WO2016091156A1

WO2016091156A1 - Node fault judgement method and device

Info

Publication number: WO2016091156A1
Application number: PCT/CN2015/096710
Authority: WO
Inventors: 王大勇; 陆伟; 付迎春
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-12-12
Filing date: 2015-12-08
Publication date: 2016-06-16
Also published as: CN105743687B; CN105743687A

Abstract

Disclosed are a node fault judgement method and device. The method comprises: processing a diagnosis packet according to a flow table possessed by a node under diagnosis, so as to obtain an expected packet; issuing the diagnosis packet to the node under diagnosis, so as to acquire an issued packet processed via the flow table of the node under diagnosis;and in the case where it is confirmed that the issued packet is inconsistent with the expected packet, judging that the node under diagnosis is a faulty node, alternatively, in the case where the issued packet is not received, judging that the node under diagnosis is a faulty node. By means of the present invention, the problem in the related art that a fault cannot be quickly and accurately positioned is solved, thereby achieving the effect of quickly and accurately positioning the fault.

Description

Method and device for judging node failure

Technical field

The present invention relates to the field of communications, and in particular, to a method and apparatus for determining a node failure.

Background technique

Software Defined Network (SDN) is a new type of data communication network technology. The SDN is characterized by separation of control and forwarding, separating the control part of the router and the switch in the original network, implemented by software, and the remaining forwarding part is implemented by the network device. The former can be called the control plane/control layer, and the latter can be called the forwarding plane or the forwarding layer.

The forwarding table implements the data packet forwarding based on the flow table. The flow table is composed of a number of flow entries with different priorities. The flow entries with high priority are matched first. As shown in Figure 1, a flow entry consists of a matching domain, a command domain, and a statistical domain. The forwarding plane device receives the inbound packet, performs matching operations on the flow table entry priority order, and switches the statistics field counter to the hit flow table entry, and performs the operation specified by the command domain. These actions may include performing datagrams on the datagram. Modify, discard, report to the control plane, forward from the specified port, and so on. Generally, the forwarding plane device supports multi-level flow tables. After all the flow tables are matched, the action set output message is executed.

Controlling the behavior of the device facing the forwarding plane is to modify the latter's flow table implementation with the OpenFlow protocol. The current OpenFlow protocol supports multi-level flow tables, group tables, supports IPv6, and supports capability negotiation.

The control plane obtains the switch port information, obtains the network connection diagram through the link detection protocol, and learns the media access control (MAC) address to obtain the connection relationship between the host and the network, thereby generating a topology and an end-to-end path, and According to this, the forwarding table is generated and sent to the forwarding plane device. For the unknown packet, the control plane can send the flow table to the control plane, generate a new path by querying the topology, and then send the forwarding table to the forwarding plane device on the corresponding path to validate the new path.

For packet exchange, the current Internet protocol-based message exchange network communication service is the communication two parties at both ends of the path, and a series of network devices in the path are completed according to the rules. When a message is sent from the originating host, it is sent to the first forwarding device in the path, which is processed and sent to the second forwarding device on the path, and thus repeated until the receiving host.

Each forwarding device and its outgoing port, and the next forwarding device on the path it points to, form a directional link. The complete path of network transmission is composed of such directional links in sequence.

The message exchange network described above is relatively traditional time-division exchange. The forwarding device in the network processes multiple packets at the same time to form multiple virtual packet paths/“circuits” and serves multiple paths of communication services.

When any forwarding device on the network generates a physical fault, such as a port fault or a logical fault, such as the path indication and other devices on the path are inconsistent, the packet cannot be sent to the receiver and the communication fails. The administrator will receive a fault report indicating what kind of service is causing the fault.

In view of the related art, it is impossible to quickly and accurately locate a fault, and an effective solution has not been proposed.

Summary of the invention

The embodiment of the invention provides a method and a device for determining a fault of a node, so as to solve at least the problem that the fault cannot be quickly and accurately located in the related art without special modification to the network forwarding device or normal service forwarding. .

According to an embodiment of the present invention, a method for judging a node fault is provided, which includes: processing a diagnostic message according to a flow table of the node to be diagnosed, to obtain an expected message; and sending the diagnostic message to the diagnostic message The diagnosed node obtains an outgoing message that has been processed by the flow table of the diagnosed node, and determines that the diagnosed node is faulty if the outgoing message is inconsistent with the expected message. The node, or in the case that the outgoing message is not received, determines that the diagnosed node is a faulty node.

In the embodiment of the present invention, the obtaining the outgoing packet that is processed by the flow table of the diagnosed node includes: sending a flow entry with the highest priority to the next node of the diagnosed node, where The flow table entry of the highest priority is used to match the outgoing message of the diagnosed node and is reported; and the outgoing message of the diagnosed node reported by the next node of the diagnosed node is obtained.

In the embodiment of the present invention, the sending the highest priority flow entry to the next node of the diagnosed node includes: if the diagnosed node is not a tail node, to the diagnosed node The next node sends the highest priority flow entry.

In the embodiment of the present invention, the obtaining the outgoing message after being processed by the flow table of the diagnosed node includes: sending the flow table to the diagnosed node if the diagnosed node is a tail node The item, wherein the delivered flow entry is used to match the outgoing message of the diagnosed node and is reported; and the outgoing message of the diagnosed node reported by the diagnosed node is obtained.

In the embodiment of the present invention, the processing of the diagnostic message according to the flow table of the diagnosed node to obtain the expected message includes: importing the flow table of the diagnosed node by using a built-in flow table processing module And processing the diagnostic packet according to the flow table of the diagnosed node to obtain the expected packet; or importing the flow of the diagnosed node by using a physical forwarding layer device or a virtual forwarding layer device And processing the diagnostic message according to the flow table of the diagnosed node to obtain the expected message.

According to another embodiment of the present invention, a device for determining a node failure is further provided, comprising: a flow table processing module, configured to process the diagnostic message according to a flow table of the node to be diagnosed, to obtain an expected message; An obtaining module, configured to send the diagnostic message to the diagnosed node, and obtain an outgoing message that is processed by the flow table of the diagnosed node; and the determining module is configured to confirm the outgoing message If the expected node is inconsistent with the expected message, it is determined that the diagnosed node is a faulty node, or if the outgoing message is not received, the diagnosed node is determined to be a faulty node.

In the embodiment of the present invention, the acquiring module includes: a first sending unit, configured to send a flow entry with the highest priority to the next node of the diagnosed node, where the highest priority flow The entry is used to match the outgoing message of the diagnosed node and is reported. The first obtaining unit is configured to obtain an outgoing message of the diagnosed node reported by the next node of the diagnosed node.

In the embodiment of the present invention, the first sending unit is further configured to send the highest priority stream to the next node of the diagnosed node if the diagnosed node is not a tail node. Entry.

In the embodiment of the present invention, the acquiring module further includes: a second sending unit, configured to send a flow entry to the diagnosed node if the diagnosed node is a tail node, where The sent flow entry is used to match the outgoing message of the diagnosed node and is reported; the second obtaining unit is configured to obtain the outgoing message of the diagnosed node reported by the diagnosed node.

In the embodiment of the present invention, the flow table processing module is built in the device, and the flow table of the diagnosed node is imported, and the diagnostic message is processed according to the flow table of the diagnosed node. Obtaining the expected packet; or the flow table processing module imports the flow table of the diagnosed node by using a physical forwarding layer device or a virtual forwarding layer device, and according to the flow table pair that the diagnosed node has The diagnostic message is processed to obtain the expected message.

According to the embodiment of the present invention, the diagnostic message is processed according to the flow table of the node to be diagnosed, and the expected message is obtained; the diagnosis message is sent to the node to be diagnosed, and the flow table of the node to be diagnosed is processed. The outgoing message; when it is confirmed that the message is inconsistent with the expected message, it is judged that the node being diagnosed is a faulty node, or if the outgoing message is not received, it is determined that the node to be diagnosed is a faulty node. The invention solves the problem that the fault cannot be quickly and accurately located in the related art, thereby realizing the effect of quickly and accurately positioning the fault.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the invention, and are intended to be a part of the invention. In the drawing:

FIG. 1 is a schematic diagram of a flow table entry structure and a flow table processing process of an Openflow switch;

2 is a flowchart of a method for determining a node failure according to an embodiment of the present invention;

3 is a structural block diagram of a device for determining a node failure according to an embodiment of the present invention;

4 is a structural block diagram 1 of a device for determining a node failure according to an embodiment of the present invention;

5 is a structural block diagram 2 of a device for determining a node failure according to an embodiment of the present invention;

6 is a flowchart of a scheduling unit selection call diagnostic unit according to an embodiment of the present invention;

7 is a flow chart of invoking a single node diagnostic unit in accordance with an embodiment of the present invention;

FIG. 8 is a flowchart of a processing procedure of a general diagnostic unit according to an embodiment of the present invention; FIG.

9 is a flow chart showing the processing of a single node diagnostic unit in accordance with an embodiment of the present invention.

detailed description

The invention will be described in detail below with reference to the drawings in conjunction with the embodiments. It should be noted that the embodiments in the present application and the features in the embodiments may be combined with each other without conflict.

In this embodiment, a method for determining a node fault is provided. FIG. 2 is a flowchart of a method for determining a node fault according to an embodiment of the present invention. As shown in FIG. 2, the process includes the following steps:

Step S202, processing the diagnostic message according to the flow table of the node to be diagnosed, to obtain an expected message;

In step S204, the diagnostic message is sent to the diagnosed node, and the outgoing message is processed after being processed by the flow table of the diagnosed node.

Step S206: When it is confirmed that the message is inconsistent with the expected message, it is determined that the diagnosed node is a faulty node, or if the outgoing message is not received, it is determined that the diagnosed node is a faulty node.

Through the above steps, the diagnostic message is set, and the expected message is generated according to the diagnostic message, and the diagnostic message is processed by the diagnosed node to generate an outgoing message, and the compared node is compared with the expected message to determine the node to be diagnosed. Whether it is a faulty node or not, in the process of determining whether a node is a faulty node, a special modification of the network forwarding device is required, and the normal service forwarding is affected. The above steps solve the problem in the related art. The problem of fast and accurate fault location is not implemented when the network forwarding device is modified or not, and the fault is quickly and accurately located.

In the case where the node being diagnosed is not a tail node, in an alternative embodiment, the highest priority flow entry is delivered to the next node of the diagnostic node. The flow table entry of the highest priority is used to match the outgoing message of the diagnosed node and is reported, so that the next node of the diagnosed node obtains the outgoing message of the diagnosed node reported by the next node of the diagnosed node.

In the case that the node to be diagnosed is a tail node, in an optional embodiment, the flow entry is sent to the node to be diagnosed, and the flow entry to be delivered is used to match the outgoing message of the node to be diagnosed and reported. Therefore, the outgoing message of the diagnosed node reported by the diagnosed node is obtained by the diagnosed node itself.

Step S202 involves processing the diagnostic message according to the flow table of the diagnosed node to obtain an expected message. In an optional embodiment, the flow table of the diagnosed node is imported through the built-in flow table processing module. And processing the diagnostic packet according to the flow table of the diagnosed node to obtain the expected packet. In another optional embodiment, the flow table of the diagnosed node is imported by using the physical forwarding layer device or the virtual forwarding layer device. And processing the diagnostic message according to the flow table of the diagnosed node to obtain the expected message.

In the embodiment, a device for determining the fault of the node is provided, and the device is used to implement the foregoing embodiment and the preferred embodiment, and details are not described herein. As used below, the term "module" may implement a combination of software and/or hardware of a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware, or The implementation of a combination of software and hardware is also possible and conceived.

3 is a structural block diagram of a device for determining a node failure according to an embodiment of the present invention. As shown in FIG. 3, the device includes: a flow table processing module 32 configured to perform a diagnostic message according to a flow table of the node being diagnosed. Processing, the expected message is obtained; the obtaining module 34 is configured to send the diagnostic message to the diagnosed node, and obtain the outgoing message after being processed by the flow table of the diagnosed node; the determining module 36 is configured to confirm the report If the text is inconsistent with the expected message, it is determined that the diagnosed node is a faulty node, or if the outgoing message is not received, it is determined that the diagnosed node is a faulty node.

4 is a structural block diagram of a node failure determining apparatus according to an embodiment of the present invention. As shown in FIG. 4, the obtaining module 34 includes: a first sending unit 342, configured to deliver the highest to the next node of the node to be diagnosed. a priority flow entry, where the highest priority flow entry is used to match the outgoing message of the diagnosed node and is reported; the first obtaining unit 344 is configured to acquire the diagnosed node reported by the next node of the diagnosed node. Outbound message.

Optionally, the first sending unit 342 is further configured to send the highest priority flow entry to the next node of the diagnosed node if the diagnosed node is not the tail node.

FIG. 5 is a structural block diagram 2 of a device for determining a fault of a node according to an embodiment of the present invention. As shown in FIG. 5, the acquiring module 34 further includes: a second sending unit 346, configured to be in a case where the node to be diagnosed is a tail node. And sending a flow entry to the diagnosed node, where the delivered flow entry is used to match the outgoing message of the diagnosed node and is reported; and the second obtaining unit 348 is configured to obtain the diagnosed report by the diagnosed node. Outbound packet of the node.

Optionally, the flow table processing module 32 is built in the node failure determining device, configured to import the flow table of the diagnosed node, and process the diagnostic message according to the flow table of the diagnosed node to obtain the The flow message processing module 32 imports the flow table of the diagnosed node through the physical forwarding layer device or the virtual forwarding layer device, and processes the diagnostic message according to the flow table of the diagnosed node to obtain the expected report. Text.

It should be noted that each of the above modules may be implemented by software or hardware. For the latter, the foregoing may be implemented by, but not limited to, the foregoing modules are all located in the same processor; or, the above modules are respectively located. In the first processor, the second processor, and the third processor.

The following is an example of the diagnosis of the faulty node of the SDN network end-to-end.

In order to solve the above technical problem, the present invention discloses an end-to-end fault diagnosis method for the SDN network. When the SDN network fails due to the forwarding device, the administrator does not need to forward the special function of the device. Specific business path or set of business paths, as well as network transport ports. For each path, simulate the actual packet sending process, and perform end-to-end accurate fault location on the network without affecting the normal service of the forwarding device. This alternative embodiment is applicable to a network transmission path having at least one forwarding node.

From the reported service information, a diagnostic message with a specific feature can be edited. The optional embodiment uses the diagnosis message as the first node, and the input diagnostic message of other nodes can use the diagnosis of the previous node on the path. Output the message, or use the output method of the previous node obtained by other methods.

This alternative embodiment can include two technical solutions, including:

plan 1:

Step 1: If the currently diagnosed node is not a tail node.

The second step is to process the diagnostic message by the flow table of the diagnosed node in the path, so as to obtain the expected outgoing message.

The third step is to deliver the highest priority flow entry to the next forwarding device of the currently diagnosed node in the path, and only match the expected outgoing diagnostic message, and send the packet.

Step 4: Issue diagnostic messages to the node being diagnosed.

Step 5: If the actual outgoing message with the specific feature is received, compare it with the expected message obtained in the second step. If not, the diagnosed node in the third step is a faulty node. The difference between the analysis and analysis can further lead to the cause of the failure. If the packet is not received within the specified time, the diagnosed node in the third step is a faulty node, and the node and its flow table may be further analyzed to obtain the fault cause.

Step 6: If you delete the flow table delivered in the third step.

Step 7: If the current diagnostic node is a tail node.

Step 8: Send a highest-priority flow entry to the currently diagnosed node, match only the diagnostic packet, copy the packet, and send it.

Step 9: The diagnostic message is processed by the flow table of the currently diagnosed node in the path, thereby obtaining the current expected outgoing message.

Step 10: Send a diagnostic message to the currently diagnosed node.

Step 11: If the received message is received, it is compared with the current expected outgoing message. If it is not met, the currently diagnosed node is a faulty node. Further comparison and analysis of the difference can result in the fault. If the packet is not received within the specified time, the currently diagnosed node is a faulty node, and the node and its flow table can be further analyzed to obtain the fault cause.

Step 12: Delete the flow table delivered in the eighth step.

Scenario 2:

The first step is to send a highest-priority flow entry to the currently diagnosed node, and only match the diagnostic packet, copy the packet, and send it.

The second step is to process the diagnostic message by the flow table of the currently diagnosed node in the path, so as to obtain the current expected outgoing message.

The third step is to send a diagnostic message to the currently diagnosed node.

Step 4: If the received message is received, it is compared with the current expected outgoing message. If it is not met, the currently diagnosed node is a faulty node. Further comparison and analysis of the difference can result in the fault. If the message is not received within the specified time, the current The node being diagnosed is a faulty node, and the node and its flow table can be further analyzed to obtain the cause of the fault.

Step 5: Delete the flow table delivered in the eighth step.

The following is a detailed explanation of the end-to-end diagnostic process based on the fault notification information.

For convenience of description, the method of the alternative embodiment is divided into the following subunits:

Ordinary diagnostic unit: If the node has subsequent nodes on the path, the diagnosis can be performed by the common diagnostic unit.

Single-node diagnostic unit: Diagnostics of nodes on a path, whether or not they have subsequent nodes.

The expected message generation unit: loads a specific flow table set, imports the data report, and obtains the outgoing message processed by the flow table.

FIG. 6 is a flowchart of a scheduling unit selection call diagnostic unit according to an embodiment of the present invention. As shown in FIG. 6, the flow includes the following steps:

Step S602, selecting one node from among the nodes on the path as the current node.

Step S602, if the current node does not have the next node on the path, then go to step S608, otherwise go to step S606.

In step S606, the normal diagnosis unit is called by the local node and the diagnostic message. This process ends.

Step S608, the single node diagnosis unit is called by the local node and the diagnostic message.

FIG. 7 is a flowchart of invoking a single node diagnostic unit according to an embodiment of the present invention. As shown in FIG. 7, the flow includes the following steps:

Step S702, selecting a node from the node on the path as the current node.

Step S704, the single node diagnosis unit is called by the local node and the diagnostic message.

FIG. 8 is a flowchart of a processing procedure of a general diagnostic unit according to an embodiment of the present invention. The processing process of the general diagnostic unit depends on the current node and the diagnostic message transmitted by the scheduling unit. As shown in FIG. 8, the process includes the following steps:

Step S802, the calling expected message generating unit generates an expected message.

In step S804, the flow table entry of the highest priority is delivered to the next node of the node, and the flow entry only matches the diagnostic message, and the diagnostic message is sent.

In step S806, the diagnostic message is sent to the current node.

In step S808, if the outgoing message sent by the next node is not received within the specified time, the process goes to step S812.

In step S810, the outgoing message is received, and the outgoing message is compared with the expected message. If not, the process goes to step S812. If yes, the process goes to step S814.

Step S812, the current node is a faulty node, and the expected message and the outgoing message (if any) are presented to the administrator. further analysis.

Step S814, deleting the flow entry sent in step S804.

This diagnostic unit processing ends.

During the processing of the single-node diagnostic unit, relying on the current node incoming by the scheduling unit and the support of the diagnostic message, FIG. 9 is a flowchart of the processing procedure of the single-node diagnostic unit according to the embodiment of the present invention, as shown in FIG. The process includes the following steps:

Step S902, the expected message generating unit is called to generate an expected message.

In step S904, the flow entry of the highest priority is delivered to the node to be diagnosed, and the flow entry only matches the diagnostic message, and the diagnostic message is copied and sent.

In step S906, the diagnostic message is sent to the current node.

In step S908, if the outgoing outgoing message is not received within the specified time, the process goes to step S912.

In step S910, the outgoing message is received, and the outgoing message is compared with the expected message. If not, the process goes to step S912. If yes, the process goes to step S914.

In step S912, the diagnosed node is a faulty node, and the expected message, and the outgoing message (if any) are presented to the administrator for further analysis.

Step S914, deleting the flow entry sent in step S904.

This diagnostic unit processing ends.

The expected message generation unit relies on the current node and the diagnostic message passed by the normal diagnosis unit or the tail node diagnosis unit.

In an optional embodiment, a flow table processing module is built in, and the flow table set of the node saved by the control plane is imported according to the specified node, and the diagnostic message is transmitted, and processed by the flow table processing module, thereby obtaining The processed expected outgoing message. In another optional embodiment, the device may be a separate forwarding layer device, which may be a physical forwarding layer device or a virtual forwarding layer device (software system). Before importing the flow table of the specified node, First clear the original flow table, then import the flow table set of the node saved by the control plane, and send the diagnostic message to obtain the expected outgoing message.

In summary, the method for implementing the fault diagnosis of the SDN network end-to-end disclosed by the present embodiment can be implemented in the SDN network because the forwarding device fails, and the administrator does not need to forward the special function of the device. A collection of business paths or business paths, as well as network transport ports. For each path, simulate the actual packet sending process, and perform end-to-end accurate fault location on the network without affecting the normal service of the forwarding device.

In another embodiment, software is also provided for performing the technical solutions described in the above embodiments and preferred embodiments.

In another embodiment, a storage medium is further provided, where the software is stored in the storage medium, and the storage Media includes, but is not limited to, optical disks, floppy disks, hard disks, rewritable memories, and the like.

It will be apparent to those skilled in the art that the various modules or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein. The steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software.

The above description is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and scope of the present invention are intended to be included within the scope of the present invention.

Industrial applicability

Claims

A method for judging a node failure includes:

The diagnosis message is processed according to the flow table of the node to be diagnosed, and the expected message is obtained;

Sending the diagnostic message to the diagnosed node, and obtaining an outgoing message after being processed by the flow table of the diagnosed node;

When it is confirmed that the outgoing message is inconsistent with the expected message, it is determined that the diagnosed node is a faulty node, or if the outgoing message is not received, determining that the diagnosed node is Faulty node.
The method according to claim 1, wherein the obtaining the outgoing message after being processed by the flow table of the diagnosed node comprises:

And sending, to the next node of the diagnosed node, a flow entry of the highest priority, where the highest priority flow entry is used to match the outgoing packet of the diagnosed node and report the packet;

Obtaining an outgoing message of the diagnosed node reported by the next node of the diagnosed node.
The method of claim 2, wherein the delivering the highest priority flow entry to the next node of the diagnosed node comprises:

In the case that the diagnosed node is not a tail node, the highest priority flow entry is delivered to the next node of the diagnosed node.
The method according to claim 1, wherein the obtaining the outgoing message after being processed by the flow table of the diagnosed node comprises:

In the case that the diagnosed node is a tail node, the flow entry to the diagnosed node is sent, and the sent flow entry is used to match the outgoing message of the diagnosed node and report the report ;

Obtaining an outgoing message of the diagnosed node reported by the diagnosed node.
The method according to any one of claims 1 to 4, wherein processing the diagnostic message according to a flow table of the node to be diagnosed to obtain the expected message comprises:

Importing, by the built-in flow table processing module, the flow table of the diagnosed node, and processing the diagnostic message according to the flow table of the diagnosed node to obtain the expected message; or

The flow table of the diagnosed node is imported by the physical forwarding layer device or the virtual forwarding layer device, and the diagnostic message is processed according to the flow table of the diagnosed node to obtain the expected packet.
A device for judging a node failure includes:

The flow table processing module is configured to process the diagnostic message according to the flow table of the diagnosed node to obtain an expected message;

Obtaining a module, configured to send the diagnostic message to the diagnosed node to obtain the diagnosed node The outgoing flow message is processed by the flow table;

The determining module is configured to: when it is confirmed that the outgoing message is inconsistent with the expected message, determine that the diagnosed node is a faulty node, or, if the outgoing message is not received, determine the location The diagnosed node is a faulty node.
The apparatus of claim 6, wherein the obtaining module comprises:

a first sending unit, configured to send a flow entry of the highest priority to the next node of the diagnosed node, where the highest priority flow entry is used to match the outgoing report of the diagnosed node And report it;

The first obtaining unit is configured to acquire an outgoing message of the diagnosed node reported by the next node of the diagnosed node.
The apparatus according to claim 7, wherein the first sending unit is further configured to issue the highest to the next node of the diagnosed node if the diagnosed node is not a tail node Priority flow entry.
The device of claim 6, wherein the obtaining module further comprises:

a second sending unit, configured to send a flow entry to the diagnosed node if the diagnosed node is a tail node, where the delivered flow entry is used to match the Diagnose the outgoing message of the node and report it;

The second obtaining unit is configured to acquire an outgoing message of the diagnosed node reported by the diagnosed node.
The apparatus according to any one of claims 6 to 9, wherein the flow table processing module is built in a node failure determining means, and is configured to import a flow table of the diagnosed node according to the predetermined node And processing the diagnostic packet according to the flow table of the diagnosed node to obtain the expected packet; or the flow label processing module importing the diagnosed by using a physical forwarding layer device or a virtual forwarding layer device The flow table of the node, and processing the diagnostic message according to the flow table of the diagnosed node to obtain the expected message.