CN116260743A

CN116260743A - Link detection method and device, intelligent network card and storage medium

Info

Publication number: CN116260743A
Application number: CN202310193156.5A
Authority: CN
Inventors: 彭剑远
Original assignee: New H3C Technologies Co Ltd
Current assignee: New H3C Technologies Co Ltd
Priority date: 2023-02-22
Filing date: 2023-02-22
Publication date: 2023-06-13

Abstract

The application provides a link detection method, a device, an intelligent network card and a storage medium, which are used for unloading a server link detection function onto the intelligent network card, and realizing the functions of autonomous link detection and isolation of a fault link of the intelligent network card. According to the method and the device, the link detection function is issued to the intelligent network card, the intelligent network card acquires the link detection capability of the opposite-end forwarding equipment, the link detection function between the intelligent network card and the forwarding equipment is enabled, and the link fault is detected and corresponding fault processing is carried out according to the sent link detection message and the link detection response message fed back by the forwarding equipment. The intelligent network card can also judge the fault generation position by comparing the link detection results of the local and forwarding equipment sides. According to the method and the device, the link detection function can be unloaded through the intelligent network card, the CPU load is reduced, the position of fault generation can be more accurately positioned, and the network fault positioning efficiency is improved.

Description

Link detection method and device, intelligent network card and storage medium

Technical Field

The present disclosure relates to the field of communications technologies, and in particular, to a link detection method, a device, an intelligent network card, and a storage medium.

Background

The Smart NIC, i.e., the intelligent network card, can transfer the virtual switching function from the server CPU to the network card of the server, thereby releasing the computing power of the expensive server CPU to the application program, so that the network card function can be better expanded and higher server performance can be provided.

The intelligent network card is characterized in that the intelligent network card is provided with the following characteristics that the FPGA is a field programmable gate array to assist a server CPU to process network load and program a network interface function:

(1) The FPGA localized programming can support the function customization of a data plane and a control plane and assist the CPU to process network load;

(2) Typically comprising multiple ports and internal virtual switches, which can rapidly forward data and intelligently map to related applications based on network packets, application sockets, etc.;

(3) Network traffic may be detected and managed.

Smart NIC can promote application program and virtualization performance, realize advantages of software defined network (Software Defined Network, SDN) and network function virtualization (Network Function Virtualization, NFV), remove network virtualization, load balancing and other low-level functions from a server CPU, and ensure maximum processing capacity for applications. At the same time, the intelligent network card can also provide distributed computing resources, so that users can develop own software or provide access services, thereby accelerating specific application programs.

In order to isolate a fault link, a scheme is generally adopted in the prior art, wherein a CPU on a server sequentially sends detection messages from each member port of an aggregation port, and after receiving the detection messages, an opposite end server sends response messages, so that the time stamps of the response messages and the detection messages are compared, and the message forwarding time delay on the member port link can be obtained. When the message forwarding delay on a certain member port link is too large or no response message is received, the member port link is considered to be faulty, and then the state of the member port is set to be an inactive state, and the message is not received from the member port later, so that the problems of large user message delay or message loss are avoided.

The above scheme has some technical defects:

firstly, the CPU on the server transmits the detection message to occupy the CPU resource of the server, which affects the user service, because the CPU is shared, and the CPU scheduling of the user service may not be timely caused when the CPU is occupied by the transmission detection message.

Second, the above solution cannot subdivide the cause of the link failure, and cannot determine whether the problem is on a forwarding device (e.g., a router or a switch) or a server.

Disclosure of Invention

In view of this, the present application provides a link detection method, device, intelligent network card and storage medium, which are used for solving the technical problem that the link detection function occupies the server CPU resource.

Based on an aspect of an embodiment of the present application, the present application provides a link detection method, where the method is applied to an intelligent network card of a server, and the method further includes:

acquiring the link detection capability of forwarding equipment;

negotiating and configuring a link detection function with the forwarding device;

periodically sending a detection message to perform link detection;

and judging the link fault and processing the fault according to the receiving and transmitting conditions of the detection message.

Further, the method further comprises:

after detecting the link fault, obtaining a link detection result of the forwarding equipment side;

and judging the fault generation position according to the link detection result of the local and forwarding equipment side and carrying out corresponding fault processing.

Further, the method for obtaining the link detection capability of the forwarding device comprises the following steps:

and obtaining the link detection capability of the forwarding equipment through an extended link discovery layer protocol LLDP message.

Further, the obtaining the link detection capability of the forwarding device includes:

acquiring whether the forwarding device supports link detection and/or acquiring a link detection protocol supported by the forwarding device, and acquiring a hash mode of the forwarding device on a message forwarded through the aggregation port.

Further, the steps of link fault judgment and fault processing in the fault processing are to isolate a fault link according to the receiving and transmitting conditions of the detection message;

and judging the fault generation position according to the link detection result of the local and forwarding equipment side, and performing fault processing steps in corresponding fault processing to call an alarm interface of the server to send an alarm, wherein the alarm comprises information of the fault generation position.

According to another aspect of the embodiments of the present application, the present application further provides a link detection device, where the device is applied to an intelligent network card of a server, and the device includes:

the capability acquisition module is used for acquiring the link detection capability of the forwarding equipment;

a negotiation configuration module, configured to negotiate with the forwarding device and configure a link detection function;

the period detection module is used for periodically sending detection messages to carry out link detection;

and the judging and processing module is used for judging the link failure according to the receiving and transmitting condition of the detection message and isolating the failed link when judging that the link fails.

Further, the apparatus further comprises:

the result acquisition module is used for acquiring a link detection result of the forwarding equipment side after the link fault is detected;

the comparison judging module is used for judging the fault generation position according to the link detection result of the local and forwarding equipment side and calling an alarm interface of the server to send an alarm, wherein the alarm comprises information of the fault generation position.

Further, the capability obtaining module obtains the link detection capability of the forwarding device through an extended link discovery layer protocol LLDP message.

Further, the capability obtaining module obtains the link detection capability of the forwarding device, including: acquiring whether the forwarding device supports link detection and/or acquiring a link detection protocol supported by the forwarding device, and acquiring a hash mode of the forwarding device on a message forwarded through the aggregation port.

Based on another aspect of the embodiments of the present application, the present application further provides an intelligent network card, where the intelligent network card includes a processor, a communication interface, a storage medium, and a communication bus, where the processor, the communication interface, and the storage medium complete communication with each other through the communication bus;

a storage medium storing a computer program;

and a processor, configured to implement method steps of the link detection method provided in the present application when executing the computer program stored on the storage medium.

Based on another aspect of the embodiments of the present application, the present application also provides a storage medium having stored thereon a computer program which, when executed by a processor, implements the method steps of the link detection method provided by the present application.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following description will briefly describe the drawings that are required to be used in the embodiments of the present application or the description in the prior art, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and other drawings may also be obtained according to these drawings of the embodiments of the present application for a person having ordinary skill in the art.

Fig. 1 is a schematic diagram of a device structure for implementing a link detection method according to an embodiment of the present application;

fig. 2 is a flowchart illustrating steps of a link detection method according to an embodiment of the present application;

fig. 3 is a schematic diagram of an intelligent network card structure for implementing a link detection method according to an embodiment of the present application.

Detailed Description

The terminology used in the embodiments of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the embodiments of the application. As used in the embodiments of the present application, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be understood that although the terms first, second, third, etc. may be used in embodiments of the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one from another or similar information, entity or step, but not to describe a particular sequence or order. For example, the first information may also be referred to as second information, and similarly, the second information may also be referred to as first information, without departing from the scope of embodiments of the present application. Furthermore, the word "if" as used may be interpreted as "at … …" or "at … …" or "in response to a determination". The term "and/or" in this application is merely an association relation describing an association object, and means that three kinds of relations may exist, for example, a and/or B may mean: there are three cases, a alone, a and B together, and B alone, wherein a, B may be singular or plural. Also, in the description of the present application, unless otherwise indicated, "a plurality" means two or more than two. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b, or c may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or plural.

In a server using a conventional network card (Network Interface Card, NIC), the conventional network card is usually plugged directly into a bus slot of a host motherboard, and when the server performs network link detection, a processor (CPU) of the server is required to receive/send and process detection messages in person, which definitely increases the load of the server.

Compared with the traditional network card, the intelligent network card (Smart NIC) not only can complete the network transmission function of the standard network card, but also can provide a built-in programmable and configurable hardware acceleration engine, thereby improving the processing performance of the server application and reducing the resource consumption of the server CPU in communication, and providing more CPU resources for the application. If the characteristics of the intelligent network card are utilized, the link detection function is unloaded to the intelligent network card, so that the occupation of the link detection function on the CPU resource of the server is reduced, and the effect is more obvious in a virtualized environment.

The application aims to provide a link detection method and related software and hardware products, which are used for unloading a server link detection function onto an intelligent network card, and realizing the functions of autonomous link detection and isolation of a fault link of the intelligent network card. The basic idea of the application is: and transmitting the link detection function to the intelligent network card, wherein the intelligent network card acquires the link detection capability of the opposite-end forwarding equipment, enables the link detection function between the intelligent network card and the forwarding equipment, detects the link fault according to the transmitted link detection message and the link detection response message fed back by the forwarding equipment, and performs corresponding fault processing. The intelligent network card can also judge the fault generation position by comparing the link detection results of the local and forwarding equipment sides. According to the technical scheme, the link detection function can be unloaded through the intelligent network card, the CPU load is reduced, the position of fault generation can be more accurately positioned, and the network fault positioning efficiency is improved.

Based on the basic ideas of the present application, the following describes the implementation of the present application in conjunction with the accompanying drawings and specific embodiments.

Fig. 1 is a schematic diagram of a device structure for implementing a link detection method according to an embodiment of the present application. Forwarding devices in this application include, but are not limited to, physical or virtual switches or routers, and this embodiment is illustrated by way of example as a switch. The link detection method is applied to a server internally provided with an intelligent network card, a server processor, namely a server CPU (Central processing Unit) is connected with the intelligent network card through an internal bus, a server can send or configure a link detection function to the intelligent network card through a management interface, after the intelligent network card on the server acquires the link detection capability of a switch, a detection message is sent to detect the link state between the intelligent network card and the switch, and whether the link fails is judged through a response message of the received detection message. Multiple ports (e.g., port A1, port A2, etc.) on the intelligent network card may form an aggregate port to connect with the aggregate ports (e.g., port S1 and port S2) on the switch side, through which the detection message is sent.

Fig. 2 is a flowchart illustrating steps of a link detection method according to an embodiment of the present application. The method is applied to the intelligent network card on the server, and comprises the following steps:

step 201, the intelligent network card obtains the link detection capability of the switch;

because of the wide variety of switch manufacturers and models, the supported network monitoring protocols are also five-in-eight, and the intelligent network card needs to know which network monitoring protocol is used by the switch, so that the link detection capability of the switch is obtained. Network monitoring protocols include, but are not limited to, SNMP, NETCONF, IPv NetStream, etc.

The present application may obtain the link detection capability of the switch by extending the link discovery layer protocol (Link Layer Discovery Protocol, LLDP) or extending other protocol messages or using a custom protocol format, so long as it can know whether the switch supports link failure detection or what network monitoring protocol is used, and the present application is not specifically limited.

Step 202, the intelligent network card negotiates with the exchanger and configures a link detection function;

after obtaining the link detection capability of the switch, the intelligent network card can configure the link detection function of the switch through a configuration interface (such as a netcon interface, a restf API interface, or other interfaces) provided by the switch, for example, configure and enable the network monitoring function supported by both sides to enable the link detection function of both sides.

Step 203, the intelligent network card periodically sends a detection message to carry out link detection;

after both sides activate the link detection function, both sides send a link detection message (detection message for short) to each other to start the link detection of each other. For example, a detection message is sent to a counterpart, the detection message carries necessary information, and after receiving the link detection message, the counterpart sends a link detection response message (abbreviated as response message) to the counterpart according to the requirement of the link detection protocol.

And 204, judging the link fault and processing the fault by the intelligent network card according to the receiving and sending condition of the detection message.

After receiving the detection response message, the intelligent network card/switch can detect whether the link fails according to the sending condition of the detection message and the condition of receiving the detection message. For example, the link delay is calculated according to the time of sending the detection message and the time of receiving the detection response message, and the packet loss rate is determined according to the number of the sent detection messages and the number of the received detection response messages.

As for determining that a link fails based on what policy or index, the link failure may be flexibly defined according to specific service requirements, for example, the time delay exceeds or the packet loss rate exceeds a certain threshold, which is determined as the link failure, and the application is not limited specifically.

After the intelligent network card judges that the link fails, a corresponding failure processing step, such as isolating the failed link, can be executed to ensure that the user service is not affected. Under the condition of setting the port, after the intelligent network card isolates the fault link, the service message can still be forwarded through the port connected with the normal link in the aggregation port, and software on the server can not sense the isolation operation completely, so that the influence on the user service is smaller.

In another embodiment of the present invention, in order to more accurately determine the cause and location of the link failure, the present application further provides the following failure location determination and processing steps:

step 205, after detecting the link fault, the intelligent network card obtains the link detection result of the exchange side;

the link detection result at the switch side is obtained through a management interface (for example, netcon interface, restul API interface, etc.) provided outside the switch.

And 206, judging the fault generation position and carrying out corresponding fault processing according to the link detection results of the local network card and the exchange side.

After the intelligent network card obtains the link detection result of the exchanger side through the management interface, the position of the fault is judged by comparing the detection results of the local and the exchanger side. For example, when the packet loss rate and the delay detected by the switch are substantially consistent with those detected by the intelligent network card itself (e.g., the results are the same or the percentage of the deviation is within the allowable deviation range), it is determined that a fault is generated on the switch link. If the switch does not detect packet loss, the failure is on a link where the server and the switch are directly connected, for example, when an optical module of an intelligent network card on the server has a failure, the switch may not detect packet loss, and the intelligent network card may detect packet loss.

The intelligent network card can basically determine whether the fault is generated on the server or the switch by comparing the link detection results at the two ends, and after the judgment result is obtained, the intelligent network card can execute corresponding fault processing steps, for example, by calling an alarm interface of the server, notifying a user of the fault position, prompting the user to discharge the fault as soon as possible, and the like.

The following describes a process of implementing the link detection method provided in the present application in conjunction with a specific embodiment.

First aspect: acquiring a portion of link detection capabilities of a switching device

After the server and the switch provided with the intelligent network card are connected by using a network cable or an optical fiber, the information of the opposite terminal can be negotiated through the LLDP message. The LLDPDU is a data unit encapsulated in the data part of the LLDP message. Before composing the LLDPDU, the device encapsulates the local information into a type length value (Type Length Value, TLV) format, and then several TLVs are combined into a data portion of the LLDPDU encapsulated in the LLDP message for transmission.

TLVs are the elements that make up the LLDPDU, each TLV representing one piece of information. The TLVs that LLDP may encapsulate include a basic TLV, an 802.1 organization definition TLV, an 802.3 organization definition TLV, and an LLDP-MED (Link Layer Discovery Protocol Media Endpoint Discovery, link layer discovery protocol media terminal discovery) TLV. The basic TLVs are a set of TLVs based on network device management, and the 802.1 organization definition TLVs, the 802.3 organization definition TLVs, and the LLDP-MED TLVs are TLVs defined by a standard organization or other institutions, so as to enhance management of network devices, and whether the TLVs are sent in the LLDPDU can be selected according to actual needs.

Among the basic TLVs, several TLVs are necessary for implementing LLDP functions, for example, a Chassis ID TLV, a Port ID TLV, a Time To Live TLV are necessary To be carried by each LLDPDU, and the remaining TLVs are optional To be carried. Each LLDPDU can carry up to 32 TLVs.

In order to acquire the link detection capability, the LLDP protocol may be extended, and a TLV carrying the link detection capability is newly added, so as to carry information about whether the switching device supports network monitoring, for example, a value of 1 indicates that network monitoring is supported, and a value of 0 indicates that network monitoring is not supported. The type of network monitoring protocol capable of supporting link detection may be further carried, including but not limited to SNMP, NETCONF, IPv NetStream, etc.

The LLDP protocol can be extended, and a TLV carrying an aggregated port message Wen Haxi is newly added, so that the TLV carrying the aggregated port message is used for carrying a hash mode of the switch device on the aggregated port message, for example, a value of 0 indicates that MAC hash is used, and a value of 1 indicates that mac+ip hash is used. The hash mode of the switching device determines how to select the member ports of the aggregation port sent by the switching device to the server intelligent network card, for example, two ports are arranged in the aggregation port group, and the hash algorithm of the switching device determines which port a certain message is sent to the server from. Likewise, the port from which the server sends to the switch is a server decision. Therefore, if both parties do not know the hash method used by the other party, there may be a case that the receiving and transmitting paths are not one link, and the detection effect may be affected. For example, a detection message is sent from the port 1, and a detection response message is received from the port 2, so if the delay is large, it cannot be judged whether the link connected to the port 1 is problematic or whether the link connected to the port 2 is problematic. By knowing the hash mode of the other party, the link detection message sent from the port 1 can be ensured, and the corresponding link detection response message is received from the port 1.

Second aspect: intelligent network card detection link quality

The intelligent network card is provided with the CPU, can directly send the detection message, and can avoid influencing user service by receiving the link detection task of the CPU of the server through the intelligent network card.

The server first negotiates and configures the switch network monitoring function directly connected with the intelligent network card through a NETCONF interface, a RESTFUL API interface or other management interfaces provided by the switch.

The intelligent network card and the switch can periodically send detection messages, and if the intelligent network card is provided with two ports eth0 and eth1, the detection messages are sent from eth0 and eth1 in sequence.

The MAC address of the detection message needs to consider the hash mode of the switch, so as to ensure that the detection message sent from eth0 can receive the detection response message from eth 0. The hash algorithm of the switch determines how to select the ports of the aggregation port members that the switch sends to the server, e.g., there are two ports in the aggregation group, and the hash algorithm of the switch determines which port a certain message sends to the server from. Likewise, the port from which the server sends to the switch is a server decision.

When the detection response message is not received or the time delay of the detection response message is overlarge, the quality of the link is considered to be poor at the moment, and whether the problem is on the server or the switch can be further judged according to the receiving and transmitting condition of the link detection message.

When detecting a fault of a certain link, the intelligent network card can isolate the faulty link first, so as to ensure that the user service is not affected. If the link quality of eth0 is not good, then eth0 is set to inactive state, and all the following messages are no longer sent from eth 0. The isolation of the failed link in the intelligent network card has the advantage that software on the server can not sense the isolation operation at all, and the influence on the user service is smaller.

After detecting the link failure, the intelligent network card can obtain the link detection result data through a NETCONF interface, a RESTFUL API interface or other interfaces provided by the switch. The same switch may be connected with a plurality of intelligent network cards, so that when the link detection result data of the opposite end is acquired, the data of which port is acquired needs to be designated, and the data of other intelligent network cards are prevented from being acquired by mistake. The network monitoring data at least comprises packet loss rate and time delay of the detection message constructed by the intelligent network card.

When the packet loss rate and the delay detected by the switch are basically consistent with those detected by the intelligent network card (for example, the result is the same or the percentage of deviation is within the allowable deviation range), the fault can be located to occur on the switch link. If the switch does not detect packet loss, a failure may occur on the link where the server and switch are directly connected. After the fault is definitely on the server or the switch, the user can be informed of the fault position by calling an alarm interface of the server, and the user is prompted to discharge the fault as soon as possible.

Third aspect: link failure recovery

Previously failed links may return to normal, such as when a user replaces a fiber optic module, or messages on the link are no longer congested resulting in reduced latency. At this time, the port that is restored to normal needs to be reset to an active state, and reliability and link bandwidth of the aggregate port are increased.

After the intelligent network card finds that the link fault is relieved by continuously sending detection messages, for example, after determining that the link quality meets the requirement, the port which has failed before can be set to be in an active state so as to restore the link.

Fig. 3 is a schematic diagram of an intelligent network card structure for implementing a link detection method according to an embodiment of the present application, where the device 300 includes: a processor 310 such as a Central Processing Unit (CPU), a communication bus 320, a communication interface 340, and a memory 330. Wherein the processor 310 and the memory 330 may communicate with each other via a communication bus 320. The memory 330 has stored therein a computer program which, when executed by the processor 310, performs the functions of one or more steps of the link detection method provided herein.

Memory refers to a device for storing computer programs and/or data based on some storage medium, which may be a Volatile Memory (VM) or a Non-Volatile Memory (NVM). The memory is an internal memory for directly exchanging data with the processor, and can read and write data at any time, and has high speed, and is used as a storage medium for temporary data of an operating system and other running programs. The memory may be synchronous dynamic random access memory (Synchronous Dynamic Random Access Memory, SDRAM), dynamic random access memory (Dynamic Random Access Memory, DRAM), or the like. The nonvolatile memory is a memory using a persistent storage medium, and has a large capacity and can store data permanently, and may be a storage class memory (Storage Class Memory, SCM), a Solid State Disk (SSD), a NAND flash memory, a magnetic Disk, or the like. SCM is a common name for new storage medium between memory and flash memory, and is a composite storage technology combining persistent storage characteristic and memory characteristic, and has access speed slower than that of DRAM and SSD hard disk.

The processor may be a general-purpose processor including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processing, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

It should be appreciated that embodiments of the present application may be implemented or realized by computer hardware, a combination of hardware and software, or by computer instructions stored in non-transitory (or referred to as non-persistent) memory. The method may be implemented in a computer program using standard programming techniques, including a non-transitory storage medium configured with a computer program, where the storage medium so configured causes a computer to operate in a specific and predefined manner. Each program may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language. Furthermore, the program can be run on a programmed application specific integrated circuit for this purpose. Furthermore, the operations of the processes described herein may be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The processes (or variations and/or combinations thereof) described herein may be performed under control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications), by hardware, or combinations thereof, collectively executing on one or more processors. The computer program includes a plurality of instructions executable by one or more processors.

Further, the method may be implemented in any type of computing platform operatively connected to a suitable computing platform, including, but not limited to, a personal computer, mini-computer, mainframe, workstation, network or distributed computing environment, separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, and so forth. Aspects of the present application may be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, an optical read and/or write storage medium, RAM, ROM, etc., such that it is readable by a programmable computer, which when read by a computer, is operable to configure and operate the computer to perform the processes described herein. Further, the machine readable code, or portions thereof, may be transmitted over a wired or wireless network. When such media includes instructions or programs that, in conjunction with a microprocessor or other data processor, implement the steps described above, the invention described herein includes these and other different types of non-transitory computer-readable storage media. The present application also includes the computer itself when programmed according to the methods and techniques described herein.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. that fall within the spirit and principles of the present application are intended to be included within the scope of the present application.

Claims

1. The link detection method is characterized by being applied to an intelligent network card of a server, and further comprises the following steps:

acquiring the link detection capability of forwarding equipment;

periodically sending a detection message to perform link detection;

2. The method according to claim 1, wherein the method further comprises:

3. The method according to claim 1, wherein the method for obtaining the link detection capability of the forwarding device is:

4. The method of claim 1, wherein the obtaining the link detection capability of the forwarding device comprises:

5. The method of claim 2, wherein the step of determining the position of the substrate comprises,

the fault processing step in the link fault judgment and fault processing according to the receiving and transmitting conditions of the detection message is to isolate a fault link;

6. A link detection device, wherein the device is applied to an intelligent network card of a server, and the device comprises:

7. The apparatus of claim 6, wherein the apparatus further comprises:

8. The apparatus of claim 6, wherein the device comprises a plurality of sensors,

the capability acquisition module acquires the link detection capability of the forwarding device through an extended link discovery layer protocol LLDP message.

9. The apparatus of claim 6, wherein the device comprises a plurality of sensors,

the capability obtaining module obtains the link detection capability of the forwarding device, including: acquiring whether the forwarding device supports link detection and/or acquiring a link detection protocol supported by the forwarding device, and acquiring a hash mode of the forwarding device on a message forwarded through the aggregation port.

10. The intelligent network card is characterized by comprising a processor, a communication interface, a storage medium and a communication bus, wherein the processor, the communication interface and the storage medium are communicated with each other through the communication bus;

a storage medium storing a computer program;

a processor for carrying out the method steps of any one of claims 1-5 when executing a computer program stored on a storage medium.

11. A storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of claims 1 to 5.