TW202145015A - Device and method for monitoring server and storage medium - Google Patents

Abstract

A server monitoring method includes: collecting SDR data of a server cluster to be monitored, wherein the server cluster to be monitored includes at least one server; storing the SDR data as a predetermined file format into a specified storage area; analyzing the SDR data which are stored in the specified storage area to determine whether abnormal SDR data is existed; outputting warning information of an abnormal component corresponding to the abnormal SDR data and if the abnormal SDR data is existed. A server monitoring device and a storage medium are also provided.

Description

Server monitoring device, method and computer readable storage medium

The present invention relates to the technical field of communication equipment, and in particular, to a server monitoring device, a method and a computer-readable storage medium.

In recent years, with the rapid development of technology and the Internet, the functions of servers have become more and more powerful. In order to ensure the stability of the server operation, a baseboard management controller (BMC) is generally used to monitor and manage the operation of the server system. The BMC can record system events in the non-volatile system event log (SEL). For example, the recorded event types include abnormal temperature, abnormal voltage, abnormal fan, etc. During the monitoring process, the BMC can also manage a non-volatile Sensor Data Record Repository (SDRR) from which system runtime information can be retrieved. Therefore, the log information automatically generated by BMC becomes an important indicator and reference for checking the running status of the server. However, based on the log information automatically generated by the BMC, the abnormal events of the sensors used to monitor the running status of the server cannot be obtained.

In view of this, it is necessary to provide a server monitoring device, method and computer-readable storage medium, which can monitor the running status of the server and its internal sensors.

An embodiment of the present invention provides a server monitoring method, the method includes: collecting SDR data of a server cluster to be monitored, wherein the server cluster to be monitored includes at least one server; Set the file format to be stored in a designated storage area; use preset analysis rules to analyze the SDR data in the designated storage area to determine whether there is abnormal SDR data; and if stored in abnormal SDR data, output and the abnormal SDR data Warning information of abnormal components corresponding to SDR data.

An embodiment of the present invention provides a server monitoring device, the device includes a processor and a memory, the memory stores a plurality of computer programs, and the processor implements the following steps when executing the computer programs stored in the memory : collect the SDR data of the server cluster to be monitored, wherein the server cluster to be monitored includes at least one server; store the collected SDR data in a preset file format in a designated storage area; use a preset analysis rule to The SDR data in the specified storage area is analyzed to determine whether the abnormal SDR data exists; and if the abnormal SDR data exists, the warning information of the abnormal element corresponding to the abnormal SDR data is output.

An embodiment of the present invention provides a computer-readable storage medium. The computer-readable storage medium stores a plurality of instructions, and a plurality of the instructions can be executed by one or more processors, so as to realize the following steps: Monitoring the SDR data of a server cluster, wherein the server cluster to be monitored includes at least one server; storing the collected SDR data in a preset file format to a designated storage area; using a preset analysis rule to store the designated storage The SDR data in the area is analyzed to determine whether the abnormal SDR data exists; and if the abnormal SDR data exists, the warning information of the abnormal element corresponding to the abnormal SDR data is output.

Compared with the prior art, the above-mentioned server monitoring device, method and computer-readable storage medium can realize the reporting of abnormal information that cannot be found by BMC by analyzing the SDR data, and find out the failure/damage of components in advance. It can locate faulty sensors and server components, and can also filter BMC log files and report abnormal logs.

Please refer to FIG. 1 , which is a schematic diagram of a preferred embodiment of the server monitoring device of the present invention.

The server monitoring device 100 can monitor multiple servers in the data center 200 . For example, the data center 200 includes at least one server cluster to be monitored, and the server cluster to be monitored may include multiple servers. It can be understood that the server monitoring device 100 can also monitor the server or server cluster designated by the user according to actual requirements, which is not limited herein.

The server monitoring device 100 may include a memory 10 , a processor 20 , and a server monitoring program 30 stored in the memory 10 and running on the processor 20 . When the processor 20 executes the server monitoring program 30, the steps in the embodiment of the server monitoring method are implemented, such as steps S300-S306 shown in FIG. 3 . Alternatively, when the processor 20 executes the server monitoring program 30 , the functions of the modules shown in FIG. 2 , such as modules 101 to 105 , are implemented.

The server monitoring program 30 can be divided into one or more modules, and the one or more modules are stored in the memory 10 and executed by the processor 20 to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the server monitoring program 30 in the server monitoring device 100 . For example, the server monitoring program 30 can be divided into a collection module 101 , a storage module 102 , an analysis module 103 , an output module 104 and a conversion module 105 in FIG. 2 . For the specific functions of each module, please refer to the function of each module in Figure 2 below.

Those skilled in the art can understand that the schematic diagram is only an example of the server monitoring device 100, and does not constitute a limitation on the server monitoring device 100, and may include more or less components than the one shown, or combine some components , or different components, for example, the server monitoring device 100 may further include an input display device, a communication module, a bus bar, and the like.

The processor 20 may be a central processing unit (CPU), other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), or an off-the-shelf program. Design Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor 20 can also be any conventional processor, etc. The processor 20 can use various interfaces and bus bars to connect various parts of the server monitoring device 100 .

The memory 10 can be used to store the server monitoring program 30 and/or modules, and the processor 20 calls the data stored in the memory 10 by running or executing the computer programs and/or modules stored in the memory 10 , to realize various functions of the server monitoring device 100 . The memory 10 may include high-speed random access memory, and may also include non-volatile memory such as hard disk drives, internal memory, plug-in hard disk drives, Smart Media Cards (SMC), secure digital (Secure Digital, SD) card, flash memory card (Flash Card), at least one disk memory device, flash memory device, or other volatile solid state memory device.

FIG. 2 is a functional module diagram of a preferred embodiment of the server monitoring program of the present invention.

Referring to FIG. 2 , the server monitoring program 30 may include a collection module 101 , a storage module 102 , an analysis module 103 , an output module 104 and a conversion module 105 . In one embodiment, the above-mentioned modules may be programmable software instructions stored in the memory 10 and invoked by the processor 20 for execution. It can be understood that, in other embodiments, the above-mentioned modules can also be program instructions or firmware solidified in the processor 20 .

The collection module 101 is used to collect the SDR data of the server cluster to be monitored.

In one embodiment, the to-be-monitored server cluster includes at least one server, and the to-be-monitored server cluster may be a server cluster designated by a user to be monitored, such as a server cluster of the data center 200 .

In one embodiment, the collection module 101 can access the SUT (System Under Test, system under test) associated with the server cluster to be monitored through a web page. The SUT can include an SDR monitor, and the server monitoring device 100 can Send a control command to the SDR monitor, and then the collection module 101 can collect the SDR data of each server in the server cluster to be monitored. In other embodiments of the present invention, the collection module 101 can also collect the to-be-monitored server by communicating with the IPMI (Intelligent Platform Management Interface) monitor of the to-be-monitored server cluster. SDR data for each server in the cluster.

It can be understood that for each server, IPMI can be used to monitor the physical characteristics of the server, such as monitoring the physical characteristics of the server through sensors scattered on the server substrate, system board, main chassis, fans, etc. feature, SDR data can be generated based on the data monitored by the sensors, and the SDR data can be stored in the sensor data record repository (SDRR). The physical features can be temperature, voltage, fan working status, power status, etc. The motherboard management controller (BMC) installed in the server can automatically monitor the management events of the server system and record the events in the non-volatile system event log (SEL). In the process of monitoring the server, the BMC can also manage the non-volatile SDRR, and can retrieve system runtime information from this library.

The storage module 102 is used for storing the collected SDR data in a designated storage area in a preset file format.

In one embodiment, the preset format can be determined according to actual needs, such as a CSV file format, and the designated storage area can also be determined according to actual needs, such as a designated file server, a designated database, etc. . The file stored in the designated storage area can be the name and number of the server or the name of the SUT corresponding to the server as the file name to distinguish, avoid file clutter, and facilitate subsequent managers to view and other operations. For example, the storage module 102 stores the collected SDR data to a designated file server in the form of SDR_SUT_1.CSV, SDR_SUT_2.CSV, SDR_SUT_3.CSV, . . . , SDR_SUT_n.CSV, etc.

The analysis module 103 is configured to analyze the SDR data in the designated storage area by using a preset analysis rule to determine whether there is abnormal SDR data.

In one embodiment, the preset analysis rules can be predefined analysis rules for different types of SDR data, and can be adjusted according to actual needs, and the different types can refer to different types of parameters such as voltage, temperature, rotational speed, etc. If the parameters are of the same type, different analysis rules exist for different components. For example, CPU temperature and hard disk temperature may have different analysis rules.

In one embodiment, if the SDR data includes CPU temperature data and power consumption data, 10 CPU temperature data and 10 power consumption data are respectively detected when the stress test is started and when the stress test is ended, and the preset analysis rules include: : When the stress test is turned on (high load), the temperature data of the CPU should be higher than the previous temperature data without the stress test, and the power consumption of the power supply should be greater than the power consumption data without the stress test. The temperature data should be lower than the temperature data during the previous stress test, and the power consumption should be lower than the power consumption data when the stress test is not turned on. Otherwise, the analysis module 103 can determine that there is an abnormality and locate the abnormal point.

It can be understood that during the operation of the server, the voltage data of the chip may need to be kept in a stable state, while the temperature or the wind speed should be dynamic and cannot be kept completely unchanged. The preset analysis rule may be: to analyze n continuous SDR data (for example, 50 data), if it is voltage data, then n continuous SDR data should be the same; if it is temperature data or wind speed data, etc., then n continuous SDR data should be Dynamic means that n consecutive SDR data are not identical. Otherwise, the analysis module 103 can determine that there is an abnormality, and locate the abnormal point.

In one embodiment, the SDR data of components belonging to a group should be substantially the same. The predetermined analysis rule may be: the SDR data difference of each component in the group at any time should be within a predetermined difference interval. For example, if a group includes multiple normal working CPUs, at any time, the difference between the temperature data of each CPU should be within 15%, and a group includes multiple normal working solid-state drives (SSD) , then at any time, the difference between the temperature data of each SSD should be within 10%, and a group includes multiple normal working fans, then at any time, the speed difference of each fan should be within 10% within. Otherwise, the analysis module 103 can determine that there is an abnormality, and locate the abnormal point.

The preset analysis rule may also be: during the server self-test, the fan speed should increase at the beginning of the self-test and decrease when the self-test ends. For example, when the self-test starts, the fan speed should increase by at least 20%. , at the end of the self-test, the fan speed should be reduced by at least 20%. Otherwise, the analysis module 103 can determine that there is an abnormality, and locate the abnormal point.

In one embodiment, the analysis module 103 can also analyze the abnormal SDR data in combination with the SEL log file, so as to determine whether the abnormal SDR data is abnormally generated by the sensor itself or a component of the server itself.

In one embodiment, the abnormal SDR data may include a first type of abnormal SDR data and a second type of abnormal SDR data. The first type of abnormal SDR data refers to the data that can trigger the BMC to generate the corresponding abnormal log file, and the second type of abnormal SDR data refers to the data that cannot trigger the BMC to generate the corresponding abnormal log file, and then the analysis module 103 can realize the error information that is not reported by BMC, and find out the signs of abnormality/deterioration of components in advance, so that the user or a certain component may be damaged.

The output module 104 is used for outputting the warning information of the abnormal element corresponding to the abnormal SDR data.

In one embodiment, the abnormal element may refer to a sensor used to monitor the component of the server and/or a component of the server, that is, it may be that the sensor used to monitor the component of the server is abnormal. , or there may be an abnormality in the components of the server itself. The warning information includes but is not limited to abnormal information (records basic abnormal information, such as component name, serial number, basic abnormal situation, etc.) and the time interval of abnormal occurrence.

In one embodiment, the server monitoring device 100 can also implement the function of filtering and reporting BMC abnormal log files. Specifically, the collection module 101 can also obtain the log files generated by the BMC of each server from the server cluster to be monitored, and the analysis module 103 can detect whether the log files generated by the BMC are stored in abnormal log files, such as the analysis module The group 103 can detect whether preset key matching information is found in the log file, whether the parameters of the components in the log file exceed a corresponding threshold, etc., so as to detect abnormal log files. The output module 104 can output and display the abnormal log file detected by the analysis module 103 .

In one embodiment, the analysis module 103 can also be used to determine whether the abnormal log file exists in the log file corresponding to the abnormal SDR data of the first type, and the double-checked BMC system log mechanism is more rigorous. If the log file corresponding to the first type of abnormal SDR data does not exist in the abnormal log file, indicating that the BMC may be stored in the system log event that is not recorded or the SDR data may be stored in the abnormal record, the output module 104 may Default prompt information is output to remind the administrator of the server cluster to be monitored to perform verification.

In one embodiment, when the analysis module 103 analyzes and determines that the abnormal SDR data is stored, it can generate a corresponding SDR abnormal log file based on the abnormal SDR data, and then the output module 104 can output the SDR abnormal log file, It is convenient for managers to view it instantly. In other embodiments of the present invention, the SDR anomaly log file can also be stored in a designated database for backup, so that the administrator can check or analyze large data later.

In one embodiment, in order to facilitate the administrator to quickly check the SDR information and understand the working status of each monitored component of each server, the conversion module 105 is used to classify the collected SDR data and convert it into an SDR curve and the output module 104 can output the SDR graph on the display interface.

It can be understood that each SDR graph can correspond to the data monitored by each sensor, and the conversion module 105 can monitor the collected SDR data in real time, so that the SDR graph can be updated in real time.

FIG. 3 is a flowchart of a server monitoring method in an embodiment of the present invention. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted.

Step S300 , collecting SDR data of the server cluster to be monitored, wherein the server cluster to be monitored includes at least one server.

Step S302, the collected SDR data is stored in a designated storage area in a preset file format.

Step S304 , using a preset analysis rule to analyze the SDR data in the designated storage area to determine whether there is abnormal SDR data.

Step S306, if the abnormal SDR data exists, output the warning information of the abnormal element corresponding to the abnormal SDR data.

The above-mentioned server monitoring device, method and computer-readable storage medium can realize the reporting of abnormal information that cannot be found by BMC by analyzing SDR data, find out signs of component deterioration/damage in advance, and locate faulty information. The problem sensor and server components can also filter BMC log files and report abnormal logs.

To sum up, the present invention complies with the requirements of an invention patent, and a patent application can be filed in accordance with the law. However, the above descriptions are only the preferred embodiments of the present invention, and the scope of the present invention is not limited to the above-mentioned embodiments, and equivalent modifications or changes made by those who are familiar with the art of the present invention according to the spirit of the present invention are all applicable. Should be covered within the scope of the following patent applications.

10: Memory 20: Processor 30: Server monitoring program 101: Collect Mods 102: Storage Module 103: Analysis Module 104: Output module 105: Conversion module 100: Server monitoring device

FIG. 1 is a functional module diagram of a server monitoring device according to an embodiment of the present invention.

FIG. 2 is a functional module diagram of a server monitoring program according to an embodiment of the present invention.

FIG. 3 is a flowchart of a server monitoring method according to an embodiment of the present invention.

Claims (10)

A server monitoring method, the method comprising: collecting SDR data of a server cluster to be monitored, wherein the server cluster to be monitored includes at least one server; Store the collected SDR data in the designated storage area in a preset file format; Analyzing the SDR data in the designated storage area using preset analysis rules to determine whether there is abnormal SDR data; and If the abnormal SDR data exists, the warning information of the abnormal element corresponding to the abnormal SDR data is output. The server monitoring method according to claim 1, further comprising: obtaining a log file generated by the baseboard management controller of the server cluster to be monitored; and Detecting whether the log file generated by the baseboard management controller is stored in the abnormal log file, and outputting the detected abnormal log file. The server monitoring method according to claim 2, wherein the abnormal SDR data includes a first type of abnormal SDR data and a second type of abnormal SDR data, wherein the first type of abnormal SDR data can trigger the The baseboard management controller generates a corresponding exception log file, and the abnormal SDR data of the second type cannot trigger the baseboard management controller to generate a corresponding exception log file. The server monitoring method according to claim 3, further comprising: determining whether a log file corresponding to the abnormal SDR data of the first type exists in the abnormal log file; and If there is no log file corresponding to the abnormal SDR data of the first type in the abnormal log file, a default prompt message is output. The server monitoring method according to claim 1, wherein if there is abnormal SDR data, the step of outputting the warning information of the abnormal element corresponding to the abnormal SDR data comprises: If stored in the abnormal SDR data, generating an SDR abnormal log file based on the abnormal SDR data; and Outputting the SDR abnormal log file and the warning information of the abnormal element corresponding to the abnormal SDR data. The server monitoring method according to claim 1, further comprising: converting the collected SDR data into an SDR graph, and outputting the SDR graph; and The collected SDR data is monitored to update the SDR graph. The server monitoring method according to claim 1, wherein the abnormal element includes a sensor for monitoring the components of the server and/or the components of the server, and the warning information includes abnormal information and abnormality The time period in which it occurred. The server monitoring method according to claim 1, further comprising: The system under test associated with the server cluster to be monitored is accessed through a web page to collect the SDR data of the server cluster to be monitored. A server monitoring device, the device includes a processor and a memory, a plurality of computer programs are stored on the memory, and the processor is used to execute any of the requirements 1 to 8 when executing the computer programs stored in the memory. A step of the described server monitoring method. A computer-readable storage medium that stores a plurality of instructions, and a plurality of the instructions can be executed by one or more processors to achieve the invention as claimed in any one of claim items 1 to 8. The steps of the server monitoring method described above.

Images