WO2019237531A1 - Network node monitoring method and system - Google Patents

Abstract

The present application relates to the technical field of communications, and provides a network node monitoring method and system. The monitoring method comprises: a server building a node monitoring task; the server issuing the built node monitoring task to message middleware; a network node being connected to a communication interface of the message middleware at every interval of a pre-set duration to acquire a corresponding node monitoring task; the network node processing the acquired node monitoring task and sending a task processing result to the message middleware; and the server receiving the task processing result by means of the message middleware. By means of this arrangement, since there is no need to establish a direct connection relationship between a server and network nodes, even though a problem occurs in the communication of the network nodes, the server does not need to wait for a reply from the network nodes. Therefore, the monitoring efficiency for the network nodes is effectively improved.

Description

Method and system for monitoring network nodes

This application claims priority from a Chinese patent application filed with the Chinese Patent Office on June 14, 2018, with application number 201810612655.2, and with the application name "A Method and System for Monitoring Network Nodes", the entire contents of which are incorporated herein by reference. Applying.

Technical field

The present application relates to the field of communication technologies, and in particular, to a method and system for monitoring network nodes.

Background technique

At present, when a server monitors a network node, it usually adopts a request / response communication mode. The processing mode of this communication mode is: the server establishes a connection with each network node; the server sends a monitoring task to each network node; after each network node has processed the task, the task processing result is returned to the server. However, most of the monitored network nodes are personal computers of the family. The communication stability of these nodes is insufficient, which is prone to problems such as network interruption, high network speed delay, computer failure or shutdown. When there is a problem with the communication between the server and the network node, the server will wait for a reply from the network node until the communication times out, which will seriously affect the monitoring efficiency of the network node.

technical problem

This application provides a method and system for monitoring network nodes, which can improve the monitoring efficiency of network nodes.

Technical solutions

A first aspect of the embodiments of the present application provides a method for monitoring a network node, including:

Server construction node monitoring task;

The server publishes the constructed node monitoring task to a message middleware, which is an integrated distributed system that monitors data transmission and reception using an asynchronous message passing mechanism;

A network node connects to the communication interface of the message middleware every preset time period, and acquires the corresponding node monitoring task;

The network node processes the acquired node monitoring task, and sends the task processing result to the message middleware;

The server receives the task processing result through the message middleware.

A second aspect of the embodiments of the present application provides a network node monitoring system, including:

A server for constructing a node monitoring task; publishing the constructed node monitoring task to a message middleware, the message middleware is an integrated distributed system that monitors data transmission and reception using an asynchronous message passing mechanism; and through the message middleware Receive task processing results;

A network node is configured to connect to the communication interface of the message middleware at a preset time interval to obtain a corresponding node monitoring task; process the acquired node monitoring task, and send the processing result of the task to the message middleware.

Beneficial effect

The method for monitoring a network node provided in the embodiment of the present application includes: a server constructs a node monitoring task; the server publishes the constructed node monitoring task to a message middleware, and the message middleware monitors data transmission and reception using an asynchronous message passing mechanism. Integrated distributed system; network nodes connect to the communication interface of the message middleware at predetermined intervals to obtain corresponding node monitoring tasks; network nodes process the obtained node monitoring tasks and send the results of the task processing to the message middle The server receives the task processing result through the message middleware. In the above process, the server publishes the constructed node monitoring task to the message middleware, and each network node actively connects the communication interface of the message middleware every preset time period to obtain the corresponding node monitoring task. With this setting, since the server and each network node do not need to establish a direct connection relationship, even if there is a problem with the communication of the network node, the server does not have to wait for a response from the network node, thereby effectively improving the monitoring efficiency of the network node.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flowchart of a first embodiment of a method for monitoring a network node according to an embodiment of the present application; FIG.

2 is a flowchart of a second embodiment of a method for monitoring a network node according to an embodiment of the present application;

3 is a flowchart of a third embodiment of a method for monitoring a network node according to an embodiment of the present application;

FIG. 4 is a structural diagram of an embodiment of a monitoring system for a network node according to an embodiment of the present application.

Embodiments of the invention

The embodiments of the present application provide a method and a system for monitoring a network node, which can improve the monitoring efficiency of the network node.

Referring to FIG. 1, a first embodiment of a method for monitoring a network node in an embodiment of the present application includes:

101. Server construction node monitoring task;

In the embodiment of the present application, the method is applied to a system composed of a server and more than one network node. First, the server constructs a node monitoring task. The node monitoring task is used to obtain the data that needs to be monitored on the network node.

102. The server publishes the constructed node monitoring task to the message middleware.

After the server constructs the node monitoring tasks, it publishes these node monitoring tasks to preset message middleware, which is an integrated distributed system that monitors data transmission and reception using an asynchronous message passing mechanism. The message middleware opens a data transmission interface. The communication process can be completed by creating a socket (socket, IP address + port number) and using related functions in the socket.

103. The network node connects to the communication interface of the message middleware at a preset time interval to obtain a corresponding node monitoring task.

Each network node actively connects the communication interface of the message middleware every preset time period to obtain the corresponding node monitoring task. For example, network node A initiates a connection to the message middleware, and obtains the monitoring tasks posted to its own node. If the task queue is empty or there are no tasks posted to itself, monitoring node A waits for 5 seconds and initiates a connection again in a loop. carried out.

Further, each network node has a unique node identifier, and each node monitoring task includes a task identifier for pointing to a task receiving node, and the acquiring the corresponding node monitoring task includes:

The network node obtains a node monitoring task with the same task ID as its own node ID.

When nodes are deployed, unique node identifiers are set for each node, such as "node 1 and node 2 ...", "node A, node B ..." and so on. When the server constructs a node monitoring task, if it wants to publish the task to node A, the task ID is set to A, and so on.

Further, the acquiring the corresponding node monitoring task may further include:

The network node queries the preset word meaning comparison table to obtain the synonyms, synonyms and superordinate words of the node identifier;

The network node acquires a task monitoring task that identifies a node that belongs to the synonyms, synonyms, or superordinate words.

When a network node obtains a task, it is determined whether the task identifier of each task and its own node identifier belong to synonyms, synonyms, or subordinate concepts. If the task identifier belongs to a synonym, a synonyms or a superordinate word of the node identifier, the corresponding task is determined as the monitoring task corresponding to the network node. Specifically, a synonym comparison table, a synonym comparison table, and a subordinate word comparison table may be set in advance, and then judged based on these comparison tables.

In addition, when the server issues a task, it can publish a certain monitoring task corresponding to multiple network nodes. For example, you can set the first-level task identifier and the second-level task identifier (generally the lower level of the first-level task identifier). The first-level task identifier and the second-level task identifier are separated by a designated label. If the second-level task identifier is set to the specified A certain character string, which indicates that the task is published to all lower positions of the first-level task identifier.

Example: When the node is deployed, the node itself is marked, for example, the node of Shenzhen City, Guangdong Province is marked "Guangdong & shenzhen". If the task sent by the server only needs to be sent to a network node in Shenzhen, Guangdong Province, then the task is marked with the task identifier "Guangdong & shenzhen", and then only the nodes in Shenzhen, Guangdong Province will obtain and execute the task after filtering. If it needs to be delivered to the network nodes of the whole Guangdong Province, then it will be marked with "Guangdong & True". True represents all nodes under this level, that is, all nodes in Guangdong Province will obtain and execute the task.

104. The network node processes the acquired node monitoring task, and sends a task processing result to the message middleware.

After each network node obtains the node monitoring task issued to itself, it processes these node monitoring tasks, and then sends the task processing results back to the message middleware.

105. The server receives the task processing result through the message middleware.

Finally, the server can receive the task processing results returned by each network node by connecting the communication interface of the message middleware, thereby implementing monitoring of each network node.

The method for monitoring a network node provided in the embodiment of the present application includes: a server constructs a node monitoring task; the server publishes the constructed node monitoring task to a message middleware, and the message middleware monitors data transmission and reception using an asynchronous message passing mechanism. Integrated distributed system; network nodes connect to the communication interface of the message middleware at predetermined intervals to obtain corresponding node monitoring tasks; network nodes process the obtained node monitoring tasks and send the results of the task processing to the message middle The server receives the task processing result through the message middleware. In the above process, the server publishes the constructed node monitoring task to the message middleware, and each network node actively connects the communication interface of the message middleware every preset time period to obtain the corresponding node monitoring task. With this setting, since the server and each network node do not need to establish a direct connection relationship, even if there is a problem with the communication of the network node, the server does not have to wait for a response from the network node, thereby effectively improving the monitoring efficiency of the network node.

Referring to FIG. 2, a second embodiment of a method for monitoring a network node according to an embodiment of the present application includes:

201. The server constructs a node monitoring task;

Step 201 is the same as step 101. For details, refer to the related description of step 101.

202. The server obtains the communication stability of the network node.

Before publishing the node monitoring task to the message middleware, the server first obtains the communication stability of the network node, which is a parameter determined according to the communication status of the network node and used to characterize the stability of the communication. If the server needs to monitor multiple network nodes, each network node has a corresponding communication stability.

Specifically, before step 202, the method may further include:

(1) The server obtains the communication status indicators of the network nodes; the communication status indicators include the number of communication interruptions per unit time between the server and the network nodes, the total duration of the communication interruptions per unit time, the average duration of the communication response, the current The length of communication delays and the frequency of equipment failures at network nodes;

(2) The server calculates the communication stability according to the communication status index.

For the above step (1), the communication status indicators of a network node include the number of communication interruptions in a unit time (such as one day) between the server and the network node, the total duration of communication interruptions in a unit time (such as one week), The average communication response time, the current communication delay time, and the frequency of equipment failures at this network node. When the server communicates with the network nodes, it will count the communication status indicators of each network node to calculate the communication stability of each network node.

For step (2) above, the formula k = k ₀ / (t ₁ * a + t ₂ * b + t ₃ * c) -x * f can be specifically used to obtain the communication stability;

Among them, k represents the communication stability, k ₀ is a preset constant, t ₁ represents the total duration of communication interruption in unit time, t ₂ represents the average duration of communication response, t ₃ represents the current communication delay duration, x Represents the number of communication interruptions per unit time, f represents the frequency of network node equipment failure, and a, b and c are preset weight coefficients. It can be seen that the smaller the above-mentioned communication status indicator, the greater the corresponding communication stability.

203: Determine whether the communication stability is less than a first threshold;

After obtaining the communication stability of the network node, it is determined whether the communication stability is less than a first threshold. If the communication stability is less than the first threshold, steps 204-207 are performed; otherwise, step 208 is performed.

204. The server publishes the constructed node monitoring task to the message middleware;

The communication stability is less than the first threshold value, which indicates that the communication quality between the server and the network node is not good. If the request / response communication mode is used, the problem of low monitoring efficiency of the network node will be caused, so the server will build a good The node monitoring task is issued to a message middleware, which is an integrated distributed system that monitors data transmission and reception using an asynchronous message delivery mechanism.

205. The network node connects to the communication interface of the message middleware at a preset time interval to obtain the corresponding node monitoring task.

206. The network node processes the acquired node monitoring task, and sends the task processing result to the message middleware.

207. The server receives the task processing result through the message middleware.

Steps 204-207 are the same as steps 102-105. For details, refer to the description of steps 102-105.

208. The server uses the request / response communication mode to monitor the network node.

The communication stability is greater than or equal to the first threshold, which indicates that the communication quality between the server and the network node is good. At this time, the server can use the request / response communication mode to monitor the network node without worrying about the communication with the network node. problem appear.

In this embodiment, for a network node with good network communication status, the server uses the request / response communication mode to monitor the network node. For a network node with poor network communication status, the server uses data forwarding through message middleware. The mode monitors network nodes. Compared with the first embodiment of the present application, the practicability of the monitoring method is further improved.

Referring to FIG. 3, a third embodiment of a method for monitoring a network node according to an embodiment of the present application includes:

301. Server construction node monitoring task;

302. The server publishes the constructed node monitoring task to the message middleware.

Steps 301-302 are the same as steps 101-102. For details, refer to the description of steps 101-102.

303: The network node counts the number of node monitoring tasks acquired in a unit time and the number of times of abnormal monitoring results.

The network node counts the number of node monitoring tasks acquired in a unit time and the number of abnormal monitoring results. For example, it counts the number of monitoring tasks issued to itself within one hour and the number of times that the network node has abnormal monitoring results within one hour. The abnormal monitoring result refers to the result that a certain data that the server needs to monitor is beyond the normal range, and generates an alarm or an error.

304. The network node calculates the probability value of the currently acquired node monitoring task according to the number of node monitoring tasks obtained in a unit time and the number of times that abnormal monitoring results occur.

After obtaining the number of node monitoring tasks and the number of times of abnormal monitoring results obtained in a unit time, the network node calculates the currently acquired node monitoring tasks according to the number of node monitoring tasks and the number of abnormal monitoring results obtained in a unit time. Probability value. For different network nodes, the calculated probability values may be different.

Further, step 304 may include:

(1) If the network node has acquired the node monitoring task within a preset time period, the formula p = m * p ₁ + n * p _{2 is} used to calculate the probability value;

(2) If the network node does not obtain a node monitoring task within a preset time period, the formula p = m * p ₃ + n * p _{4 is} used to calculate the probability value;

Among them, p represents the probability value, m represents the number of node monitoring tasks obtained by the network node in a unit time, n represents the number of times the network node has abnormal monitoring results, and p ₁ , p ₂ , p _3, and p ₄ are presets. the default probability value, and _{p 1 <p 3, p 2} <p 4.

For a network node, the more the number of node monitoring tasks acquired in a unit time, and the more times the network node has abnormal monitoring results, the greater the probability that the network node currently acquires the node monitoring tasks; and, If the node monitoring task has been acquired within a preset time period (for example, within 5 seconds), the probability of currently acquiring the node monitoring task will be relatively small, so set p ₁ <p ₃ , p ₂ <p ₄ .

305. The network node sets a target duration according to the probability value, and the probability value is negatively related to the target duration.

After the network node calculates the probability value of the node monitoring task currently obtained, the target duration is set according to the probability value, and the probability value is negatively correlated with the target duration, that is, the larger the probability value, the set target The shorter the duration. For example, if the probability value is 90%, the target duration is set to 3 seconds; if the probability value is 30%, the target duration is set to 30 seconds.

306. The network node connects to the communication interface of the message middleware every the target duration to obtain a corresponding node monitoring task.

After determining the target duration, the network node connects to the communication interface of the message middleware every the target duration to obtain the corresponding node monitoring task.

307. The network node processes the acquired node monitoring task, and sends a task processing result to the message middleware.

308. The server receives the task processing result through the message middleware.

Steps 307-308 are the same as steps 104-105. For details, refer to the related descriptions of steps 104-105.

Compared with the first embodiment of the present application, the interval at which each network node connects to the message middleware in this embodiment is set according to the probability value that the network node currently obtains the node monitoring task. If the probability value is greater, the setting is set. The interval is about short. By setting in this way, it is possible to avoid a network node that has obtained a node monitoring task from a small probability to frequently connect to the communication interface of the message middleware, and alleviate the communication pressure between the message middleware and the network node.

It should be understood that the size of the sequence numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

The above mainly describes a method for monitoring a network node, and a monitoring system for a network node will be described below.

Referring to FIG. 4, an embodiment of a monitoring system for a network node in the embodiment of the present application includes:

The server 401 is configured to construct a node monitoring task; post the constructed node monitoring task to a message middleware, which is an integrated distributed system that monitors data transmission and reception by using an asynchronous message passing mechanism; and through the message intermediate Receive task processing results;

The network node 402 is configured to connect to the communication interface of the message middleware at a preset time interval to obtain the corresponding node monitoring task; process the acquired node monitoring task, and send the task processing result to the message middleware.

Further, the message middleware opens a data transmission interface and uses a socket to complete a communication process.

Further, the server publishing the constructed node monitoring task to the message middleware may include: obtaining communication stability of a network node, where the communication stability is used to characterize the degree of communication stability determined according to the communication status of the network node. If the communication stability is less than the first threshold value, the server publishes the constructed node monitoring task to the message middleware.

Further, after the server obtains the communication stability of the network node, it may further include:

If the communication stability is greater than or equal to the first threshold, the server monitors the network node using a request / response communication mode.

Further, the server may be further configured to:

Obtain a communication status indicator of a network node, where the communication status indicator includes the number of communication interruptions in a unit time between the server and the network node, the total duration of communication interruptions in a unit time, the average duration of a communication response, and the current communication delay duration And the frequency of equipment failures at network nodes;

The communication stability is calculated according to the communication status index.

Furthermore, the communication stability calculated by the server according to the communication status indicator is specifically:

The server uses the formula k = k ₀ / (t ₁ * a + t ₂ * b + t ₃ * c) -x * f to calculate the communication stability;

Among them, k represents the communication stability, k ₀ is a preset constant, t ₁ represents the total duration of communication interruption in unit time, t ₂ represents the average duration of communication response, t ₃ represents the current communication delay duration, x Represents the number of communication interruptions per unit time, f represents the frequency of network node equipment failure, and a, b and c are preset weight coefficients.

Further, the communication interface for the network node to connect to the message middleware every preset time period may include:

The network node counts the number of node monitoring tasks acquired in a unit time and the number of abnormal monitoring results;

The network node calculates the probability value of the currently acquired node monitoring task according to the number of node monitoring tasks obtained in a unit time and the number of times an abnormal monitoring result occurs;

A network node sets a target duration according to the probability value, and the probability value is negatively related to the target duration;

The network node is connected to the communication interface of the message middleware every the target duration.

Furthermore, the network node calculating the probability value of the currently acquired node monitoring task according to the number of node monitoring tasks obtained in a unit time and the number of times of abnormal monitoring results may include:

If the network node has acquired the node monitoring task within a preset time period, the probability value is calculated by using the formula p = m * p ₁ + n * p ₂ ;

If the network node does not obtain a node monitoring task within a preset time period, the probability value is calculated by using the formula p = m * p ₃ + n * p ₄ ;

Among them, p represents the probability value, m represents the number of node monitoring tasks obtained by the network node in a unit time, n represents the number of times the network node has abnormal monitoring results, and p ₁ , p ₂ , p _3, and p ₄ are presets. the default probability value, and _{p 1 <p 3, p 2} <p 4.

Further, each network node has a unique node identifier, and each node monitoring task includes a task identifier for pointing to a task receiving node, and the acquiring the corresponding node monitoring task includes:

The network node obtains a node monitoring task with the same task ID as its own node ID.

Furthermore, the acquiring the corresponding node monitoring task may further include:

The network node queries the preset word meaning comparison table to obtain the synonyms, synonyms and superordinate words of the node identifier;

The network node acquires a task monitoring task that identifies a node that belongs to the synonyms, synonyms, or superordinate words.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and are not repeated here.

Claims (20)

A method for monitoring a network node, comprising: Server construction node monitoring task; The server publishes the constructed node monitoring task to a message middleware, which is an integrated distributed system that monitors data transmission and reception using an asynchronous message passing mechanism; A network node connects to the communication interface of the message middleware every preset time period, and acquires the corresponding node monitoring task; The network node processes the acquired node monitoring task, and sends the task processing result to the message middleware; The server receives the task processing result through the message middleware. The method for monitoring a network node according to claim 1, wherein the server publishing the constructed node monitoring task to a message middleware comprises: The server acquires communication stability of a network node, where the communication stability is a parameter determined according to the communication status of the network node and used to characterize the degree of communication stability; If the communication stability is less than the first threshold, the server publishes the constructed node monitoring task to the message middleware. The method for monitoring a network node according to claim 2, after the server obtains the communication stability of the network node, further comprising: If the communication stability is greater than or equal to the first threshold, the server monitors the network node using a request / response communication mode. The method for monitoring a network node according to claim 2, before the server obtains the communication stability of the network node, further comprising: The server obtains a communication status indicator of the network node, and the communication status indicator includes the number of communication interruptions in a unit time between the server and the network node, a total duration of the communication interruption in a unit time, an average response response time, and a current communication delay. Duration and frequency of equipment failures at network nodes; The server calculates the communication stability according to the communication status indicator. The method for monitoring a network node according to claim 4, wherein the communication stability calculated by the server according to the communication status indicator is specifically: The server uses the formula k = k ₀ / (t ₁ * a + t ₂ * b + t ₃ * c) -x * f to calculate the communication stability; Among them, k represents the communication stability, k ₀ is a preset constant, t ₁ represents the total duration of communication interruption in unit time, t ₂ represents the average duration of communication response, t ₃ represents the current communication delay duration, x Represents the number of communication interruptions per unit time, f represents the frequency of network node equipment failure, and a, b and c are preset weight coefficients. The method for monitoring a network node according to claim 1, wherein the communication interface for the network node to connect to the message middleware every preset time period comprises: The network node counts the number of node monitoring tasks acquired in a unit time and the number of abnormal monitoring results; The network node calculates the probability value of the currently acquired node monitoring task according to the number of node monitoring tasks obtained in a unit time and the number of times an abnormal monitoring result occurs; A network node sets a target duration according to the probability value, and the probability value is negatively related to the target duration; The network node is connected to the communication interface of the message middleware every the target duration. The method for monitoring a network node according to claim 6, wherein the network node calculates a probability value of the currently acquired node monitoring task according to the number of node monitoring tasks obtained in a unit time and the number of times that abnormal monitoring results occur include: If the network node has acquired the node monitoring task within a preset time period, the probability value is calculated by using the formula p = m * p ₁ + n * p ₂ ; If the network node does not obtain a node monitoring task within a preset time period, the probability value is calculated by using the formula p = m * p ₃ + n * p ₄ ; Among them, p represents the probability value, m represents the number of node monitoring tasks obtained by the network node in a unit time, n represents the number of times the network node has abnormal monitoring results, and p ₁ , p ₂ , p _3, and p ₄ are presets. the default probability value, and _{p 1 <p 3, p 2} <p 4. The method for monitoring a network node according to claim 1, wherein the message middleware opens a data transmission interface and uses a socket to complete a communication process. The method for monitoring a network node according to any one of claims 1 to 8, wherein each network node has a unique node identifier, and each node monitoring task includes a task identifier for pointing to a task receiving node The acquiring the corresponding node monitoring task includes: The network node obtains a node monitoring task with the same task ID as its own node ID. The method for monitoring a network node according to claim 9, wherein the acquiring the corresponding node monitoring task further comprises: The network node queries the preset word meaning comparison table to obtain the synonyms, synonyms and superordinate words of the node identifier; The network node acquires a task monitoring task that identifies a node that belongs to the synonyms, synonyms, or superordinate words. A monitoring system for a network node is characterized in that it includes: A server for constructing a node monitoring task; publishing the constructed node monitoring task to a message middleware, the message middleware is an integrated distributed system that monitors data transmission and reception using an asynchronous message passing mechanism; and through the message middleware Receive task processing results; A network node is configured to connect to the communication interface of the message middleware at a preset time interval to obtain a corresponding node monitoring task; process the acquired node monitoring task, and send the processing result of the task to the message middleware. The monitoring system for a network node according to claim 11, wherein the server publishing the constructed node monitoring task to a message middleware comprises: The server acquires communication stability of a network node, where the communication stability is a parameter determined according to the communication status of the network node and used to characterize the degree of communication stability; If the communication stability is less than the first threshold, the server publishes the constructed node monitoring task to the message middleware. The monitoring system for a network node according to claim 12, after the server obtains the communication stability of the network node, further comprising: If the communication stability is greater than or equal to the first threshold, the server monitors the network node using a request / response communication mode. The monitoring system for a network node according to claim 12, before the server obtains the communication stability of the network node, further comprising: The server obtains a communication status indicator of the network node, and the communication status indicator includes the number of communication interruptions in a unit time between the server and the network node, a total duration of the communication interruption in a unit time, an average response response time, and a current communication delay. Duration and frequency of equipment failures at network nodes; The server calculates the communication stability according to the communication status indicator. The monitoring system of a network node according to claim 14, wherein the communication stability calculated by the server according to the communication status indicator is specifically: The server uses the formula k = k ₀ / (t ₁ * a + t ₂ * b + t ₃ * c) -x * f to calculate the communication stability; Among them, k represents the communication stability, k ₀ is a preset constant, t ₁ represents the total duration of communication interruption in unit time, t ₂ represents the average duration of communication response, t ₃ represents the current communication delay duration, x Represents the number of communication interruptions per unit time, f represents the frequency of network node equipment failure, and a, b and c are preset weight coefficients. The monitoring system of a network node according to claim 11, wherein the communication interface for the network node to connect to the message middleware every preset time period comprises: The network node counts the number of node monitoring tasks acquired in a unit time and the number of abnormal monitoring results; The network node calculates the probability value of the currently acquired node monitoring task according to the number of node monitoring tasks obtained in a unit time and the number of times an abnormal monitoring result occurs; A network node sets a target duration according to the probability value, and the probability value is negatively related to the target duration; The network node is connected to the communication interface of the message middleware every the target duration. The monitoring system of a network node according to claim 16, wherein the network node calculates the probability value of the currently acquired node monitoring task according to the number of node monitoring tasks acquired in a unit time and the number of times that abnormal monitoring results occur include: If the network node has acquired the node monitoring task within a preset time period, the probability value is calculated by using the formula p = m * p ₁ + n * p ₂ ; If the network node does not obtain a node monitoring task within a preset time period, the probability value is calculated by using the formula p = m * p ₃ + n * p ₄ ; Among them, p represents the probability value, m represents the number of node monitoring tasks obtained by the network node in a unit time, n represents the number of times the network node has abnormal monitoring results, and p ₁ , p ₂ , p _3, and p ₄ are presets. the default probability value, and _{p 1 <p 3, p 2} <p 4. The monitoring system for a network node according to claim 11, wherein the message middleware opens a data transmission interface and uses a socket to complete a communication process. The monitoring system for a network node according to any one of claims 11 to 18, wherein each network node has a unique node identifier, and each node monitoring task includes a task identifier for pointing to a task receiving node The acquiring the corresponding node monitoring task includes: The network node obtains a node monitoring task with the same task ID as its own node ID. The monitoring system for a network node according to claim 19, wherein the task of acquiring the corresponding node monitoring further comprises: The network node queries the preset word meaning comparison table to obtain the synonyms, synonyms and superordinate words of the node identifier; The network node acquires a task monitoring task that identifies a node that belongs to the synonyms, synonyms, or superordinate words.