Multi-controller hot-swap recording method, device and storage medium

Abstract

The present invention relates to a multi-controller hot-plug recording method, device and storage medium, which can realize recording the hot plug of the controller when the controller is hot-plugged. In the present invention, multiple controllers use a backplane to share their own hot-plug status with other controllers. When the target controller is hot-plugged, the other active controllers set the hot-plug flag of the target controller and trigger the data management platform event repair strategy; after monitoring the target controller to be re-plugged, the data management platform event repair strategy controls the target controller to obtain the hot-plug flag saved in the other active controllers, and uses the value of the obtained hot-plug flag to analyze whether the target controller has been hot-plugged; save the hot-plug action analysis result, and report the analysis result to the upper-level management program through the baseboard management controller, triggering the management platform repair strategy to make the controller cancel the hot-plug flag setting of the target controller and end.

Description

Technical Field

[0001] The present invention relates to the technical field of hot-plug recording and repairing, and in particular to a multi-controller hot-plug recording method, device and storage medium. Background Art

[0002] The baseboard management controller (BMC) in a server serves as the core of board-level hardware management and often undertakes important functions such as server hardware status monitoring and fault logging. During operation, the BMC typically provides the hot-swap status of each field-replaceable unit (FRU) of the server to the server management software for data management platform event repair in operation and maintenance or customer service technical support.

[0003] Multi-controller redundancy is a common design in storage servers. Each controller is equipped with a separate baseboard management controller (BMC), which records the hot-swap status of its field-replaceable units (FRUs). The connections between the controllers are mutually redundant, providing data disaster recovery capabilities. However, due to the specific characteristics of the controllers themselves, hot-swapping a controller as a whole causes the associated data to disappear, and the BMC in the controller cannot monitor and record the hot-swap status of its own controller. Summary of the Invention

[0004] In order to solve the above technical problem or at least partially solve the above technical problem, the present invention provides a multi-controller hot plug recording method, device and storage medium.

[0005] In a first aspect, the present invention provides a multi-controller hot-swap recording method, comprising: multiple controllers interconnected via a backplane share their own hot-swap status with other controllers using the backplane; when a target controller is hot-swapped, the other active controllers set a hot-swap flag of the target controller; the active controller notifies an upper-layer management program, triggering a data management platform event repair strategy of the upper-layer management program;

[0006] After monitoring that the target controller is re-plugged, the data management platform event repair strategy controls the target controller to access the hot-plug flags saved in the remaining active controllers, and uses the obtained hot-plug flag value to analyze whether the target controller has been hot-plugged; save the hot-plug action analysis results, and report the analysis results to the upper-level management program through the baseboard management controller, triggering the management platform repair strategy to cause the controller to cancel the hot-plug flag setting of the target controller and end.

[0007] Furthermore, multiple controllers interconnected through a backplane share their own presence status with the remaining controllers using the backplane; when the active controller detects that the presence status of the target controller changes from being in place to being not in place, it determines that the target controller has been hot-swapped; when the presence status of the target controller changes from being not in place to being in place, the data management platform event repair strategy controls the target controller to access the hot-swap identifiers saved in the remaining active controllers.

[0008] Furthermore, the in-place status of each controller is encoded so that the controller has a unique in-place status code; when a target controller is hot-plugged, the identity of the target controller is determined by the change of the in-place status code.

[0009] Furthermore, when the active controller notifies the upper-level management program of the hot-swap event of the target controller, the upper-level management program records the corresponding active controller that issued the notification in the controller list, and the data management platform event repair strategy controls the target controller to traverse the hot-swap identifiers saved in the active controllers in the access controller list.

[0010] Furthermore, using the obtained hot-swap identifier value to analyze whether a hot-swap action has occurred in the target controller includes: analyzing whether the obtained hot-swap identifier values are consistent; if the hot-swap identifier values are inconsistent, counting the number of hot-swap identifier values, and taking the result represented by the majority of hot-swap identifier values as the analysis result; if the hot-swap identifier values are consistent, taking the result represented by the hot-swap identifier value as the analysis result.

[0011] Furthermore, the restart controller obtains the hot-swap identifiers of each controller from the remaining active controllers, and analyzes whether the obtained hot-swap identifier values are consistent according to the controller. If the hot-swap identifier values are inconsistent, the number of hot-swap identifier values is counted, and the result represented by the majority of hot-swap identifier values is used as the hot-swap identifier value of the corresponding controller for synchronization; if the hot-swap identifier values are consistent, the hot-swap identifier value is used as the hot-swap identifier value of the corresponding controller for synchronization.

[0012] Furthermore, the data management platform event repair strategy detects that the target controller is re-plugged and starts timing. If the timing exceeds the set time threshold and no analysis results are fed back from the baseboard management controller to trigger the end of the data management platform event repair strategy, a timeout prompt will be given, a timeout log will be recorded, and the hot plug flag will be canceled.

[0013] In the second aspect, the present invention provides a multi-controller hot-swap recording device, comprising: several controllers connected via a backplane, multiple controllers connected to each other via the backplane share their own hot-swap status with the remaining controllers using the backplane, when the target controller is hot-swapped, the remaining active controllers set the hot-swap flag of the target controller; the active controller notifies the upper-level management program, triggering the data management platform event repair strategy of the upper-level management program; after monitoring that the target controller is re-plugged, the data management platform event repair strategy controls the target controller to access the hot-swap flags saved in the remaining active controllers, and uses the obtained hot-swap flag value to analyze whether the target controller has undergone a hot-swap action; the hot-swap action analysis result is saved in the data area of the processing unit of the target controller, and the analysis result is reported to the upper-level management program through the baseboard management controller, triggering the management platform repair strategy to cause the controller to cancel the hot-swap flag setting of the target controller and then end.

[0014] Furthermore, the controller includes: a CPU, a PCH south bridge, a baseboard management controller and a processing unit. The baseboard management controller and the processing unit on the same controller are connected, and the processing units on different controllers are interconnected through a bus on the backplane. The processing unit is used to store a status register of a hot plug identification. The number of bits of the status register is not less than the number of controllers. Different storage bits of the status register correspond to different controllers for recording the hot plug identification of different controllers. The processing unit realizes the in-place status perception of the controller, communicates with the other processing units to obtain the value of the hot plug identification, and uses the obtained hot plug identification value to analyze whether the target controller has undergone a hot plug action; the hot plug action analysis result is saved in the data area of the processing unit of the target controller, and the analysis result is reported to the upper-level management program through the baseboard management controller.

[0015] In a third aspect, the present invention provides a storage medium for implementing a multi-controller hot-swap recording method, wherein the storage medium stores a computer program, and the computer program implements the multi-controller hot-swap recording method when executed by a processor.

[0016] The above technical solution provided by the embodiment of the present invention has the following advantages compared with the prior art:

[0017] In the present invention, multiple controllers share their hot-swap status with other controllers via a backplane. When a target controller is hot-swapped, the remaining active controllers set the target controller's hot-swap flag and trigger a data management platform event repair strategy. After detecting that the target controller has been re-plugged, the data management platform event repair strategy controls the target controller to obtain the hot-swap flags stored in the remaining active controllers and use the obtained hot-swap flags to analyze whether the target controller has been hot-swapped. The hot-swap action analysis results are saved and reported to the upper-level management program via the baseboard management controller, triggering the management platform repair strategy to cause the controller to unset the hot-swap flag of the target controller before terminating. This achieves the function of recording hot-swap actions of the controllers themselves in a multi-controller structure. The design adopts a distributed hot-swap action recording system, monitoring the hot-swap of hot-swappable controllers through multiple controllers to avoid omissions. Furthermore, the hot-swap flags of the majority of controllers are used as the basis for analysis, avoiding inaccurate hot-swap analysis caused by omissions of abnormal controllers between the hot-swap occurrence and recording. Timeout control prevents the data management platform event repair strategy from being stuck due to events where hot-swap recording is not possible. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 A flowchart of a multi-controller hot-swap recording method provided by an embodiment of the present invention;

[0021] Figure 2 A flowchart of an embodiment of the present invention providing a method for controlling a target controller to access hot-swap identifiers stored in other active controllers after detecting that the target controller has been re-plugged, and for the target controller to analyze whether a hot-swap action has occurred using the obtained hot-swap identifier value;

[0022] Figure 3 A flowchart of synchronizing a hot-swap flag in a processing unit of a restart controller provided by an embodiment of the present invention;

[0023] Figure 4 A schematic diagram of a multi-controller hot-swap recording device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0025] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0026] Example 1

[0027] See Figure 1 As shown, the present invention provides a multi-controller hot plug recording method, comprising:

[0028] S100, multiple controllers interconnected through a backplane share their own hot-swap status with other controllers using the backplane. In the specific implementation process, the controller includes a CPU, a PCH south bridge, a baseboard management controller and an additionally configured processing unit. The processing units in the same controller are connected to the baseboard management controller. The processing units of each controller are interconnected through the I2C link on the backplane to share data access. The processing units use CPLD. Each controller shares its own in-place status with the other controllers, that is, any controller can sense the change in the in-place status of other controllers. The detection of the in-place status is completed by the processing unit in the controller. Specifically, the in-place status of each controller is encoded so that each controller has a unique in-place status code. When any controller on the backplane is the target controller and hot-plugged, the processing unit determines that the target controller is hot-plugged by the change in the in-place status code, and determines the identity of the target controller based on the in-place status code. When the processing unit of the active controller detects that the in-place status of the target controller changes from in-place to not-in-place, it determines that the target controller is hot-plugged.

[0029] S200 , when a target controller is hot-plugged, an active controller that detects that the target controller is hot-plugged sets a hot-plug flag of the target controller.

[0030] During a specific implementation, the processing unit is configured to configure a status register for storing a hot-swap flag. The number of bits in the status register is no less than the number of controllers. Different storage bits in the status register correspond to different controllers and are used to record the values of the hot-swap flags of different controllers. When a target controller is hot-swapped, the processing unit in the active controller that detects the hot-swap of the target controller sets the hot-swap flag of the target controller.

[0031] S300: The active controller that detects a hot-swap of the target controller notifies the upper-level hypervisor, triggering the upper-level hypervisor's data management platform event repair strategy. As a feasible implementation, when a processing unit in the active controller detects a change in the in-place status of the target controller, the processing unit notifies the baseboard management controller (BMC), which then issues a notification to the upper-level hypervisor, triggering the upper-level hypervisor's data management platform event repair strategy. This notification is sent to the upper-level hypervisor using the BMC's communication resources.

[0032] S400, after detecting that the target controller is re-plugged, the data management platform event repair strategy controls the target controller to access the hot plug identifiers saved in the other active controllers. The target controller uses the obtained hot plug identifier value to analyze whether the target controller has been hot plugged. For specific implementation, please refer to Figure 2 As shown, step S400 includes:

[0033] S401: The data management platform event repair strategy monitors the in-place status of the target controller through other controllers.

[0034] S402, detect whether the in-position state changes from not in position to in-position, if yes, execute S403,

[0035] S403: The data management platform event repair strategy controls the processing unit of the target controller through the baseboard management controller to access and obtain the value of the hot plug identifier stored in the processing units of other active controllers.

[0036] During the specific implementation process, when the active controller notifies the upper-level management program of the hot-swap event of the target controller, the upper-level management program records the corresponding active controller that issued the notification in the controller list; when the target controller is reinserted, the data management platform event repair strategy controls the processing unit of the target controller to traverse the hot-swap identifier stored in the active controller in the access controller list through the I2C link.

[0037] S404 , analyzing whether the obtained values of the hot-swap flags of the controllers are consistent. If the values of the hot-swap flags are inconsistent, executing S405 ; if the values of the hot-swap flags are consistent, executing S406 .

[0038] S405: Count the number of hot-plug identification values, and use the result represented by the majority of hot-plug identification values as the analysis result.

[0039] S406: Use the result represented by the hot-swap identification value as the analysis result.

[0040] S500: Save the hot-swap action analysis results in the data area of the processing unit of the target controller, and report the analysis results to the upper-level management program via the baseboard management controller, triggering the management platform repair strategy to cause the controller to cancel the hot-swap flag of the target controller, and then end. During the specific implementation process, the processing unit of the target controller saves the hot-swap action analysis results in the data area and sends them to the baseboard management controller. The baseboard management controller reports the analysis results to the upper-level management program. Upon receiving the analysis results, the upper-level management program triggers the management platform repair strategy to cause the currently active controller to cancel the hot-swap flag of the target controller via the baseboard management controller, and then the management platform repair strategy ends.

[0041] As a feasible implementation method, after the controller is restarted, the hot plug flag in the processing unit of the restarted controller is synchronized, see Figure 3 Shown, including:

[0042] S1: The processing unit of the restarted controller obtains the hot plug identifier of each controller from the processing units of the remaining active controllers.

[0043] S2: Analyze the obtained hot-swap flag values according to the controller to see whether they are consistent. If the hot-swap flag values are inconsistent, execute S3; if the hot-swap flag values are consistent, execute S4.

[0044] S3, counting and comparing the number of hot-plug identification values, and using the result represented by the majority of hot-plug identification values as the hot-plug identification value of the corresponding controller to synchronize the processing unit of the restarted controller.

[0045] S4: Synchronize the processing unit of the restarted controller by using the hot-swap identification value as the hot-swap identification value of the corresponding controller.

[0046] As a feasible implementation method, the data management platform event repair strategy detects that the target controller is re-plugged and starts timing. If the timing time exceeds the set time threshold and no analysis result is fed back from the baseboard management controller, triggering the end of the data management platform event repair strategy, a timeout prompt is given, a timeout log is recorded, and the hot plug flag is canceled.

[0047] Example 2

[0048] See Figure 4As shown, an embodiment of the present invention provides a multi-controller hot-swap recording device, including: several controllers connected via a backplane, multiple controllers connected to each other via the backplane share their own hot-swap status with the remaining controllers using the backplane, when the target controller is hot-swapped, the remaining active controllers set the hot-swap flag of the target controller; the active controller notifies the upper-level management program, triggering the data management platform event repair strategy of the upper-level management program; after monitoring that the target controller is re-plugged, the data management platform event repair strategy controls the target controller to access the hot-swap flags saved in the remaining active controllers, and uses the obtained hot-swap flag value to analyze whether the target controller has undergone a hot-swap action; the hot-swap action analysis result is saved in the data area of the processing unit of the target controller, and the analysis result is reported to the upper-level management program through the baseboard management controller, triggering the management platform repair strategy to cause the controller to cancel the hot-swap flag setting of the target controller and then end.

[0049] During the specific implementation process, the controller includes: a CPU, a PCH south bridge, a baseboard management controller and a processing unit. The baseboard management controller and the processing unit on the same controller are connected, and the processing units on different controllers are interconnected through the I2C bus on the backplane. The processing unit is used to store a status register of the hot plug identification. The number of bits of the status register is not less than the number of controllers. Different storage bits of the status register correspond to different controllers for recording the hot plug identification of different controllers. The processing unit realizes the in-place status perception of the controller, communicates with the other processing units to obtain the value of the hot plug identification, and uses the obtained hot plug identification value to analyze whether the target controller has undergone a hot plug action; the hot plug action analysis result is saved in the data area of the processing unit of the target controller, and the analysis result is reported to the upper management program through the baseboard management controller.

[0050] Example 3

[0051] An embodiment of the present invention provides a storage medium for implementing a multi-controller hot plug recording method. The storage medium stores a computer program. When the computer program is executed by a processor, the multi-controller hot plug recording method is implemented.

[0052] In the several embodiments provided by the present invention, it should be understood that the disclosed modules and units can be implemented in other ways. For example, the structural embodiments described above are merely illustrative. For example, the division of the units is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the system or unit, which can be electrical, mechanical or other forms.

[0053] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0054] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0055] Although the present invention has been described in detail with reference to the accompanying drawings and in combination with preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and such modifications or substitutions shall be within the scope of the present invention. Any person skilled in the art who is familiar with the present invention may easily conceive of changes or substitutions within the technical scope disclosed in the present invention, and such changes or substitutions shall be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A multi-controller hot-swap recording method, characterized in that: include: Multiple controllers connected to each other via a backplane share their hot-swap status with other controllers. Each controller is connected to each other via a backplane link for data access sharing. Each controller shares its presence status with other controllers, allowing any controller to sense changes in the presence status of other controllers. When the target controller is hot-swapped, the remaining active controllers set the hot-swappable flag of the target controller; the active controller notifies the upper-layer management program, triggering the upper-layer management program's data management platform event repair strategy; After monitoring that the target controller is re-plugged, the data management platform event repair strategy controls the target controller to access the hot-plug flags saved in the remaining active controllers, and uses the obtained hot-plug flag value to analyze whether the target controller has been hot-plugged; save the hot-plug action analysis results, and report the analysis results to the upper-level management program through the baseboard management controller, triggering the management platform repair strategy to cause the controller to cancel the hot-plug flag setting of the target controller and end.

2. The multi-controller hot-swap recording method according to claim 1, wherein: Multiple controllers connected to each other through a backplane share their own presence status with other controllers using the backplane; when the active controller detects that the presence status of the target controller changes from being in place to not being in place, it determines that the target controller has been hot-swapped; when the presence status of the target controller changes from not being in place to being in place, the data management platform event repair strategy controls the target controller to access the hot-swap identifiers saved in the other active controllers.

3. The multi-controller hot-swap recording method according to claim 2, wherein: The in-place status of each controller is encoded so that the controller has a unique in-place status code; when a target controller is hot-plugged, the identity of the target controller is determined by the change in the in-place status code.

4. The multi-controller hot-swap recording method according to claim 1, wherein: When the active controller notifies the upper-level management program of the hot-swap event of the target controller, the upper-level management program records the corresponding active controller that issued the notification in the controller list, and the data management platform event repair strategy controls the target controller to traverse the hot-swap identifiers saved in the active controllers in the access controller list.

5. The multi-controller hot-swap recording method according to claim 1, characterized in that: Analyzing whether a hot-plug action has occurred in a target controller using the obtained hot-plug identifier value includes: analyzing whether the obtained hot-plug identifier values of the controller are consistent; if the hot-plug identifier values are inconsistent, counting the number of hot-plug identifier values, and using the result represented by the majority of hot-plug identifier values as the analysis result; if the hot-plug identifier values are consistent, using the result represented by the hot-plug identifier value as the analysis result.

6. The multi-controller hot-swap recording method according to claim 1, characterized in that: After the controller is restarted, the restarting controller obtains the hot-swap flags of each controller from the remaining active controllers, and analyzes the obtained hot-swap flag values to see if they are consistent. If the hot-swap flag values are inconsistent, the number of hot-swap flag values is counted, and the result represented by the majority of hot-swap flag values is used as the hot-swap flag value of the corresponding controller for synchronization; If the hot-swap flag values are consistent, the hot-swap flag value is used as the hot-swap flag value of the corresponding controller for synchronization.

7. The multi-controller hot-swap recording method according to claim 1, characterized in that: The data management platform event repair strategy detects that the target controller is re-plugged and starts timing. If the timing exceeds the set time threshold and no analysis results are received from the baseboard management controller to trigger the end of the data management platform event repair strategy, a timeout prompt will be given, a timeout log will be recorded, and the hot plug flag will be canceled.

8. A multi-controller hot-swap recording device, implementing the multi-controller hot-swap recording method according to any one of claims 1 to 7, characterized in that: include: Several controllers connected via a backplane, and multiple controllers interconnected via the backplane, use the backplane to share their own hot-swap status with other controllers. Each controller is interconnected via a backplane link for data access sharing, and each controller shares its own in-place status with other controllers. Any controller can sense changes in the in-place status of other controllers. When a target controller is hot-swapped, the other active controllers set the hot-swap flag of the target controller. The active controller notifies the upper-level management program, triggering the upper-level management program's data management platform event repair strategy. After monitoring that the target controller is re-plugged, the data management platform event repair strategy controls the target controller to access the hot-plug flags saved in the remaining active controllers, and uses the obtained hot-plug flag value to analyze whether the target controller has undergone a hot-plug action; the hot-plug action analysis results are saved in the data area of the processing unit of the target controller, and the analysis results are reported to the upper-level management program through the baseboard management controller, triggering the management platform repair strategy to cause the controller to cancel the hot-plug flag setting of the target controller and end.

9. The multi-controller hot-swap recording device according to claim 8, characterized in that: The controller includes: a CPU, a PCH south bridge, a baseboard management controller and a processing unit. The baseboard management controller and the processing unit on the same controller are connected, and the processing units on different controllers are interconnected via a bus on the backplane. The processing unit is used to store a status register of a hot-swap identifier. The number of bits of the status register is not less than the number of controllers. Different storage bits of the status register correspond to different controllers and are used to record the hot-swap identifiers of different controllers. The processing unit realizes the in-place status perception of the controller, communicates with the other processing units to obtain the value of the hot-swap identifier, and uses the obtained value of the hot-swap identifier to analyze whether a hot-swap action has occurred in the target controller; the hot-swap action analysis result is saved in the data area of the processing unit of the target controller, and the analysis result is reported to the upper-level management program through the baseboard management controller.

10. A storage medium for implementing a multi-controller hot-swap recording method, wherein the storage medium stores a computer program, characterized in that: When the computer program is executed by a processor, the multi-controller hot-swap recording method according to any one of claims 1 to 7 is implemented.

Images