Storage link failure cause localization method, storage subsystem, and computer device

Abstract

The present application relates to the technical field of servers, and discloses a storage link failure cause localization method, a storage subsystem, and a computer device. The method comprises: acquiring an on / off state of a serial log recording module of an extended backplane; when the serial log recording module of the extended backplane is in an on state, using the serial log recording module to record command execution information of a target command issued by a server; when a failure occurs in a storage link, acquiring first log information corresponding to a RAID card, second log information corresponding to a hard disk, and the command execution information of the target command recorded by the serial log recording module; and sending the first log information, the second log information, and the command execution information of the target command to a server host side, so that a technician performs storage link failure cause localization on the basis of the first log information, the second log information, and the command execution information of the target command. The present application improves the accuracy of a storage link failure cause localization.

Description

Storage link fault cause positioning method, storage subsystem and computer device

[0001] Cross-reference to Related Applications

[0002] The present application claims priority to the Chinese patent application No. 202411303460.1, filed on September 19, 2024, and entitled "Storage link fault cause positioning method, storage subsystem and computer device", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] The present application relates to the technical field of servers, and in particular to a storage link fault cause positioning method, a storage subsystem and a computer device. BACKGROUND

[0004] With the continuous development of the server industry, users have higher and higher requirements for the security and reliability of servers. The storage subsystem composed of disk array cards, hard disks, expansion backplanes, cables and other components is an important part of the server whole machine, and therefore, the security and reliability of the storage subsystem are crucial.

[0005] In the related art, in the case that a storage link of a storage subsystem fails, the cause of the failure of the storage link can be determined by using log files of each component on the storage link.

[0006] However, the inventors have realized that when the cause of the failure of the storage link is determined by using the log files of each component on the storage link, since the log file of the expansion backplane can only record basic abnormal events of the expansion backplane and cannot record information related to command execution, the accuracy of the determined cause of the failure of the storage link is low. SUMMARY

[0007] Therefore, the present application provides a storage link fault cause positioning method, a storage subsystem and a computer device to solve the problem that the accuracy of the determined cause of the failure of the storage link is low due to the fact that the log file of the expansion backplane can only record basic abnormal events of the expansion backplane and cannot record information related to command execution in the related art.

[0008] In a first aspect, the present application provides a storage link fault cause positioning method, comprising:

[0009] obtaining a switch state of a serial log recording module of an expansion backplane;

[0010] in the case that the switch state of the serial log recording module of the expansion backplane is an open state, recording command execution information of a target command issued by a server by using the serial log recording module;

[0011] In the case of a storage link failure, the first log information corresponding to the disk array card, the second log information corresponding to the hard disk, and the command execution information of the target command recorded by the serial log recording module are obtained;

[0012] The first log information, the second log information, and the command execution information of the target command are sent to the server host end, so that the technical personnel can locate the cause of the storage link failure based on the first log information, the second log information, and the command execution information of the target command received by the server host end.

[0013] In some embodiments, the command execution information of the target command issued by the server is recorded by using the serial log recording module, including:

[0014] The command issued by the server is received;

[0015] In the case of a target command type, the command is determined to be a target command, wherein the target command type is a read-write type command, a monitoring type command, a self-monitoring and analysis type command, an energy-saving type command, and a link reset type command;

[0016] The command execution information of the target command issued by the server is recorded by using the serial log recording module in an overwrite manner.

[0017] In some embodiments, in the case of a storage link failure, the command execution information of the target command recorded by the serial log recording module is obtained, including:

[0018] In the case of a storage link failure, it is determined whether the serial log recording module is set with a keyword trigger condition;

[0019] In the case of the serial log recording module being set with a keyword trigger condition, the command execution information recorded by the serial log recording module is filtered by using the keywords contained in the keyword trigger condition to obtain the command execution information of the target command.

[0020] In some embodiments, the command execution information of the target command is obtained by filtering the command execution information recorded by the serial log recording module by using the keywords contained in the keyword trigger condition, including:

[0021] For any command execution information, if the command execution information includes at least one keyword, the command execution information is determined to be the command execution information of the target command.

[0022] In some embodiments, the method for locating the cause of the storage link failure further includes:

[0023] In a case where the serial log recording module is not set with the keyword trigger condition, the command execution information of a target command recorded by the serial log recording module currently is taken as the command execution information of the target command.

[0024] In some embodiments, the method for locating the cause of the storage link failure further comprises:

[0025] In a case where the storage link fails, the serial log recording module is paused from recording the command execution information of the target command issued by the server.

[0026] In some embodiments, the method for locating the cause of the storage link failure further comprises:

[0027] After the first log information, the second log information and the command execution information of the target command are sent to the server host end, the serial log recording module continues to record the command execution information of the target command issued by the server.

[0028] In some embodiments, the method for locating the cause of the storage link failure further comprises:

[0029] In a case where the storage link fails, the basic log information corresponding to the extended backboard is acquired;

[0030] The first log information, the second log information and the command execution information of the target command are sent to the server host end, so that a technician locates the cause of the storage link failure based on the first log information, the second log information and the command execution information of the target command received by the server host end, comprising:

[0031] The first log information, the second log information, the basic log information and the command execution information of the target command are sent to the server host end, so that a technician locates the cause of the storage link failure based on the first log information, the second log information, the basic log information and the command execution information of the target command received by the server host end.

[0032] In some embodiments, the method for locating the cause of the storage link failure further comprises:

[0033] In a case where the switch state of the serial log recording module of the extended backboard is the off state, the serial log recording module of the extended backboard is used to record the basic log information corresponding to the extended backboard.

[0034] In some embodiments, the command execution information comprises: command response time, command type, command mode, execution state information and error state information.

[0035] In a second aspect, the application provides an extended backboard, comprising: a serial log recording module, wherein the serial log recording module comprises a switch control module and a serial log recording subunit;

[0036] The switch control module is configured to control the switch state of the serial log recording subunit in response to the operation of the user.

[0037] The serial log recording subunit is configured to record the command execution information of the target command issued by the server host in the on state.

[0038] In a third aspect, the present application provides a storage subsystem, comprising: a disk array card, an expansion backplane and a hard disk, wherein the storage subsystem is configured to locate the cause of the storage link fault based on the method for locating the cause of the storage link fault according to the first aspect or any one of the corresponding embodiments thereof.

[0039] In a fourth aspect, the present application provides a computer device, comprising: a memory and one or more processors, the memory storing computer readable instructions, and the computer readable instructions being executed by the one or more processors to cause the one or more processors to perform the method for locating the cause of the storage link fault according to the first aspect or any one of the corresponding embodiments thereof.

[0040] In a fifth aspect, the present application provides a computer readable storage medium, the computer readable storage medium storing computer instructions, and the computer instructions being configured to cause a computer to perform the method for locating the cause of the storage link fault according to the first aspect or any one of the corresponding embodiments thereof.

[0041] In a sixth aspect, the present application provides a computer readable instruction product, comprising computer readable instructions, and the computer readable instructions being configured to cause a computer to perform the method for locating the cause of the storage link fault according to the first aspect or any one of the corresponding embodiments thereof. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the specific embodiments or the related art, the drawings needed in the specific embodiments or the related art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0043] FIG. 1 is a flowchart of a method for locating the cause of a storage link fault according to some embodiments of the present application;

[0044] FIG. 2 is a flowchart of another method for locating the cause of a storage link fault according to some embodiments of the present application;

[0045] FIG. 3 is a schematic diagram of the data structure of the command execution information recorded in the serial log recording module according to some embodiments of the present application;

[0046] FIG. 4 is a schematic diagram of command execution information recorded by a serial logging module according to some embodiments of the present application;

[0047] FIG. 5 is a schematic diagram of the structure of an extended backplane according to some embodiments of the present application;

[0048] FIG. 6 is a flowchart of obtaining command execution information of a target command according to some embodiments of the present application;

[0049] FIG. 7 is a schematic diagram of the structure of a storage subsystem according to some embodiments of the present application;

[0050] FIG. 8 is a schematic diagram of the hardware structure of a computer device according to some embodiments of the present application. DETAILED DESCRIPTION

[0051] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0052] With the continuous development of the server industry, the server market gradually covers various fields, especially the mainstream fields such as finance, Internet, communication, etc.

[0053] With the rapid development of the Internet, users have higher and higher requirements for the high security of information and the high reliability of the system. Memory, hard disk, solid state disk (SSD), redundant array of independent disks (RAID) card, network card and other key components are the most important components of the server, and their stability and reliability directly affect the user and business application and data security. Therefore, the security of the above components needs to be ensured.

[0054] The storage subsystem composed of a RAID card / Serial Attached SCSI (SAS) card, a hard disk / SSD, an Expander card / backboard, a cable and other components is one of the most important components of a server, and its stability and reliability directly affect the security of the server. It should be noted that the storage link includes the connection path between the storage subsystem and the components of the storage subsystem. Due to the diversity of the components of the storage subsystem and the complexity of the interconnection of the components of the storage subsystem, the probability of compatibility failure is relatively high. According to statistics, the failure of the storage link accounts for more than 60% of the total server failure on average every year, and therefore, accurately locating the root cause of the storage link failure and quickly providing a solution are crucial to ensuring the security of the storage link.

[0055] Accurately locating the root cause of the storage link failure and quickly providing a solution not only require rich experience and professional tools, but also require innovative thinking and diversified analysis methods.

[0056] In the related art, for common fault types of the server storage link, such as disk drop, command timeout, speed reduction, high delay, error and the like, the industry and component manufacturers have rich debugging methods and standard processing methods.

[0057] The common method for locating the cause of the storage link failure includes locating the cause of the failure of the storage link by means of the storage link related log. It also includes locating the cause of the failure of the storage link by means of a SAS protocol analyzer.

[0058] The method for locating the cause of the failure of the storage link by means of the storage link related log includes: first, according to the fault phenomenon and the error log, the time point of the failure is determined, the problem is analyzed, and the suspected direction is determined through software and hardware comparison, model comparison, customer application, failure rule and cross verification analysis; second, the components are checked from the firmware and the driver, and it is checked whether the new release version solves the similar problem; and third, the problem is analyzed and located by combining the RAIS card log, the hard disk log, the baseboard management controller log and the operating system log.

[0059] The method for locating the cause of the failure of the storage link by means of the SAS protocol analyzer includes: the trace of the link at the time of the failure is captured by the SAS protocol analyzer for further positioning analysis until the failure is solved. Generally, as long as the trace at the time of the failure is successfully captured, the cause of the failure of the storage link can be located.

[0060] However, due to the functional positioning of the components themselves, technical capabilities, and limitations of log storage space size, etc., the log files of the components cannot record all types of failures and information communication processes, especially the log files of the expansion backboard, which can only record some hardware firmware information, backboard SAS address, physical (phy) topology, firmware fatal event, phy error technology, backboard transmission / reception (Tx / Rx) signal, rate, etc., and cannot record detailed information such as other failure types and command processing abnormalities. As one of the most important components of the storage link, the log recording function of the Expander backboard directly affects the storage link failure cause positioning. The Expander log recording function in the related art is relatively weak, only records the above basic information, and the accuracy of the determined cause of the storage link failure is low.

[0061] The SAS protocol analyzer can be connected in series between the Expander backboard and the hard disk, and can confirm whether the problem is on the host side or the device side by capturing the communication of the link at the time of failure. However, the SAS protocol analyzer supports at most 4 interfaces, and if the failure can be stably reproduced at a fixed hard disk slot position, the trace can be captured to locate the cause of the storage link failure. However, for problems that cannot be stably reproduced or cannot be fixed, the current means cannot be used to locate the cause of the storage link failure.

[0062] The present application provides a storage link failure cause positioning method, which records the command execution information of the target command issued by the server by using the serial log recording module when the switch state of the serial log recording module of the expansion backboard is in the open state, and sends the first log information corresponding to the disk array card, the second log information corresponding to the hard disk, and the command execution information of the target command recorded by the serial log recording module to the server host side in the case of storage link failure, so that the technical personnel can locate the cause of the storage link failure based on the first log information, the second log information, and the command execution information of the target command received by the server host side, thereby improving the accuracy of the storage link failure cause positioning, and recording and analyzing the problems that cannot be stably reproduced or cannot be fixed to achieve the effect of storage link failure cause positioning.

[0063] The present application provides a storage link failure cause positioning method, and it should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0064] In some embodiments, a storage link fault cause positioning method is provided, which can be used in a storage subsystem. FIG. 1 is a flowchart of a storage link fault cause positioning method according to the present application. As shown in FIG. 1, the flow includes the following steps:

[0065] In step S101, a switch state of a serial log recording module of an expansion backplane is acquired.

[0066] In the embodiment, the switch state of the serial log recording module of the expansion backplane is in an open state by default.

[0067] It should be noted that the opening of the serial log recording module consumes part of the resources for additional data statistics and recording, which will inevitably affect the input and output performance of the system. The user can select to close according to the actual business situation, and then select to use the command to open when the storage link fault occurs.

[0068] In step S102, in a case where the switch state of the serial log recording module of the expansion backplane is in an open state, the command execution information of a target command issued by a server is recorded by using the serial log recording module.

[0069] In the embodiment, the switch state of the serial log recording module of the expansion backplane is in an open state, which means that the serial log recording module can be used for log recording. Therefore, the command execution information of the target command issued by the server is recorded by using the serial log recording module.

[0070] It should be noted that the serial log recording module of the Expander backplane is directly connected to the Expander chip. The Serial Advanced Technology Attachment (SATA) protocol command and the SAS protocol command issued by the server host are both forwarded by the Expander chip.

[0071] It should be noted that the serial log recording module is designed to record only the most critical and most likely to fail protocol commands and key events.

[0072] In step S103, in a case where a storage link fault occurs, the first log information corresponding to a disk array card, the second log information corresponding to a hard disk, and the command execution information of the target command recorded by the serial log recording module are acquired.

[0073] The disk array card is a RAID card. The RAID card and the hard disk component have powerful and complete log recording functions, and can record abnormal events such as various errors and timeouts in detail. Therefore, in the case of a storage link failure, the first log information corresponding to the RAID card, the second log information corresponding to the hard disk, and the command execution information of the target command recorded by the serial log recording module can be obtained directly from the RAID card, so that the cause of the storage link failure can be located.

[0074] In step S104, the first log information, the second log information, and the command execution information of the target command are sent to the server host end, so that a technician can locate the cause of the storage link failure based on the first log information, the second log information, and the command execution information of the target command received by the server host end.

[0075] In some embodiments, the method for locating the cause of the storage link failure comprises the following steps:

[0076] In some embodiments, the method for locating the cause of the storage link failure comprises the following steps:

[0077] In some embodiments, the method for locating the cause of the storage link failure comprises the following steps:

[0078] In step S201, the switch state of the serial log recording module of the expansion backplane is obtained. For details, see step S101 of the embodiment shown in FIG. 1, which will not be described here.

[0079] Step S202, in a case where the switch state of the serial logging module of the expansion backboard is in an open state, the serial logging module is used to record the command execution information of the target command issued by the server.

[0080] Specifically, the above step S202 includes:

[0081] Step S2021, receiving the command issued by the server.

[0082] The storage subsystem receives various types of commands issued by the server.

[0083] Step S2022, in a case where the type of the command is a target command type, determining that the command is a target command, wherein the target command type is a read-write type command, a monitoring type command, a self-monitoring analysis type command, a power saving type command and a link reset type command.

[0084] After receiving the command issued by the server, it is determined whether the type of the command is a target command type. In response to the type of the command being a target command type, it is determined that the command is a target command, and the serial logging module is used to record the command execution information of the target command.

[0085] It should be noted that the read-write type command includes a read (read) command and a write (write) command. The monitoring type command can be a Read Log Ext command. The self-monitoring analysis type command is a self-monitoring, analysis and reporting technology (Self-Monitoring Analysis and Reporting Technology, abbreviated as: SMART) command set (SMART Command Set). The power saving type command is a Check Power Mode command. The link reset type command is a Reset command.

[0086] Step S2023, using the serial logging module to record the command execution information of the target command issued by the server in an overwrite manner.

[0087] Due to the limitations of chip size and cost, the capacity of the serial logging module can be set to a minimum of 64MB, which is expected to record about an hour of logs, and the serial logging module is used to record the command execution information of the target command issued by the server in an overwrite manner.

[0088] It should be noted that the serial logging module is also designed to prevent loss of power logs.

[0089] Step S203, in a case where the storage link fails, obtaining first log information corresponding to the disk array card, second log information corresponding to the hard disk, and command execution information of the target command recorded by the serial logging module.

[0090] Specifically, the step S203 includes:

[0091] In the case that the storage link fails, it is judged whether the serial log recording module is set with the keyword trigger condition.

[0092] The keyword trigger condition is set by a user, and in the case that the storage link fails, it is judged whether the serial log recording module is set with the keyword trigger condition, so as to determine whether all logs or part of logs recorded by the current serial log recording module are acquired.

[0093] In the case that the serial log recording module is set with the keyword trigger condition, the command execution information recorded by the serial log recording module is filtered by using the keyword contained in the keyword trigger condition, so as to acquire the command execution information of the target command.

[0094] The serial log recording module is set with the keyword trigger condition, which indicates that in the case that the storage link fails, the command execution information recorded in the serial log recording module needs to be filtered, so as to acquire the command execution information of the target command.

[0095] The keyword trigger condition contains a keyword, which can be suspend, discard, fail, etc. The command execution information currently recorded in the serial log recording module is filtered by using the keyword, so as to acquire the command execution information of the target command.

[0096] In the step S204, the first log information, the second log information and the command execution information of the target command are sent to the server host end, so that a technician locates the cause of the storage link failure based on the first log information, the second log information and the command execution information of the target command received by the server host end. For details, please refer to the step S104 of the embodiment shown in FIG. 1, which will not be described here.

[0097] The storage link failure cause positioning method provided by some embodiments of the present application records the command execution information of the target command issued by the server through the serial log recording module, which improves the efficiency of determining the cause of the storage link failure.

[0098] The log is recorded in the way of overwrite, which can effectively utilize the limited storage space to achieve the maximum performance, avoids the problem of exhausting the storage space caused by the unlimited accumulation of logs, and ensures the stable operation of log recording.

[0099] By introducing the keyword trigger condition in the case of storage link failure, the command execution information of the target command with high correlation degree to the failure cause positioning is screened from the command execution information recorded by the serial log recording module according to the keyword trigger condition, which not only reduces the analysis burden of the command execution information in the later stage, but also improves the efficiency and accuracy of the storage link failure cause positioning.

[0100] In some embodiments, the step S2032 comprises:

[0101] Step a1, for any command execution information, if the command execution information includes at least one keyword, it is determined that the command execution information is the command execution information of the target command.

[0102] Among them, the keyword trigger condition can include at least one keyword, and in the case that any command execution information includes at least one keyword in the keyword trigger condition, the command execution information is determined as the command execution information of the target command, and the screening of the command execution information is realized.

[0103] Some embodiments of the present application provide a storage link failure cause positioning method, which can quickly identify the command execution information related to the failure cause positioning from a large amount of log data through keyword matching, thereby improving the efficiency and accuracy of the storage link failure cause positioning.

[0104] In some embodiments, the storage link failure cause positioning method further comprises:

[0105] Step b1, in the case that the serial log recording module does not set the keyword trigger condition, the command execution information currently recorded by the serial log recording module is taken as the command execution information of the target command.

[0106] Among them, the serial log recording module does not set the keyword trigger condition, so it is not necessary to screen the command execution information recorded by the serial log recording module, and all the command execution information currently recorded by the serial log recording module is taken as the command execution information of the target command to participate in the subsequent positioning of the storage link failure cause.

[0107] Some embodiments of the present application provide a storage link failure cause positioning method, which, by taking the command execution information currently recorded by the serial log recording module as the command execution information of the target command in the case that the serial log recording module does not set the keyword trigger condition, ensures the stability and business continuity of the storage link failure cause positioning.

[0108] In some embodiments, the storage link failure cause positioning method further comprises:

[0109] Step c1, in the case of a failure of the storage link, suspending the recording of the command execution information of the target command issued by the server by using the serial log recording module.

[0110] In the case of a failure of the storage link, to avoid the key log being overwritten, the recording of the command execution information of the target command issued by the server by using the serial log recording module is suspended.

[0111] The storage link failure cause positioning method provided by some embodiments of the present application suspends the recording of the command execution information of the target command issued by the server by using the serial log recording module in the case of a failure of the storage link, prevents the key log from being overwritten, and improves the accuracy of the storage link failure cause positioning.

[0112] In some embodiments, the storage link failure cause positioning method further comprises:

[0113] Step d1, after the first log information, the second log information and the command execution information of the target command are sent to the server host end, the recording of the command execution information of the target command issued by the server by using the serial log recording module is continued.

[0114] It can be understood that, after the first log information, the second log information and the command execution information of the target command are sent to the server host end, the key log has been sent to the server host end, and therefore the recording of the command execution information of the target command issued by the server by using the serial log recording module can be continued.

[0115] That is, after the first log information, the second log information and the command execution information of the target command are sent to the server host end, the serial log recording module continues to record the command execution information of the target command issued by the server.

[0116] The storage link failure cause positioning method provided by some embodiments continues to record the command execution information of the target command issued by the server by using the serial log recording module after the first log information, the second log information and the command execution information of the target command are sent to the server host end, and ensures the continuity of the log recording.

[0117] In some embodiments, the storage link failure cause positioning method further comprises:

[0118] Step e1, in the case of a failure of the storage link, acquiring the basic log information corresponding to the extended backboard.

[0119] The basic log information corresponding to the expansion backboard is obtained by using an expansion backboard log in the related technology, and the recording function of the expansion backboard log is relatively weak, and only records basic abnormal events such as link topology basic information, firmware fatal events, phy error count, backboard Tx / Rx rate and the like. That is, the basic log information is the information recorded by the expansion backboard log.

[0120] The step S204 includes:

[0121] In the case that the storage link fails and the first log information, the second log information, the basic log information and the command execution information of the target command are obtained, the first log information, the second log information, the basic log information and the command execution information of the target command are sent to the server host end, so that the technical personnel can locate the cause of the storage link failure.

[0122] In the case that the storage link fails and the first log information, the second log information, the basic log information and the command execution information of the target command are obtained, the first log information, the second log information, the basic log information and the command execution information of the target command are sent to the server host end, so that the technical personnel can locate the cause of the storage link failure.

[0123] The storage link failure cause positioning method provided by some embodiments of the present application determines the cause of the storage link failure based on the first log information, the second log information, the basic log information and the command execution information of the target command, so that the determined cause of the storage link failure is more accurate, and the accuracy of the storage link failure cause positioning is improved.

[0124] In some embodiments, the storage link failure cause positioning method further includes:

[0125] In the case that the switch state of the serial log recording module of the expansion backboard is in the off state, the basic log information corresponding to the expansion backboard is recorded by using the expansion backboard log recording module.

[0126] It can be understood that, in the case that the switch state of the serial log recording module of the expansion backboard is in the off state, the serial log recording module cannot be used to record the command execution information of the target command sent by the server, and only the expansion backboard log recording module can be used to record the basic log information corresponding to the expansion backboard. The expansion backboard log recording module is an expansion backboard log in the related technology.

[0127] The storage link failure cause positioning method provided by some embodiments of the present application records only the basic log information corresponding to the expansion backboard by using the expansion backboard log recording module in the case that the switch state of the serial log recording module of the expansion backboard is in the off state, which improves the stability of the system and realizes efficient use of resources.

[0128] In some embodiments, the command execution information includes: command response time, command type, command mode, execution status information and error status information.

[0129] Figure 3 is a schematic diagram of the data structure of the command execution information recorded in the serial log module according to an embodiment of the present application. The serial log module is set to be accessible and recordable, and is designed to be able to be captured and parsed by the expansion backplane or RAID card tool. As shown in Figure 3, the command execution information includes: log generation time, timestamp, logical address, command tag, command type, data block size, command response time, command mode, execution status information and error status information.

[0130] The log generation time is the specific time when the serial log is generated, and the format is: Y / M / D_H:M:S. The serial log is the command execution information of the target command.

[0131] The timestamp is used to record the time of each command from receiving, executing and completing. That is, the time of receiving the target command, executing the target command and completing the target command is recorded.

[0132] The logical address is the logical block address (LBA), which is used to record the logical address space of each command operation.

[0133] The command tag is a unique identification tag of the command, which is used to record the execution process of each command.

[0134] The command type is used to record the type of the command, which covers six categories of protocol class commands, including Read, Write, Read log Ext, SMART Command Set, Check Power Mode and Reset.

[0135] The data block size is used to record the data block size of the read and write commands, such as 4K, 64K, 128K, etc.

[0136] The command response time is used to count the time used from receiving to completing the execution of each command, with the unit of milliseconds ms.

[0137] The command mode is mainly used to record the order and randomness of command execution, which is mainly divided into sequential read, sequential write, random read, random write and other commands.

[0138] The execution status information is used to record the final execution status of each command, which is divided into three types: completed, suspended and discarded, which respectively represent the completion of command execution, the suspension of command without execution completion and the discarding of command.

[0139] Error status information is used to record whether there is an error in the command execution process, mainly Pass and Fail two states.

[0140] Figure 4 is a schematic diagram of command execution information recorded by the serial log recording module according to some embodiments of the application. As shown in Figure 4, the time stamp, command type, command response time, logical address, data block size, execution status information and error status information in the command execution information are exemplarily shown.

[0141] According to the command execution information as shown in Figure 4, the execution of each target instruction can be clearly known, so that the technician can analyze the typical problems of the storage link such as disc problem, performance problem and command timeout according to the command execution information.

[0142] For example, by analyzing the log, i.e. the command execution information, it is determined which commands are timed out, which commands are suspended, and which commands are discarded to determine the command timeout problem.

[0143] For example, by analyzing the log, i.e. the command execution information, it is determined which command or which commands take a long time, which LBA address is concentrated, whether there is a frequent Read log temperature monitoring influence or SMART log monitoring influence, Check power energy saving mode influence, or whether there is an abnormal reset event influence to determine the performance problem.

[0144] The storage link fault cause positioning method provided by some embodiments of the application records the command execution information of the target command issued by the server by using the serial log recording module, so that the storage link fault cause determined based on the first log information, the second log information and the command execution information of the target command is more accurate, and the accuracy of the storage link fault cause positioning is improved.

[0145] In some embodiments, the storage link fault cause positioning method further comprises:

[0146] In response to the keyword update operation, the keywords in the keyword trigger condition are updated.

[0147] Among them, the technician can update the keywords after positioning the cause of the storage link fault, so as to obtain more command execution information which is more conducive to positioning the cause of the storage link fault.

[0148] In the case of setting the keyword trigger condition in the serial log recording module, the command execution information recorded by the serial log recording module is screened by using the keywords contained in the keyword trigger condition to obtain the command execution information of the target command, comprising:

[0149] In the case that the serial logging module is set with the keyword trigger condition, the command execution information recorded by the serial logging module is filtered by the keyword contained in the updated keyword trigger condition to obtain the command execution information of the target command.

[0150] The storage link fault cause positioning method provided by some embodiments of the present application can update the keyword in the keyword trigger condition in real time according to the storage link fault cause, so that the obtained command execution information is more conducive to storage link fault cause positioning, and the accuracy of storage link fault cause positioning is improved.

[0151] In some embodiments, an expansion backboard is also provided, and FIG. 5 is a structural schematic diagram of the expansion backboard according to an embodiment of the present application. As shown in FIG. 5, the expansion backboard includes an expansion backboard chip (Expander chip) and a serial logging module.

[0152] The serial logging module includes a switch control module and a serial logging subunit. The switch control module is configured to control the switch state of the serial logging subunit in response to the operation of a user.

[0153] The serial logging subunit is configured to record the command execution information of the target command issued by the server host in the on state. The serial logging subunit is a serial logging chip.

[0154] That is, the switch control module is mainly responsible for the switching of the serial logging function. The switch control module can also pause the recording of the log by setting the keyword according to the log recording content, so as to prevent the key log from being covered. That is, the switch control module is also configured to set the keyword trigger condition.

[0155] Generally, the storage link fails when the keyword appears in the command execution information.

[0156] FIG. 6 is a flowchart of obtaining the command execution information of the target command according to an embodiment of the present application. As shown in FIG. 6, the process of obtaining the command execution information of the target command includes the following steps:

[0157] Step 1: Determine whether the switch control module turns on the log recording function, that is, determine whether the switch state of the serial logging module of the expansion backboard is in the on state. It should be noted that determining whether the switch state of the serial logging module of the expansion backboard is in the on state is equivalent to determining whether the switch state of the serial logging subunit is in the on state.

[0158] Second step: in the case that the switch control module opens the log recording function, serial log recording is performed. In the case that the switch control module opens the serial log recording module on the extension backboard, the command execution information of the target command issued by the server is recorded by using the serial log recording module.

[0159] In the case that the switch control module does not open the log recording function, the process ends.

[0160] Third step: whether to read the log is judged, that is, whether the storage link fails is judged, and the command execution information in the serial log recording module needs to be read.

[0161] Fourth step: in the case that the log is not read, the log is cyclically recorded in real time, that is, in the case that the storage link does not fail, the command execution information of the target command issued by the server is recorded by using the serial log recording module in the way of overwriting.

[0162] Fifth step: in the case that the log is read, whether the switch control module sets the trigger condition is judged, that is, in the case that the storage link fails, whether the serial log recording module sets the keyword trigger condition is judged.

[0163] Sixth step: in the case that the switch control module sets the trigger condition, the simplified log is generated.

[0164] In the case that the switch control module does not set the trigger condition, the serial log file is generated.

[0165] In the case that the keyword trigger condition is set in the serial log recording module, the command execution information recorded by the serial log recording module is filtered by using the keyword contained in the keyword trigger condition to obtain the command execution information of the target command. In the case that the keyword trigger condition is not set in the serial log recording module, the command execution information currently recorded by the serial log recording module is taken as the command execution information of the target command. For details, refer to the corresponding description, which is not described here.

[0166] Some embodiments of the application provide an extension backboard, which can obtain the command type, data block size, logical address, command tag, command response time, command mode, execution state, error state and other information of the command passing through the extension backboard in real time when the server runs by using the built-in serial log recording module, realizes real-time monitoring of read-write type commands, monitoring type commands, SMART type commands, energy-saving type commands and link reset type commands and command execution information statistics, so that in the case that the storage link fails, the typical problems of the storage link such as disc drop, command timeout, performance and error can be quickly located by analyzing the command execution information, and the time efficiency of solving the storage link failure is improved.

[0167] The embodiment of the present application further provides a storage subsystem. FIG. 7 is a structural schematic diagram of the storage subsystem according to the embodiment of the present application. As shown in FIG. 7, the storage subsystem is connected with a server mainboard through a Peripheral Component Interconnect Express (PCIE) protocol interface.

[0168] The storage subsystem includes a RAID card, an expander backplane and hard disks. Each component follows a SATA / SAS protocol, and communication between links is realized through the protocol. The storage subsystem is used to locate a storage link fault cause based on the storage link fault cause locating method shown in the above embodiment.

[0169] The expander backplane in the server is one of the most important components of the storage link, is connected in series between the RAID card / SAS card and the hard disk (including an SSD), and plays a role of a link and a bridge. The RAID card and the hard disk component have a powerful and complete log recording function, can record abnormal events such as various errors and timeouts in detail, and can quickly locate a storage link fault cause in combination with a serial log recording module in the expander backplane.

[0170] The storage subsystem provided by the present application can acquire, in real time during server running, information such as a command type, a data block size, a logical address, a command tag, a command response time, a command mode, an execution state and an error state of a command passing through the expander backplane, simultaneously realizes a switching function and a specified keyword triggering condition of a serial log recording module, quickly locates typical problems such as a disk drop, a command timeout, performance and errors related to storage link compatibility by analyzing command execution information, improves accuracy of storage link fault cause locating, and improves a storage link fault solving capability and a solving timeliness.

[0171] The present application further provides a computer device including a memory and one or more processors, the memory storing computer readable instructions, and the computer readable instructions, when executed by the one or more processors, causing the one or more processors to perform the storage link fault cause locating method provided by one or more technical solutions.

[0172] In some embodiments, the computer device can be a server, and an internal structure diagram of the computer device can be as shown in FIG. 8. The computer device includes a processor, a memory and a network interface connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer readable instructions. The internal memory provides an environment for running the operating system and the computer readable instructions in the non-volatile storage medium. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer readable instructions are executed by the processor to implement a storage link fault cause positioning method.

[0173] The computer readable storage medium can be the computer readable storage medium in the computer device shown in FIG. 8.

[0174] The computer readable storage medium can be the computer readable storage medium in the computer device shown in FIG. 8.

[0175] Part of the present application can be applied as a computer readable instruction product, for example, computer readable instruction instructions, when executed by a computer, through the operation of the computer, the method and / or technical solutions according to the present application can be called or provided. Those skilled in the art should understand that the form of computer readable instruction instructions in computer readable medium includes but is not limited to source file, executable file, installation package file and the like, and accordingly, the way of executing computer readable instruction instructions by computer includes but is not limited to: the computer directly executes the instructions, or the computer executes the corresponding compiled program after compiling the instructions, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Here, the computer readable medium can be any available computer readable storage medium or communication medium accessible to the computer.

[0176] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims.

Claims

1. A storage link fault cause localization method, characterized by, The method comprises: acquiring the switch state of the serial log recording module of the extension backplane; in the case where the switch state of the serial log recording module of the extension backplane is in the on state, recording the command execution information of the target command issued by the server by using the serial log recording module; in the case where the storage link fails, acquiring the first log information corresponding to the disk array card, the second log information corresponding to the hard disk, and the command execution information of the target command recorded by the serial log recording module; and sending the first log information, the second log information, and the command execution information of the target command to the host end of the server, so that the technical personnel can locate the cause of the storage link failure based on the first log information, the second log information, and the execution information of the target command received by the server host end.

2. The method of claim 1, wherein, The recording of the command execution information of the target command issued by the server by using the serial log recording module comprises: receiving the command issued by the server; in the case where the type of the command is the target command type, determining that the command is the target command, wherein the target command type is a read-write type command, a monitoring type command, a self-monitoring analysis type command, an energy-saving type command, and a link reset type command; and recording the command execution information of the target command issued by the server by using the serial log recording module in an overwriting manner.

3. The method of claim 1, wherein, In the case where the storage link fails, acquiring the command execution information of the target command recorded by the serial log recording module comprises: in the case where the storage link fails and the serial log recording module is set with a keyword trigger condition, screening the command execution information recorded by the serial log recording module by using the keyword contained in the keyword trigger condition, to obtain the command execution information of the target command.

4. The method of claim 3, wherein, The screening of the command execution information recorded by the serial log recording module by using the keyword contained in the keyword trigger condition, to obtain the command execution information of the target command, comprises: for any command execution information, if the command execution information includes at least one keyword, it is determined that the command execution information is the command execution information of the target command.

5. The method of claim 3, wherein, The method further comprises: in the case where the serial log recording module is not set with a keyword trigger condition, taking the command execution information currently recorded by the serial log recording module as the command execution information of the target command.

6. The method of claim 1, wherein, The method further comprises: in the case where the storage link fails, pausing the recording of the command execution information of the target command issued by the server by using the serial log recording module.

7. The method of claim 6, wherein, The method further comprises: after sending the first log information, the second log information, and the command execution information of the target command to the server host end, continuing to record the command execution information of the target command issued by the server by using the serial log recording module.

8. The method of claim 1, wherein, The method further comprises: in the case where the storage link fails, acquiring the basic log information corresponding to the extension backplane; The first log information, the second log information and the command execution information of the target command are sent to a server host end, so that a technician performs storage link fault cause positioning based on the first log information, the second log information and the command execution information of the target command received by the server host end. The first log information, the second log information, the basic log information and the command execution information of the target command are sent to a server host end, so that a technician performs storage link fault cause positioning based on the first log information, the second log information, the basic log information and the command execution information of the target command received by the server host end.

9. The method of claim 1, wherein, The method further comprises: In a case where the switch state of the serial log recording module of the extension backboard is in an off state, the basic log information corresponding to the extension backboard is recorded by the extension backboard log recording module.

10. The method of claim 1, wherein, The command execution information includes a command response time, a command type, a command mode, execution state information and error state information.

11. The method of claim 10, wherein, The command mode is used to record the order and randomness of command execution, the command mode includes sequential reading, sequential writing, random reading and random writing, and the execution state information is used to record the final execution state of each command, and the execution state information includes completion, suspension and discard.

12. The method of claim 1, wherein, The serial log recording module is designed to not lose power-off logs.

13. The method of claim 3, wherein, The keywords include suspension, discard and error.

14. An extended backplane, characterized by The extension backboard comprises a serial log recording module, wherein the serial log recording module comprises a switch control module and a serial log recording subunit; The switch control module is used to control the switch state of the serial log recording subunit in response to the operation of a user; and The serial log recording subunit is used to record the command execution information of the target command issued by the server host in an on state.

15. The extended backplane of claim 14, wherein, The extension backboard further comprises an extension backboard chip, and the extension backboard chip is connected with the serial log recording module.

16. The extended backplane of claim 14, wherein, The switch control module is further used to set a keyword trigger condition to pause the recording of logs by keywords in the keyword trigger condition.

17. A memory sub-system, comprising: The storage subsystem comprises a disk array card, an extension backboard and a hard disk, and the storage subsystem is used to perform storage link fault cause positioning based on the storage link fault cause positioning method in any one of claims 1-13.

18. A computer device, comprising: Comprise: A memory and one or more processors, the memory stores computer readable instructions, when the computer readable instructions are executed by the one or more processors, the one or more processors execute the steps of the method in any one of claims 1 to 13.

19. A non-transitory computer-readable storage medium having stored thereon computer-readable instructions, wherein, The computer readable instructions are executed by the processor to implement the storage link fault cause positioning method in any one of claims 1 to 13.

20. A computer readable instruction product, characterized in that, Comprise computer readable instructions for causing a computer to execute the storage link fault cause positioning method in any one of claims 1 to 13.

Images