Hard disk health monitoring method and device, electronic equipment and storage medium

Abstract

The application provides a hard disk health monitoring method and device, electronic equipment and storage medium, and belongs to the technical field of computers. The method comprises the following steps: obtaining hard disk related information of a server, wherein the hard disk related information comprises a model configuration file corresponding to the server, information of a redundant array of independent disks card of the server, hard disk backplane information of the server, and hard disk information of all hard disks of the server; performing slow disk detection on each hard disk of the server based on the hard disk information; performing sub-health detection on each hard disk of the server based on the information of the redundant array of independent disks card and the hard disk information; in the case that a target hard disk is detected to be in a slow disk state or a sub-health state, issuing a prompt information, and based on the model configuration file and the hard disk backplane information, lighting up a fault light of the target hard disk, so as to realize automatic health monitoring of the hard disks assembled in the server, and improve the universality, implementability, accuracy and positioning of the server hard disk health monitoring.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a hard disk health monitoring method, apparatus, electronic device, and storage medium. Background Technology

[0002] Hard drives store vast amounts of data; however, with increasing usage time, they may experience various failures such as slow read speeds and file loss, increasing the risk of server malfunctions. Storage systems often employ independent redundant RAID cards from different manufacturers and use different hard drives from various manufacturers, each with different interface types. This complexity makes monitoring the health of each hard drive in a server extremely difficult. Therefore, effectively monitoring the health status of each hard drive in a large number of servers is crucial. Currently, hard drive health checks are relatively simple and often misdiagnose healthy drives as unhealthy, leading to a high false alarm rate and the spread of problems. Summary of the Invention

[0003] This invention provides a hard disk health monitoring method, device, electronic device, and storage medium to solve the shortcomings of existing technologies, such as the inability to effectively monitor the health status of each hard disk in a large number of servers and the high false judgment rate.

[0004] In a first aspect, the present invention provides a hard disk health monitoring method, comprising:

[0005] Obtain the server's hard drive related information, which includes: the server's corresponding model configuration file, the server's independent redundant disk array card information, the server's hard drive backplane information, and the hard drive information of all the server's hard drives;

[0006] Based on the hard drive information, slow disk detection is performed on each hard drive of the server;

[0007] Based on the independent redundant disk array card information and the hard disk information, sub-health detection is performed on each hard disk of the server;

[0008] If the target hard drive is detected to be in a slow or unhealthy state, a prompt message will be issued, and the fault indicator light of the target hard drive will be illuminated based on the model configuration file and hard drive backplane information.

[0009] In some embodiments, obtaining the server's hard drive-related information includes:

[0010] Obtain the server's model information, and obtain the corresponding model configuration file based on the model information;

[0011] Obtain the high-speed serial bus PCIe device information of the server, and obtain the independent redundant disk array card information of the server based on the PCIe device information. The independent redundant disk array card information includes the number of independent redundant disk array cards and manufacturer and model information.

[0012] Obtain the Small Computer System Interface (SCSI) device information of the server, and obtain the hard disk backplane information of the server based on the SCSI device information. The hard disk backplane information includes the number of hard disk backplanes and their manufacturer and model information.

[0013] The hard drive information of all hard drives of the server is obtained based on the SCSI device information, wherein the hard drive information includes at least one of the following: number of hard drives, manufacturer, model, drive letter, protocol type, and media type.

[0014] In some embodiments, the step of performing slow disk detection on each hard drive of the server based on the hard drive information includes:

[0015] Based on the hard drive information, determine whether the server is equipped with a hard drive;

[0016] If the server is equipped with hard drives, iterate through all the hard drives of the server and perform slow disk detection on each hard drive in turn.

[0017] The step of sequentially performing slow disk detection on each hard drive includes:

[0018] Obtain the input / output performance parameters of the hard drive under test;

[0019] Determine whether there exists a second target count period within the first consecutive target count period that the input / output performance index parameters satisfy the slow disk condition;

[0020] If it exists, then the hard drive under test is determined to be in a slow disk state.

[0021] In some embodiments, the step of issuing a prompt message and illuminating the fault indicator light of the target hard drive based on the model configuration file and hard drive backplane information when the target hard drive is detected to be in a slow or unhealthy state includes:

[0022] If the target hard drive is detected to be in a slow state, the first prompt message will be issued;

[0023] By parsing the model configuration file, the backplane identification information and slot identification information of the target hard drive are obtained;

[0024] Based on the hard drive backplane information, backplane identification information, and slot identification information, determine the hard drive light status band setting instructions.

[0025] Execute the hard drive light status setting instruction to set the hard drive light status of the target hard drive to a fault state.

[0026] In some embodiments, the method further includes:

[0027] During the slow disk detection cycle, the input and output performance index parameters are recorded, analyzed, and processed. The changing trend of the input and output performance index parameters is predicted based on the gradient boosting decision tree algorithm.

[0028] In some embodiments, the sub-health detection of each hard drive of the server based on the independent redundant disk array card information and the hard drive information includes:

[0029] Based on the hard drive information, determine whether the server is equipped with a hard drive;

[0030] If the server is equipped with hard drives, iterate through all the hard drives of the server and perform sub-health checks on each hard drive in turn;

[0031] The step of sequentially performing sub-health checks on each hard drive includes:

[0032] Based on the independent redundant disk array card information and the application form of the hard drive under test, the key parameter information of the SMART self-monitoring analysis and reporting technology of the hard drive under test is obtained;

[0033] Determine whether the SMART key parameter information meets the conditions for hard drive sub-health;

[0034] If the conditions are met, the hard drive under test is determined to be in a sub-healthy state.

[0035] In some embodiments, the step of issuing a prompt message and illuminating the fault indicator light of the target hard drive based on the model configuration file and hard drive backplane information when the target hard drive is detected to be in a slow or unhealthy state includes:

[0036] If the target hard drive is detected to be in a sub-healthy state, a second prompt message will be issued;

[0037] By parsing the model configuration file, the backplane identification information and slot identification information of the target hard drive are obtained;

[0038] Based on the hard drive backplane information, backplane identification information, and slot identification information, determine the hard drive light status band setting instructions.

[0039] Execute the hard drive light status setting instruction to set the hard drive light status of the target hard drive to a fault state.

[0040] Secondly, the present invention also provides a hard disk health monitoring device, comprising:

[0041] The acquisition unit is used to acquire hard drive related information of the server. The hard drive related information includes: the model configuration file corresponding to the server, the independent redundant disk array card information of the server, the hard drive backplane information of the server, and the hard drive information of all hard drives of the server.

[0042] The slow disk detection unit is used to perform slow disk detection on each hard disk of the server based on the hard disk information.

[0043] The sub-health detection unit is used to perform sub-health detection on each hard drive of the server based on the independent redundant disk array card information and the hard drive information.

[0044] The fault light unit is used to issue a prompt message when the target hard drive is detected to be in a slow or sub-healthy state, and to illuminate the fault light of the target hard drive based on the model configuration file and hard drive backplane information.

[0045] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the hard disk health monitoring method as described in any of the first aspects.

[0046] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the hard disk health monitoring method as described in any of the first aspects.

[0047] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the hard disk health monitoring method as described in any of the first aspects.

[0048] This invention provides a hard drive health monitoring method, device, electronic device, and storage medium. By acquiring hard drive-related information from a server, it performs slow drive detection on each hard drive based on this information; it also performs sub-health detection on each hard drive based on independent redundant disk array card information and hard drive information; and then, upon detecting that a target hard drive is in a slow or sub-healthy state, it issues a prompt message and illuminates the target hard drive's fault indicator light based on the machine configuration file and hard drive backplane information. This achieves automatic health monitoring of the hard drives installed in the server, improving the universality, feasibility, accuracy, and location of server hard drive health monitoring. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0050] Figure 1 This is one of the flowcharts illustrating a hard disk health monitoring method according to an embodiment of the present invention;

[0051] Figure 2 This is a schematic diagram of a process for obtaining hard disk-related information of a server according to an embodiment of the present invention;

[0052] Figure 3 This is a schematic diagram illustrating the process of sequentially performing slow disk detection on each hard drive according to an embodiment of the present invention;

[0053] Figure 4 This is a second schematic flowchart of a hard disk health monitoring method provided in one embodiment of the present invention;

[0054] Figure 5 This is a schematic diagram illustrating the process of sequentially performing sub-health checks on each hard drive according to an embodiment of the present invention;

[0055] Figure 6 This is a third flowchart illustrating a hard disk health monitoring method according to an embodiment of the present invention.

[0056] Figure 7 This is a schematic diagram of the structure of a hard disk health monitoring device provided in one embodiment of the present invention;

[0057] Figure 8 This is a schematic diagram of the physical structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0059] The terms "first," "second," etc., used in the specification and claims of this invention are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of the invention can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0060] In view of the shortcomings of related technologies, such as the inability to effectively monitor the health status of each hard drive in a large number of servers and the high false judgment rate, this invention provides a hard drive health monitoring method, device, electronic device and storage medium.

[0061] The hard disk health monitoring method provided by this invention can be implemented by a hard disk health monitoring device, which can be implemented by software and / or hardware. The device can be integrated into an electronic device, which can be a terminal device (such as a smartphone, personal computer, learning machine, etc.), a server (such as a local server or cloud server, or a server cluster, etc.), a processor, or a chip, etc.

[0062] The following example uses a hard drive health monitoring device as the execution entity, combined with... Figures 1-8 The present invention will now be described.

[0063] Figure 1 This is one of the flowcharts illustrating a hard disk health monitoring method provided in an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps: step 110, step 120, step 130, and step 140. These method steps are merely one possible implementation of the invention. The method includes:

[0064] Step 110: Obtain the hard drive related information of the server. The hard drive related information includes: the model configuration file corresponding to the server, the independent redundant disk array card information of the server, the hard drive backplane information of the server, and the hard drive information of all hard drives of the server.

[0065] In this invention, obtaining information about the server's hard drive is the foundation for detecting slow hard drives and sub-optimal hard drive health.

[0066] The server's corresponding configuration file includes commands for turning on hard drive lights, querying hard drive light status, querying hard drive presence status, the relationship between the server's hard drive slot silkscreen and the hard drive's slot_id and backplane_id, and the relationship between the hard drive slot silkscreen and the hard drive's phy_id. The Redundant Array of Independent Disks (RAID) card information includes the number of RAID cards and their manufacturer and model information; the hard drive backplane information includes the number of hard drive backplanes and their manufacturer and model information; the hard drive information includes at least one of the following: number of hard drives, manufacturer, model, drive letter, protocol type, and media type.

[0067] In some embodiments, Figure 2 This is a schematic diagram of a process for obtaining hard disk-related information of a server according to an embodiment of the present invention, as shown below. Figure 2 As shown, step 110 includes:

[0068] Step 210: Obtain the server model information, and obtain the corresponding server model configuration file based on the model information;

[0069] Optionally, the server model information can be obtained through computer program instructions. For example, the server model information can be obtained through the ipmitool fru print0 instruction. This invention does not specifically limit this method.

[0070] Furthermore, based on the obtained server model information, a file named after the model is located to obtain the server model configuration file.

[0071] Step 220: Obtain the high-speed serial bus (Peripheral Component Interface Express, PCIe) device information of the server, and obtain the independent redundant disk array card information of the server based on the PCIe device information. The independent redundant disk array card information includes the number of independent redundant disk array cards and manufacturer and model information.

[0072] Optionally, the high-speed serial bus PCIe device information of the server can be obtained through computer program instructions. For example, the high-speed serial bus PCIe device information of the server can be obtained through the lspci instruction. This invention does not specifically limit this.

[0073] Furthermore, Table 1 is a list of RAID card manufacturers and models. Based on the obtained high-speed serial bus PCIe device information of the server, the number of RAID cards is obtained by searching for manufacturer keywords (the second column in Table 1). For example, the manufacturer keyword for PMC is Smart, and the manufacturer keyword for LSI is LSI.

[0074] Furthermore, by parsing the RAID card's configuration file, the RAID card's model keyword is obtained (the third column in Table 1), and the manufacturer's model information of the RAID card is obtained by sequentially searching the model keyword (the first column in Table 1).

[0075] Table 1. RAID Card Manufacturer Model List

[0076] SAS9361 LSI 3108 SAS9440 LSI 3408 SAS9460 LSI 3508 PM8204 Smart 8024 PM8222 Smart 8222

[0077] Step 230: Obtain the Small Computer System Interface (SCSI) device information of the server, and obtain the hard disk backplane (Expander) information of the server based on the SCSI device information. The hard disk backplane information includes the number of hard disk backplanes and manufacturer and model information.

[0078] Optionally, the SCSI device information of the server can be obtained through computer program instructions. For example, the SCSI device information of the server can be obtained through the lsscsi instruction. This invention does not specifically limit this method.

[0079] Furthermore, Table 2 is a list of hard drive backplane manufacturers and models. Based on the obtained SCSI device information of the server, the number of hard drive backplanes can be obtained by searching for the keyword (enclsou).

[0080] Furthermore, by parsing the configuration file of the hard drive backplane, the model keywords of the hard drive backplane are obtained (the third column in Table 2), and the manufacturer model information of the hard drive backplane is obtained by searching the model keywords in turn (the first column in Table 2).

[0081] Table 2. Manufacturer Model List of Hard Drive Backplanes

[0082] PM8043 PMC 24Sx12G PM8044 PMC 26Sx12G SAS35X24 LSI SAS35X24 SAS35X36 LSI SAS35X36 SAS35X48 LSI SAS35X48

[0083] Step 240: Obtain the hard drive information of all hard drives of the server based on the SCSI device information, wherein the hard drive information includes at least one of the following: number of hard drives, manufacturer, model, drive letter, protocol type, and media type.

[0084] Optionally, based on the SCSI device information, information such as the manufacturer, quantity, model, and drive letter of all data disks and system disks installed on the server can be obtained by searching for hard disk keywords (disk); for example, the server NVMe device information can be obtained by using the nvme list command to obtain information such as the model and drive letter of all NVMe hard disks installed on the server; based on the obtained hard disk drive letter, the protocol type and media type of the corresponding hard disk can be obtained by using the smartctl -a <drive letter> command. This invention does not make specific limitations in this regard.

[0085] In this embodiment of the invention, the server model information is first obtained, and then the corresponding model configuration file is obtained based on the model information. Next, the server's high-speed serial bus PCIe device information is obtained, and then the server's independent redundant disk array card information is obtained based on the PCIe device information. Then, the server's small computer system interface (SCSI) device information is obtained, and then the server's hard drive backplane information is obtained based on the SCSI device information. Finally, the hard drive information of all hard drives on the server is obtained based on the SCSI device information. This enables automatic identification of the server model, RAID card, and hard drive backplane, as well as automatic identification of hard drive manufacturers, models, protocol types, etc., preparing for hard drive health monitoring.

[0086] Step 120: Based on the hard drive information, perform slow disk detection on each hard drive of the server;

[0087] In practice, based on the number of hard drives, manufacturer, model, drive letter, protocol type, media type, etc. in the hard drive information, it is determined whether the server is equipped with hard drives. If the server is equipped with hard drives, all hard drives of the server are traversed and slow disk detection is performed on each hard drive in turn.

[0088] Step 130: Based on the independent redundant disk array card information and the hard disk information, perform sub-health detection on each hard disk of the server;

[0089] In practice, based on the hard drive information, it is determined whether the server is equipped with hard drives. If the server is equipped with hard drives, all hard drives on the server are traversed, and based on the independent redundant disk array card information and the application type of the hard drive under test, sub-health checks are performed on each hard drive in turn. The application type of the hard drive under test refers to whether it is a data disk or a RAID system disk.

[0090] Step 140: If the target hard drive is detected to be in a slow or unhealthy state, issue a prompt message and, based on the model configuration file and hard drive backplane information, illuminate the fault indicator light of the target hard drive.

[0091] In practice, if the target hard drive is detected to be in a slow state, a prompt message will be issued indicating that the hard drive is in a slow state; if the target hard drive is detected to be in a sub-healthy state, a prompt message will be issued indicating that the hard drive is in a sub-healthy state.

[0092] Furthermore, based on the model configuration file and hard drive backplane information, a hard drive light status setting instruction is determined and executed to set the hard drive light status of the target hard drive to a fault state, thereby illuminating the fault light of the target hard drive.

[0093] In this embodiment of the invention, by acquiring the server's hard drive related information, slow disk detection is performed on each hard drive of the server based on the hard drive information; sub-health detection is performed on each hard drive of the server based on the independent redundant disk array card information and hard drive information; then, when the target hard drive is detected to be in a slow disk state or a sub-health state, a prompt message is issued, and the fault light of the target hard drive is lit based on the machine configuration file and hard drive backplane information, thereby realizing automatic health monitoring of the hard drives installed in the server, which can improve the universality, feasibility, accuracy and location of server hard drive health monitoring.

[0094] In some embodiments, the step of performing slow disk detection on each hard drive of the server based on the hard drive information includes:

[0095] Based on the hard drive information, determine whether the server is equipped with a hard drive;

[0096] If the server is equipped with hard drives, iterate through all the hard drives of the server and perform slow disk detection on each hard drive in turn.

[0097] In practice, the server is determined to have a hard drive based on whether the hard drive information is empty. If the server does not have a hard drive, a reason is given and the server hard drive slow disk detection process is not executed. If the server has a hard drive, all hard drives on the server are traversed and slow disk detection is performed on each hard drive in turn.

[0098] in, Figure 3 This is a schematic diagram illustrating the process of sequentially performing slow disk detection on each hard drive according to an embodiment of the present invention, as shown below. Figure 3 As shown, the step of sequentially performing slow disk detection on each hard drive includes:

[0099] Step 310: Obtain the input / output performance parameters of the hard drive under test;

[0100] In practice, the input / output performance parameters of the hard drive under test are obtained through computer program instructions. For example, the input / output performance parameters of the target hard drive are obtained through the iostat command, whereby the input / output performance parameters of the target hard drive include:

[0101] Metric 1: util (proportion of non-empty input / output queues);

[0102] Metric 2: svctm (average number of milliseconds to process each input / output request, excluding queuing time);

[0103] Metric 3: await (average number of milliseconds per input / output request processed, including processing time and waiting time);

[0104] Metric 4: IOPS (the number of input / output operations per second).

[0105] Step 320: Determine whether the input / output performance index parameters of the second target count period satisfy the slow disk condition within the consecutive first target count period;

[0106] It can be understood that the first target count period refers to the total number of periods for which the input / output performance index parameters need to be statistically analyzed, and the second target count period refers to the number of periods within the total number of periods for which the input / output performance index parameters meet the slow disk condition. Each period is 1 second. For example, the first target count period is 30 and the second target count period is 6. It should be noted that the values ​​of the first and second target count periods can be adjusted according to actual needs, and this invention does not impose specific limitations on them.

[0107] To improve the accuracy of server hard drive slow disk detection, the input and output performance index parameters are statistically analyzed within a consecutive first target number of cycles, and it is further determined whether the input and output performance index parameters meet the hard drive slow disk condition within a second target number of cycles within the consecutive first target number of cycles.

[0108] The slow disk condition refers to a situation where util (non-empty ratio of input / output queues) is ≥ 50%, and then the following indicators are queried. If any of the following indicators are met:

[0109] Indicator A: svctm ≥ 90ms;

[0110] Metric B: await ≥ 600ms and iops ≤ 50;

[0111] For example, within 30 consecutive cycles, the input / output performance parameters are statistically analyzed, and it is further determined whether there are 6 cycles within 30 consecutive cycles in which the input / output performance parameters meet the above-mentioned slow hard disk condition.

[0112] Step 330: If it exists, then determine that the hard drive under test is in a slow disk state.

[0113] In practice, if the input / output performance parameters meet the slow disk condition within a second target number of consecutive periods of the first target number of periods, then the hard disk under test is determined to be in a slow disk state.

[0114] For example, if the input / output performance parameters do not meet the above-mentioned slow disk condition for 6 out of 30 consecutive cycles, then the hard disk under test is determined to be in a slow disk state.

[0115] In this embodiment of the invention, based on hard drive information, it is determined whether the server is equipped with a hard drive; if the server is equipped with a hard drive, all hard drives on the server are traversed, and slow disk detection is performed on each hard drive in turn; wherein, the slow disk detection of each hard drive in turn includes: obtaining the input / output performance index parameters of the hard drive under test; determining whether there are input / output performance index parameters that meet the slow disk condition for a second target number of periods within a continuous first target number of periods; if so, it is determined that the hard drive under test is in a slow disk state, thereby realizing automatic slow disk detection of the hard drives equipped on the server, improving the universality, feasibility and accuracy of server hard drive slow disk detection.

[0116] In some embodiments, Figure 4 This is a second schematic flowchart of a hard disk health monitoring method according to an embodiment of the present invention, wherein step 140 includes:

[0117] Step 410: If the target hard drive is detected to be in a slow state, issue the first prompt message;

[0118] In practice, when the target hard drive is detected to be in a slow state, a first prompt message is issued to inform relevant personnel that the target hard drive is in a slow state.

[0119] Step 420: Obtain the backplane identification information and slot identification information of the target hard drive by parsing the model configuration file;

[0120] Furthermore, by parsing the model configuration file, the backplane identification information (backplane_id) and slot identification information (slot_id) of the target hard drive are obtained.

[0121] Step 430: Based on the hard drive backplane information, backplane identification information, and slot identification information, determine the hard drive light status band setting instruction;

[0122] Furthermore, based on the hard drive backplane information, backplane_id, and slot_id, the hard drive light status in-band setting instruction is determined. For example, the hard drive light status in-band setting instruction is ipmitool raw 0x3c 0xef 0x01 0x<backplane_id> 0x<slot_id>

[0123] 0x<led_state> Here, `led_state` represents the lamp state, indicating whether the lamp is on or off, and can be set to 0 or 1.

[0124] Step 440: Execute the hard disk light status setting instruction to set the hard disk light status of the target hard disk to a fault state.

[0125] Furthermore, after determining the setting instruction within the hard drive light status band, the instruction is executed to set the hard drive light status of the corresponding slot to a fault state, thereby quickly locating the hard drive in a slow state.

[0126] In this embodiment of the invention, when a target hard drive is detected to be in a slow state, a first prompt message is issued. First, the backplane identification information and slot identification information of the target hard drive are obtained by parsing the model configuration file. Then, based on the hard drive backplane information, backplane identification information, and slot identification information, a setting instruction within the hard drive light status band is determined. Finally, the setting instruction within the hard drive light status band is executed to set the hard drive light status of the target hard drive to a fault state. This improves the accuracy and locability of server hard drive slow disk detection when the target hard drive is in a slow state.

[0127] In some embodiments, the method further includes:

[0128] During the slow disk detection cycle, the input and output performance index parameters are recorded, analyzed, and processed. The changing trend of the input and output performance index parameters is predicted based on the gradient boosting decision tree algorithm.

[0129] It should be noted that the Gradient Boosting Decision Tree (XGBoost) algorithm internally uses Newton's method, expands the internal function to second-order Taylor expansion, and adds a regularization term to the internal function. By using the Gradient Boosting Decision Tree (XGBoost) algorithm to predict the changing trends of input and output performance index parameters, it can more efficiently, quickly, and accurately determine the changing trends of slow disk status and predict the trend of slow disks. It can truly achieve the detection and prediction of hard disk health status, increasing the reliability of server hard disks.

[0130] In practical implementation, for the four input / output performance metrics of the target hard drive—util (non-empty IO queue ratio), svctm (average milliseconds per IO request processing, excluding queuing time), await (average milliseconds per IO request processing, including processing time and waiting time), and iops (number of input / output operations per second)—a deviation function (the deviation function refers to the deviation relative to the metric) is constructed after each slow disk detection cycle (exemplarily, 30 consecutive cycles, each cycle being 1 second).

[0131] util deviation function: y u =f u (s,a,i);

[0132] svctm deviation function: y s =f s (u,a,i);

[0133] await deviation function: y a =f a (u,s,i);

[0134] IOPS deviation function: y i =f i (u,s,a);

[0135] Where, parameter u represents the input / output performance index parameter util; parameter s represents the input / output performance index parameter svctm; parameter a represents the input / output performance index parameter await; and parameter i represents the input / output performance index parameter iops.

[0136] The results for the deviation function in the first period are as follows:

[0137] Util deviation results for period 1: y u1 =f u (s1,a1,i1);

[0138] SVCTM deviation results for cycle 1: y s1 =f s (u1,a1,i1);

[0139] Await bias result for period 1: y a1 =f a (u1,s1,i1);

[0140] IOPS bias results for period 1: y i1 =f i (u1,s1,a1);

[0141] The results for the 30th periodic function are as follows:

[0142] Util deviation result for cycle 30: y u30 =f u (s 30 ,a 30 i 30 );

[0143] SVCTM deviation result for cycle 30: y s30 =f s (u 30 ,a 30 i 30 );

[0144] Await bias result for cycle 30: y a30 =f a (u 30 ,s 30 i 30 );

[0145] IOPS bias result for cycle 30: y i30 =f i (u 30 ,s 30 ,a 30 );

[0146] Therefore, the final function results for 30 periods are as follows:

[0147] Final util deviation results after 30 cycles:

[0148] Final SVCTM bias results after 30 cycles:

[0149] Final await bias results after 30 cycles:

[0150] Final IOPS deviation results after 30 cycles:

[0151] Based on the above, the final function results for the nth slow disk detection cycle are as follows:

[0152] Final util deviation result after n slow disk detection cycles:

[0153] Final SVCTM deviation results after n slow disk detection cycles:

[0154] Final await deviation result after n slow disk detection cycles:

[0155] Final IOPS deviation results after n slow disk detection cycles:

[0156] After the above process, the objective function can be constructed:

[0157] The objective function of util:

[0158] svctm objective function:

[0159] await target function:

[0160] IOPS target function:

[0161] in, The loss function; Ω(f) uk ) is the complexity control function; n and K are consistent representations, representing the nth slow disk detection cycle; k represents the count from 1 to K; i represents the count from 1 to n.

[0162] The following section will elaborate on the objective function of util as an example:

[0163] (1) Perform a Taylor expansion:

[0164]

[0165] Among them, g i and h i It can be viewed as the residuals of n-1 slow disk detection cycles (the first and second derivatives of the loss function, respectively); t represents the current slow disk detection cycle, i.e., the nth slow disk detection cycle.

[0166] (2) Constants:

[0167]

[0168] Complexity:

[0169] Where γ and λ are hyperparameters of the complexity function; T represents the number of slow disk detection cycles executed; ω j This represents the predicted value for the j-th slow disk detection cycle.

[0170] (3) Combining (1) and (2) into the objective function yields:

[0171]

[0172] Therefore, the optimal solution can be obtained:

[0173]

[0174] Among them, G j g represents the sample of the j-th slow disk detection cycle. i Total; H j h represents the sample of the j-th slow disk detection cycle. i sum.

[0175] In this embodiment of the invention, input and output performance index parameters are recorded during the slow disk detection cycle, and the input and output performance index parameters are analyzed and processed. The changing trend of the input and output performance index parameters is predicted based on the gradient boosting decision tree algorithm, which can improve the universality of hard disk status prediction and obtain the changing trend of the slow disk status more efficiently, quickly and accurately.

[0176] In some embodiments, the sub-health detection of each hard drive of the server based on the independent redundant disk array card information and the hard drive information includes:

[0177] Based on the hard drive information, determine whether the server is equipped with a hard drive;

[0178] If the server is equipped with hard drives, iterate through all the hard drives of the server and perform sub-health checks on each hard drive in turn;

[0179] In practice, the server is determined to have a hard drive based on whether the hard drive information is empty. If the server does not have a hard drive, the reason is displayed and the server hard drive sub-health detection process is not executed. If the server has a hard drive, all hard drives of the server are traversed and each hard drive is checked for sub-health in turn.

[0180] in, Figure 5 This is a schematic diagram illustrating the process of sequentially performing sub-health checks on each hard drive according to an embodiment of the present invention, as shown below. Figure 5 As shown, the step of sequentially performing sub-health checks on each hard drive includes:

[0181] Step 510: Based on the independent redundant disk array card information and the application form of the hard disk under test, obtain the key parameter information of the SMART self-monitoring analysis and reporting technology of the hard disk under test;

[0182] In practical implementation, based on the independent redundant disk array card information and the application form of the hard disk under test, the key parameter information of the self-monitoring analysis and reporting technology SMART of the hard disk under test is obtained through computer program instructions. The key parameter information of SMART is different for hard disks with different interface forms and different media. Table 3 shows the key parameter information of SMART.

[0183] Table 3 SMART Key Parameter Information

[0184]

[0185]

[0186] Different applications of the hard drive under test require different methods (instructions) for obtaining SMART key parameter information. Several methods for obtaining SMART key parameter information are illustrated below.

[0187] a. System disk: Different RAID card manufacturers use different commands.

[0188] LSI RAID card: smartctl-all-device = sat+megaraid, <did><Disk letter>, where <did>Obtained from the vendor's tools.

[0189] PMC RAID card: smartctl-d cciss, <num>-a<drive letter>, where <num>Obtained from the vendor's tools;

[0190] b. NVMe drive:

[0191] nvme smart-log<drive letter);

[0192] c. Data disk:

[0193] smartctl -a <drive letter>;

[0194] Step 520: Determine whether the SMART key parameter information meets the conditions for hard drive sub-health.

[0195] Furthermore, after obtaining the SMART key parameter information of the hard drive under test, it is determined whether each of the SMART key parameter information in Table 3 meets the sub-health conditions of the hard drive. Table 4 shows the SMART key parameter information that meets the sub-health conditions of the hard drive.

[0196] Table 4 shows the key SMART parameter information, indicating whether the hard drive meets the sub-health conditions.

[0197]

[0198]

[0199]

[0200] Step 530: If the condition is met, then the hard drive under test is determined to be in a sub-healthy state.

[0201] Furthermore, if the SMART key parameter information meets the sub-health conditions of the hard drive in Table 4, then the hard drive under test is determined to be in a sub-healthy state.

[0202] In this embodiment of the invention, based on hard drive information, it is determined whether the server is equipped with a hard drive; if the server is equipped with a hard drive, all hard drives on the server are traversed, and each hard drive is sequentially subjected to sub-health detection; wherein, sequentially performing sub-health detection on each hard drive includes: obtaining the SMART key parameter information of the hard drive under test based on the independent redundant disk array card information and the application form of the hard drive under test; determining whether the SMART key parameter information meets the sub-health conditions of the hard drive; if it does, the hard drive under test is determined to be in a sub-healthy state, thereby realizing automatic sub-health detection of the hard drives equipped on the server, improving the universality, feasibility and accuracy of server hard drive sub-health detection.

[0203] In some embodiments, Figure 6 This is a third flowchart illustrating a hard disk health monitoring method according to an embodiment of the present invention. Step 140 includes:

[0204] Step 610: If the target hard drive is detected to be in a sub-healthy state, issue a second prompt message;

[0205] In practice, if the target hard drive is detected to be in a sub-healthy state, a second prompt message is issued to notify relevant personnel that the target hard drive is in a sub-healthy state.

[0206] Step 620: Obtain the backplane identification information and slot identification information of the target hard drive by parsing the model configuration file;

[0207] Furthermore, by parsing the model configuration file, the backplane identification information (backplane_id) and slot identification information (slot_id) of the target hard drive are obtained.

[0208] Step 630: Based on the hard drive backplane information, backplane identification information, and slot identification information, determine the hard drive light status band setting instruction;

[0209] Furthermore, based on the hard drive backplane information, backplane_id, and slot_id, the hard drive light status in-band setting instruction is determined. For example, the hard drive light status in-band setting instruction is ipmitool raw 0x3c 0xef 0x01 0x<backplane_id> 0x<slot_id>

[0210] 0x<led_state> Here, `led_state` represents the lamp state, indicating whether the lamp is on or off, and can be set to 0 or 1.

[0211] Step 640: Execute the hard disk light status setting instruction to set the hard disk light status of the target hard disk to a fault state.

[0212] Furthermore, after determining the setting instruction within the hard drive light status band, the instruction is executed to set the hard drive light status of the corresponding slot to a fault state, thereby quickly locating the hard drive in a sub-healthy state.

[0213] In this embodiment of the invention, when a target hard drive is detected to be in a sub-healthy state, a second prompt message is issued. First, the backplane identification information and slot identification information of the target hard drive are obtained by parsing the model configuration file. Then, based on the hard drive backplane information, backplane identification information, and slot identification information, a setting instruction within the hard drive light status band is determined. Finally, the setting instruction within the hard drive light status band is executed to set the hard drive light status of the target hard drive to a fault state. This improves the accuracy and locability of server hard drive sub-health detection when the target hard drive is in a sub-healthy state.

[0214] The hard disk health monitoring device provided by the present invention is described below. The hard disk health monitoring device described below can be referred to in correspondence with the hard disk health monitoring method described above.

[0215] Figure 7 This is a schematic diagram of the structure of a hard disk health monitoring device provided in one embodiment of the present invention, as shown below. Figure 7 As shown, the hard drive health monitoring device 700 includes:

[0216] The acquisition unit 710 is used to acquire hard disk related information of the server. The hard disk related information includes: the model configuration file corresponding to the server, the independent redundant disk array card information of the server, the hard disk backplane information of the server, and the hard disk information of all hard disks of the server.

[0217] The slow disk detection unit 720 is used to perform slow disk detection on each hard disk of the server based on the hard disk information.

[0218] The sub-health detection unit 730 is used to perform sub-health detection on each hard drive of the server based on the independent redundant disk array card information and the hard drive information.

[0219] The fault light unit 740 is used to issue a prompt message when the target hard drive is detected to be in a slow or sub-healthy state, and to light up the fault light of the target hard drive based on the model configuration file and hard drive backplane information.

[0220] In some embodiments, obtaining the server's hard drive-related information includes:

[0221] Obtain the server's model information, and obtain the corresponding model configuration file based on the model information;

[0222] Obtain the high-speed serial bus PCIe device information of the server, and obtain the independent redundant disk array card information of the server based on the PCIe device information. The independent redundant disk array card information includes the number of independent redundant disk array cards and manufacturer and model information.

[0223] Obtain the Small Computer System Interface (SCSI) device information of the server, and obtain the hard disk backplane information of the server based on the SCSI device information. The hard disk backplane information includes the number of hard disk backplanes and their manufacturer and model information.

[0224] The hard drive information of all hard drives of the server is obtained based on the SCSI device information, wherein the hard drive information includes at least one of the following: number of hard drives, manufacturer, model, drive letter, protocol type, and media type.

[0225] In some embodiments, the step of performing slow disk detection on each hard drive of the server based on the hard drive information includes:

[0226] Based on the hard drive information, determine whether the server is equipped with a hard drive;

[0227] If the server is equipped with hard drives, iterate through all the hard drives of the server and perform slow disk detection on each hard drive in turn.

[0228] The step of sequentially performing slow disk detection on each hard drive includes:

[0229] Obtain the input / output performance parameters of the hard drive under test;

[0230] Determine whether there exists a second target count period within the first consecutive target count period that the input / output performance index parameters satisfy the slow disk condition;

[0231] If it exists, then the hard drive under test is determined to be in a slow disk state.

[0232] In some embodiments, the step of issuing a prompt message and illuminating the fault indicator light of the target hard drive based on the model configuration file and hard drive backplane information when the target hard drive is detected to be in a slow or unhealthy state includes:

[0233] If the target hard drive is detected to be in a slow state, the first prompt message will be issued;

[0234] By parsing the model configuration file, the backplane identification information and slot identification information of the target hard drive are obtained;

[0235] Based on the hard drive backplane information, backplane identification information, and slot identification information, determine the hard drive light status band setting instructions.

[0236] Execute the hard drive light status setting instruction to set the hard drive light status of the target hard drive to a fault state.

[0237] In some embodiments, the apparatus further includes:

[0238] The prediction unit is used to record input and output performance index parameters during the slow disk detection cycle, analyze and process the input and output performance index parameters, and predict the changing trend of input and output performance index parameters based on the gradient boosting decision tree algorithm.

[0239] In some embodiments, the sub-health detection of each hard drive of the server based on the independent redundant disk array card information and the hard drive information includes:

[0240] Based on the hard drive information, determine whether the server is equipped with a hard drive;

[0241] If the server is equipped with hard drives, iterate through all the hard drives of the server and perform sub-health checks on each hard drive in turn;

[0242] The step of sequentially performing sub-health checks on each hard drive includes:

[0243] Based on the independent redundant disk array card information and the application form of the hard drive under test, the key parameter information of the SMART self-monitoring analysis and reporting technology of the hard drive under test is obtained;

[0244] Determine whether the SMART key parameter information meets the conditions for hard drive sub-health;

[0245] If the conditions are met, the hard drive under test is determined to be in a sub-healthy state.

[0246] In some embodiments, the step of issuing a prompt message and illuminating the fault indicator light of the target hard drive based on the model configuration file and hard drive backplane information when the target hard drive is detected to be in a slow or unhealthy state includes:

[0247] If the target hard drive is detected to be in a sub-healthy state, a second prompt message will be issued;

[0248] By parsing the model configuration file, the backplane identification information and slot identification information of the target hard drive are obtained;

[0249] Based on the hard drive backplane information, backplane identification information, and slot identification information, determine the hard drive light status band setting instructions.

[0250] Execute the hard drive light status setting instruction to set the hard drive light status of the target hard drive to a fault state.

[0251] It should be noted that the hard disk health monitoring device provided in this embodiment of the invention can implement all the method steps implemented in the above-described hard disk health monitoring method embodiment, and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0252] It should be noted that each implementation method of this application can be freely combined, rearranged, or executed individually, and does not need to rely on or depend on a fixed execution order.

[0253] Figure 8 This is a schematic diagram of the physical structure of an electronic device provided in one embodiment of the present invention, as shown below. Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other through the communication bus 840. The processor 810 can call logical instructions in the memory 830 to execute a hard disk health monitoring method, which includes: acquiring hard disk related information of the server, including: the server's corresponding model configuration file, the server's independent redundant disk array card information, the server's hard disk backplane information, and hard disk information of all hard disks of the server; performing slow disk detection on each hard disk of the server based on the hard disk information; performing sub-health detection on each hard disk of the server based on the independent redundant disk array card information and the hard disk information; and issuing a prompt message when a target hard disk is detected to be in a slow disk state or a sub-health state, and illuminating the fault light of the target hard disk based on the model configuration file and the hard disk backplane information.

[0254] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0255] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the hard disk health monitoring method provided by the above methods. The method includes: acquiring hard disk related information of a server, the hard disk related information including: the server's corresponding model configuration file, the server's independent redundant disk array card information, the server's hard disk backplane information, and hard disk information of all hard disks of the server; performing slow disk detection on each hard disk of the server based on the hard disk information; performing sub-health detection on each hard disk of the server based on the independent redundant disk array card information and the hard disk information; and issuing a prompt message when a target hard disk is detected to be in a slow disk state or a sub-health state, and illuminating the fault light of the target hard disk based on the model configuration file and the hard disk backplane information.

[0256] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the hard disk health monitoring method provided by the above methods. The method includes: acquiring hard disk related information of a server, the hard disk related information including: the server's model configuration file, the server's independent redundant disk array card information, the server's hard disk backplane information, and hard disk information of all hard disks of the server; performing slow disk detection on each hard disk of the server based on the hard disk information; performing sub-health detection on each hard disk of the server based on the independent redundant disk array card information and the hard disk information; and issuing a prompt message when a target hard disk is detected to be in a slow disk state or a sub-health state, and illuminating the fault light of the target hard disk based on the model configuration file and the hard disk backplane information.

[0257] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0258] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0259] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.< / num> < / num> < / did> < / did>

Claims

1. A method for monitoring the health of a hard drive, characterized in that, include: Obtain the server's hard drive related information, which includes: the server's corresponding model configuration file, the server's independent redundant disk array card information, the server's hard drive backplane information, and the hard drive information of all the server's hard drives; Based on the hard drive information, slow disk detection is performed on each hard drive of the server; Based on the independent redundant disk array card information and the hard disk information, sub-health detection is performed on each hard disk of the server; If the target hard drive is detected to be in a slow or sub-healthy state, a prompt message is issued, and based on the model configuration file and hard drive backplane information, the hard drive light status setting instruction is determined and executed to set the hard drive light status of the target hard drive to a fault state. The steps for determining the hard disk light status setting instruction include: By parsing the model configuration file, the backplane identification information and slot identification information of the target hard drive are obtained, and based on the hard drive backplane information, backplane identification information and slot identification information, the setting instructions within the hard drive light status band are determined; The method further includes: During the slow disk detection cycle, input and output performance index parameters are recorded, and these parameters are analyzed and processed. The changing trend of the input and output performance index parameters is predicted based on the gradient boosting decision tree algorithm. The gradient boosting decision tree algorithm uses Newton's method, expands the internal function to second order Taylor expansion, and adds a regularization term to the internal function.

2. The hard disk health monitoring method according to claim 1, characterized in that, The process of obtaining server hard drive related information includes: Obtain the server's model information, and obtain the corresponding model configuration file based on the model information; Obtain the high-speed serial bus PCIe device information of the server, and obtain the independent redundant disk array card information of the server based on the PCIe device information. The independent redundant disk array card information includes the number of independent redundant disk array cards and manufacturer and model information. Obtain the Small Computer System Interface (SCSI) device information of the server, and obtain the hard disk backplane information of the server based on the SCSI device information. The hard disk backplane information includes the number of hard disk backplanes and their manufacturer and model information. The hard drive information of all hard drives of the server is obtained based on the SCSI device information, wherein the hard drive information includes at least one of the following: number of hard drives, manufacturer, model, drive letter, protocol type, and media type.

3. The hard disk health monitoring method according to claim 1, characterized in that, The step of performing slow disk detection on each hard drive of the server based on the hard drive information includes: Based on the hard drive information, determine whether the server is equipped with a hard drive; If the server is equipped with hard drives, iterate through all the hard drives of the server and perform slow disk detection on each hard drive in turn. The step of sequentially performing slow disk detection on each hard drive includes: Obtain the input / output performance parameters of the hard drive under test; Determine whether there exists a second target count period within the first consecutive target count period that the input / output performance index parameters satisfy the slow disk condition; If it exists, then the hard drive under test is determined to be in a slow disk state.

4. The hard disk health monitoring method according to claim 3, characterized in that, The method of issuing a prompt message when the target hard drive is detected to be in a slow or unhealthy state includes: If the target hard drive is detected to be in a slow state, the first warning message will be issued.

5. The hard disk health monitoring method according to claim 1, characterized in that, The sub-health detection of each hard drive of the server based on the independent redundant disk array card information and the hard drive information includes: Based on the hard drive information, determine whether the server is equipped with a hard drive; If the server is equipped with hard drives, iterate through all the hard drives of the server and perform sub-health checks on each hard drive in turn; The step of sequentially performing sub-health checks on each hard drive includes: Based on the independent redundant disk array card information and the application form of the hard drive under test, the key parameter information of the SMART self-monitoring analysis and reporting technology of the hard drive under test is obtained; Determine whether the SMART key parameter information meets the conditions for hard drive sub-health; If the conditions are met, the hard drive under test is determined to be in a sub-healthy state.

6. The hard disk health monitoring method according to claim 5, characterized in that, The method of issuing a prompt message when the target hard drive is detected to be in a slow or unhealthy state includes: If the target hard drive is detected to be in a sub-healthy state, a second prompt message will be issued.

7. A hard disk health monitoring device, characterized in that, include: The acquisition unit is used to acquire hard drive related information of the server. The hard drive related information includes: the model configuration file corresponding to the server, the independent redundant disk array card information of the server, the hard drive backplane information of the server, and the hard drive information of all hard drives of the server. The slow disk detection unit is used to perform slow disk detection on each hard disk of the server based on the hard disk information. The sub-health detection unit is used to perform sub-health detection on each hard drive of the server based on the independent redundant disk array card information and the hard drive information. The fault light unit is used to issue a prompt message when the target hard drive is detected to be in a slow or sub-healthy state, and to determine the hard drive light status setting instruction based on the model configuration file and hard drive backplane information, and execute the hard drive light status setting instruction to set the hard drive light status of the target hard drive to a fault state. The fault light unit is also used to: obtain the backplane identification information and slot identification information of the target hard drive by parsing the model configuration file, and determine the hard drive light status band setting instruction based on the hard drive backplane information, backplane identification information and slot identification information; The device is further configured to: record input and output performance index parameters during the slow disk detection cycle, analyze and process the input and output performance index parameters, and predict the changing trend of the input and output performance index parameters based on the gradient boosting decision tree algorithm; the gradient boosting decision tree algorithm internally uses Newton's method, expands the internal function to second order Taylor expansion, and adds a regularization term to the internal function.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the hard disk health monitoring method as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the hard disk health monitoring method as described in any one of claims 1 to 6.

Images