Abnormality processing method and device

By directly sending communication link fault notifications in the storage network and performing link switching, the problem of low efficiency in communication link fault handling in the prior art is solved, efficient exception handling and resource management are realized, and the processing flow is simplified.

CN118945037APending Publication Date: 2024-11-12HUAWEI TECH CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202310543932.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-12
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

When the communication link failure of the existing storage network, the host needs to send heartbeat detection packets multiple times, resulting in inefficient exception handling and additional installation of specific plug-ins, which increases complexity and resource consumption.

Method used

The host directly sends a communication link failure notification to the first control node by detecting a communication interface failure, uses an unfailed communication link to access the service data in the storage network, avoids multiple heartbeat detection and plug-in installation, and realizes direct communication link switching. and resource release.

Benefits of technology

It improves the efficiency and convenience of exception handling, reduces resource consumption and complexity, broadens application scenarios, and ensures the access efficiency and reliability of the storage network.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118945037A_ABST
    Figure CN118945037A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides an exception handling method and device, and relates to the technical field of computers, in the exception handling method, a plurality of communication links based on an NVMe-oF protocol are established between a storage network and a plurality of communication interfaces of a host, the plurality of communication links includes a first communication link in communication with the first communication interface and a second communication link in communication with the second communication interface, and the storage network includes one or more control nodes including the first control node, the host performs communication based on an NVMe-oF protocol with the first control node through the first communication link to access service data of the host in the storage network, and sends a fault notification to the first control node through the second communication link under the condition that a fault of the first communication interface is detected; and after receiving a response notification of the control node to the fault notification, communicating with the control node in the storage network through a second communication link. According to the embodiment of the invention, the exception handling efficiency and convenience in the storage network access process can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, and in particular, to an exception handling method and device. Background Art

[0002] A storage network is a parallel computer cluster containing flash memory that can be used to store data transmitted by hosts accessing the storage network. Hosts can access the storage network through multiple paths, i.e., communication links, to improve the reliability and access continuity of the storage network.

[0003] In the related art, the host is connected to the switch in communication, and the switch is connected to the control node in the storage network in communication, so that the host communicates with the control node through the communication network to access the storage network. On this basis, in the case of a failure in the communication link between the host and the communication network, the host determines the failure through the heartbeat detection mechanism between the host and the control node, and then the host performs path switching through a pre-installed specific plug-in. Among them, path switching means that the host accesses the storage network through a path other than the failed path among the above multiple paths.

[0004] However, in the above heartbeat detection mechanism, the host needs to send heartbeat detection packets to the control node multiple times, which is time-consuming and requires the host to install additional specific plug-ins, resulting in low exception handling efficiency. Summary of the invention

[0005] In order to solve the above technical problems, the present application provides an exception handling method and device. In the exception handling method, the host directly sends a notification of a communication link failure to the first control node by detecting a communication interface failure, and then the first control node feeds back a response notification to instruct the host to obtain the service data of the host in the storage medium managed by the first control node through a non-faulty communication link, avoiding multiple heartbeat detection packets between the first control node and the host and the host additionally installing a specific plug-in, thereby improving the efficiency of exception handling.

[0006] In a first aspect, an embodiment of the present application provides an exception handling method, wherein multiple communication links based on the NVMe-oF protocol are established between a storage network and multiple communication interfaces of a host, the multiple communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, the storage network includes one or more control nodes including a first control node, the method comprising: the host communicates with the first control node based on the NVMe-oF protocol through the first communication link to access the business data of the host in the storage network, and when a failure of the first communication interface is detected, sends a fault notification to the first control node through the second communication link; after receiving a response notification of the control node to the fault notification, the host communicates with the control node in the storage network through the second communication link.

[0007] In the embodiment of the present application, the host directly sends a notification of the communication link failure to the first control node by detecting the communication interface failure, and then the first control node, based on the failure notification sent by the host, can instruct the host to access the service data of the host in the storage medium managed by the first control node through a non-faulty communication link such as the second communication link when the first communication interface fails, that is, the first communication link cannot be used for communication, through a response notification, thereby avoiding the sending of heartbeat detection packets between the first control node and the host multiple times, thereby improving the efficiency of exception handling. In addition, the first control node feedback response notification can ensure that the host directly switches the communication link based on the protocol stack, so that the problem that the host pre-installs a specific plug-in to perform exception handling can be avoided, thereby further improving the efficiency and convenience of exception handling, and broadening the application scenarios of the embodiments of the present application.

[0008] According to the first aspect, the above method also includes: the first control node receives a fault notification; the first control node clears the first target resource based on the fault notification; wherein the first target resource includes the computing resources used by the first control node to obtain business data through the first communication link; when the first control node completes the clearing, the first control node sends a response notification for notifying the host to communicate through the second communication link.

[0009] In an embodiment of the present application, when the first communication link fails, the first control node clears the computing resources used by the first control node to obtain business data through the first communication link based on the fault notification, thereby releasing the computing resources of the first control node (such as one or more of CPU resources, memory resources, and network resources), which is conducive to improving performance. In addition, when the first control node completes the clearing, it sends a response notification for notifying the host to communicate through the second communication link, which can avoid the first target resource, such as the business data obtained through the first communication link, and the resources created for the first communication link to access the business data (such as the identification of the first control node in the storage network, etc.) in the process of the host accessing the business data through the second communication link. Abnormal business data access caused by the resources (such as the identification of the first control node in the storage network, etc.) can further improve the access efficiency to the storage network.

[0010] According to the first aspect, the storage network also includes a second control node that communicates with the host through a second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace, and the target namespace is used to store business data; sending a fault notification to the first control node through the second communication link includes: the host sends a fault notification to the second control node through the second communication link; the second control node transmits the fault notification to the first control node.

[0011] In an embodiment of the present application, there are multiple control nodes (such as a first control node and a second control node), which are respectively used to establish multiple communication links (such as a first communication link and a second communication link). The first control node and the second control node are used to manage the same storage area and share a target namespace (namespace), so that the second control node transmits the fault notification to the first control node to ensure timely and accurate delivery of the fault notification, and can also ensure that the host can access the business data in the target namespace through the second communication link between the second control node, thereby ensuring the efficiency of exception handling and expanding application scenarios.

[0012] According to the first aspect, or any implementation of the first aspect above, the storage network also includes a second control node that communicates with the host through a second communication link, and the second control node and the first control node are used to manage different storage areas and share a target namespace; sending a fault notification to the first control node through the second communication link includes: the host sends a fault notification to the second control node through the second communication link; the second control node sends a fault notification to the first control node through a third communication link between the control nodes, or the second control node transmits the fault notification to the first control node. In addition, in an embodiment of the present application, similar to the sending of a fault notification, a response notification of the control node to the fault notification can be fed back, the difference being that the feedback direction is opposite to the sending direction of the fault notification. In this way, timely and accurate feedback of the response notification can be ensured, further ensuring the efficiency of exception handling and expanding application scenarios.

[0013] In an embodiment of the present application, when there are multiple control nodes (such as a first control node and a second control node), they are respectively used to establish multiple communication links (such as a first communication link and a second communication link), and the first control node and the second control node are used to manage different storage areas and share a target namespace. Not only can the second control node directly transmit or send a fault notification to the first control node through a third communication link between the control nodes to ensure timely and accurate delivery of the fault notification, but it can also ensure that the host can access the service data in the target namespace through the second communication link between the second control node, thereby ensuring the efficiency of exception handling and expanding application scenarios. In addition, in an embodiment of the present application, similar to the sending of the fault notification, the response notification of the control node to the fault notification can be fed back, except that the feedback direction is opposite to the sending direction of the fault notification. In this way, it can be ensured that the response notification is fed back in a timely and accurate manner, further ensuring the efficiency of exception handling and expanding application scenarios.

[0014] According to the first aspect, or any implementation method of the first aspect above, the first control node clears the first target resource based on the fault notification, including: the first control node parses the fault notification to obtain resource clearing information; wherein the resource clearing information is used to indicate the first target resource; the first control node clears the first target resource indicated by the resource clearing information.

[0015] In an embodiment of the present application, the fault notification includes resource clearing information, which is used to indicate the computing resources that need to be cleared by the first control node that communicates with the faulty communication link, thereby ensuring that the first control node clears the first target resources more accurately and further ensuring the efficiency of exception handling.

[0016] According to the first aspect, or any implementation of the first aspect above, the method also includes: after the host receives a response notification from the control node to the fault notification, the second target resource is cleared; wherein the second target resource includes the computing resources used by the host to obtain business data through the first communication link.

[0017] Exemplarily, the second target resource may be, for example, a CPU resource, a network resource, a storage resource, etc. used in the host to establish the first communication link.

[0018] In the embodiment of the present application, the host clears the computing resources used by the host to obtain business data through the first communication link, which is beneficial to releasing host resources and improving host performance to further improve efficiency.

[0019] In a second aspect, an embodiment of the present application provides an exception handling method, wherein multiple communication links based on the NVMe-oF protocol are established between a storage network and multiple communication interfaces of a host, the multiple communication links include a first communication link for communicating with a first communication interface, and a second communication link for communicating with a second communication interface, the storage network includes one or more control nodes including a first control node, and the method includes: in a process in which the first control node communicates with the host based on the NVMe-oF protocol through a first communication link, if a fault notification sent by the host is received, a response notification to the fault notification is fed back to the host through the control node; wherein the fault notification is a notification sent by the host through the second communication link when a fault in the first communication interface is detected; the response notification is used to instruct the host to communicate with the control node in the storage network through the second communication link to access the business data of the host in the storage network.

[0020] According to the second aspect, the method also includes: executing the following steps through the first control node: receiving a fault notification; clearing the first target resource based on the fault notification; wherein the first target resource includes the computing resources used by the first control node to obtain business data through the first communication link; and when the clearing is completed, sending a response notification for notifying the host to communicate through the second communication link.

[0021] According to the second aspect, or any implementation of the second aspect above, the storage network also includes a second control node that communicates with the host through a second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace, and the target namespace is used to store business data; a response notification to the fault notification is fed back to the host through the control node, including: generating a response notification to the fault notification through the first control node and transmitting it to the second control node; and feeding back the response notification to the first control node through the second control node.

[0022] According to the second aspect, or any implementation of the second aspect above, the storage network also includes a second control node that communicates with the host through a second communication link, and the second control node and the first control node are used to manage different storage areas and share a target namespace; a response notification to the fault notification is fed back to the host through the control node, including: generating a response notification to the fault notification through the first control node; the first control node transmits the response notification to the second control node, or the first control node sends a response notification to the second control node through a third communication link between the control nodes; and the response notification is fed back to the first control node through the second control node.

[0023] The second aspect and any implementation of the second aspect correspond to the first aspect and any implementation of the first aspect respectively. The technical effects corresponding to the second aspect and any implementation of the second aspect can refer to the technical effects corresponding to the above-mentioned first aspect and any implementation of the first aspect, which will not be repeated here.

[0024] In a third aspect, an embodiment of the present application provides an exception handling method, which is applied to a host, wherein multiple communication interfaces of the host and a storage network establish multiple communication links based on the NVMe-oF protocol, the multiple communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, and the storage network includes one or more control nodes including a first control node. The method includes: communicating with the first control node based on the NVMe-oF protocol through the first communication link to access the business data of the host in the storage network, and sending a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; after receiving a response notification of the control node to the fault notification, communicating with the control node in the storage network through the second communication link.

[0025] The third aspect and any implementation of the third aspect correspond to the first aspect and any implementation of the first aspect, respectively. The technical effects corresponding to the third aspect and any implementation of the third aspect can refer to the technical effects corresponding to the first aspect and any implementation of the first aspect, which will not be repeated here.

[0026] In a fourth aspect, an embodiment of the present application provides an exception handling device, wherein multiple communication links based on the NVMe-oF protocol are established between the storage network and multiple communication interfaces of the host, the multiple communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, the storage network includes one or more control nodes including a first control node, and the device includes: a fault notification module, which is used for the host to communicate with the first control node based on the NVMe-oF protocol through the first communication link to access the business data of the host in the storage network, and send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; a path switching module, which is used for the host to communicate with the control node in the storage network through the second communication link after receiving a response notification of the control node to the fault notification.

[0027] According to the fourth aspect, the device also includes: a first resource clearing module, which is used for the first control node to receive a fault notification and clear the first target resource based on the fault notification; wherein the first target resource includes the computing resources used by the first control node to obtain business data through the first communication link; an information feedback module, which is used for the first control node to send a response notification for notifying the host to communicate through the second communication link when the clearing is completed.

[0028] According to the fourth aspect, or any implementation of the fourth aspect above, the storage network also includes a second control node that communicates with the host through a second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace (namespace), and the target namespace is used to store business data; a fault notification module, specifically used to: the host sends a fault notification to the second control node through the second communication link; the second control node transmits the fault notification to the first control node.

[0029] According to the fourth aspect, or any implementation of the fourth aspect above, the storage network also includes a second control node that communicates with the host through a second communication link, and the second control node and the first control node are used to manage different storage areas and share a target namespace; a fault notification module, specifically used to: the host sends a fault notification to the second control node through the second communication link; the second control node sends a fault notification to the first control node through a third communication link between the control nodes, or the second control node transmits the fault notification to the first control node.

[0030] According to the fourth aspect, or any implementation method of the fourth aspect above, the first resource clearing module is specifically used for: the first control node parses the fault notification to obtain resource clearing information; wherein the resource clearing information is used to indicate the first target resource; the first control node clears the first target resource indicated by the resource clearing information.

[0031] According to the fourth aspect, or any implementation of the fourth aspect above, the apparatus further includes:

[0032] The second resource clearing module is used for clearing the second target resource after the host receives the response notification of the control node to the fault notification; wherein the second target resource includes the computing resource used by the host to obtain business data through the first communication link.

[0033] The fourth aspect and any implementation of the fourth aspect correspond to the first aspect and any implementation of the first aspect, respectively. The technical effects corresponding to the fourth aspect and any implementation of the fourth aspect can refer to the technical effects corresponding to the above-mentioned first aspect and any implementation of the first aspect, which will not be repeated here.

[0034] In a fifth aspect, an embodiment of the present application provides an exception handling device, wherein multiple communication links based on the NVMe-oF protocol are established between the storage network and multiple communication interfaces of the host, the multiple communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, the storage network includes one or more control nodes including a first control node, and the device includes: an information feedback module, which is used to feedback a response notification to the host through the control node if a fault notification sent by the host is received during the process in which the first control node communicates with the host based on the NVMe-oF protocol through the first communication link; wherein the fault notification is a notification sent by the host through the second communication link when the fault of the first communication interface is detected; the response notification is used to instruct the host to communicate with the control node in the storage network through the second communication link to access the business data of the host in the storage network.

[0035] According to the fifth aspect, the device also includes: a first resource clearing module, which is used to: receive a fault notification through a first control node; clear a first target resource based on the fault notification; wherein the first target resource includes a computing resource used by the first control node to obtain business data through a first communication link; and when the clearing is completed, send a response notification for notifying the host to communicate through a second communication link.

[0036] According to the fifth aspect, or any implementation of the fifth aspect above, the storage network also includes a second control node that communicates with the host through a second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace (namespace), and the target namespace is used to store business data; an information feedback module, specifically used to: generate a response notification to the fault notification through the first control node and transmit it to the second control node; and feedback the response notification to the first control node through the second control node.

[0037] According to the fifth aspect, or any implementation of the fifth aspect above, the storage network also includes a second control node that communicates with the host through a second communication link, and the second control node and the first control node are used to manage different storage areas and share a target namespace; an information feedback module, specifically used to: generate a response notification to the fault notification through the first control node; the first control node transmits the response notification to the second control node, or the first control node sends a response notification to the second control node through a third communication link between the control nodes; and feedback the response notification to the first control node through the second control node.

[0038] The fifth aspect and any implementation of the fifth aspect correspond to the first aspect and any implementation of the first aspect, respectively. The technical effects corresponding to the fifth aspect and any implementation of the fifth aspect can refer to the technical effects corresponding to the first aspect and any implementation of the first aspect, which will not be repeated here.

[0039] In a sixth aspect, an embodiment of the present application provides an exception handling device, which is applied to a host, and multiple communication interfaces of the host establish multiple communication links based on the NVMe-oF protocol with the storage network, the multiple communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, the storage network includes one or more control nodes including a first control node, and the device includes: a fault notification module, which is used to communicate with the first control node based on the NVMe-oF protocol through the first communication link to access the business data of the host in the storage network, and send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; a path switching module, which is used to communicate with the control node in the storage network through the second communication link after receiving a response notification of the control node to the fault notification.

[0040] The sixth aspect and any implementation of the sixth aspect correspond to the first aspect and any implementation of the first aspect, respectively. The technical effects corresponding to the fifth and sixth aspects and any implementation of the sixth aspect can refer to the technical effects corresponding to the first aspect and any implementation of the first aspect, which will not be repeated here.

[0041] In the seventh aspect, an embodiment of the present application provides a storage system, which includes: a host and a storage network, wherein multiple communication links based on the NVMe-oF protocol are established between the storage network and multiple communication interfaces of the host, the multiple communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, and the storage network includes one or more control nodes including a first control node; the host is used to communicate with the first control node based on the NVMe-oF protocol through the first communication link to access the business data of the host in the storage network, and send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; the storage network is used to send a response notification to the fault notification to the host through the second communication link when a fault notification is received through the control node; the host is also used to communicate with the control node in the storage network through the second communication link after receiving a response notification to the fault notification from the control node.

[0042] The seventh aspect and any implementation of the seventh aspect correspond to the first aspect and any implementation of the first aspect, respectively. The technical effects corresponding to the seventh aspect and any implementation of the seventh aspect can refer to the technical effects corresponding to the above-mentioned first aspect and any implementation of the first aspect, which will not be repeated here.

[0043] In the eighth aspect, an embodiment of the present application provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by one or more processors, the one or more processors implement a method such as the first aspect, the second aspect, the third aspect, any one of the implementations of the first aspect, any one of the implementations of the second aspect, and any one of the implementations of the third aspect.

[0044] In the ninth aspect, an embodiment of the present application provides a computer-readable medium for storing a computer program, wherein the computer program includes instructions for executing the methods in the first aspect, the second aspect, the third aspect, any implementation of the first aspect, any implementation of the second aspect, and any implementation of the third aspect.

[0045] In the tenth aspect, an embodiment of the present application provides a computer program, which includes instructions for executing the methods in the first aspect, the second aspect, the third aspect, any implementation of the first aspect, any implementation of the second aspect, and any implementation of the third aspect.

[0046] In an eleventh aspect, an embodiment of the present application provides a chip, the chip comprising a processing circuit and a transceiver pin. The transceiver pin and the processing circuit communicate with each other through an internal connection path, and the processing circuit executes the method in the first aspect, the second aspect, the third aspect, any one of the implementations of the first aspect, any one of the implementations of the second aspect, and any one of the implementations of the third aspect to control the receiving pin to receive a signal and control the sending pin to send a signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0048] Figure 1 is a structural example diagram showing an exception handling architecture;

[0049] Figure 2a is a structural block diagram of a storage system 100 provided in an embodiment of the present application;

[0050] Figure 2b is a structural block diagram of an electronic device 200 provided in an embodiment of the present application;

[0051] Figure 3This is one of the structural block diagrams of a storage system provided in an embodiment of the present application;

[0052] Figure 4 It is a flowchart of an exception handling method provided in an embodiment of the present application;

[0053] Figure 5 This is one of the structural block diagrams of a storage system provided in an embodiment of the present application;

[0054] Figure 6 This is one of the example diagrams of the processing process of an exception handling method provided in an embodiment of the present application;

[0055] Figure 7 This is one of the structural block diagrams of a storage system provided in an embodiment of the present application;

[0056] Figure 8 This is one of the example diagrams of the processing process of an exception handling method provided in an embodiment of the present application;

[0057] Fig. 9 It is one of the structural block diagrams of an exception handling device 900 provided in an embodiment of the present application;

[0058] Fig.10 This is one of the structural block diagrams of an exception handling device 1000 provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to facilitate understanding of this embodiment, some technical terms and related technologies involved in this embodiment are first introduced:

[0060] NVMe-oF: A large parallel computer cluster containing flash memory that communicates using fabrics such as Ethernet and Fibre Channel, i.e., a storage network. In NVMe-oF, storage targets are presented in the form of namespaces.

[0061] NVMe-oF protocol: The protocol used in NVMe-oF specifies how to access storage targets in NVMe-oF.

[0062] nvmeSubsystem subsystem: used to indicate the storage area in NVMe-oF, which is equivalent to the sub-area obtained by dividing the storage area of ​​the storage network, and can specifically include at least one of the following storage areas: one or more physical storage devices, and one or more storage spaces in a physical storage device.

[0063] Namespace: The nvmeSubsystem subsystem may include at least one controller, i.e., a control node. Each controller may access several namespaces indicated by the nvmeSubsystem subsystem. Each namespace is composed of several logic blocks. A logic block is the smallest data unit that can be read or written by a controller.

[0064] Network interface card (NIC): Usually used to connect different types of devices to a network to access the Internet or an intranet.

[0065] Namespace Identifier (NSID): An identifier used to indicate the access location of a controller to access a namespace.

[0066] Figure 1 FIG. 1 is a diagram showing an example structure of an exception handling architecture. Figure 1 As shown, host Host1 and host Host2 may include physical devices or virtual devices such as computers, mobile terminals, virtual machines, wearable devices, etc. for providing services (such as instant messaging, cloud computing, etc.) to users; switches 1 and switches 2 are components of the communication network and are network switching nodes. Control node Ctrl A and control node Ctrl B are used to manage the storage end, such as the nvmeSubsystem subsystem. Host Host1 or host Host2 can access the storage end through 4 paths, that is, access the business data of the business run by the host in the storage end. For example, the 4 paths corresponding to host Host1 may include: path P1P5, path P1P6, path P2P7 and path P2P8; the 4 paths corresponding to host Host2 may include: path P3P5, path P3P6, path P4P7 and path P4P8. The storage end stores the business data of the host through a namespace, and different hosts correspond to different namespaces. For any host, the storage end pre-establishes a mapping relationship between the namespace corresponding to the host and the four paths corresponding to the host and sends it to the host to ensure that the host accesses the corresponding namespace in the storage end such as the nvmeSubsystem subsystem through any of the four paths corresponding to the host. In addition, the above services depend on the specific application scenario, such as mailbox services, instant messaging services, cloud storage services, etc., which are not limited in the embodiments of the present application.

[0067] For example, see Figure 1, when the communication link between the host and switch 1 fails, the host determines the failure through the heartbeat detection mechanism between the host and the control node Ctrl A, and then the host switches the path through the pre-installed specific plug-in. Path switching means that the host determines a new path other than the failed path from the above multiple paths through the program pre-set in the specific plug-in, and then the host accesses the storage network through the new path.

[0068] However, the above heartbeat detection requires sending multiple detection packets to the control node, which is time-consuming, and the host needs to install additional specific plug-ins, resulting in low exception handling efficiency.

[0069] In response to the above problems, an embodiment of the present application provides an exception handling method. The host directly sends a notification of a communication link failure to the first control node by detecting a communication interface failure, and then the first control node, based on the fault notification sent by the host, can instruct the host to access the service data of the host in the storage medium managed by the first control node through a non-faulty communication link such as the second communication link when the first communication interface fails, that is, the first communication link cannot be used for communication, through a response notification, thereby avoiding multiple heartbeat detection packets between the first control node and the host, thereby improving the efficiency of exception handling. In addition, the first control node feedback response notification can ensure that the host directly switches the communication link based on the protocol stack, so that the problem that the host needs to pre-install a specific plug-in to perform exception handling can be avoided, thereby further improving the efficiency and convenience of exception handling, and broadening the application scenarios of the embodiments of the present application.

[0070] Before describing the technical solution of the embodiment of the present application, the storage system provided by the embodiment of the present application is first described in conjunction with the accompanying drawings. Figure 2a 1 is a structural block diagram of a storage system 100 provided in an embodiment of the present application. Figure 2a As shown, the storage system 100 may include a host (HOST) and a storage network. Among them, one or more control nodes of the storage network including the first control node establish multiple communication links corresponding to multiple communication interfaces (such as communication interface HP0 and communication interface HP1) of the host. For example, in the case of one control node, the storage network establishes two communication links with the communication interface HP0 and the communication interface HP1 respectively through the network interface card (NIC) P0 and the network interface card (NIC) P1 corresponding to the control node (such as the first control node) via a switch (switch, SW). The host can access the control node through one of the multiple communication links. In order to facilitate understanding and reasonable layout, the process of interaction between the host and the control node in the storage system to solve the abnormality of the host accessing the storage network, as well as the case where the number of control nodes is multiple, are explained in conjunction with different structural diagrams of the storage system in the subsequent method embodiments.

[0071] In one example, the host may include a virtual machine running on an electronic device. The control node may be a virtual node at the logical level in NVMe-oF for managing storage areas. For example, the control node may be an object named "controller" created in the management logic of NVMe-oF. Among them, an object is an entity used to describe objective things in an application system, and is a basic operating unit with specific attributes and behaviors (methods).

[0072] In one example, the host and / or control node may be an electronic device. The electronic device provided by the embodiment of the present application is described below. The embodiment of the present application may be applied to electronic devices such as computers, mobile terminals, wearable devices, Internet televisions, servers, etc. The electronic device may be used as a host or control node in the storage system provided by the embodiment of the present application. For example, Figure 2b 2 is a structural block diagram of an electronic device 200 provided in an embodiment of the present application. Figure 2b As shown, the electronic device 200 may include: a processor 201 , a communication line 205 , a memory 203 , a storage medium 204 , and some or all of the components in at least one communication interface 202 .

[0073] Among them, the processor 201 is used to execute the exception handling method provided in the subsequent embodiments of this application.

[0074] Specifically, the processor 201 may include a general-purpose central processing unit (CPU), and the processor 201 may also include a microprocessor, a field programmable gate array (Field Programmable Gate Array, FPGA), a digital signal processor (digital signal processing, DSP) or an application-specific integrated circuit (application-specific integrated circuit, ASIC), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0075] The communication line 205 is used to connect various components in the electronic device 200. Specifically, the communication line 205 may include a data bus, a power bus, a control bus, a status signal bus, etc. However, for the sake of clarity, various buses are marked as communication lines 205 in the figure.

[0076] The memory 203 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The memory 203 may be independent and connected to the processor 201 via communication lines 207205. The memory 203 may also be integrated with the processor 201. The memory 203 stores computer instructions. The processor 201 may implement the exception handling method provided in the subsequent embodiments of the present application by executing the computer instructions stored in the memory 203. Optionally, the computer execution instructions in the present embodiment may also be referred to as application code, which is not specifically limited in the present embodiment.

[0077] The storage medium 204 is used to store computer instructions and various data for implementing the technical solutions of the embodiments of the present application, so that when the electronic device 200 executes the exception handling method provided in the subsequent embodiments of the present application, the computer instructions and various data stored in the storage medium 204 are loaded into the memory 203, so that the processor 201 can implement the exception handling method provided in the embodiments of the present application by executing the computer instructions stored in the memory 203.

[0078] The communication interface 202 uses any electronic device such as a transceiver for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0079] It should be understood that Figure 2b The electronic device 200 shown is only one example of an electronic device, and the electronic device 200 may have more or fewer components than shown in the figure, may combine two or more components, or may have a different configuration of components. Figure 2b The various components shown in the EMBODIMENTS 2000 may be implemented in hardware, software, or a combination of hardware and software, including one or more signal processing and / or application specific integrated circuits.

[0080] Combine the following Figures 3 to 8 The exception handling method and storage system provided in the embodiments of the present application are exemplarily described.

[0081] Figure 3 This is one of the structural block diagrams of a storage system provided in an embodiment of the present application. Figure 3As shown, for a storage network with multiple nodes, in one example, the storage network may include multiple controllers, that is, control nodes. For example, a first control node NodeA and a second control node NodeB. Multiple communication interfaces (such as a first communication interface: communication interface HP0, and a second communication interface: communication interface HP1) between the storage network and the host (HOST) establish multiple communication links based on the NVMe-oF protocol. The multiple communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface. The first communication link is used for communication between the host and the first control node, and the second communication link is used for communication between the host and the second control node. Exemplarily, taking the first communication link as an example, the host establishes a communication connection with the controller1 (controller object) representing the first control node created by the storage network based on the NVMe command request for establishing a communication connection in the NVMe-oF protocol at the application layer, and the controller1 of the storage network establishes an association with the host in response to the request, and forms multiple communication links (paths) at the physical layer and the network layer through the communication medium between the host and the storage network.

[0082] In addition, in NVMe-oF, the communication medium between the host and the storage network may include one or more of Ethernet, Fibre Channel, RoCE (a transmission medium for distributed storage networks) or InfiniBand (a transmission medium for distributed storage networks). After the application layer of the host instructs to establish a communication link, the specific communication devices involved in the network layer and the physical layer, such as switches, establish a communication link according to the instructions of the application layer and the communication technology of the corresponding communication medium. For example, Ethernet's TCP / IP technology, RoCE and InfiniBand's RDMA technology. Among them, RDMA technology can directly access memory data through the network interface without the intervention of the operating system kernel.

[0083] Figure 4 is a flowchart of an exception handling method provided in an embodiment of the present application. Figure 4 As shown, the exception handling method provided in the embodiment of the present application may include but is not limited to the following steps:

[0084] S401, when a host accesses a control node through a first communication link among multiple communication links, the host detects whether a first communication interface corresponding to the first communication link is faulty.

[0085] For example, Figure 3As shown, in an embodiment of the present application, the control node may include one or more control nodes including a first control node in the storage network. The first communication link may be, for example, a communication link between a communication interface HP0, a switch, a network interface card P0 and a first control node. The host accesses the control node through the first communication link among multiple communication links, that is, the host communicates with the first control node based on the NVMe-oF protocol through the first communication link to access the service data of the host in the storage network. For example, the host sends a read / write instruction to the service data of the host in the storage network through the first communication link, and the first control node performs a read / write operation on the service data of the host in the storage network according to the received read / write instruction. In addition, the host can perform fault detection on the HP0 port of the host through the fault detection function of the communication interface, such as the network port.

[0086] S402: When a first communication interface fails, the host sends a failure notification to a control node of the storage network through a second communication link corresponding to a second communication interface.

[0087] The host detects a first communication interface failure (such as an interface downtime or other communication interface failure), indicating that the first communication link cannot be used for communication, and can send a fault notification to the control node of the storage network through a second communication link other than the first communication link among multiple communication links established with the storage network. For example, Figure 3 The second communication link corresponding to the second communication interface shown is the communication link for communicating with the communication interface HP1: the communication link between the communication interface HP1, the switch and the network interface card P1 and the second control node. Figure 3 As shown, the host notifies the second control node NodeB of the fault information through the path of the communication interface HP1, that is, the second communication link. The fault information can be, for example, Figure 3 The notification message “①” is shown.

[0088] It can be understood that, when the number of the control node is one, the second communication link may be a communication link between the communication interface HP1, the switch and the network interface card P1 and the first control node.

[0089] In one example, Figure 5 This is one of the structural block diagrams of a storage system provided in an embodiment of the present application. Figure 5As shown, two communication links are established between the host and an nvmeSubsystem subsystem through two control nodes through a communication network (such as Ethernet, optical fiber, etc.). The two communication links work in the master-slave mode, the path (that is, the communication link) between the host and the first control node (controller1) is the primary path, and the path between the host and the second control node (controller2) is the backup path. The first control node (controller1) and the second control node (controller2) manage the same storage area nvmeSubsystem subsystem n1, and share the namespace (namespace) ns1 for storing the service data of the host. In other words, the first control node (controller1) and the second control node (controller2) run in the same operating domain: Domain 1.

[0090] Figure 6 FIG. 1 is one of the example diagrams of the processing process of an exception handling method provided in an embodiment of the present application. Figure 6 As shown, this method is applied to Figure 5 The storage system shown, and Figure 4 Compared with the process shown in the figure, the difference is that it can specifically include the following steps:

[0091] S4021, when the first communication interface fails, the host sends a fault notification to the nvmeSubsystem subsystem through the second communication link, and the second control node transmits the fault notification to the first control node of the NvmeSubsystem subsystem.

[0092] The host sends the extended command in the NVMe-oF protocol stack to the second control node of the nvmeSubsystem subsystem through the backup path, that is, the second communication link, that is, sends the fault notification to the nvmeSubsystem subsystem. In this way, within the nvmeSubsystem subsystem, the second control node can directly pass the fault notification to the first control node. Exemplarily, the second control node and the first control node can correspond to different processes respectively, so that the second control node can pass the fault notification to the first control node through inter-process communication.

[0093] The above-mentioned extended command may include a new NVM set Feature command interface extended on the basis of the host protocol stack in the NVMe-oF protocol of the embodiment of the present application, through which the host Host sends a control message, i.e., a fault notification. Based on this, the NVMe host protocol stack provided in the embodiment of the present application may include the command interface information shown in Table 1 below:

[0094]

[0095] Among them, Feature Identifier (FID) refers to a function identifier; set_features refers to a command requesting the setting or enabling of a feature value of a USB device, interface, or endpoint. Uses MemoryBuffer for Attributes refers to whether to allocate a memory buffer for the instruction according to the attributes of the instruction. Here, it is set to allocate according to the attributes of the instruction, that is, yes. FeatureName is the function name of the instruction. In the embodiment of the present application, it is set to clear the computing resources of the control node (Clearcontrollers resource). It can be understood that the above specific settings are only examples and can be set according to application requirements in specific applications.

[0096] In an embodiment of the present application, a second control node that manages the same storage area as the first control node and shares a target namespace can be used to achieve rapid switching of a host access path when a host access path of an nvmeSubsystem subsystem fails.

[0097] In one example, Figure 7 This is one of the structural block diagrams of a storage system provided in an embodiment of the present application. Figure 7 As shown, the host and two nvmeSubsystem subsystems (such as nvmeSubsystem subsystem n1 and nvmeSubsystem subsystem n2) communication networks (such as Ethernet, optical fiber, etc.) respectively establish a path, that is, a communication link. Assuming that these two paths work in the master-slave mode, the path between the host and the first control node (controller1) is the master path, and the path between the host and the second control node (controller2) is the backup path. The second control node (controller2) and the first control node (controller1) are used to manage different storage areas nvmeSubsystem subsystem n2 and nvmeSubsystem subsystem n1, respectively, and share the namespace (namespace) 1. In other words, the first control node (controller1) can access the namespace (namespace) 1 of the host through the namespace identifier NSID1, and the second control node (controller2) can access the namespace (namespace) 1 used to store the service data of the host through the namespace identifier NSID2. In this way, the control nodes of different nvmeSubsystem subsystems can access the same namespace through different namespace identifiers.

[0098] Figure 8FIG. 1 is one of the example diagrams of the processing process of an exception handling method provided in an embodiment of the present application. Figure 8 As shown, this method is applied to Figure 7 The storage system shown, and Figure 4 Compared with the process shown in the figure, the difference is that the method can specifically include the following steps:

[0099] S4022: When the first communication interface fails, the host sends a fault notification to the nvmeSubsystem subsystem n2 through the second communication link corresponding to the second communication interface.

[0100] The host sends an extended command to NVM subsystem 2 using the backup path, i.e., the second communication link, to notify NVM subsystem 2 of the fault information: the host sends a fault notification to the second control node via the second communication link (e.g., Figure 3 Notification message “②” shown).

[0101] In an embodiment of the present application, between multiple nvmeSubsystem subsystems (such as greater than or equal to 2), the above-mentioned second communication link is used to achieve rapid switching of the host access path when a host access path across nvmeSubsystem subsystems fails.

[0102] S403: The control node of the storage network sends a response notification to the fault notification through the second communication link.

[0103] After receiving the fault notification, the control node of the storage network may perform resource processing using the second communication link as the primary path, such as adjusting the primary-backup working mode between the first control node and the second control node from the first control node Controller1 being the primary node to the second control node Controller2 being the primary node, etc. After the resource processing is completed, the control node of the storage network may reply to the host through the second communication link with a response notification of a successful response.

[0104] In an optional embodiment, if Figure 6 The exception handling method shown may further include the following steps:

[0105] S4051, the nvmeSubsystem subsystem clears the target resources in the first control node.

[0106] After receiving the fault notification, the first control node Controller1 of the nvmeSubsystem subsystem parses the fault notification and obtains resource clearing information; wherein, the resource clearing information is used to indicate the first target resource; the target resource, i.e., the first target resource, is cleared according to the fault notification. After the clearing is completed, the nvmeSubsystem subsystem sends a response notification to the fault notification to the host through the second communication link. Specifically, after the clearing of the target resource is completed, the first control node Controller1 transmits a response notification of the clearing completion to the second control node Controller2, that is, a response notification of a successful response is replied to the host.

[0107] In one example, the first target resource may include a computing resource for establishing the first communication link, for example, one or more of a CPU resource, a memory resource, and a network resource for establishing the first communication link. And / or, the target resource may include service data of the host obtained by the control node through the first communication link. The embodiment of the present application is not limited to this, and any computing resource used to access the service data of the host through the first communication link can be used as a target resource.

[0108] In one example, the embodiment of the present application expands the pre-established data structure (Data Structure Entry) of resource clearing information in the NVMe-oF protocol to indicate the computing resources (Clear controllers resource) that the control node corresponding to the failed communication link needs to clear, that is, the first target resource. The host can generate a fault notification based on the data structure of resource clearing information pre-stored in the host protocol stack of the NVMe-oF protocol.

[0109] Exemplarily, the data structure of resource clearing information is shown in Table 2 below:

[0110]

[0111] In an optional embodiment, if Figure 8 The exception handling method shown may further include the following steps:

[0112] S4052, nvmeSubsystem subsystem n2 notifies nvmeSubsystem subsystem n1 to clear the target resource in the first control node

[0113] S4052 and Figure 6 The same parts are not described here, see the above Figure 6Description of the embodiment. The difference is that the two control nodes are used to manage different nvmeSubsystem subsystems. Specifically: After receiving the fault notification, the second control node (controller2) of the nvmeSubsystem subsystem n2:

[0114] In the case where the first control node and the second control node belong to different physical storage devices (such as disks, hard disks, etc.), the second control node may send a fault notification to the first control node (controller1) via a third communication link between the second control node and the control node. The third communication link may include a wireless or wired (hardware connection, etc.) communication connection established between different physical storage devices.

[0115] Alternatively, when the first control node and the second control node belong to the same physical storage device, the second control node transmits the fault notification to the first control node. For example, the second control node and the first control node may correspond to different processes respectively, so that the second control node can transmit the fault notification to the first control node through inter-process communication.

[0116] With the above Figure 6 Similarly, the first control node (controller1) can clear the target resource according to the fault notification, that is, the nvmeSubsystem subsystem sends a response notification to the fault notification to the host through the second communication link after the clearing is completed. Figure 6 Description of the embodiment. The difference is that the control nodes in this embodiment belong to different nvmeSubsystem subsystems. Specifically: After the first control node (controller1) clears the target resource:

[0117] In the case where the first control node and the second control node belong to different physical storage devices (such as disks, hard disks, etc.), the first control node may send a response notification to the second control node (controller1) via a third communication link between the first control node and the control node. The third communication link may include a wireless or wired (hardware connection, etc.) communication connection established between different physical storage devices.

[0118] Alternatively, when the first control node and the second control node belong to the same physical storage device, the first control node transmits the response notification to the second control node. For example, the second control node and the first control node may correspond to different processes respectively, so that the first control node may transmit the response notification to the second control node through inter-process communication.

[0119] After receiving the response notification from the first control node, the control node may send a response notification to the fault notification to the host through the second communication link.

[0120] S404: After receiving a response notification from the control node to the fault notification, the host accesses the control node through the second communication link.

[0121] Exemplarily, after receiving the extended command, i.e., the notification of successful execution of the fault notification (such as the response notification to the fault notification), the host switches the service to the path of the communication interface HP1, that is, the host accesses the control node through the second communication link. For example, when there is one control node, the host accesses the first control node through the path of the communication interface HP1; when there are multiple control nodes, the host accesses the second control node through the path of the communication interface HP1.

[0122] It is understandable that, when the number of control nodes is greater than or equal to 3, the second control node may be, for example, Figure 5 and Figure 7 The second control node (controller2) shown is a control node used to transmit fault notifications and response notifications, or it can be a control node that can communicate with the host among the control nodes (controller3, ...controller, M is the number of control nodes) that are not shown in the figure and do not participate in the exception handling process.

[0123] In an exemplary embodiment, when the host receives the clearing completion notification, i.e., the response notification, from the control node, the host may also clean up the second target resource on the host side to release the local storage space, computing resources, etc. of the host. The second target resource may include the computing resource used by the host to obtain the service data through the first communication link. For example, one or more of the CPU resources, memory resources, and network resources used to establish the first communication link.

[0124] It should be understood that Figure 5 and Figure 7 The storage system shown is merely one example of a storage system, and the storage system may have more or fewer components than shown in the figures, may combine two or more components, or may have a different configuration of components. Figure 5 and Figure 7 The various components shown in the EMBODIMENTS 2000 may be implemented in hardware, software, or a combination of hardware and software, including one or more signal processing and / or application specific integrated circuits.

[0125] In one example, Fig. 9 This is one of the structural block diagrams of an exception handling device 900 provided in an embodiment of the present application. Fig. 9As shown, a plurality of communication links based on the NVMe-oF protocol are established between the storage network and the plurality of communication interfaces of the host, the plurality of communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, the storage network includes one or more control nodes including the first control node, and the exception handling device 900 may include:

[0126] A fault notification module 901 is used for the host to communicate with the first control node based on the NVMe-oF protocol through the first communication link to access the service data of the host in the storage network, and to send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected;

[0127] The path switching module 902 is used for the host to communicate with the control node in the storage network through the second communication link after receiving the response notification of the control node to the fault notification.

[0128] In one example, Fig.10 This is one of the structural block diagrams of an exception handling device 1000 provided in an embodiment of the present application. Fig.10 As shown, a plurality of communication links based on the NVMe-oF protocol are established between the storage network and the plurality of communication interfaces of the host, the plurality of communication links include a first communication link for communicating with the first communication interface, and a second communication link for communicating with the second communication interface, the storage network includes one or more control nodes including the first control node, and the exception handling device 1000 may include:

[0129] The information feedback module 1001 is used to feedback a response notification to the host through the control node if a fault notification sent by the host is received during the process in which the first control node communicates with the host based on the NVMe-oF protocol through the first communication link;

[0130] Among them, the fault notification is a notification sent by the host through the second communication link when the host detects a fault in the first communication interface; the response notification is used to instruct the host to communicate with the control node in the storage network through the second communication link to access the service data of the host in the storage network.

[0131] All relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and will not be repeated here.

[0132] This embodiment further provides a computer storage medium, in which computer instructions are stored. When the computer instructions are executed on an electronic device, the electronic device executes the above-mentioned related method steps to implement the exception handling method in the above-mentioned embodiment.

[0133] This embodiment further provides a computer program product. When the computer program product is run on a computer, the computer is caused to execute the above-mentioned related steps to implement the exception handling method in the above-mentioned embodiment.

[0134] Among them, the electronic device, computer storage medium, computer program product or chip provided in this embodiment is used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above and will not be repeated here.

[0135] Any content of each embodiment of the present application, as well as any content of the same embodiment, can be freely combined. Any combination of the above content is within the scope of the present application.

[0136] Those skilled in the art should be aware that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented with hardware, software, firmware, or any combination thereof. When implemented using software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein the communication media include any media that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that a general or special-purpose computer can access.

[0137] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. An exception handling method, characterized in that: A storage network and a plurality of communication interfaces of a host establish a plurality of communication links based on the NVMe-oF protocol, wherein the plurality of communication links include a first communication link for communicating with a first communication interface and a second communication link for communicating with a second communication interface, the storage network includes one or more control nodes including a first control node, and the method includes: The host communicates with the first control node based on the NVMe-oF protocol through the first communication link to access service data of the host in the storage network, and sends a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; After receiving the response notification from the control node to the fault notification, the host communicates with the control node in the storage network through the second communication link.

2. The method according to claim 1, characterized in that The method further comprises: The first control node receives the fault notification; The first control node clears a first target resource based on the fault notification; wherein the first target resource includes a computing resource used by the first control node to obtain the service data through the first communication link; When the clearing is completed, the first control node sends a response notification for notifying the host to communicate through the second communication link.

3. The method according to claim 1 or 2, characterized in that: The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace, and the target namespace is used to store the service data; The sending a fault notification to the first control node through the second communication link includes: The host sends a fault notification to the second control node through the second communication link; The second control node transmits the fault notification to the first control node.

4. The method according to claim 1 or 2, characterized in that: The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node being used to manage different storage areas and share a target namespace; The sending a fault notification to the first control node through the second communication link includes: The host sends a fault notification to the second control node through the second communication link; The second control node sends the fault notification to the first control node through a third communication link between the second control node and the first control node, or, The second control node transmits the fault notification to the first control node.

5. The method according to claim 2, characterized in that: The first control node clearing the first target resource based on the fault notification includes: The first control node parses the fault notification to obtain resource clearing information; wherein the resource clearing information is used to indicate the first target resource; The first control node clears the first target resource indicated by the resource clearing information.

6. The method according to any one of claims 1 to 5, characterized in that The method further comprises: After receiving a response notification from the control node to the fault notification, the host clears the second target resources; wherein the second target resources include computing resources used by the host to obtain the service data through the first communication link.

7. An exception handling method, characterized in that: A storage network and a plurality of communication interfaces of a host establish a plurality of communication links based on the NVMe-oF protocol, wherein the plurality of communication links include a first communication link for communicating with a first communication interface and a second communication link for communicating with a second communication interface, the storage network includes one or more control nodes including a first control node, and the method includes: During the process in which the first control node communicates with the host based on the NVMe-oF protocol through the first communication link, if a fault notification sent by the host is received, a response notification to the fault notification is fed back to the host through the control node; Among them, the fault notification is a notification sent by the host through the second communication link when the host detects a fault in the first communication interface; the response notification is used to instruct the host to communicate with the control node in the storage network through the second communication link to access the business data of the host in the storage network.

8. The method according to claim 7, characterized in that The method further includes: executing the following steps by the first control node: receiving the fault notification; Clearing a first target resource based on the fault notification; wherein the first target resource includes a computing resource used by the first control node to obtain the service data through the first communication link; When the clearing is completed, a response notification is sent to notify the host to communicate through the second communication link.

9. The method according to claim 7 or 8, characterized in that: The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace, and the target namespace is used to store the service data; The feeding back a response notification to the fault notification to the host through the control node includes: generating, by the first control node, a response notification to the fault notification and transmitting it to the second control node; The response notification is fed back to the first control node through the second control node.

10. The method according to claim 7 or 9, characterized in that: The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node being used to manage different storage areas and share a target namespace; The feeding back a response notification to the fault notification to the host through the control node includes: generating, by the first control node, a response notification to the fault notification; The first control node transmits the response notification to the second control node, or the first control node sends the response notification to the second control node through a third communication link between the first control node and the control node; The response notification is fed back to the first control node through the second control node.

11. An exception handling method, characterized in that: Applied to a host, multiple communication interfaces of the host and a storage network establish multiple communication links based on the NVMe-oF protocol, the multiple communication links include a first communication link for communicating with a first communication interface, and a second communication link for communicating with a second communication interface, the storage network includes one or more control nodes including a first control node, and the method includes: Communicate with the first control node based on the NVMe-oF protocol through the first communication link to access the service data of the host in the storage network, and send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; After receiving a response notification from the control node to the fault notification, communicating with the control node in the storage network through the second communication link.

12. An exception handling device, characterized in that: A plurality of communication links based on the NVMe-oF protocol are established between the storage network and the plurality of communication interfaces of the host, the plurality of communication links including a first communication link for communicating with the first communication interface and a second communication link for communicating with the second communication interface, the storage network including one or more control nodes including a first control node, and the device including: a fault notification module, configured for the host to communicate with the first control node based on the NVMe-oF protocol through the first communication link to access the service data of the host in the storage network, and to send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; A path switching module is used for the host to communicate with the control node in the storage network through the second communication link after receiving a response notification from the control node to the fault notification.

13. The device according to claim 12, characterized in that The device also includes: A first resource clearing module, configured for the first control node to receive the fault notification and clear a first target resource based on the fault notification; wherein the first target resource includes a computing resource used by the first control node to obtain the service data through the first communication link; An information feedback module is used for the first control node to send a response notification for notifying the host to communicate through the second communication link when the clearing is completed.

14. The device according to claim 12 or 13, characterized in that The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace, and the target namespace is used to store the service data; The fault notification module is specifically used for: The host sends a fault notification to the second control node through the second communication link; and the second control node transmits the fault notification to the first control node.

15. The device according to claim 12 or 13, characterized in that The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node being used to manage different storage areas and share a target namespace; The fault notification module is specifically used for: The host sends a fault notification to the second control node through the second communication link; The second control node sends the fault notification to the first control node via a third communication link between the second control node and the first control node, or the second control node transmits the fault notification to the first control node.

16. The device according to claim 13, characterized in that The first resource clearing module is specifically used to: The first control node parses the fault notification to obtain resource clearing information; wherein the resource clearing information is used to indicate the first target resource; The first control node clears the first target resource indicated by the resource clearing information.

17. The device according to any one of claims 12 to 16, characterized in that The device also includes: The second resource clearing module is used for the host to clear the second target resource after receiving the response notification of the control node to the fault notification; wherein the second target resource includes the computing resource used by the host to obtain the business data through the first communication link.

18. An exception handling device, characterized in that: A plurality of communication links based on the NVMe-oF protocol are established between the storage network and the plurality of communication interfaces of the host, the plurality of communication links including a first communication link for communicating with the first communication interface and a second communication link for communicating with the second communication interface, the storage network including one or more control nodes including a first control node, and the device including: An information feedback module, configured to feedback a response notification to the fault notification to the host through the control node if a fault notification sent by the host is received during the process in which the first control node communicates with the host based on the NVMe-oF protocol through the first communication link; Among them, the fault notification is a notification sent by the host through the second communication link when the host detects a fault in the first communication interface; the response notification is used to instruct the host to communicate with the control node in the storage network through the second communication link to access the business data of the host in the storage network.

19. The device according to claim 18, characterized in that The device further includes: a first resource clearing module, configured to, through the first control node: receiving the fault notification; and clearing a first target resource based on the fault notification; wherein the first target resource comprises a computing resource used by the first control node to obtain the service data through the first communication link; When the clearing is completed, a response notification is sent to notify the host to communicate through the second communication link.

20. The device according to claim 18 or 19, characterized in that The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node are used to manage the same storage area and share a target namespace, and the target namespace is used to store the service data; The information feedback module is specifically used for: generating, by the first control node, a response notification to the fault notification and transmitting it to the second control node; The response notification is fed back to the first control node through the second control node.

21. The device according to claim 18 or 20, characterized in that The storage network further includes a second control node communicating with the host through the second communication link, the second control node and the first control node being used to manage different storage areas and share a target namespace; The information feedback module is specifically used for: generating, by the first control node, a response notification to the fault notification; The first control node transmits the response notification to the second control node, or the first control node sends the response notification to the second control node through a third communication link between the first control node and the control node; The response notification is fed back to the first control node through the second control node.

22. An exception handling device, characterized in that: Applied to a host, multiple communication interfaces of the host and a storage network establish multiple communication links based on the NVMe-oF protocol, the multiple communication links include a first communication link for communicating with a first communication interface, and a second communication link for communicating with a second communication interface, the storage network includes one or more control nodes including a first control node, and the device includes: a fault notification module, configured to communicate with the first control node based on the NVMe-oF protocol through the first communication link to access the service data of the host in the storage network, and send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; A path switching module is used to communicate with the control node in the storage network through the second communication link after receiving a response notification from the control node to the fault notification.

23. A storage system, characterized in that: The system includes: a host and a storage network, wherein the storage network and multiple communication interfaces of the host establish multiple communication links based on the NVMe-oF protocol, the multiple communication links include a first communication link for communication with a first communication interface, and a second communication link for communication with a second communication interface, and the storage network includes one or more control nodes including a first control node; The host is configured to communicate with the first control node based on the NVMe-oF protocol through the first communication link to access service data of the host in the storage network, and send a fault notification to the first control node through the second communication link when a fault of the first communication interface is detected; The storage network is configured to send a response notification to the fault notification to the host through the second communication link when the fault notification is received through the control node; The host is further configured to communicate with the control node in the storage network through the second communication link after receiving a response notification from the control node to the fault notification.

24. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 11.

25. A computer-readable storage medium, characterized in that: The method comprises a computer program, wherein when the computer program is run on an electronic device, the electronic device executes the method according to any one of claims 1 to 11.

26. A chip, characterized in that: It comprises one or more interface circuits and one or more processors; the interface circuit is used to receive a signal from a memory of an electronic device and send the signal to the processor, the signal comprising a computer instruction stored in the memory; when the processor executes the computer instruction, the electronic device executes the method described in any one of claims 1 to 11.

27. A computer program product, characterized in that The invention comprises a computer program, which, when executed by an electronic device, enables the electronic device to execute the method according to any one of claims 1 to 11.

Citation Information

Cited By

  • Computing device communication system and server

    CN119739570A

  • Edge server management method and device, electronic equipment and storage medium

    CN120407349A