CN107566098B - Method and system for generating software-defined network application test sequences - Google Patents

Abstract

The invention discloses a method and system for generating a software-defined network application test sequence. The generation method includes constructing an information table extended finite state machine model for the software-defined network application to be tested, and the information table extended finite state machine model includes a Multiple component state machines for the behavior of software-defined network applications; based on multiple component state machines, form a combined finite state machine for only one specified switch in the network topology; test generation methods using finite state machine models, based on combined The finite state machine generates a single-node test sequence for a specified switch; the single-node test sequence is simulated in the network topology to expand the single-node test sequence into a multi-node test sequence. The method eliminates the defect in the prior art that when testing a software-defined network application, only a single network entity can be tested, and it is separated from the network topology of the network where it is located, which is not conducive to testing.

Description

Method and system for generating software-defined network application test sequences

technical field

The invention belongs to the technical field of network protocol testing, and in particular relates to a method and system for generating a software-defined network application test sequence.

Background technique

Protocol testing technology is an important means to ensure the correct implementation of network communication protocols and the correct interconnection between different network devices. Conformance testing is a basic protocol testing method, which is mainly used to detect whether the implementation of the protocol is consistent with the specification of the protocol.

Formal method-based test case generation is an important issue in the field of protocol testing technology. The test case generation method starts from the formalized model of the protocol specification to generate a test set (or test sequence) for testing activities. In most of the proposed test generation techniques, the basic idea is to model the system under test (SUT) as a finite state machine or a system extending the finite state machine, and then generate test sequences.

A software-defined network application is an application protocol closely related to the network topology in which it resides. Generally, a software-defined network application has a global view of the network where it is located, and modifies the configuration of the network during its working process, thereby dynamically affecting the functions and behaviors of the network. Since the generation of test sequences based on formal methods in the prior art only considers a single network entity as the system under test, and does not involve the association of multiple network entities in the network, it is difficult to effectively define Web applications are tested.

Contents of the invention

One of the technical problems to be solved by the present invention is to provide a method that can effectively test the software-defined network application by considering the network topology relationship where the software-defined network application is located.

In order to solve the above technical problems, the embodiments of the present application firstly provide a method for generating a test sequence of a software-defined network application, including: constructing an information table extension finite state machine model for the software-defined network application to be tested, the information table extension is limited The state machine model includes a plurality of component state machines for describing the behavior of the software-defined networking application, channels describing connections between the component state machines, and describing the network topology of the network for which the software-defined networking application is directed;

forming a combined finite state machine for only one specified switch in the network topology based on the plurality of component state machines;

A test generation method using a finite state machine model generates a single-node test sequence for the specified switch based on the combined finite state machine, and the single-node test sequence includes the specified switch for the software to be tested Define the input and output messages of the network application;

The single-node test sequence is simulated in the network topology to expand the single-node test sequence into a multi-node test sequence.

Preferably, the plurality of component state machines include a plurality of extended finite state machines for describing a plurality of parallel processes that correspond one-to-one to a plurality of information entries stored in the software-defined network application, and the extended finite state machines The machine includes initial state, state set, variable set, input symbol set, output symbol set and transition set.

Preferably, the description of the network topology of the network targeted by the software-defined network application includes an array for describing each network entity in the network, and each element in the array corresponds to a host in the network A collection of switches, a collection of controllers, and a collection of links connecting hosts, switches, and controllers.

Preferably, based on the plurality of component state machines, forming a combined finite state machine for only one specified switch in the network topology includes:

specifying a switch in said network topology;

Select at least two hosts that communicate with each other from the network topology, and find at least two component states corresponding to the hosts that communicate with each other and for the specified switch in the plurality of component state machines machine;

The at least two component state machines are combined to form a combined finite state machine, each state of the combined finite state machine being a combination of a set of states and a set of variables of the at least two component state machines.

Preferably, the simulated execution of the single-node test sequence in the network topology to expand the single-node test sequence into a multi-node test sequence includes:

Taking out each input message in the single-node test sequence in turn, and generating an extended sequence for the taken-out input message;

Spreading sequences corresponding to each input message in the single-node test sequence are concatenated to form a multi-node test sequence.

Preferably, said generating an extension sequence for the fetched input message includes:

Step 1. Initialize storage space for the extended sequence to be generated;

Step 2, extracting the data packet and port information of the host from the input message, and adding the data packet and port information to the extended sequence;

Step 3, according to the order of storage, sequentially obtain data packets or control packets and corresponding port information from the extended sequence;

Step 4, based on the type of the port where the data message or control message will arrive and its own type, determine the processing of the data message or control message to obtain a new data message or control message and Corresponding port information, and adding the obtained new data message or control message and corresponding port information into the extended sequence;

Step 5. Repeat step 3 and step 4 until the processing of the data packets or control packets stored in the spreading sequence is completed, and all the content finally stored in the spreading sequence is used as a spreading sequence.

Preferably, said step 4 specifically includes:

If the data packet or the control packet will arrive at the port of the switch, then judge the type of the data packet or the control packet:

If it is a flood instruction type control message, extracting a data message from the control message, and adding the data message and each data port of the switch to the extended sequence;

If it is a control message of the installation rule type, record the installation rule, and extract the data message from the control message at the same time, and add the data message and the data message specified by the installation rule to the extended sequence The data port of the switch;

If it is a data packet matching the recorded rule, adding the data packet and the data port of the switch specified by the recorded rule to the extended sequence;

If it is a data message that does not match the recorded rule, construct a control message of the reporting type based on the data message, and add the constructed control message and the control port of the switch to the extended sequence;

If the data message or control message will arrive at the port of the controller, the component state machine in the finite state machine model is expanded according to the information table to find the corresponding control message, and the control message is added to the extended sequence Messages and corresponding controller ports.

Preferably, the step 4 further includes: if the data message or the control message will arrive at the port of the host, then do not process the data message or the control message, and directly follow the stored sequence from the extended sequence Obtain the next data packet or control packet and the corresponding port information.

Preferably, in the step 3, a first-in-first-out queue is used to synchronously back up the extended sequence, and data packets or control packets and corresponding port information are sequentially obtained from the first-in-first-out queue.

The embodiment of the present application also provides a generation system of a software-defined network application test sequence, including a model generation module, which is configured to build an information table extension finite state machine model for the software-defined network application to be tested, and the information table extension is limited The state machine model includes a plurality of component state machines for describing the behavior of the software-defined networking application, channels describing connections between the component state machines, and describing the network topology of the network for which the software-defined networking application is directed;

A combined module, configured to form a combined finite state machine for only one specified switch in the network topology based on the plurality of component state machines;

A single-node test sequence generation module, which is configured to adopt a test generation method of a finite state machine model, generates a single-node test sequence for the specified switch based on the combined finite state machine, and the single-node test sequence Including the input message and output message of the designated switch to the software-defined network application to be tested;

A multi-node test sequence generation module is configured to simulate and execute the single-node test sequence in the network topology, so as to expand the single-node test sequence into a multi-node test sequence.

Compared with the prior art, one or more embodiments in the above solutions may have the following advantages or beneficial effects:

Based on the single-node test sequence generated for software-defined network applications and the network topology of the network where the application is located, the single-node test sequence is simulated and executed to obtain a multi-node test sequence for software-defined network applications, which eliminates the existing In the existing technology, when testing software-defined network applications, only a single network entity can be tested, and it is separated from the network topology of the network where it is located, which is not conducive to testing.

Other advantages, objects, and features of the present invention will be set forth in the ensuing description to some extent, and to some extent, will be obvious to those skilled in the art based on the investigation and research below, or can be Learn from the practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Description of drawings

The accompanying drawings are used to provide a further understanding of the technical solutions of the present application or the prior art, and constitute a part of the description. Wherein, the drawings expressing the embodiments of the present application are used together with the embodiments of the present application to explain the technical solutions of the present application, but do not constitute limitations on the technical solutions of the present application.

FIG. 1 is a schematic flow diagram of a method for generating a software-defined network application test sequence according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a network where a software-defined network application is located according to an embodiment of the present invention;

3 is a schematic structural diagram of an information table extended finite state machine model according to an embodiment of the present invention;

FIG. 4 is a schematic flow diagram of simulating and executing an input message in a single-node test sequence according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a system for generating a software-defined network application test sequence according to another embodiment of the present invention.

Detailed ways

The implementation of the present invention will be described in detail below in conjunction with the accompanying drawings and examples, so as to fully understand and implement the implementation process of how to apply technical means to solve technical problems and achieve corresponding technical effects in the present invention. The embodiments of the present application and the various features in the embodiments can be combined with each other under the premise of no conflict, and the formed technical solutions are all within the protection scope of the present invention.

FIG. 1 is a schematic flow diagram of a method for generating a software-defined network application test sequence according to an embodiment of the present invention. As shown in the figure, the method for generating includes the following steps:

Step S110, constructing an information table extended finite state machine model for the SDN application to be tested.

Step S120, based on the obtained information table, expand multiple component state machines in the finite state machine model to form a combined finite state machine for only one specified switch in the network topology where the application is located.

Step S130, using the test generation method of the finite state machine model to generate a single-node test sequence for the specified switch based on the combined finite state machine.

Step S140, simulating and executing a single-node test sequence in the network topology where the application is located, so as to expand the single-node test sequence into a multi-node test sequence.

In the prior art, an extended finite state machine (Extended Finite State Machine, EFSM for short) is widely used to describe network protocols. However, a software-defined network application is closely related to the topology of the network where it is located. It has a global view of the network, and it will modify the configuration of the network during its work, which will affect the functions and behaviors of the network. Therefore, in the embodiments of the present invention, it is proposed An information-table extended finite state machine model that combines extended finite state machines and network topologies to describe software-defined network applications.

The information table extended finite state machine model includes a plurality of component state machines for describing the behavior of the software defined network application, channels describing the connection between the various component state machines, and describing the network topology of the network targeted by the software defined network application.

The information table extends the finite state machine model with IT. IT is a triplet, IT=(M, Ch, To). Wherein, M represents a set composed of n component state machines, M={M ₀ , M ₁ ,M ₂ ,...,M _n }, which includes an information used to describe the input message of the software-defined network application A state machine for extraction and distribution (denoted as M ₀ ), and multiple extended finite state machines for describing multiple parallel processes that correspond one-to-one to information entries stored in the software-defined network application. The information entry in the software-defined network application receives the information sent by its internal distributor, and outputs the message forwarding message and the installation rule message to the port connected to the switch. Ch represents a finite set of channels used to connect the state machines of each component. To represents the network topology of the network where M resides.

Each extended finite state machine M _i , where i=1,...,n, corresponds to a plurality of information entries stored in the software-defined network application. M _i is a six-element array, M _i ={S,s ₀ ,V,I,O,T}.

In the six-tuple M _i , S is the finite state set of M _i ;

s ₀ ∈ S, is the initial state of _Mi ;

V=(V ₁ ,V ₂ ,...V _k ,...,V _n ), where V _k is the variable of M _i , k=1,2,...,n, n is a positive integer, and V is the variable of all the variables of M _i gather;

I=(I ₁ ,I ₂ ,...I _k ,...,In ), where I _k is an input symbol of _Mi , k=1,2,..., _n , _n is a positive integer, and I is all A set of input symbols, I is a non-empty set;

O=(O ₁ ,O ₂ ,...O _k ,...,On ), where O _k is an output symbol of _Mi , k=1,2,..., _n , _n is a positive integer, O is all A set of output symbols, O is a non-empty set;

T=(T ₁ ,T ₂ ,...T _k ,...,T _n ) is the set of all transitions of _Mi , T is a non-empty set, where T _k is a transition of _Mi , k=1,2,..., n, n is a positive integer, T _k is a six-tuple, T _k = (s _start , s _end , I _k , O _k , P, A);

In the six-tuple of T _k , s _{start ∈} S is the initial state before the execution of T _k ; send ∈ S is the final state after the execution of T _k ; I _k ∈ I is the input required for the execution of T _k Parameters; O _k ∈ O, is the output result produced by the execution of T _k ; P is the predicate expression about the internal variables in V _I , the external variables in V _E , the input I _k and some constants, P explains T _k Execution conditions; A is the behavior sequence when T _k is executed, A includes one or more of the variable assignment behavior and output behavior, and multiple behaviors in A are executed in sequence.

Each channel in the finite set Ch of channels connects a component state machine with the external environment or other state machines.

To is a quaternion array, each element of which corresponds to each network entity in the network, To=(H, SW, C, LK).

In the quadruple To, H is a collection of hosts, each host has a unique address, is connected to the switch, and sends or receives data packets.

SW is a collection of switches, and each switch has several ports, which are respectively connected to hosts, controllers and other switches. The switch records the installation rules issued by the controller, forwards the data packets according to the rules, and sends the data packets to the controller if it receives a data packet that cannot be matched by the rules.

C is a collection of controllers, the controllers are connected to the switch, and control packets are exchanged between the two. The controller handles the interaction of control packets with the switch, including sending installation rules to the switch and receiving data packets that the switch cannot match the rules.

LK is a collection of links, and each link can be used to connect two topological nodes of the same or different types, such as hosts, switches or controllers.

Figure 2 shows a schematic diagram of the structure of the network where the software-defined network application is located. This topology is an exemplary topology structure. The actual topology may include various combinations of software-defined network applications, controllers, switches, hosts, and links connecting them. one. As shown in the figure, 21 in the figure represents a host, 22 represents a switch, and each host 21 is correspondingly connected to a switch 22. 23 represents a controller on which a software-defined network application runs. In practice, a software-defined network application may work on one or more controllers, and only one controller is used as an example for illustration in FIG. 2 , but this does not constitute a limitation of the present invention.

In FIG. 2 , the solid lines between the host 21 and the switches 22 and between the switches 22 , and the dotted lines between the controller 23 and the switches 22 are all used to represent links. Wherein, the links between the controller 23 and each switch 22 are used to transmit control messages, mainly for the interaction of control messages. The links between the host 21 and the switch 22 and between the switches 22 are used to transmit data packets, mainly for the interaction of data messages.

Fig. 3 shows a schematic structural diagram of an information table extended finite state machine model according to an embodiment of the present invention. As shown in the figure, the distributor is modeled as a state machine M ₀ , which is mainly used to process messages sent by the switch to the controller Turn in the message. Each information entry is modeled as an extended finite state machine M _i , i=1,...,n. It can also be seen from the figure that each information table item actually corresponds to a different label. The meaning of the label is that the first number indicates the number of the switch in the network to which it belongs, and the second number indicates the number of the switch in the network to which it belongs. The number of the host. When each information entry is modeled as an extended finite state machine, the above label information will also correspond to different extended finite state machines.

The distributor first receives the report message of the switch through an input port (logical port), extracts information from it, and distributes it to one or more information table items. The solid line pointing to each information entry from the distributor indicates the channel, and the arrow at the end of the solid line indicates the transmission direction of the information data. The solid line from the information table item to the output port (logical port) also represents the channel, that is, the channel connecting the component state machine with the external environment, and the arrow at the end of the solid line represents the transmission of information data such as message forwarding and installation rules direction. All of the above two types of channels constitute a finite set Ch of channels.

The extended finite state machine model of the information table in the embodiment of the present invention includes the topology of the network where the software-defined network application is located, which can more accurately describe the software-defined network application to be tested, and facilitates the testing.

Next, in step S120, first a switch is arbitrarily designated in the network topology, and at least two hosts communicating with each other are selected from the network topology. Then, in the extended finite state machine model of the information table, at least two component state machines respectively corresponding to the selected hosts communicating with each other and aimed at the designated switch are found. The found at least two component state machines are then combined to form a combined finite state machine. Specifically, it can be searched through the label corresponding to the extended finite state machine.

For example, if switch 1 is specified from the network topology shown in Figure 2, and host 1 and host 4 that communicate with each other are selected from it, then two component state machines are found from the extended finite state machine model of the information table, and The component state machine corresponding to the switch (1,1) information table item and the component state machine corresponding to the switch (1,4) information table item, and then combine the two found component state machines to obtain the combined finite state machine . Each state of the combined finite state machine is a combination of the state sets and variable sets of the above two component state machines.

It should be noted that when selecting hosts that communicate with each other, in order to alleviate the problem of state space explosion, it is preferable to select two hosts that communicate with each other.

The generated combined finite state machine is a model for only one designated switch. In step S130, a test generation method for the finite state machine model is used to generate a single-node test sequence covering the combined finite state machine. A single-node test sequence consists of incoming and outgoing messages from the same switch for the SDN application under test. Wherein, the finite state machine test generation method used may be any existing or future method capable of generating a test sequence based on a finite state machine model, which is not limited in this embodiment of the present invention.

In step S140 , extending the single-node test sequence to obtain the multi-node test sequence further includes sequentially extracting each input message in the single-node test sequence, and generating extended sequences for the extracted input messages respectively. The spreading sequences corresponding to each input message in the single-node test sequence are concatenated to form a multi-node test sequence. The process of simulating and executing an input message in a single-node test sequence is shown in FIG. 4 , and will be described in detail below in conjunction with FIG. 4 .

First, initialize an extended input sequence storage space for an input message in the single-node test sequence, and extract the data packet and port information of the host from the input message, and add the data packet and port information to the extended sequence , as shown in step S410 in FIG. 4 . Then, according to the order of storage, the data message or control message and the corresponding port information are sequentially obtained from the extended sequence, and the extended sequence is updated by judging the data message or control message and the corresponding port information .

Specifically, the extended sequence can be synchronously backed up by using the first-in-first-out queue, and the extended sequence can be simulated by sequentially obtaining data packets or control packets and corresponding port information by using its first-in-first-out feature.

Let the first-in-first-out queue be Q, and put the elements in the extended sequence into the first-in-first-out queue Q. Next in step S420, it is judged whether the queue Q is empty, and if the queue Q is empty, the simulation execution process ends. If queue Q is not empty, then take out a data message or control message from Q, and the type of the port that data message or control message will arrive and its own type are judged, as shown in step S430 among Fig. 4 Show. By judging the type of the port where the data message or the control message will arrive and its own type, the specific processing of the data message or the control message is determined, and the corresponding processing result (usually a new data message) or control packets and corresponding port information) are added to the extended sequence.

First judge the type of the port where the data packet or control packet will arrive, and then judge the type of the data packet or control packet when the data packet or control packet will arrive at the port of the switch:

When the message is a control message of the flood command type, extract the data message from the control message, and add the extracted data message and each data port of the switch to the extended sequence, as shown in the steps in Figure 4 S440 shown. In addition, for items newly added to the extended sequence, a synchronous backup needs to be performed in the queue Q.

When the message is a control message of the installation rule type, record the installation rule, extract the data message from the control message at the same time, and add the extracted data message and the data of the switch specified by the installation rule to the extended sequence port, as shown in step S450 in FIG. 4 . Synchronous backups are performed in queue Q for items newly added to the extended sequence.

When the message is a data message matching the recorded rule, add the data message and the data port of the switch specified by the recorded rule to the extended sequence, as shown in step S460 in FIG. 4 . Synchronous backups are performed in queue Q for items newly added to the extended sequence.

When the message is a data message that does not match the recorded rules, a control message of the reported type is constructed based on the data message, and the constructed control message and the control port of the switch are added to the extended sequence, as shown in the figure 4 in step S470. Synchronous backups are performed in queue Q for items newly added to the extended sequence.

When the port where the data message or control message will arrive is not the port of the switch but the port of the controller, the component state machine in the extended finite state machine model of the information table is used to find the corresponding control message, and in the extended sequence Add the control message and the corresponding controller port, as shown in step S480 in FIG. 4 . Synchronous backups are performed in queue Q for items newly added to the extended sequence.

After executing one of the above steps, return to judge whether the queue Q is empty, if not, repeat the above simulation process until the queue Q is empty.

In addition, when the port where the data packet or control packet will arrive is not the port of the switch or the port of the controller, but directly arrives at the port of the host, there is no need to process the data packet or control packet, and the The order of storage is to obtain the next data packet or control packet and the corresponding port information from the queue Q, as shown in step S490 in FIG. 4 .

After the processing of all data packets or control packets stored in the extended sequence is completed, all the content finally stored in the extended sequence is used as an extended sequence corresponding to an input message.

In the embodiment of the present invention, the extended finite state machine model accurately describes the topology of the software-defined network application and its target network by introducing an information table, and performs extended multi-node testing by using topology simulation based on a single-node test sequence The sequence realizes the test coverage of the network topology nodes.

The method for generating a test sequence for a software-defined network application in the embodiment of the present invention can generate a test sequence covering multiple network nodes, and at the same time, can alleviate the state space explosion problem.

In another embodiment of the present invention, a system for generating a software-defined network application test sequence is provided, as shown in FIG. For performing the action of step S120; the single-node test sequence generation module 53, which is used for performing the action of step S130; the multi-node test sequence generation module 54, for performing the action of step S140. The functions of the relevant modules can be obtained according to the foregoing embodiments, and will not be repeated here.

Although the embodiments disclosed in the present invention are as above, the described content is only an embodiment adopted for the convenience of understanding the present invention, and is not intended to limit the present invention. Anyone skilled in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed by the present invention, but the patent protection scope of the present invention, The scope defined by the appended claims must still prevail.

Claims (8)

1. A method for generating a software-defined network application test sequence, comprising: An information table extended finite state machine model is constructed for the software-defined network application to be tested, and the information table extended finite state machine model includes a plurality of component state machines for describing the behavior of the software-defined network application, and a state machine describing each component state machine and a network topology describing the network for which the software-defined network application is targeted; forming a combined finite state machine for only one specified switch in the network topology based on the plurality of component state machines; A test generation method using a finite state machine model generates a single-node test sequence for the specified switch based on the combined finite state machine, and the single-node test sequence includes the specified switch for the software to be tested Define the input and output messages of the network application; Simulating and executing the single-node test sequence in the network topology to expand the single-node test sequence into a multi-node test sequence, including: Taking out each input message in the single-node test sequence in turn, and generating an extended sequence for the taken-out input message; concatenating the spreading sequences corresponding to each input message in the single-node test sequence to form a multi-node test sequence; Wherein, the generation of the extended sequence for the fetched input message includes: Step 1. Initialize storage space for the extended sequence to be generated; Step 2, extracting the data packet and port information of the host from the input message, and adding the data packet and port information to the extended sequence; Step 3, according to the order of storage, sequentially obtain data packets or control packets and corresponding port information from the extended sequence; Step 4, based on the type of the port where the data message or control message will arrive and its own type, determine the processing of the data message or control message to obtain a new data message or control message and Corresponding port information, and adding the obtained new data message or control message and corresponding port information into the extended sequence; Step 5. Repeat step 3 and step 4 until the processing of the data packets or control packets stored in the spreading sequence is completed, and all the content finally stored in the spreading sequence is used as a spreading sequence. 2. The generation method according to claim 1, wherein the plurality of component state machines include a plurality of parallel processes used to describe a one-to-one correspondence with a plurality of information entries stored in the software-defined network application A plurality of extended finite state machines, the extended finite state machine includes an initial state, a state set, a variable set, an input symbol set, an output symbol set, and a transition set. 3. The generating method according to claim 1 or 2, wherein the description of the network topology of the network targeted by the software-defined network application includes an array for describing each network entity in the network, the Each element in the array respectively corresponds to a set of hosts, a set of switches, a set of controllers, and a set of links used to connect the hosts, switches, and controllers in the network. 4. The generating method according to claim 3, wherein, based on the plurality of component state machines, forming a combined finite state machine only for a specified switch in the network topology comprises: specifying a switch in said network topology; Select at least two hosts that communicate with each other from the network topology, and find at least two component states corresponding to the hosts that communicate with each other and for the specified switch in the plurality of component state machines machine; The at least two component state machines are combined to form a combined finite state machine, each state of the combined finite state machine being a combination of a set of states and a set of variables of the at least two component state machines. 5. generation method according to claim 1, is characterized in that, described step 4 specifically comprises: If the data packet or the control packet will arrive at the port of the switch, then judge the type of the data packet or the control packet: If it is a flood instruction type control message, extracting a data message from the control message, and adding the data message and each data port of the switch to the extended sequence; If it is a control message of the installation rule type, record the installation rule, and extract the data message from the control message at the same time, and add the data message and the data message specified by the installation rule to the extended sequence The data port of the switch; If it is a data packet matching the recorded rule, adding the data packet and the data port of the switch specified by the recorded rule to the extended sequence; If it is a data message that does not match the recorded rule, construct a control message of the reporting type based on the data message, and add the constructed control message and the control port of the switch to the extended sequence; if The data message or the control message will arrive at the port of the controller, then find the corresponding control message according to the component state machine in the extended finite state machine model of the information table, and add the control message to the extended sequence file and the corresponding controller port. 6. The generating method according to claim 5, characterized in that, said step 4 further comprises: if said data message or control message will arrive at the port of the host computer, then the data message or control message will not be processed. Processing, directly according to the order of storage, obtain the next data message or control message and corresponding port information from the extended sequence. 7. The generating method according to claim 1, characterized in that, in said step 3, said extended sequence is synchronously backed up by using a first-in-first-out queue, and datagrams are sequentially obtained from said first-in-first-out queue text or control packets and the corresponding port information. 8. A system for generating software-defined network application test sequences, comprising: A model generation module, which is configured to construct an information table extended finite state machine model for the software-defined network application to be tested, and the information table extended finite state machine model includes a plurality of component state machines for describing the behavior of the software-defined network application , describing the channel of the connection between the state machines of each component and describing the network topology of the network targeted by the software-defined network application; A combined module, configured to form a combined finite state machine for only one specified switch in the network topology based on the plurality of component state machines; A single-node test sequence generation module, which is configured to adopt a test generation method of a finite state machine model, generates a single-node test sequence for the specified switch based on the combined finite state machine, and the single-node test sequence Including the input message and output message of the designated switch to the software-defined network application to be tested; A multi-node test sequence generation module, which is configured to simulate and execute the single-node test sequence in the network topology, to expand the single-node test sequence into a multi-node test sequence, including: Taking out each input message in the single-node test sequence in turn, and generating an extended sequence for the taken-out input message; concatenating the spreading sequences corresponding to each input message in the single-node test sequence to form a multi-node test sequence; Wherein, the generation of the extended sequence for the fetched input message includes: Step 1. Initialize storage space for the extended sequence to be generated; Step 2, extracting the data packet and port information of the host from the input message, and adding the data packet and port information to the extended sequence; Step 3, according to the order of storage, sequentially obtain data packets or control packets and corresponding port information from the extended sequence; Step 4, based on the type of the port where the data message or control message will arrive and its own type, determine the processing of the data message or control message to obtain a new data message or control message and Corresponding port information, and adding the obtained new data message or control message and corresponding port information into the extended sequence; Step 5. Repeat step 3 and step 4 until the processing of the data packets or control packets stored in the extension sequence is completed, and all the content finally stored in the extension sequence is used as an extension sequence.