, PXE installation stall or performance drop is Network Card or Gbic problem, Blue Screen (0x00004f) is motherboard / BIOS / disk / memory failure, Blue Screen is motherboard / BIOS / disk failure, OSBooting failures are motherboard / BIOS / disk failures, process downs and panics during OS installation are motherboard / BIOS / disk failures, smells coming from the server are fan / motherboard / PSU problems, solutions to prevent NAS connection are network / OS setting problems, motherboard / KVM cable / KVM problems when KVM connection is not possible, Disk Amber LED is a disk failure, delays at post boot are motherboard / fan / PCI / memory problems, solutions to power failures are PSU failures, reduced Teaming performance is a network / OS setting problem, VD Bad Block is a disk failure, HBA Loop is an HBA failure, firmware / disk driver problems that prevent Raid configuration information from being seen, firmware / disk driver problems that prevent Volume from being recognized, Kernel Panic is an OS / App problem, server reboots when using maximum performance are CPU / PSU / motherboard / memory problems, significantly slower server processing is a CPU / PSU / motherboard / memory / disk problem, and server not turning on is caused by a PSU failure.

[0158] Although the present invention has been described above using several preferred embodiments, these embodiments are illustrative and not limiting. It will be understood that those skilled in the art to which the present invention pertains can make various changes and modifications without departing from the spirit of the present invention and the scope of rights presented in the appended claims. [Explanation of symbols]

[0159] 110: Management server 112: Database 120: Administrator terminal 130: Customer terminal 10, 20, 30, 40: Managed Servers

Claims

1. In a server management system for managing two or more managed servers, a database for storing the managed server related data; a management server that collects hardware-related data and software-related data from the managed servers, grasps and manages the current status of each managed server, and provides various server management information including related management service statistical data and management service reports to a management terminal used by the manager and to a customer terminal that has requested the managed servers; The management server uses AI technology to analyze the managed servers and predict the status and failures of the managed servers, and if a problem occurs through the prediction, transmits an alarm message to the administrator terminal and the customer terminal by text message and email. Server management system.

2. 2. The server management system according to claim 1, The management server collects formatted log data and unformatted log data from each managed server, classifies the collected data, performs data pre-processing, and then performs learning through an AI learning data model to predict the status and failure of the managed server.

3. 3. The server management system according to claim 2, The management server provides an AI analysis function using Redfish API, learns what normal traffic is in each managed server, finds abnormal traffic, and provides an AI analysis function that sets necessary risk priorities and analyzes problems to support users.

4. 2. The server management system according to claim 1, The management server checks the backup battery unit (BBU) period of the managed server, and when a predetermined period is reached, transmits the information to the client terminal of the managed server. The management server includes: Check the BBU charging capacity of the managed server, and if the battery charging efficiency falls below a predetermined value, notify the customer terminal of the managed server of this information, Check the remaining capacity of the BBU of the managed server, and if the remaining battery capacity is below a predetermined value, notify the customer terminal of the managed server of this information, A server management system that checks the BBU write policy of a managed server, and if the write policy is changed, notifies the customer terminal of the managed server of the change. The server management system checks the battery full charging efficiency (%) by checking the log of the managed server, and notifies the customer terminal of the managed server of the change to a battery for devices whose full charging efficiency is less than a predetermined value.

5. 2. The server management system according to claim 1, A server management system, wherein the management server collects and stores multi-vendor hardware inventory information from a plurality of registered managed servers.

6. 6. The server management system according to claim 5, A server management system, wherein the management server proceeds with firmware updates for all managed servers when there is a firmware update event, including an emergency firmware update.

7. 2. The server management system according to claim 1, The management server analyzes logs and patterns when a fault occurs in any of the devices of the managed servers, stores the analyzed data, and when the fault is resolved, classifies devices similar to the faulty device and performs pre-failure response processing for the classified similar devices. This is a server management system.

8. 2. The server management system according to claim 1, The management server classifies similar devices having a high probability of failure as dangerous devices by referring to a classification table when a hardware-related problem occurs in the managed server, sends a warning message regarding the classified dangerous devices, and takes countermeasures in advance of a failure.

9. 9. The server management system according to claim 8, The classification table includes specific similarity judgment criteria for system devices, and includes classification of devices of the same class, classification of devices of the same CPU, classification of devices of the same memory, classification of devices of the same NIC, classification of devices of the same Disk, classification of devices of the same HBA, classification of devices of the same BIOS, classification of devices of the same Driver version, classification of devices of the same OS, and classification of devices of the same Firmware version, in a server management system.

10. 10. The server management system according to claim 9, When a hardware-related problem occurs in a managed server, the management server grasps the failure symptoms, refers to a list including failure causes corresponding to symptom codes for each failure symptom, checks the symptom code corresponding to the failure symptom, checks the cause corresponding to the symptom code, sends a countermeasure report accordingly, performs the failure countermeasure corresponding to the cause of the failure, and if there is no symptom code corresponding to the failure symptom, generates a new symptom code and adds it to the list.

11. 11. The server management system according to claim 10, The above list is RAC1198 is an iDrac firmware issue, The connectable memory failure is caused by a memory problem and a BIOS firmware problem. Link failure can occur due to NIC failure or firmware problem. A high number of Link Failure Counts may be due to NIC driver and firmware issues. NIC Link is Down is caused by NIC driver and firmware issues. Link status and server inspection requests can help troubleshoot NIC driver and firmware issues, HOST_DOWN occurs due to NIC driver and firmware issues. The yellow light on the front of the server is a problem with the iDrac firmware. SWC5008:critical message output indicates iDrac firmware issue, The NO_PARTITION alarm occurs when a disk failure occurs. Reset adapte can fix BIOS firmware issues, Correctable memory error is a problem with memory and BIOS firmware. The CPU performance degradation is due to a BIOS firmware problem. If Memory and Slot are not displayed, this may be a memory problem or a BIOS firmware problem. Disk fault error means disk failure, Disk predicted fail is a failure caused by a disk bad block. Periodic FAN 6 recognition problems may indicate FAN 6 failure, Fault due to light intensity below 400 is Gbic failure, NIC DBIC communication failure is a Gbic failure. System infinite reboots are caused by BIOS firmware issues, LCD Panel specific message output indicates iDrac firmware issue, Repeated error messages on iDRAC can be caused by iDrac firmware issues, The vCenter agent and sync error may be due to EXSi version, OS version issues, The server reboot phenomenon is due to a BIOS firmware problem. The slowdown in HBA Write speed is caused by problems with the HBA firmware and driver. The slowness of HBA Read speed is caused by problems with the HBA firmware and driver. HBA link down is a problem with the HBA Gbic and card, HBA dual redundant transfer failure is a problem with HBA Gbic and Card. Riser1 recognition failure is due to a Riser Card problem. Riser2 recognition failure is due to a Riser Card problem. Network duplication failure is caused by a network card problem, PSU Alert Yellow LED on indicates PSU failure. Abnormalities caused by low voltage include PSU failure, PXE boot failure BIOS setting, NIC firmware / driver problem, POST boot failure motherboard failure, LifeCycle connection failure motherboard failure, iDRAC Hang symptoms can be caused by iDrac firmware issues, iDRAC Network disconnection can be caused by a motherboard failure, an iDrac firmware problem, iDRAC SNMP service failure may be caused by iDrac firmware issue, If the server suddenly shuts down while in use, it is likely due to a motherboard problem. A medium error occurs when a disk failure occurs. The ERROR Event confirmation request is for problems caused by Error Events, problems with CMC firmware, or problems with CMC connection. DSET analysis requests are made based on the following: TSR Log analysis requests are for issues caused by the analysis, The failure to start the NFS service may be due to the following: Check the NFS settings and OS settings, EXSi version and OS version issues that cannot connect to the vCenter, NIC Reset can fix network card problems, If the GPU cannot be recognized, it may be due to a GPU card failure. When OS crash occurs, OS dump analysis, Network error / dropped packets may be caused by a problem with the network card. CRC errors occur due to network card problems. The server switch is turned off due to a network card problem. Problems with smooth communication to the network (Bonding) may be due to network card problems, The occurrence of the same slot event after memory replacement may indicate a memory failure or motherboard failure, a disk failure that is inaccessible due to a disk read-only state, or a problem with the RAID configuration. If the switch hangs 3-4 times a month, it could be due to a problem with the motherboard or OS version. LACP Network Speed ​​problem occurs due to Network Card problem, Cluster failover occurs due to cluster configuration problems or HW failure, RTSP synchronization failure, OS settings or network failure, The occurrence of session degradation may be due to network card or Gbic problems. Unknown power outages can be caused by PSU failure, Server slowness and hangs are caused by application or hardware failures, Network ping loss may indicate a network card or Gbic problem. If LoadAvg increases, check your CPU. Fatal Error occurs due to a problem with the PCI Card or Riser Card. Stalls or slow performance during PXE installation can be caused by Network Card or Gbic issues, Blue Screen (0x00004f) may be caused by a motherboard / BIOS / disk / memory failure. Blue Screen can be caused by motherboard / BIOS / disk failure, OS Booting failure is caused by motherboard / BIOS / disk failure, Process down and panic during OS installation may be caused by motherboard / BIOS / disk failure, The smell from the server could be a fan / motherboard / PSU problem, NAS connection failure countermeasures include network / OS setting problems, KVM connection failure motherboard / KVM cable / KVM problems, Disk Amber LED indicates a disk failure. Post boot delays can be caused by motherboard / fan / PCI / memory issues, Power supply failure countermeasures include PSU failure, The degradation of Teming performance may be due to network / OS configuration issues, VD Bad Block indicates a disk failure. HBA loops can occur due to HBA failure, firmware / disk driver issues that prevent RAID configuration information from being seen, firmware / disk driver issues that prevent volume recognition, Kernel Panic is a problem with the OS / apps, a problem with the CPU / PSU / motherboard / memory that causes the server to reboot when using maximum performance, a problem with the CPU / PSU / motherboard / memory / disk that causes server processing to slow down significantly, Server not powering on is caused by PSU failure, including: Server management system.

Citation Information

Patent Citations

  • Failure monitoring apparatus and failure monitoring method

    JP2010049551A

  • Electronic apparatus

    JP2017224090A

  • Method of determining operation data from network device and method of transmitting operation data to network device

    JP2019036939A

  • Monitoring system, monitoring method, and monitoring program

    JP2020030628A

  • Deduction of state of medica analyzer

    JP2022018100A

Cited By

  • Automatic setting device and automatic setting method

    JP7839352B1