CN110581784B

CN110581784B - Node health check method, device and equipment

Info

Publication number: CN110581784B
Application number: CN201810588880.7A
Authority: CN
Inventors: 吴梦麒
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2018-06-08
Filing date: 2018-06-08
Publication date: 2023-01-10
Anticipated expiration: 2038-06-08
Also published as: CN110581784A

Abstract

The application provides a node health check method, a node health check device and node health check equipment, wherein the method comprises the following steps: selecting a first load balancing device from a load balancing device set according to the identification information of the node, wherein the first load balancing device is used for transferring the health check request; sending a health check request carrying the identification information to the first load balancing device; receiving a health check response corresponding to the identification information returned by the first load balancing equipment; and determining the health check result of the node according to the health check response. By the technical scheme, the number of health check requests received by the nodes is reduced, so that the processing pressure of the nodes is relieved, and the processing resources of the nodes are saved.

Description

Node health check method, device and equipment

Technical Field

The present application relates to the field of internet, and in particular, to a node health check method, apparatus, and device.

Background

With the rapid growth of internet users, the node pressure is higher and higher, and in order to cope with the huge business pressure, a large number of nodes can be deployed, and the nodes provide services for the users together. In order to share the service request of the user to each node in a balanced manner, load balancing equipment can be deployed, and after receiving the service request, the load balancing equipment distributes the service request to the nodes according to a load balancing algorithm to realize load balancing.

In order to know whether the node is abnormal or not, the load balancing device may periodically send a health check request to each node, and if a normal response message for the health check request is received, it is determined that the node is not abnormal, otherwise, it is determined that the node is abnormal, and the service request is no longer allocated to the abnormal node.

With the increase of the number of nodes, a plurality of load balancing devices can be deployed, each load balancing device performs health check on the node, that is, sends a health check request to the node, so that the node needs to process a large number of health check requests, the processing pressure of the node is high, and a large number of resources of the node are occupied. For example, if there are 10 load balancing devices, the 10 load balancing devices all periodically send health check requests to each node, that is, each node needs to process 10 health check requests in each period.

Disclosure of Invention

The application provides a node health check method, which comprises the following steps:

selecting a first load balancing device from a load balancing device set according to the identification information of the node, wherein the first load balancing device is used for relaying the health check request;

sending a health check request carrying the identification information to the first load balancing device;

receiving a health check response corresponding to the identification information returned by the first load balancing equipment;

and determining the health check result of the node according to the health check response.

receiving a health check request carrying node identification information and sent by second load balancing equipment;

inquiring whether a health check response corresponding to the identification information exists in a specified storage medium;

if a health check response exists and the health check response is a normal response message, sending the normal response message to the second load balancing equipment; wherein the normal response message is used for the second load balancing device to determine that the health check result of the node is that the node is normal.

sending a health check request to first load balancing equipment, wherein the health check request carries identification information of a node, so that the first load balancing equipment acquires a health check response corresponding to the identification information;

receiving a health check response returned by the first load balancing equipment; if the health check response is a normal response message, determining that the health check result of the node is normal; or if the health check response is an abnormal response message, determining that the health check result of the node is abnormal.

The present application provides a node health check device, the device includes:

the selection module is used for selecting first load balancing equipment from a load balancing equipment set according to the identification information of the node, wherein the first load balancing equipment is used for transferring the health check request;

a sending module, configured to send a health check request to the first load balancing device; wherein the health check request carries identification information of the node;

a receiving module, configured to receive a health check response corresponding to the identification information, where the health check response is returned by the first load balancing device;

and the determining module is used for determining the health examination result of the node according to the health examination response.

the receiving module is used for receiving a health check request sent by the second load balancing equipment; wherein the health check request carries identification information of a node;

the query module is used for querying whether a health check response corresponding to the identification information exists in a specified storage medium;

a sending module, configured to send a normal response message to the second load balancing device when a health check response exists and the health check response is a normal response message; the normal response message is used for enabling the second load balancing equipment to determine that the health check result of the node is that the node is normal.

a sending module, configured to send a health check request to a first load balancing device, where the health check request carries identification information of a node, so that the first load balancing device obtains a health check response corresponding to the identification information;

a receiving module, configured to receive a health check response returned by the first load balancing device;

a determining module, configured to determine that a health check result of the node is node normal when the health check response is a normal response message; or when the health check response is an abnormal response message, determining that the health check result of the node is abnormal.

The application provides a load balancing device, which comprises a processor and a machine-readable storage medium, wherein a plurality of computer instructions are stored on the machine-readable storage medium, and the processor executes the computer instructions to perform the following processing: selecting a first load balancing device from a load balancing device set according to the identification information of the node, wherein the first load balancing device is used for relaying the health check request; sending a health check request carrying the identification information to the first load balancing device; receiving a health check response corresponding to the identification information returned by the first load balancing equipment; and determining the health check result of the node according to the health check response.

The application provides a load balancing device, which comprises a processor and a machine-readable storage medium, wherein a plurality of computer instructions are stored on the machine-readable storage medium, and the processor executes the computer instructions to perform the following processing: receiving a health check request carrying the identification information of the node and sent by second load balancing equipment; inquiring whether a health check response corresponding to the identification information exists in a specified storage medium or not; if a health check response exists and the health check response is a normal response message, sending the normal response message to the second load balancing equipment; wherein the normal response message is used for the second load balancing device to determine that the health check result of the node is that the node is normal.

The application provides a load balancing device, which comprises a processor and a machine-readable storage medium, wherein the machine-readable storage medium is stored with a plurality of computer instructions, and the processor executes the computer instructions to perform the following processing: sending a health check request to first load balancing equipment, wherein the health check request carries identification information of a node, so that the first load balancing equipment acquires a health check response corresponding to the identification information; receiving a health check response returned by the first load balancing equipment; if the health check response is a normal response message, determining that the health check result of the node is that the node is normal; or if the health check response is an abnormal response message, determining that the health check result of the node is abnormal.

Based on the above technical solution, in the embodiment of the application, when sending the health check request, the second load balancing device sends the health check request to the first load balancing device instead of sending the health check request to the node, and after receiving the health check request, the first load balancing device queries whether a corresponding health check response exists in the specified storage medium; if so, the health check reply is sent directly to the second load balancing device without sending a health check request to the node. In summary, the first load balancing device may cache the health check response of the node locally, and when receiving the health check request sent by the second load balancing device, may send the health check response of the local cache to the second load balancing device, so that the health check request does not need to be sent to the node, thereby reducing the number of health check requests received by the node, further reducing the processing pressure of the node, and saving the processing resources of the node.

Drawings

FIGS. 1A and 1B are schematic diagrams of an application scenario in an embodiment of the present application;

FIG. 2 is a flow diagram of a node health check method in one embodiment of the present application;

FIG. 3 is a flow chart of a node health check method in another embodiment of the present application;

FIG. 4 is a flow diagram of a node health check method in another embodiment of the present application;

FIG. 5 is a block diagram of a node health check device according to an embodiment of the present application;

fig. 6 is a block diagram of a node health check device according to another embodiment of the present application;

fig. 7 is a block diagram of a node health check device according to another embodiment of the present application.

Detailed Description

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein is meant to encompass any and all possible combinations of one or more of the associated listed items.

It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present application. Depending on the context, moreover, the word "if" may be used to be interpreted as "at 8230; \8230when" or "when 8230; \8230when" or "in response to a determination".

An embodiment of the present application provides a node health check method, which may be applied to a system including multiple load balancing devices, and is shown in fig. 1A, which is an application scenario diagram of the embodiment of the present application. In fig. 1A, 4 clients, 3 load balancing devices, and 5 nodes are taken as an example for explanation, and in practical application, the number of the clients, the number of the load balancing devices, and the number of the nodes may be all more, which is not limited to this. In addition, the system may also include other types of devices, which are not limited in this regard, and will be described with reference to fig. 1A as an example.

The client may be an APP (Application) included in a terminal device (e.g., a Personal Computer (PC), a notebook Computer, a mobile terminal, etc.), or may be a browser included in the terminal device, and the client may access the node without limitation to the type of the client.

The load balancing equipment is used for receiving a service request sent by a client and distributing the service request to the nodes according to a load balancing algorithm to realize load balancing. For example, after receiving a large number of service requests, the service requests may be load balanced to each node, and the load balancing process is not limited.

The node in this embodiment may include, but is not limited to, a server, such as a data storage server, an application server, a service server, and a data query server; other types of nodes are also possible, such as a PC, etc., without limitation to this type.

In one example, in order to know whether a node is abnormal, each load balancing device periodically sends a health check request to each node, determines that the node is not abnormal if a normal response message for the health check request is received, and determines that the node is abnormal if an abnormal response message for the health check request is received or a normal response message and an abnormal response message for the health check request are not received, so that the abnormal node is removed, and the service request is no longer distributed to the abnormal node.

Along with the transverse capacity expansion of the load balancing cluster, the number of the load balancing devices is increased, and when each load balancing device performs health check on the nodes, a large number of health check requests are sent to the nodes, so that the nodes need to process a large number of health check requests, the processing pressure is high, and a large number of processing resources are occupied.

Different from the foregoing manner, in this embodiment, a load balancing device for relaying the health check request may be selected from the load balancing device set, and subsequently, the load balancing device is referred to as a first load balancing device, and each load balancing device except the first load balancing device is a second load balancing device.

In an example, the load balancing device set may include all load balancing devices, or may include some load balancing devices, which is not limited to this.

Referring to fig. 1B, it is assumed that the first load balancing device for relaying the health check request is the load balancing device 111, and the second load balancing devices are the load balancing device 112 and the load balancing device 113.

Based on this, each second load balancing device sends the health check request to the first load balancing device instead of sending the health check request to the node when sending the health check request. After receiving the health check request, the first load balancing device may send the health check request to the node 121 for a first health check request of a certain node (e.g., the node 121), and after receiving a health check response for the health check request, may cache a corresponding relationship between the identification information of the node 121 and the health check response locally, for example, for a preset time (e.g., 10 seconds). Within 10 seconds, if a health check request for node 121 is received, a health check response is directly queried from the local cache and sent to the second load balancing device, instead of sending the health check request to node 121.

Obviously, after the above processing, in a preset time, since the corresponding relationship between the identification information of the node 121 and the health check response is locally cached, the health check request is no longer sent to the node 121, even if the number of the load balancing devices in the load balancing cluster is large, the node does not receive a large number of health check requests, the influence on the node is reduced, and the lateral expansion capability of the load balancing cluster is greatly improved.

In one example, each load balancing device may employ a consistent hashing algorithm to select a first load balancing device for relaying the health check request. Of course, each load balancing device may also use another type of hash algorithm to select the first load balancing device for relaying the health check request, for example, use a remainder hash algorithm, which is not limited herein. For example, referring to fig. 1B, the first load balancing device selected by the load balancing device 111 by using the consistent hash algorithm is the load balancing device 111, the first load balancing device selected by the load balancing device 112 by using the consistent hash algorithm is the load balancing device 111, and the first load balancing device selected by the load balancing device 113 by using the consistent hash algorithm is the load balancing device 111.

The consistent hash algorithm is a special hash algorithm, after the first load balancing device is selected by using the consistent hash algorithm, the load balancing devices can be made to be decentralized, the load balancing device set is not dependent on centralization, health check can be performed autonomously, and when the first load balancing device fails, other load balancing devices can be migrated automatically by using the consistent hash algorithm, so that the risk of single-point failure is reduced.

In one example, the load balancing device selects the first load balancing device using a consistent hashing algorithm, which may include but is not limited to: when health check is carried out on a certain node, a hash algorithm input value is obtained according to identification information (such as an IP address and the like) of the node, the hash algorithm input value is processed by adopting a consistent hash algorithm to obtain a processing result, and first load balancing equipment is selected from a load balancing equipment set according to the processing result. Similarly, the load balancing device selects the first load balancing device by using a remainder hash algorithm, which may include but is not limited to: when a certain node is subjected to health check, a hash algorithm input value is obtained according to the identification information of the node, the hash algorithm input value is processed by adopting a remainder hash algorithm to obtain a processing result, and first load balancing equipment is selected from the load balancing equipment set according to the processing result.

For example, taking the complementary hash algorithm as an example, assuming that the node 121 needs to be subjected to health check, each load balancing device may obtain a hash algorithm input value according to the IP address of the node 121, for example, the IP address 192.168.1.31 of the node 121 is converted into the hash algorithm input value 192168131, which is only an example and is not limited thereto. Then, a remainder hash algorithm may be used to perform a remainder processing on the hash algorithm input value 192168131, for example, a remainder of the hash algorithm input value 192168131 divided by the number of load balancing devices may also be calculated, and the first load balancing device may be selected from the load balancing device set according to the remainder. For example, if the remainder is 1 and the remainder 1 corresponds to the load balancing device 112, each load balancing device determines that the first load balancing device is the load balancing device 112. If the number of the load balancing devices is changed, when the remainder of dividing the hash algorithm input value 192168131 by the number of the load balancing devices is calculated, the remainder is changed, so that a new first load balancing device can be selected from the load balancing device set according to the changed remainder, and if the remainder is 2 and the remainder 2 corresponds to the load balancing device 113, each load balancing device determines that the first load balancing device is the load balancing device 113.

For another example, a consistent hash algorithm is taken as an example for description, and assuming that the node 121 needs to be subjected to health check, each load balancing device may obtain a hash algorithm input value according to the IP address of the node 121, for example, the IP address 192.168.1.31 of the node 121 may be converted into the hash algorithm input value 192168131, which is only an example and is not limited thereto. Then, consistent hashing may be performed on the hash algorithm input value 192168131 by using a consistent hashing algorithm, without limitation to the processing procedure, and then, a first load balancing device may be selected from the load balancing device set according to the processing result. For example, assuming that the number of the load balancing devices is 10, the hash algorithm input value 192168131 may be divided by 80, and if the remainder is any value from 0 to 7, the remainder corresponds to the first load balancing device, that is, the first load balancing device may be used as the first load balancing device; if the remainder is any value in 8-15, the remainder corresponds to the second load balancing device, and the second load balancing device can be used as the first load balancing device; and so on. In this application scenario, assuming that the remainder corresponds to the load balancing device 111, each load balancing device may determine that the first load balancing device is the load balancing device 111.

If the number of the load balancing devices is changed, if the number of the load balancing devices is changed from 10 to 8, the hash algorithm input value 192168131 can be calculated to be divided by 80, if the remainder is any value from 0 to 9, the remainder corresponds to the first load balancing device, and the first load balancing device can be used as the first load balancing device; if the remainder is any value in 10-19, the remainder corresponds to the second load balancing device, and the second load balancing device can be used as the first load balancing device; and so on. Obviously, since the hash algorithm input value 192168131 is not changed and the denominator 80 is not changed, the remainder is not changed, if the remainder is any value from 0 to 7, the first load balancing device does not change before and after the change of the number of the load balancing devices, if the remainder is any value from 8 to 9, the first load balancing device does not change before and after the change of the number of the load balancing devices, if the remainder is any value from 10 to 15, the first load balancing device does not change before and after the change of the number of the load balancing devices, and so on. In summary, after the consistent hash algorithm is adopted, the first load balancing device may or may not be changed. In this application scenario, taking the first load balancing device as an example, assuming that the remainder still corresponds to the load balancing device 111, each load balancing device may determine that the first load balancing device is the load balancing device 111.

In the above embodiment, the first load balancing device is determined based on the identification information (e.g., IP address) of the node, but the identification information of different nodes is different, and for different nodes, the first load balancing device may be the same or different. For example, for node 121, the first load balancing device is load balancing device 111, i.e., the health check request is sent by load balancing device 111 to node 121; for node 122, the first load balancing device is load balancing device 112, i.e., a health check request is sent by load balancing device 112 to node 122; for the node 123, the first load balancing device is the load balancing device 111, that is, the load balancing device 111 sends a health check request to the node 123; and so on.

In summary, for different nodes, different load balancing devices may send health check requests, so that one load balancing device is prevented from sending health check requests to all nodes, and thus the load balancing device is not centralized and the load balancing device set is not dependent on the centralization. When a certain load balancing device fails or load balancing devices are dynamically added, each load balancing device can adopt a consistent hash algorithm to determine the first load balancing device again, so that automatic migration is realized, and the risk of single-point failure is reduced.

In the application scenario, referring to fig. 2, a schematic flow diagram of a node health check method provided in the embodiment of the present application may be applied to a first load balancing device, and the method may include:

step 201, receiving a health check request sent by the second load balancing device, where the health check request carries identification information (such as an IP address) of a node, and the node is a node that needs to perform health check.

Step 202, inquiring whether a health check response corresponding to the identification information exists in the specified storage medium; wherein the health check response may be a normal response message or an abnormal response message.

Step 203, if there is a health check response, and the health check response is a normal response message, the normal response message may be sent to the second load balancing device; wherein the normal response message is used for the second load balancing device to determine that the health check result of the node is that the node is normal.

In one example, after querying whether the health check response corresponding to the identification information exists in the specified storage medium, if the health check response exists and the health check response is an abnormal response message, the abnormal response message may be sent to the second load balancing device; wherein the exception response message is used for the second load balancing device to determine that the health check result of the node is a node exception; or, the second load balancing device may also refuse to send an abnormal response message; and when the second load balancing equipment does not receive the normal response message or the abnormal response message, determining that the health check result of the node is abnormal.

In one example, after querying whether the specified storage medium has the health check response corresponding to the identification information, if the health check response does not exist, the health check request may be further sent to the node corresponding to the identification information. If a health check response (such as a normal response message or an abnormal response message) returned by the node for the health check request is received, the received health check response may be sent to the second load balancing device. And if the health check response returned by the node for the health check request is not received, the health check response for the health check request is refused to be sent to the second load balancing equipment.

In the above embodiment, after the health check request is sent to the node, if a health check response returned by the node for the health check request is received, the corresponding relationship between the identification information and the health check response may be recorded in a designated storage medium; wherein the health check response may be a normal response message or an abnormal response message. If the health check response returned by the node for the health check request is not received, an abnormal response message (for example, an abnormal response message is simulated) may be generated, and the corresponding relationship between the identification information and the abnormal response message may be recorded in a designated storage medium.

Further, after the correspondence relation is recorded in the designated storage medium, an aging timer may be set for the correspondence relation, and an aging time of the aging timer may be configured empirically, for example, may be 10 seconds. When the aging timer is overtime, the correspondence may be deleted from the specified storage medium.

In an example, when the first load balancing device needs to perform health check on a node, it may also be queried whether a health check response corresponding to the identification information of the node exists in a specified storage medium; if a health check answer exists and the health check answer is a normal answer message, the health check result of the node can be determined to be that the node is normal; if a health check reply exists and the health check reply is an exception reply message, it may be determined that the health check result for the node is a node exception.

If no health check reply exists, a health check request may also be sent to the node. If a health check response aiming at the health check request is received, if the health check response is a normal response message, determining that the health check result of the node is normal; if the health check response is an abnormal response message, determining that the health check result of the node is abnormal; and if the health check response aiming at the health check request is not received, determining that the health check result of the node is abnormal.

Further, if a health check response to the health check request is received, the corresponding relationship between the identification information of the node and the health check response may be recorded in a designated storage medium; if a health check response to the health check request is not received, an abnormal response message may be generated, and a correspondence between the identification information and the abnormal response message may be recorded in a designated storage medium.

In the above embodiments, the specified storage medium may be any storage medium used for storing data, such as a buffer, a memory, a hard disk, a magnetic disk, and the like, and the type of the specified storage medium is not limited.

In an example, the execution sequence is only an example given for convenience of description, and in practical applications, the execution sequence between steps may also be changed, and the execution sequence is not limited. Moreover, in other embodiments, the steps of the respective methods do not have to be performed in the order shown and described herein, and the methods may include more or less steps than those described herein. Moreover, a single step described in this specification may be broken down into multiple steps for description in other embodiments; multiple steps described in this specification may be combined into a single step in other embodiments.

In the above application scenario, referring to fig. 3, a flowchart of the node health check method provided in the embodiment of the present application may be applied to a second load balancing device, and the method may include:

step 301, a health check request is obtained, where the health check request may carry identification information (e.g., an IP address) of a node, and the node may be a node that needs to perform health check.

Step 302, sending the health check request to the first load balancing device, so that the first load balancing device obtains a health check response corresponding to the identification information and aiming at the health check request.

Step 303, receiving a health check response returned by the first load balancing device; if the health check response is a normal response message, determining that the health check result of the node is normal; or if the health check response is an abnormal response message, determining that the health check result of the node is abnormal.

In one example, after the second load balancing device sends the health check request to the first load balancing device (i.e., step 302), if a health check response returned by the first load balancing device for the health check request is not received, it may be determined that the health check result of the node is a node exception.

In the application scenario, referring to fig. 4, a schematic flow diagram of a node health check method provided in the embodiment of the present application may be applied to a second load balancing device, and the method may include:

step 401, selecting a first load balancing device from the load balancing device set according to the identification information of the node, where the first load balancing device is a load balancing device for relaying the health check request.

The selecting the first load balancing device from the load balancing device set according to the identification information of the node may include, but is not limited to: and based on a specific hash algorithm, selecting a first load balancing device from the load balancing device set according to the identification information of the node. Further, a hash algorithm input value may be obtained according to the identification information of the node, a specific hash algorithm is used to process the hash algorithm input value to obtain a processing result, and a first load balancing device may be selected from the load balancing device set according to the processing result.

Wherein, the specific hash algorithm includes but is not limited to: a consistent hash algorithm or a remainder hash algorithm.

When the total number of the load balancing devices in the load balancing device set changes, the first load balancing device can be reselected from the load balancing device set according to the identification information of the node.

Step 402, sending the health check request carrying the identification information to the first load balancing device.

And step 403, receiving a health check response corresponding to the identification information, which is returned by the first load balancing device.

And step 404, determining the health check result of the node according to the health check response.

Wherein, determining the health check result of the node according to the health check response may include, but is not limited to: if the health check response is a normal response message, determining that the health check result of the node is normal according to the health check response; alternatively, if the health check response is an exception response message, it may be determined from the health check response that the health check result of the node is a node exception.

In addition, after the health check request carrying the identification information is sent to the first load balancing device, if a health check response corresponding to the identification information and returned by the first load balancing device is not received, it may be determined that the health check result of the node is a node anomaly.

The node health check method is described below with reference to several specific application scenarios. In this application scenario, the node to be checked for health is the node 121 for example, and the processing flow for other nodes is the same, which is not described herein again. For the node 121, the first load balancing device is the load balancing device 111, and the second load balancing device is the load balancing device 112 and the load balancing device 113.

Application scenario 1: the load balancing device 111 checks the health of the node 121.

When the node 121 needs to be checked for health, the load balancing device 111 may obtain a hash algorithm input value according to the IP address of the node 121, process the hash algorithm input value by using a hash algorithm (e.g., a consistent hash algorithm), obtain a processing result, and select a first load balancing device from the load balancing device set according to the processing result, which is not described in detail herein, and refer to the above embodiments.

Since the first load balancing device is the load balancing device 111, the load balancing device 111 may query whether a health check reply corresponding to the IP address of the node 121 exists in the specified storage medium.

If a health check reply exists and the health check reply is a normal reply message, the load balancing device 111 may determine that the health check result of node 121 is that node 121 is normal; if a health check answer exists and is an exception answer message, load balancing device 111 may determine that the health check result of node 121 is that node 121 is abnormal, i.e., node 121 needs to be culled.

If no health check reply exists, load balancing device 111 may send a health check request to node 121. If the load balancing device 111 receives the health check reply for the health check request, if the health check reply is a normal reply message, it is determined that the health check result of the node 121 is that the node 121 is normal; if the health check reply is an exception reply message, then it is determined that the health check result of node 121 is node 121 abnormal. If load balancing device 111 does not receive a health check reply to the health check request, it is determined that the health check result of node 121 is an exception for node 121.

Further, if the load balancing device 111 receives a health check response to the health check request, the corresponding relationship between the identification information of the node 121 and the health check response may also be recorded in a designated storage medium; if the load balancing device 111 does not receive the health check response to the health check request, the corresponding relationship between the identification information and the abnormal response message may also be recorded in a designated storage medium.

Application scenario 2: the health check process of the load balancing device 112 on the node 121.

When the node 121 is subjected to health check, the load balancing device 112 selects a first load balancing device, such as the load balancing device 111, from the load balancing device set. The load balancing device 112 then sends a health check request to the load balancing device 111.

After receiving the health check request (carrying the IP address of the node 121), the load balancing device 111 queries whether a health check response corresponding to the IP address of the node 121 exists in a designated storage medium.

If a health check reply exists and the health check reply is a normal reply message, the load balancing device 111 sends the normal reply message to the load balancing device 112. Load balancing device 112, upon receiving the normality answer message, may determine that node 121 is normal as a result of the health check of node 121.

If there is a health check reply and the health check reply is an exception reply message, the load balancing device 111 sends the exception reply message to the load balancing device 112. Load balancing device 112, upon receiving the exception reply message, may then determine that the health check result of node 121 is an exception for node 121. Alternatively, when the health check response is an abnormal response message, the load balancing device 111 may refuse to send the abnormal response message to the load balancing device 112, and when the load balancing device 112 does not receive the normal response message or the abnormal response message, it may also determine that the health check result of the node 121 is that the node 121 is abnormal.

If no health check reply exists, load balancing device 111 may send a health check request to node 121. If the load balancing device 111 receives the health check reply, the health check reply may be sent to the load balancing device 112; after receiving the health check response, the load balancing device 112 determines that the health check result of the node 121 is that the node 121 is normal if the health check response is a normal response message, and determines that the health check result of the node 121 is that the node 121 is abnormal if the health check response is an abnormal response message. If the load balancing apparatus 111 does not receive the health check response, it may refuse to send the health check response to the load balancing apparatus 112, and when the load balancing apparatus 112 does not receive the health check response, it may also determine that the health check result of the node 121 is an exception for the node 121.

Further, if the load balancing device 111 receives the health check response, the corresponding relationship between the identification information of the node 121 and the health check response (such as a normal response message or an abnormal response message) may be recorded in a specified storage medium; if the load balancing apparatus 111 does not receive the health check response, the correspondence between the identification information of the node 121 and the abnormal response message may be recorded in a designated storage medium.

In the application scenarios 1 and 2, after the correspondence relation is recorded in the designated storage medium, an aging timer may be further set for the correspondence relation, and the aging time of the aging timer may be configured according to experience, for example, 10 seconds. And when the aging timer is overtime, deleting the corresponding relation from the specified storage medium. Thus, the health check response may be cached locally, and if a health check request is received for node 121, the health check response is queried directly from the local cache and the health check request is no longer sent to node 121.

In application scenarios 1 and 2, the health check request may be unrelated to the service provided by the node, such as a PING health check request, or may be related to the service provided by the node, such as when the node provides HTTP (HyperText Transfer Protocol) service, the health check request may be an HTTP health check request, and the type of the health check request is not limited. The health examination response is a response to the health examination request, and the type of the health examination response is not limited.

Based on the same application concept as the method described above, an embodiment of the present application further provides a node health check apparatus, which is applied to a first load balancing device, as shown in fig. 5, and is a structural diagram of the apparatus, including:

a selecting module 501, configured to select a first load balancing device from a load balancing device set according to the identification information of the node, where the first load balancing device is a load balancing device for relaying a health check request;

a sending module 502, configured to send a health check request to the first load balancing device; wherein the health check request carries identification information of the node;

a receiving module 503, configured to receive a health check response corresponding to the identification information, where the health check response is returned by the first load balancing device;

a determining module 504, configured to determine a health check result of the node according to the health check response.

The selecting module 501 is specifically configured to, when selecting a first load balancing device from the load balancing device set according to the identification information of the node: based on a specific Hash algorithm, selecting first load balancing equipment from a load balancing equipment set according to the identification information of the nodes; the specific hashing algorithm comprises a consistent hashing algorithm.

The selecting module 501 is specifically configured to, when selecting a first load balancing device from the load balancing device set according to the identification information of the node: acquiring a hash algorithm input value according to the identification information of the node; processing the hash algorithm input value by adopting a specific hash algorithm to obtain a processing result; and selecting a first load balancing device from the load balancing device set according to the processing result.

Based on the same application concept as the method, an embodiment of the present application further provides a load balancing device (such as the second load balancing device described above), which may include a processor and a machine-readable storage medium, where the machine-readable storage medium has a plurality of computer instructions stored thereon, and when the processor executes the computer instructions, the processor performs the following processes: selecting a first load balancing device from a load balancing device set according to the identification information of the node, wherein the first load balancing device is used for transferring the health check request; sending a health check request carrying the identification information to the first load balancing device; receiving a health check response corresponding to the identification information returned by the first load balancing equipment; and determining the health check result of the node according to the health check response.

Based on the same application concept as the method, the embodiment of the present application further provides a machine-readable storage medium, where a plurality of computer instructions are stored on the machine-readable storage medium, and when executed, the computer instructions perform the following processes: selecting a first load balancing device from a load balancing device set according to the identification information of the node, wherein the first load balancing device is used for relaying the health check request; sending a health check request carrying the identification information to the first load balancing device; receiving a health check response corresponding to the identification information returned by the first load balancing equipment; and determining the health check result of the node according to the health check response.

Based on the same application concept as the method described above, an embodiment of the present application further provides a node health check apparatus, which is applied to a first load balancing device, as shown in fig. 6, and is a structural diagram of the apparatus, including:

a receiving module 601, configured to receive a health check request sent by a second load balancing device; the health examination request carries identification information of a node; an inquiry module 602, configured to inquire whether a health check response corresponding to the identification information exists in a specified storage medium; a sending module 603, configured to send a normal response message to the second load balancing device when a health check response exists and the health check response is a normal response message; the normal response message is used for the second load balancing device to determine that the health check result of the node is that the node is normal.

The sending module 603 is further configured to send, when there is a health check response and the health check response is an abnormal response message, the abnormal response message to the second load balancing device; wherein the exception response message is used for the second load balancing device to determine that the health check result of the node is a node exception; or refusing to send the abnormal response message to the second load balancing equipment.

In an example, the sending module 603 is further configured to send the health check request to the node when there is no health check response; the device further comprises (not shown in the figures): a recording module, configured to record, in a specified storage medium, a correspondence between the identification information and a health check response when a health check response returned by a node for the health check request is received, where the health check response includes a normal response message or an abnormal response message; and when a health check response returned by the node aiming at the health check request is not received, generating an abnormal response message, and recording the corresponding relation between the identification information and the abnormal response message in a specified storage medium.

Based on the same application concept as the method described above, an embodiment of the present application further provides a load balancing device (such as the first load balancing device described above), which may include a processor and a machine-readable storage medium, where the machine-readable storage medium has a plurality of computer instructions stored thereon, and when the processor executes the computer instructions, the processor performs the following processes: receiving a health check request carrying the identification information of the node and sent by second load balancing equipment; inquiring whether a health check response corresponding to the identification information exists in a specified storage medium; if a health check response exists and the health check response is a normal response message, sending the normal response message to the second load balancing equipment; wherein the normal response message is used for the second load balancing device to determine that the health check result of the node is that the node is normal.

Based on the same application concept as the method, the embodiment of the present application further provides a machine-readable storage medium, where a plurality of computer instructions are stored on the machine-readable storage medium, and when executed, the computer instructions perform the following processes: receiving a health check request carrying the identification information of the node and sent by second load balancing equipment; inquiring whether a health check response corresponding to the identification information exists in a specified storage medium or not; if a health check response exists and the health check response is a normal response message, sending the normal response message to the second load balancing equipment; wherein the normal response message is used for the second load balancing device to determine that the health check result of the node is that the node is normal.

Based on the same application concept as the method described above, an embodiment of the present application further provides a node health check apparatus, which is applied to a second load balancing device, as shown in fig. 7, and is a structural diagram of the apparatus, including:

a sending module 701, configured to send a health check request to a first load balancing device, where the health check request carries identification information of a node, so that the first load balancing device obtains a health check response corresponding to the identification information; a receiving module 702, configured to receive a health check response returned by the first load balancing device; a determining module 703, configured to determine that the health check result of the node is that the node is normal when the health check response is a normal response message; or when the health check response is an abnormal response message, determining that the health check result of the node is abnormal.

Based on the same application concept as the method, an embodiment of the present application further provides a load balancing device (such as the second load balancing device described above), which may include a processor and a machine-readable storage medium, where the machine-readable storage medium has a plurality of computer instructions stored thereon, and when the processor executes the computer instructions, the processor performs the following processes: sending a health check request to first load balancing equipment, wherein the health check request carries identification information of a node, so that the first load balancing equipment acquires a health check response corresponding to the identification information; receiving a health check response returned by the first load balancing equipment; if the health check response is a normal response message, determining that the health check result of the node is normal; or if the health check response is an abnormal response message, determining that the health check result of the node is abnormal.

Based on the same application concept as the method, the embodiment of the present application further provides a machine-readable storage medium, where a plurality of computer instructions are stored on the machine-readable storage medium, and when executed, the computer instructions perform the following processes: sending a health check request to first load balancing equipment, wherein the health check request carries identification information of a node, so that the first load balancing equipment acquires a health check response corresponding to the identification information; receiving a health check response returned by the first load balancing equipment; if the health check response is a normal response message, determining that the health check result of the node is normal; or if the health check response is an abnormal response message, determining that the health check result of the node is abnormal.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. A typical implementation device is a computer, which may take the form of a personal computer, laptop computer, cellular telephone, camera phone, smart phone, personal digital assistant, media player, navigation device, email messaging device, game console, tablet computer, wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and so forth) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Furthermore, these computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A node health check method, wherein a load balancing device set comprises a plurality of load balancing devices, and the method is applied to each second load balancing device in the load balancing device set, and the method comprises:

selecting a first load balancing device from a load balancing device set according to the identification information of the node, wherein the first load balancing device is used for relaying the health check request; wherein, the health check request is used for learning whether the node is abnormal;

receiving a health check response corresponding to the identification information returned by the first load balancing equipment; if the specified storage medium of the first load balancing device has a health check response corresponding to the identification information, returning the health check response in the specified storage medium;

2. The method of claim 1, wherein selecting the first load balancing device from the set of load balancing devices according to the identification information of the node comprises: and based on a specific hash algorithm, selecting a first load balancing device from the load balancing device set according to the identification information of the node.

3. The method according to claim 1, wherein selecting the first load balancing device from the set of load balancing devices according to the identification information of the node comprises:

acquiring a hash algorithm input value according to the identification information of the node;

processing the hash algorithm input value by adopting a specific hash algorithm to obtain a processing result;

and selecting a first load balancing device from the load balancing device set according to the processing result.

4. The method of claim 2 or 3,

the particular hash algorithm comprises a consistent hash algorithm.

5. The method of claim 1, further comprising:

and when the total number of the load balancing devices in the load balancing device set changes, reselecting the first load balancing device from the load balancing device set according to the identification information of the node.

6. The method of claim 1,

the determining a health check result of the node according to the health check response includes:

if the health check response is a normal response message, determining that the health check result of the node is normal according to the health check response; or,

and if the health check response is an abnormal response message, determining that the health check result of the node is abnormal according to the health check response.

7. The method of claim 1, wherein after sending the health check request carrying the identification information to the first load balancing device, the method further comprises:

and if the health check response corresponding to the identification information returned by the first load balancing equipment is not received, determining that the health check result of the node is abnormal.

8. A node health check method is characterized in that a load balancing device set comprises a first load balancing device and a plurality of second load balancing devices, and the method is applied to the first load balancing device and comprises the following steps:

receiving a health check request carrying the identification information of the node and sent by each second load balancing device; wherein, the health check request is used for learning whether the node is abnormal;

inquiring whether a health check response corresponding to the identification information exists in a specified storage medium or not;

9. The method of claim 8, wherein after the query specifies whether a health check reply corresponding to the identification information exists in the storage medium, the method further comprises:

if a health check response exists and the health check response is an abnormal response message, sending the abnormal response message to the second load balancing equipment; wherein the exception response message is used for the second load balancing device to determine that the health check result of the node is a node exception; or,

refusing to send the exception response message to the second load balancing device.

10. The method of claim 8, wherein after the query specifies whether a health check reply corresponding to the identification information exists in the storage medium, the method further comprises:

if the health check response does not exist, the health check request is sent to the node;

if a health examination response returned by the node aiming at the health examination request is received, recording the corresponding relation between the identification information and the health examination response in a specified storage medium;

and if a health check response returned by the node aiming at the health check request is not received, generating an abnormal response message, and recording the corresponding relation between the identification information and the abnormal response message in a specified storage medium.

11. The method of claim 10, further comprising:

setting an aging timer for the corresponding relation;

and when the aging timer is overtime, deleting the corresponding relation.

12. The method of claim 8, further comprising:

when the first load balancing device needs to perform health check on the node, inquiring whether a specified storage medium has a health check response corresponding to the identification information of the node;

if the health check response exists and the health check response is a normal response message, determining that the health check result of the node is normal; if a health check answer exists and the health check answer is an exception answer message, determining that the health check result of the node is a node exception.

13. A node health check method, wherein a load balancing device set comprises a plurality of load balancing devices, and the method is applied to each second load balancing device in the load balancing device set, and the method comprises:

sending a health check request to first load balancing equipment, wherein the health check request carries identification information of a node, so that the first load balancing equipment acquires a health check response corresponding to the identification information; wherein, the health check request is used for learning whether the node is abnormal; if the specified storage medium of the first load balancing device has a health check response corresponding to the identification information, the first load balancing device acquires the health check response corresponding to the identification information from the specified storage medium;

14. A node health check apparatus, wherein a set of load balancing devices comprises a plurality of load balancing devices, the apparatus being applied to each second load balancing device in the set of load balancing devices, the apparatus comprising:

the selection module is used for selecting first load balancing equipment from a load balancing equipment set according to the identification information of the node, wherein the first load balancing equipment is used for transferring the health check request; wherein, the health check request is used for learning whether the node is abnormal;

a receiving module, configured to receive a health check response corresponding to the identification information, where the health check response is returned by the first load balancing device; if the specified storage medium of the first load balancing device has a health check response corresponding to the identification information, returning the health check response in the specified storage medium;

15. The apparatus according to claim 14, wherein the selecting module, when selecting the first load balancing device from the set of load balancing devices according to the identification information of the node, is specifically configured to:

based on a specific Hash algorithm, selecting first load balancing equipment from a load balancing equipment set according to the identification information of the nodes; wherein the particular hash algorithm comprises a consistent hash algorithm.

16. The apparatus according to claim 14, wherein the selecting module is specifically configured to, when selecting the first load balancing device from the set of load balancing devices according to the identification information of the node:

17. A node health check apparatus, wherein a load balancing device set comprises a first load balancing device and a plurality of second load balancing devices, and the apparatus is applied to the first load balancing device, and the apparatus comprises:

the receiving module is used for receiving the health check request sent by each second load balancing device; the health examination request carries identification information of a node; wherein, the health check request is used for learning whether the node is abnormal;

18. The apparatus of claim 17,

the sending module is further configured to send the abnormal response message to the second load balancing device when a health check response exists and the health check response is an abnormal response message; wherein the exception response message is used for the second load balancing device to determine that the health check result of the node is a node exception; or refusing to send the abnormal response message to the second load balancing equipment.

19. The apparatus of claim 17, wherein the sending module is further configured to send the health check request to the node if there is no health check reply;

the device further comprises: a recording module, configured to record, in a specified storage medium, a correspondence between the identification information and a health check response when a health check response returned by a node for the health check request is received, where the health check response includes a normal response message or an abnormal response message; and when a health check response returned by the node aiming at the health check request is not received, generating an abnormal response message, and recording the corresponding relation between the identification information and the abnormal response message in a specified storage medium.

20. A node health check apparatus, wherein a set of load balancing devices comprises a plurality of load balancing devices, the apparatus being applied to each second load balancing device in the set of load balancing devices, the apparatus comprising:

a sending module, configured to send a health check request to a first load balancing device, where the health check request carries identification information of a node, so that the first load balancing device obtains a health check response corresponding to the identification information; wherein, the health check request is used for learning whether the node is abnormal; if the specified storage medium of the first load balancing device has a health check response corresponding to the identification information, the first load balancing device acquires the health check response corresponding to the identification information from the specified storage medium;

the receiving module is used for receiving a health check response returned by the first load balancing equipment;

21. A load balancing device, wherein a set of load balancing devices comprises a plurality of load balancing devices, and the load balancing device is each second load balancing device in the set of load balancing devices, and the load balancing device comprises a processor and a machine-readable storage medium, the machine-readable storage medium having stored thereon a plurality of computer instructions, which when executed by the processor performs the following processes: selecting a first load balancing device from a load balancing device set according to the identification information of the node, wherein the first load balancing device is used for transferring the health check request; wherein, the health check request is used for learning whether the node is abnormal; sending a health check request carrying the identification information to the first load balancing device; receiving a health check response corresponding to the identification information returned by the first load balancing equipment; if the specified storage medium of the first load balancing device has a health check response corresponding to the identification information, returning the health check response in the specified storage medium; and determining the health check result of the node according to the health check response.

22. A load balancing device, wherein a set of load balancing devices includes a first load balancing device and a plurality of second load balancing devices, the load balancing device is the first load balancing device, the load balancing device includes a processor and a machine-readable storage medium, the machine-readable storage medium has a number of computer instructions stored thereon, and the processor executes the computer instructions to perform the following processes: receiving a health check request carrying the identification information of the node and sent by each second load balancing device; wherein, the health check request is used for learning whether the node is abnormal; inquiring whether a health check response corresponding to the identification information exists in a specified storage medium or not; if a health check response exists and the health check response is a normal response message, sending the normal response message to the second load balancing equipment; wherein the normal response message is used for the second load balancing device to determine that the health check result of the node is that the node is normal.

23. A load balancing device, wherein a set of load balancing devices comprises a plurality of load balancing devices, and the load balancing device is a second load balancing device in the set of load balancing devices, and the load balancing device comprises a processor and a machine-readable storage medium, the machine-readable storage medium has a number of computer instructions stored thereon, and the processor executes the computer instructions to perform the following processes: sending a health check request to first load balancing equipment, wherein the health check request carries identification information of a node, so that the first load balancing equipment acquires a health check response corresponding to the identification information; wherein, the health check request is used for learning whether the node is abnormal; if the specified storage medium of the first load balancing device has a health check response corresponding to the identification information, the first load balancing device acquires the health check response corresponding to the identification information from the specified storage medium; receiving a health check response returned by the first load balancing equipment; if the health check response is a normal response message, determining that the health check result of the node is that the node is normal; or if the health check response is an abnormal response message, determining that the health check result of the node is abnormal.