A fault processing method, device and system of an application server cluster

Abstract

This application provides a method, apparatus, and system for handling application server cluster faults. In this method, a monitoring system sends health check messages to the application server cluster, determines the health rate of the application server cluster based on the response messages from the cluster, and when the health rate is determined to be lower than a preset health rate, it determines whether the application server cluster meets preset fading-out conditions based on the number of virtual servers fading in from the link load balancer. If the preset fading-out conditions are met, the application server cluster is determined to be faulty, and fading-out control information is sent to the link load balancer to control its fading-out of the application server cluster. This solves the problem in existing technologies where maintenance personnel need to input commands on management devices based on alarm information to identify and fade-out faulty application server clusters, resulting in low efficiency and accuracy in fault handling.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a fault handling method, apparatus and system for application server clusters. Background Technology

[0002] With the development of computer technology, the access volume of application servers has exploded, and the requirements for the processing power and computing intensity of application servers have also increased accordingly. Therefore, multiple application servers can be connected together by setting up an application server cluster to improve their external computing power.

[0003] In existing technologies, for application server clusters, a load balancer with virtual servers can be configured and deployed. The load balancer can receive request packets sent by clients and, based on a load balancing algorithm, send the request packets to a real application server in the application server cluster referenced by the virtual server, so as to achieve fast processing of request packets.

[0004] However, when an application server cluster fails, existing technology requires maintenance personnel who are on duty to log into the management device connected to the load balancer after receiving the alarm information and input a series of commands based on their personal experience in order to identify the faulty application server cluster from multiple application server clusters and perform a fade-out process on the faulty application server cluster. This results in low efficiency and accuracy in handling application server cluster failures. Summary of the Invention

[0005] This application provides a method, apparatus, and system for troubleshooting application server clusters, which addresses the problem that existing technologies require maintenance personnel to log into a management device connected to a load balancer after receiving alarm information and input a series of commands based on their personal experience in order to identify the faulty application server cluster from multiple application server clusters and perform fade-out processing on the faulty application server cluster, resulting in low efficiency and accuracy in troubleshooting application server clusters.

[0006] Firstly, this application provides a fault handling method for an application server cluster, comprising:

[0007] The monitoring system sends health check messages to the application servers in the application server cluster, and determines the health rate of the application server cluster based on the response messages returned by the application server cluster.

[0008] When the monitoring system determines that the health rate of the application server cluster is less than the preset health rate, it determines whether the application server cluster meets the preset exit conditions based on the number of virtual servers being phased in fed back by the link load balancer.

[0009] When the monitoring system determines that the application server cluster meets the preset fade-out conditions, it identifies a fault in the application server cluster and sends fade-out control information to the link load balancer so that the link load balancer can perform fade-out processing on the application server cluster according to the fade-out control information.

[0010] In the preferred embodiment of the above-mentioned fault handling method for application server clusters, when the monitoring system determines that the health rate of the application server cluster is less than a preset health rate, it determines whether the application server cluster meets the preset exit conditions based on the number of virtual servers being phased in fed back by the link load balancer, including:

[0011] When the monitoring system determines that the health rate of the application server cluster is less than the preset health rate, it sends a query message to the link load balancer so that the link load balancer can determine the number of virtual servers in the link load balancer resource pool in the gradual entry state based on the query message, and then feed back the number of virtual servers in the gradual entry state to the monitoring system.

[0012] The monitoring system determines whether the number of virtual servers being gradually introduced is greater than a preset threshold based on the number of virtual servers being gradually introduced as reported by the link load balancer.

[0013] When the monitoring system determines that the number of virtual servers entering the system is greater than the preset entry threshold, it determines that the application server cluster meets the preset exit condition.

[0014] In the preferred embodiment of the above-mentioned fault handling method for application server clusters, the monitoring system sends health check messages to the application servers in the application server cluster, and determines the health rate of the application server cluster based on the response messages returned by the application server cluster, including:

[0015] The proxy server in the monitoring system sends health check messages to each application server in the application server cluster, and determines the health rate of the application server cluster based on whether each application server in the cluster responds with a response message within a preset time period; or,

[0016] The proxy server in the monitoring system sends health check messages to each application server in the application server cluster, and determines the health rate of the application server cluster based on the status code in the response message from each application server and a preset status code; or,

[0017] The proxy server in the monitoring system sends health check messages to each application server in the application server cluster, and determines the health rate of the application server cluster based on whether each application server in the application server cluster sends a response message within the preset time period, and the status code in the response message and the preset status code.

[0018] The proxy server in the monitoring system obtains the request message sent by the monitoring platform in the monitoring system, and sends the health rate corresponding to the application server cluster to the monitoring platform.

[0019] In the preferred embodiment of the above-mentioned fault handling method for application server clusters, the monitoring system sends health check messages to the application servers in the application server cluster, and determines the health rate of the application server cluster based on the response messages returned by the application server cluster, including:

[0020] The monitoring platform in the monitoring system identifies one of the application servers in the application server cluster as a proxy server and triggers the proxy server to send a health check message to other servers in the application server cluster.

[0021] The monitoring platform determines the health rate of the application server cluster based on the response messages received and sent by the proxy server from other application servers.

[0022] The preferred technical solution for the above-mentioned application server cluster fault handling method also includes:

[0023] The monitoring system sends fault alarm information to the terminal device and forwards the fade-out processing results generated and sent by the link load balancer to the terminal device.

[0024] Secondly, this application provides a monitoring system, comprising:

[0025] The proxy server is used to send health check messages to the application servers in the application server cluster, and determine the health rate of the application server cluster based on the response messages returned by the application server cluster.

[0026] The monitoring platform is used to determine whether the application server cluster meets the preset exit conditions based on the number of virtual servers being phased in, as fed back by the link load balancer, when it is determined that the health rate of the application server cluster is less than the preset health rate.

[0027] The monitoring platform is also used to determine that the application server cluster is faulty when it is determined that the application server cluster meets the preset fade-out conditions, and to send fade-out control information to the link load balancer so that the link load balancer performs fade-out processing on the application server cluster according to the fade-out control information.

[0028] In the preferred technical solution of the above-mentioned monitoring system, the monitoring platform is specifically used for:

[0029] When it is determined that the health rate of the application server cluster is less than the preset health rate, a query message is sent to the link load balancer so that the link load balancer can determine the number of virtual servers in the link load balancer resource pool in the gradual entry state based on the query message, and report the gradual entry number of the virtual servers to the monitoring platform.

[0030] Based on the number of virtual servers gradually introduced as reported by the link load balancer, determine whether the number of virtual servers gradually introduced is greater than a preset gradual introduction threshold.

[0031] When it is determined that the number of virtual servers entering the application server cluster is greater than the preset entry threshold, the application server cluster is determined to meet the preset exit condition.

[0032] In the preferred technical solution of the above monitoring system, the proxy server is specifically used for:

[0033] A health check message is sent to each application server in the application server cluster, and the health rate of the application server cluster is determined based on whether each application server in the cluster responds with a response message within a preset time period; or,

[0034] A health check message is sent to each application server in the application server cluster, and the health rate of the application server cluster is determined based on the status code in the response message from each application server and a preset status code; or,

[0035] A health check message is sent to each application server in the application server cluster, and the health rate of the application server cluster is determined based on whether each application server in the application server cluster sends a response message within the preset time period, and the status code in the response message and the preset status code.

[0036] Obtain the request message sent by the monitoring platform in the monitoring system, and send the health rate corresponding to the application server cluster to the monitoring platform.

[0037] In the preferred embodiment of the above-mentioned monitoring system, the monitoring platform is further used for:

[0038] The fault alarm information is sent to the terminal device, and the fade-out processing result generated and sent by the link load balancer is forwarded to the terminal device.

[0039] Thirdly, this application provides a monitoring platform, including:

[0040] The transceiver module is used to identify an application server in the application server cluster as a proxy server and trigger the proxy server to send health check messages to other application servers in the application server cluster.

[0041] The processing module is used to determine the health rate of the application server cluster based on the response messages received and sent by the proxy server from the other application servers.

[0042] The processing module is also used to determine whether the application server cluster meets the preset phase-out condition based on the number of virtual servers phased in fed back by the link load balancer when it is determined that the health rate of the application server cluster is less than the preset health rate.

[0043] The transceiver module is further configured to determine that the application server cluster is faulty when it is determined that the application server cluster meets the preset fade-out conditions, and send fade-out control information to the link load balancer so that the link load balancer performs fade-out processing on the application server cluster according to the fade-out control information.

[0044] In the preferred technical solution of the above-mentioned monitoring platform, the processing module is specifically used for:

[0045] When it is determined that the health rate of the application server cluster is less than the preset health rate, a query message is sent to the link load balancer so that the link load balancer can determine the number of virtual servers in the link load balancer resource pool in the gradual entry state based on the query message, and report the gradual entry number of the virtual servers to the monitoring platform.

[0046] Based on the number of virtual servers gradually introduced as reported by the link load balancer, determine whether the number of virtual servers gradually introduced is greater than a preset gradual introduction threshold.

[0047] When it is determined that the number of virtual servers entering the application server cluster is greater than the preset entry threshold, the application server cluster is determined to meet the preset exit condition.

[0048] In the preferred technical solution of the above-mentioned monitoring platform, the transceiver module is further used for:

[0049] The fault alarm information is sent to the terminal device, and the fade-out processing result generated and sent by the link load balancer is forwarded to the terminal device.

[0050] Fourthly, this application provides a fault handling system for an application server cluster, comprising:

[0051] Link load balancers, application server clusters, and monitoring systems as described in the second aspect; or,

[0052] The link load balancer, the application server cluster, and the monitoring platform as described in the third aspect.

[0053] This application provides a fault handling method, apparatus, and system for an application server cluster. In this method, a monitoring system sends health check messages to the application servers in the application server cluster, determines the health rate of the application server cluster based on the response messages from the application server cluster, and when it is determined that the health rate of the application server cluster is less than a preset health rate, it determines whether the application server cluster meets the preset fade-out conditions based on the number of virtual servers faded in from the link load balancer. When it is determined that the application server cluster meets the preset fade-out conditions, the application server cluster is determined to be faulty, and fade-out control information is sent to the link load balancer so that the link load balancer performs fade-out processing on the application server cluster according to the fade-out control information. Compared to existing technologies that require maintenance personnel to log into the management device connected to the load balancer after receiving alarm information and input a series of commands based on personal experience to identify the faulty application server cluster from multiple application server clusters and perform wiping out the faulty application server cluster, resulting in low efficiency and accuracy in application server cluster fault handling, the monitoring system in this application can generate wiping out control information when it is determined that the health rate of the corresponding application server cluster is greater than the preset health rate and the application server cluster meets the preset wiping out conditions. This information controls the link load balancer to perform wiping out the application server cluster, thus eliminating the need for maintenance personnel during working hours to perform wiping out the faulty application server cluster based on the reminder information. It can automatically achieve timely detection and accurate handling of faulty application server clusters, solving the problem of low efficiency and accuracy in application server cluster fault handling caused by the existing technology that requires maintenance personnel to log into the management device connected to the load balancer after receiving alarm information and input a series of commands based on personal experience. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0055] Figure 1 A structural diagram of an application service system provided in this application;

[0056] Figure 2 A flowchart illustrating an embodiment of a fault handling method for an application server cluster provided in this application;

[0057] Figure 3 A flowchart illustrating a second embodiment of a fault handling method for an application server cluster provided in this application;

[0058] Figure 4 A flowchart illustrating a third embodiment of a fault handling method for an application server cluster provided in this application;

[0059] Figure 5 A flowchart illustrating Embodiment 4 of a fault handling method for an application server cluster provided in this application;

[0060] Figure 6 A flowchart illustrating Embodiment 5 of a fault handling method for an application server cluster provided in this application;

[0061] Figure 7 A flowchart illustrating Embodiment Six of a fault handling method for an application server cluster provided in this application;

[0062] Figure 8 A flowchart illustrating Embodiment Seven of a fault handling method for an application server cluster provided in this application;

[0063] Figure 9 A flowchart illustrating an eighth embodiment of a fault handling method for an application server cluster provided in this application;

[0064] Figure 10 A flowchart illustrating Embodiment Nine of a fault handling method for an application server cluster provided in this application;

[0065] Figure 11 A flowchart illustrating Embodiment 10 of the application server cluster fault handling method provided in this application;

[0066] Figure 12 A schematic diagram of the structure of a monitoring system provided in this application;

[0067] Figure 13 A schematic diagram of the structure of a monitoring platform provided in this application;

[0068] Figure 14A schematic diagram of the structure of an application server cluster fault handling system provided in this application;

[0069] Figure 15 A schematic diagram of the structure of another application server cluster fault handling system provided in this application. Detailed Implementation

[0070] To make the technical solutions and advantages of the embodiments of this application clearer, the technical solutions in the embodiments will be clearly and completely described below with reference to the accompanying drawings. The described embodiments are only some embodiments of this application, not all embodiments. Other embodiments made by those skilled in the art based on the embodiments of this application and under the guidance of these embodiments are all within the scope of protection of this application.

[0071] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0072] First, let me explain the terms used in this application:

[0073] Load balancing is the process of balancing and distributing workloads across multiple operational units (such as web application servers) to complete the tasks together.

[0074] Load Balancer: Typically existing as a standalone hardware component, load balancers offer excellent load balancing performance and employ numerous load balancing strategies to ensure request distribution in a relatively optimal manner. Common load balancers include F5 and BIGIP.

[0075] Server load balancer: It can be deployed directly in front of the web application server to achieve load balancing for the web application server.

[0076] Link load balancer: can be deployed in front of server load balancer to achieve simple Layer 4 server load balancing functionality.

[0077] Virtual Server (VS): Deployed on a load balancer. Client requests are not sent to the IP address and port of the application deployed on the actual application server, but rather to the IP address and port of the virtual server. A virtual server deployed on a server load balancer is called a Local Traffic Management Virtual Server (LTM-VS), which references multiple application servers in the application server cluster. A virtual server deployed on a link load balancer is called a Link Controller Virtual Server (LC-VS).

[0078] A Server Load Balancing Resource Pool (LTM-pool) is a collection of IP addresses and ports corresponding to application servers (or applications deployed on application servers) within an application server cluster. It has a one-to-one mapping relationship with the LTM-VS. When a client's request packet arrives at the LTM-VS, the server load balancer, based on the established VS-pool mapping, translates the destination IP address and port of the request packet into the IP address and port of a specific LTM-pool member, and then forwards the request packet to that member.

[0079] Link Load Balancing Resource Pool (LC-pool): This pool contains all the virtual servers corresponding to the application server clusters in the application service system, and is a collection of LTM-VSs. Based on the established VS-pool mapping relationship, the link load balancer translates the destination IP address and port of a request packet into the IP address and port of a specific LC-pool member, and then sends the request packet to that member (LTM-VS).

[0080] Gradual access: The link load balancer forwards the request packets sent by the client to the application server (or the application deployed on the application server) in the application server cluster through the server load balancer, so that the user can access the server (or the application deployed on the application server) in the application server cluster.

[0081] Emerging: The link load balancer fails to forward the client's request packet to the application server (or the application deployed on the application server) through the server load balancer, making it impossible for the user to access the application server (or the application deployed on the application server). In other words, the link load balancer isolates the application server (or application).

[0082] Socket connection: Establishing a socket connection requires at least one pair of sockets, one running on the client side and the other on the server side. The connection process between sockets consists of three steps: server listening, client request, and connection confirmation.

[0083] Existing technology requires maintenance personnel to receive a reminder message during working hours, log in to the management device connected to the link load balancer, and identify the faulty application server cluster from multiple application server clusters based on their personal experience. Then, they need to perform a fade-out process on the faulty application server cluster, which results in low efficiency and accuracy in handling application server cluster faults.

[0084] Based on the above-mentioned technical problems, the technical concept of this application is: how to realize the automatic detection and handling of application server cluster failures.

[0085] Figure 1 This is a structural diagram of an application service system provided in this application. Figure 1 As shown, the application service system includes a monitoring system 101, a link load balancer 102, and a server load balancer (…). Figure 1 Two server load balancers are shown, namely server load balancer 103 and server load balancer 104, and an application server cluster ( Figure 1 Two application server clusters are shown, each application server cluster including two application servers, namely application server 105, application server 106, application server 107 and application server 108. Server load balancer 103 is communicatively connected to application servers 105 and 106 belonging to the same application server cluster, and server load balancer 104 is communicatively connected to application servers 107 and 108 belonging to the other application server cluster.

[0086] It should be noted that, Figure 1 This is merely a structural diagram of an emergency service system provided in this application; this application does not constitute an indication of any intention to provide a specific solution. Figure 1 The actual form and interaction method of the various devices included are limited.

[0087] The following section provides a detailed description of the fault handling scheme for the application server cluster in this application. The following specific embodiments can be combined with each other, and similar concepts or processes may not be repeated in some embodiments.

[0088] Figure 2 This is a flowchart illustrating an embodiment of a fault handling method for an application server cluster provided in this application. See also... Figure 2 The troubleshooting method for this application server cluster specifically includes the following steps:

[0089] S201: The monitoring system sends health check messages to the application servers in the application server cluster, and determines the health rate of the application server cluster based on the response messages returned by the application server cluster.

[0090] In this embodiment, since each application server in the application server cluster corresponds to a port and IP address, the monitoring system can send health check messages to each application server based on its port and IP address. It should be noted that the monitoring system can periodically send health check messages to the application servers on a cluster-by-cluster basis.

[0091] The monitoring system can determine that the health check result of the application server that responded to the message is healthy when it receives a response message containing the same status code as a preset status code, or when it receives a response message containing the same status code as a preset status code within a preset time period, and increment the number of application servers in a healthy state by 1. After obtaining the health check results of all application servers, the monitoring system can determine the health rate of the application server cluster based on the number of application servers in the cluster and the number of application servers in a healthy state.

[0092] S202: The monitoring system determines whether the health rate of the application server cluster is less than the preset health rate.

[0093] In this embodiment, after obtaining the health rate of the application server cluster, the monitoring system can determine whether the health rate of the application server cluster is less than a preset health rate. If the monitoring system determines that the health rate of the application server cluster is less than the preset health rate, it executes step S203; if it determines that the health rate of the application server cluster is greater than or equal to the preset health rate, it terminates the process.

[0094] S203: The monitoring system determines whether the application server cluster meets the preset exit conditions based on the number of virtual servers entering the network as reported by the link load balancer.

[0095] In this embodiment, when the monitoring system determines that the health rate of the application server cluster is less than the preset health rate, it can determine whether the application server cluster meets the preset exit conditions based on the number of virtual servers entering the network as reported by the link load balancer based on query information. If the monitoring system determines that the application server cluster meets the preset exit conditions, it executes step S204; if it determines that the application server cluster does not meet the preset exit conditions, it terminates the process.

[0096] S204: The monitoring system determines that the application server cluster is faulty and sends fading control information to the link load balancer so that the link load balancer can perform fading out of the application server cluster according to the fading control information.

[0097] In this embodiment, when the monitoring system determines that the application server cluster meets the preset fading conditions, it identifies the application server cluster as a faulty application server cluster. The monitoring system can send fading control information to the link load balancer, so that the link load balancer, according to the fading control information, fades out the virtual servers to be faded from the link load balancing resource pool and isolates the application server cluster in the registry center; wherein, the virtual servers to be faded are virtual servers deployed on the server load balancer corresponding to the application server cluster.

[0098] In this embodiment, the monitoring system sends health check messages to the application servers in the application server cluster, and determines the health rate of the application server cluster based on the response messages returned by the application server cluster. When the monitoring system determines that the health rate of the application server cluster is less than the preset health rate, it determines whether the application server cluster meets the preset fade-out conditions based on the number of virtual servers faded in by the link load balancer. When it determines that the preset fade-out conditions are met, it determines that the application server cluster is faulty and sends fade-out control information to the link load balancer to control the fade-out of the application server cluster. Compared to existing technologies that require maintenance personnel to log into the device connected to the load balancer after receiving alarm information and input a series of commands based on personal experience in order to identify the faulty application server cluster from multiple application server clusters and perform fade-out processing on the faulty application server cluster, resulting in low efficiency and accuracy in application server cluster fault handling, the monitoring system in this application can generate fade-out control information when it is determined that the health rate of the corresponding application server cluster is greater than the preset health rate and the application server cluster meets the preset fade-out conditions. This information controls the link load balancer to perform fade-out processing on the application server cluster, thereby enabling automatic discovery and handling of application server cluster faults. This solves the problem of low efficiency and accuracy in application server cluster fault handling caused by existing technologies that require maintenance personnel to log into the device connected to the load balancer after receiving alarm information and input a series of commands based on personal experience in order to identify the faulty application server cluster from multiple application server clusters and perform fade-out processing on the faulty application server cluster.

[0099] Based on the above method embodiment one, the following method embodiment two will further explain S203.

[0100] Figure 3This is a flowchart illustrating a second embodiment of a fault handling method for an application server cluster provided in this application. See also... Figure 3 The troubleshooting method for this application server cluster specifically includes the following steps:

[0101] S301: The monitoring system sends a query to the link load balancer so that the link load balancer can determine the number of virtual servers in the link load balancer resource pool that are in the gradual entry state based on the query information, and report the gradual entry number of virtual servers back to the monitoring system.

[0102] In this embodiment, when the link load balancer sets the application state of a certain application server cluster to a progressive state, the virtual server LTM-VS corresponding to that application server cluster in the link load balancer's link load balancing resource pool (LC-pool) is also in a progressive state. At this time, the link load balancer can send request packets sent by clients to that application server cluster. It should be noted that during the construction of the application server cluster, all virtual servers LTM-VS corresponding to the application server clusters are in a progressive state.

[0103] When the monitoring system determines that the health rate of the application server cluster is lower than the preset health rate, it can send a query to the link load balancer. The link load balancer can then query the number of virtual servers in the link load balancer's resource pool that are in a lag state. After obtaining the number of lag-in virtual servers, the link load balancer can send this number back to the monitoring system.

[0104] S302: The monitoring system determines whether the number of virtual servers entering the network exceeds the preset threshold based on the number of virtual servers entering the network as reported by the link load balancer.

[0105] In this embodiment, to ensure that the application service system can provide a certain level of service capability, i.e., handle a large number of request packets sent by clients, and to prevent a large number of application server clusters from being rolled out and affecting external service capabilities, the monitoring system pre-sets a minimum number of application server clusters that can handle request packets, i.e., sets a preset rollout threshold for virtual servers. After obtaining the number of virtual servers rolled in from the link load balancer, the monitoring system determines whether the number of virtual servers rolled in is greater than the preset rollout threshold. If the number of virtual servers rolled in is greater than the preset rollout threshold, S303 is executed; if the number of virtual servers rolled in is less than or equal to the preset rollout threshold, the process ends.

[0106] S303: The monitoring system determines that the application server cluster meets the preset fade-out conditions.

[0107] In this embodiment, when the monitoring system determines that the number of virtual servers entering the application server cluster is greater than the preset entry threshold, it can determine that the application server cluster meets the preset exit conditions, and at this time the application server cluster can be exited.

[0108] In this embodiment, the monitoring system sends query information to the link load balancer, enabling the link load balancer to determine the number of virtual servers in the link load balancing resource pool that are in a fading-in state based on the query information, and then reports the number of fading-in virtual servers back to the monitoring system. The monitoring system, based on the number of fading-in virtual servers reported by the link load balancer, determines whether the number of fading-in virtual servers exceeds a preset fading-in threshold. If the number of fading-in virtual servers exceeds the preset fading-in threshold, the monitoring system determines that the application server cluster meets the preset fading-out conditions. Compared to the prior art, which only determines whether the application server cluster has failed after obtaining its health rate, potentially leading to a large number of application server clusters being faded out and affecting the external service capabilities of the application service system, this application can determine whether the application server cluster meets the preset fading-out conditions based on the number of fading-in virtual servers when the health rate of the application server cluster is determined to be less than the preset health rate. This improves the fading-out limit of the application server cluster and avoids the problem of a large number of application server clusters being faded out, thus affecting the external service capabilities of the application service system.

[0109] Based on the above method embodiment one, the following describes a specific implementation of S201 through method embodiment three.

[0110] Figure 4 This is a flowchart illustrating a third embodiment of a fault handling method for an application server cluster provided in this application. See also... Figure 4 The troubleshooting method for this application server cluster specifically includes the following steps:

[0111] S401: The agent server in the monitoring system sends health check messages to the application servers in the application server cluster, and determines the health rate of the application server cluster based on the response messages returned by the application server cluster.

[0112] In this embodiment, the monitoring system includes a proxy server and a monitoring platform, wherein a health check program runs on the proxy server and a monitoring program runs on the monitoring platform.

[0113] The proxy server sends health check messages to each application server based on its port and IP address. When the proxy server receives a response message containing the same status code as a preset status code, or receives a response message within a preset time period, it determines that the application server that responded with the response message has a healthy health check result and increments the number of healthy application servers by 1.

[0114] After obtaining the health check results of all application servers, the proxy server can determine the health rate of the application server cluster based on the number of application servers in the cluster and the number of application servers in a healthy state.

[0115] S402: The proxy server in the monitoring system obtains the request message sent by the monitoring platform in the monitoring system and sends the health rate of the application server cluster to the monitoring platform.

[0116] When the proxy server in the monitoring system receives a request message from the monitoring platform, it can report the health rate of the application server cluster to the monitoring platform so that the monitoring platform can determine whether the health rate of the application server cluster is less than the preset health rate.

[0117] In this embodiment, the monitoring system may include a monitoring platform and a proxy server. The proxy server sends health check messages to the application servers in the application server cluster and determines the health rate of the application server cluster based on the response messages from the application server cluster. After receiving a request message from the monitoring platform, the proxy server can send the health rate of the application server cluster to the monitoring platform. Based on the previous embodiments, this embodiment further divides the monitoring system into a monitoring platform and a proxy server, thereby distributing the fault handling tasks for the application server cluster to different devices for execution, thus improving the fault handling efficiency of the application server cluster.

[0118] Based on the above method embodiment three, the following method embodiment four will describe a specific implementation of the proxy server determining the health rate of the application server in S401.

[0119] Figure 5 This is a flowchart illustrating Embodiment 4 of a fault handling method for an application server cluster provided in this application. See also... Figure 5 The troubleshooting method for this application server cluster specifically includes the following steps:

[0120] S501: The agent server in the monitoring system sends health check messages to each application server in the application server cluster.

[0121] S502: The agent server in the monitoring system determines the health rate of the application server cluster based on whether each application server in the application server cluster sends a response message within a preset time period.

[0122] In this embodiment, the proxy server can start timing when sending a health check message and determine whether it receives a response message with the source IP address and source port of the application server within a preset time period. When the proxy server receives a response message with the source IP address and source port of the application server within the preset time period, it can determine that the health check result of the application server is healthy and increment the number of application servers in a healthy state by 1.

[0123] After obtaining the health check results of all application servers, the proxy server can calculate the health rate of the application server cluster based on the number of application servers in a healthy state and the total number of application servers in the application server cluster.

[0124] Taking an application server cluster consisting of 6 application servers as an example, when the proxy server receives a response packet within a preset time period whose source IP address and source port are the same as those of an application server, it determines that the health check result of the application server corresponding to that source IP address and source port is healthy, and increments the number of application servers in a healthy state by 1. After obtaining the health check results of the 6 application servers, the proxy server can calculate the health rate of the application server cluster as 50%, based on the fact that the number of application servers in a healthy state is 3 and the application server cluster consists of 6 application servers.

[0125] In this embodiment, the proxy server sends health check messages to each application server in the application server cluster, and determines the health rate of the application server cluster based on whether each application server in the application server cluster responds with a response message within a preset time period, thereby improving the efficiency and accuracy of determining the health rate of application servers.

[0126] Based on the above method embodiment three, the following describes another specific implementation of the proxy server determining the health rate of the application server in S401 through method embodiment five.

[0127] Figure 6 This is a flowchart illustrating Embodiment 5 of a fault handling method for an application server cluster provided in this application. See also... Figure 6 The troubleshooting method for this application server cluster specifically includes the following steps:

[0128] S601: The agent server in the monitoring system sends health check messages to each application server in the application server cluster.

[0129] S602: The agent server in the monitoring system determines the health rate of the application server cluster based on the status code in the response message returned by each application server in the application server cluster and the preset status code.

[0130] In this embodiment, after receiving the response message from each application server, the proxy server can determine whether the status code in each response message matches the preset status code. When it is determined that the status code in the response message matches the preset status code, the health check result of the application server that sent the response message is healthy, and the number of application servers in a healthy state is incremented by 1.

[0131] After obtaining the health check results of all application servers, the proxy server can calculate the health rate of the application server cluster based on the number of application servers in a healthy state and the total number of application servers in the application server cluster.

[0132] In this embodiment, the proxy server in the monitoring system sends health check messages to each application server in the application server cluster, and determines the health rate of the application server cluster based on the status code in the response message returned by each application server and the preset status code, thereby improving the accuracy of determining the health rate of the application server cluster.

[0133] Based on the above method embodiment three, the following method embodiment six will describe another specific implementation of the proxy server determining the health rate of the application server cluster in S401.

[0134] Figure 7 A flowchart illustrating Embodiment Six of a fault handling method for an application server cluster provided in this application. See also... Figure 7 The troubleshooting method for this application server cluster specifically includes the following steps:

[0135] S701: The agent server in the monitoring system sends health check messages to each application server in the application server cluster.

[0136] S702: The agent server in the monitoring system determines the health rate of the application server cluster based on whether each application server in the application server cluster sends a response message within a preset time period, and the status code in the response message matches the preset status code.

[0137] In this embodiment, the proxy server can send health check messages to the application server based on the application server's IP address and port, and start timing when sending the health check message. The proxy server determines whether it has received a response message whose source IP address and source port are the application server's IP address and port within a preset time period. When the proxy server receives a response message whose source IP address and source port are the application server's IP address and port within the preset time period, it can determine whether the status code in the response message matches a preset status code. When the proxy server determines that the status code in the response message matches the preset status code, it can determine that the application server's health check result is healthy and increment the number of application servers in a healthy state by 1. After obtaining the health check results of all application servers, the proxy server can calculate the health rate corresponding to the application server cluster based on the number of application servers in a healthy state and the total number of application servers in the application server cluster.

[0138] In this embodiment, the proxy server in the monitoring system sends health check messages to each application server in the application server cluster. Based on whether each application server in the application server cluster responds with a response message within a preset time period, and the status code in the response message and the preset status code, the health rate of the application server cluster is determined, thereby improving the accuracy and efficiency of determining the health rate of the application server cluster.

[0139] Based on the above method embodiment one, the following describes another specific implementation of S201 through method embodiment seven.

[0140] Figure 8 This is a flowchart illustrating Embodiment Seven of a fault handling method for an application server cluster provided in this application. See also... Figure 8 The troubleshooting method for this application server cluster specifically includes the following steps:

[0141] S801: The monitoring platform in the monitoring system identifies one of the application servers in the application server cluster as a proxy server and triggers the proxy server to send health check messages to other servers in the application server cluster.

[0142] In this embodiment, the monitoring platform can identify any application server in the application server cluster as a proxy server and send a health check request to the proxy server. This allows the proxy server to send health check messages to other servers in the application server cluster, excluding the proxy server, based on the health check request. After receiving the response messages from the other servers, the proxy server forwards the response messages to the monitoring platform.

[0143] S802: The monitoring platform in the monitoring system determines the health rate of the application server cluster based on the response messages from other application servers obtained and sent by the proxy server.

[0144] In this embodiment, the response message sent by each application server includes the port and IP address of the application server that generated and sent the response message. When the monitoring platform receives a response message containing the same status code as a preset status code, or receives a response message within a preset time period, it can determine that the health check result of the application server that sent the response message is healthy based on the port and IP address in the response message, and increment the number of application servers in a healthy state by 1. After obtaining the health check results of all application servers, the monitoring platform determines the health rate corresponding to the application server cluster based on the number of application servers in the application server cluster and the number of application servers in a healthy state.

[0145] In this embodiment, the monitoring platform in the monitoring system identifies one application server in the application server cluster as a proxy server and triggers the proxy server to send health check messages to other servers in the application server cluster. The monitoring platform determines the health rate of the application server cluster based on the response messages received and sent by the proxy server from other application servers. This embodiment can separate the monitoring system into a monitoring platform and a proxy server, and set the proxy server as any application server in the application server cluster to distribute the monitoring load. That is, each application service system can include one proxy server, and multiple application service systems can share one monitoring platform, reducing the number of monitoring platforms deployed while improving the utilization rate of the monitoring platform.

[0146] Figure 9 This is a flowchart illustrating an eighth embodiment of a fault handling method for an application server cluster provided in this application. See also... Figure 9 The troubleshooting method for this application server cluster specifically includes the following steps:

[0147] S901: The monitoring system sends health check messages to the application servers in the application server cluster, and determines the health rate of the application server cluster based on the response messages returned by the application server cluster.

[0148] S902: The monitoring system determines whether the health rate of the application server cluster is less than the preset health rate.

[0149] In this embodiment, when the monitoring system determines that the health rate of the application server cluster is less than the preset health rate, it executes S903; when it determines that the health rate of the application server cluster is greater than or equal to the preset health rate, it terminates.

[0150] S903: The monitoring system determines whether the application server cluster meets the preset exit conditions based on the number of virtual servers entering the network as reported by the link load balancer.

[0151] In this embodiment, when the monitoring system determines that the health rate of the application server cluster is less than the preset health rate, it can determine whether the application server cluster meets the preset exit conditions based on the number of virtual servers entering the network as reported by the link load balancer based on query information. If the monitoring system determines that the application server cluster meets the preset exit conditions, it executes step S904; if it determines that the application server cluster does not meet the preset exit conditions, it terminates the process.

[0152] S904: The monitoring system has determined that the application server cluster is faulty.

[0153] S905: The monitoring system sends fault alarm information to the terminal equipment.

[0154] In this embodiment, after determining that the application server cluster is a faulty application server cluster, the monitoring system can generate fault alarm information and send the fault alarm information to the terminal device so that maintenance personnel are aware that the application server cluster has failed.

[0155] S906: The monitoring system sends fade-out control information to the link load balancer so that the link load balancer can perform fade-out processing on the application server cluster according to the fade-out control information.

[0156] It should be noted that S905 and S906 are different steps that are executed simultaneously.

[0157] S907: The monitoring system forwards the fade-out processing results generated and sent by the link load balancer to the terminal device.

[0158] In this embodiment, the monitoring system can obtain the fade-out processing results generated and sent by the link load balancer, and forward the fade-out processing results to the terminal device so that maintenance personnel know that the link load balancer has performed fade-out processing on the faulty application server cluster.

[0159] Based on Method Embodiment 1, this embodiment allows the monitoring system to generate fault alarm information and send it to terminal devices after determining that the application server cluster is faulty. This alerts maintenance personnel or the application server cluster itself to the fault, enabling them to promptly investigate and resolve the issue. Furthermore, this embodiment can also acquire the fading-out result generated by the link load balancer after performing fading-out processing on the application server cluster and forward it to the terminal devices. This informs maintenance personnel that the link load balancer has performed fading-out processing on the faulty application server cluster, thus avoiding the problem of the link load balancer repeatedly performing fading-out processing due to maintenance personnel sending fading-out control information to the link load balancer via the terminal devices after receiving fault alarm information.

[0160] Based on the aforementioned method embodiment six, the following method embodiment nine will describe the specific implementation of determining the health rate of the application server cluster based on the application-level proxy server.

[0161] Figure 10 A flowchart illustrating Embodiment Nine of the application server cluster fault handling method provided in this application is shown below. Figure 10 As shown, the fault handling method for this application server cluster specifically includes the following steps:

[0162] S1001: The proxy server sets the number of applications in a healthy state and the number of applications in a faulty state to zero.

[0163] In this embodiment, each application server cluster may include multiple application servers, and each application server may deploy at least one application. Each application corresponds to a unique IP address and port. Initially, the proxy server can set both the number of applications in a healthy state and the number of applications in a faulty state to zero.

[0164] S1002: The proxy server identifies an application deployed on an application server in the application server cluster as the target application, establishes a socket connection to the target application, and sends a health check message to the target application.

[0165] In this embodiment, the proxy server can designate an application as the target application and establish a socket connection with that application. The proxy server can also send health check messages to the target application based on its IP address and port.

[0166] S1003: The proxy server determines whether a socket connection has been successfully established with the target application and receives a response message from the target application within a preset time period.

[0167] In this embodiment, the proxy server determines whether a socket connection has been successfully established with the target application and receives a response message from the target application within a preset time period. When the proxy server determines that a socket connection has been successfully established with the target application and receives a response message from the target application within the preset time period, it executes S1004; when it determines that a socket connection has not been successfully established with the target application and / or no response message has been received from the target application within the preset time period, it executes S1005.

[0168] S1004: The proxy server determines that the health check result of the target application is healthy, and increments the number of applications in a healthy state by 1.

[0169] In this embodiment, when the proxy server determines that a socket connection has been successfully established with the target application and receives a response message from the target application within a preset time period, it can intercept the success flag, determine that the health check result of the target application is healthy, and increment the number of applications in a healthy state by 1.

[0170] S1005: The proxy server determines that the health check result of the target application is faulty, and increments the number of applications in a faulty state by 1.

[0171] S1006: The proxy server determines whether to obtain the health check results of all applications.

[0172] In this embodiment, the proxy server can determine whether to obtain the health check results of all applications in the application server cluster. If it determines that the health check results of all applications have been obtained, it executes S1007; if it determines that the health check results of all applications have not been obtained, it returns to S1002.

[0173] S1007: The proxy server determines the health rate of the application server cluster based on the number of applications in a healthy state.

[0174] In this embodiment, after determining that the health check results of all applications have been obtained, the proxy server can calculate the health rate of the application server cluster based on the number of applications in a healthy state and the total number of applications in the application server cluster.

[0175] In this embodiment, the proxy server can determine the health check result of the application based on whether the proxy server has successfully established a socket connection with the application and received the response message from the application, thereby determining the health rate of the application server cluster and improving the accuracy of determining the health rate of the application server cluster.

[0176] Based on Method Implementation Example 3, Method Implementation Example 10 will further explain the specific implementation of the proxy server determining the health rate of the application server cluster in S401.

[0177] Figure 11 A flowchart illustrating Embodiment 10 of the application server cluster fault handling method provided in this application is shown below. Figure 11 As shown, the fault handling method for this application server cluster specifically includes the following steps:

[0178] S1101: The proxy server sets the number of health checks to zero.

[0179] S1102: The proxy server sends a health check message to the application server in the application server cluster, and determines the health rate corresponding to this health check based on the response message returned by the application server cluster, and increments the health check count by 1.

[0180] S1103: The proxy server determines whether the number of health checks has reached the preset number of health checks.

[0181] In this embodiment, the proxy server determines whether the current number of health checks has reached the preset number of health checks. If it determines that the current number of health checks of the proxy server has reached the preset number of health checks, it executes S1104; if it determines that the current number of health checks of the proxy server has not reached the preset number of health checks, it returns to S1102.

[0182] S1104: The proxy server determines whether the health rate corresponding to each health check is less than the preset health rate.

[0183] When the proxy server determines that the health rate corresponding to each health check is not always less than the preset health rate, it executes S1105; when it determines that the health rate corresponding to each health check is always less than the preset health rate, it executes S1106.

[0184] S1105: The proxy server determines whether the health rate corresponding to each health check is greater than or equal to the preset health rate.

[0185] When the proxy server determines that the health rate corresponding to each health check is greater than or equal to the preset health rate, it executes S1106; when it determines that the health rate corresponding to each health check is not greater than or equal to the preset health rate, it returns S1101.

[0186] S1106: The proxy server determines the health rate of the application server cluster based on the health rate corresponding to each health check.

[0187] In this embodiment, the proxy server can calculate the health rate of the application server cluster based on the health rate obtained from each health check.

[0188] In this embodiment, multiple health checks can be performed on the application server cluster, and the health rate of the application server cluster can be determined based on the health rate obtained from the multiple health checks. This avoids the problem of incorrect health check results and improves the accuracy of determining the health rate of the application server cluster.

[0189] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0190] Figure 12 A schematic diagram of the structure of a monitoring system provided in this application; such as Figure 12 As shown, the monitoring system includes a proxy server 1201 and a monitoring platform 1202. The proxy server 1201 sends health check messages to the application servers in the application server cluster and determines the health rate of the application server cluster based on the response messages from the application server cluster. The monitoring platform 1202, when determining that the health rate of the application server cluster is less than a preset health rate, determines whether the application server cluster meets preset fading-out conditions based on the number of virtual servers fading in fed back by the link load balancer. The monitoring platform 1202 is also used to determine that the application server cluster is faulty when it is determined that the application server cluster meets the preset fading-out conditions, and sends fading-out control information to the link load balancer so that the link load balancer can perform fading-out processing on the application server cluster according to the fading-out control information.

[0191] The monitoring system provided in this application embodiment can execute the technical solutions shown in the above method embodiments. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0192] In one possible implementation, the monitoring platform 1202 is specifically configured to: when it is determined that the health rate of the application server cluster is less than a preset health rate, send query information to the link load balancer, so that the link load balancer determines the number of virtual servers in the link load balancer resource pool in the easing state based on the query information, and reports the easing number of virtual servers to the monitoring platform; determine whether the easing number of virtual servers is greater than a preset easing threshold based on the easing number of virtual servers reported by the link load balancer; and determine that the application server cluster meets the preset easing conditions when it is determined that the easing number of virtual servers is greater than the preset easing threshold.

[0193] The monitoring system provided in this application embodiment can execute the technical solutions shown in the above method embodiments. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0194] In one possible implementation, the proxy server 1201 is specifically configured to: send health check messages to each application server in the application server cluster, and determine the health rate of the application server cluster based on whether each application server in the application server cluster sends a response message within a preset time period; or, send health check messages to each application server in the application server cluster, and determine the health rate of the application server cluster based on the status code in the response message sent by each application server in the application server cluster and a preset status code; or, send health check messages to each application server in the application server cluster, and determine the health rate of the application server cluster based on whether each application server in the application server cluster sends a response message within a preset time period, and the status code in the response message sent and the preset status code.

[0195] The monitoring system provided in this application embodiment can execute the technical solutions shown in the above method embodiments. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0196] In one possible implementation, the monitoring platform 1202 is also used to: send fault alarm information to the terminal device and forward the fade-out processing results generated and sent by the link load balancer to the terminal device.

[0197] The monitoring system provided in this application embodiment can execute the technical solutions shown in the above method embodiments. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0198] Figure 13 A schematic diagram of the structure of a monitoring platform provided in this application; such as Figure 13As shown, the monitoring platform includes a transceiver module 1301 and a processing module 1302. The transceiver module 1301 is used to identify an application server in the application server cluster as a proxy server and trigger the proxy server to send health check messages to other application servers in the application server cluster. The processing module 1302 is used to determine the health rate of the application server cluster based on the response messages received and sent by the proxy server from other application servers. The processing module 1302 is also used to determine whether the application server cluster meets preset fading-out conditions based on the number of virtual servers fading in fed back by the link load balancer when the health rate of the application server cluster is determined to be less than a preset health rate. Furthermore, the transceiver module 1301 is used to determine an application server cluster failure when the application server cluster meets the preset fading-out conditions and send fading-out control information to the link load balancer so that the link load balancer performs fading-out processing on the application server cluster according to the fading-out control information.

[0199] The monitoring platform provided in this application embodiment can execute the technical solutions shown in the above method embodiments. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0200] In one possible implementation, the processing module 1302 is specifically used for: when it is determined that the health rate of the application server cluster is less than a preset health rate, sending query information to the link load balancer, so that the link load balancer determines the number of virtual servers in the link load balancer resource pool in the easing state based on the query information, and reports the easing number of virtual servers to the monitoring platform; judging whether the easing number of virtual servers is greater than a preset easing threshold based on the easing number of virtual servers reported by the link load balancer; and determining that the application server cluster meets the preset easing conditions when it is determined that the easing number of virtual servers is greater than the preset easing threshold.

[0201] The monitoring platform provided in this application embodiment can execute the technical solutions shown in the above method embodiments. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0202] In one possible implementation, the transceiver module 1301 is further configured to: send fault alarm information to the terminal device and forward the fade-out processing results generated and sent by the link load balancer to the terminal device.

[0203] The monitoring platform provided in this application embodiment can execute the technical solutions shown in the above method embodiments. Its implementation principle and beneficial effects are similar, and will not be described again here.

[0204] Figure 14 This application provides a schematic diagram of the structure of an application server cluster fault handling system; as shown below. Figure 14As shown, the fault handling system of the application server cluster includes: link load balancer 1401, application server cluster 1402, and monitoring system 1403.

[0205] The application server cluster fault handling system provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0206] Figure 15 A schematic diagram of the structure of another application server cluster fault handling system provided in this application; as shown. Figure 15 As shown, the fault handling system of the application server cluster includes: link load balancer 1501, application server cluster 1502, and monitoring platform 1503.

[0207] The application server cluster fault handling system provided in this application embodiment can execute the technical solution shown in the above method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0208] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for failure handling of an application server cluster, characterized in that, The application comprises the following steps: The monitoring system sends a health check message to an application server in an application server cluster, and determines a health rate corresponding to the application server cluster according to a response message fed back by the application server cluster; When the monitoring system determines that the health rate corresponding to the application server cluster is less than a preset health rate, the monitoring system determines whether the application server cluster meets a preset gradual exit condition according to a gradual entry number of a virtual server fed back by a link load balancer; When the monitoring system determines that the application server cluster meets the preset gradual exit condition, the monitoring system determines that the application server cluster is faulty, and sends gradual exit control information to the link load balancer, so that the link load balancer performs gradual exit processing on the application server cluster according to the gradual exit control information; When the monitoring system determines that the health rate corresponding to the application server cluster is less than a preset health rate, the monitoring system determines whether the application server cluster meets a preset gradual exit condition according to a gradual entry number of a virtual server fed back by a link load balancer, comprising: When the monitoring system determines that the health rate corresponding to the application server cluster is less than a preset health rate, the monitoring system sends query information to the link load balancer, so that the link load balancer determines a number of virtual servers in a gradual entry state in a link load balancing resource pool according to the query information, and feeds back the gradual entry number of the virtual servers to the monitoring system; wherein the link load balancing resource pool comprises all virtual servers corresponding to the application server cluster in an application service system; The monitoring system determines whether the gradual entry number of the virtual servers is greater than a preset gradual entry threshold according to the gradual entry number of the virtual servers fed back by the link load balancer; When the monitoring system determines that the gradual entry number of the virtual servers is greater than the preset gradual entry threshold, the monitoring system determines that the application server cluster meets the preset gradual exit condition.

2. The failure handling method of a cluster of application servers according to claim 1, characterized in that, The monitoring system sends a health check message to an application server in an application server cluster, and determines a health rate corresponding to the application server cluster according to a response message fed back by the application server cluster, comprising: The proxy server in the monitoring system sends a health check message to each application server in the application server cluster respectively, and determines a health rate corresponding to the application server cluster according to whether each application server in the application server cluster feeds back a response message within a preset time period; or, The proxy server in the monitoring system sends a health check message to each application server in the application server cluster respectively, and determines a health rate corresponding to the application server cluster according to a status code in a response message fed back by each application server in the application server cluster and a preset status code; or, The proxy server in the monitoring system sends a health check message to each application server in the application server cluster respectively, and determines a health rate corresponding to the application server cluster according to whether each application server in the application server cluster feeds back a response message within the preset time period and a status code in the feedback response message and the preset status code. The proxy server in the monitoring system obtains a request message sent by a monitoring platform in the monitoring system, and sends the health rate corresponding to the application server cluster to the monitoring platform.

3. The failure handling method of a cluster of application servers according to claim 1, characterized in that, The monitoring system sends a health check message to an application server in an application server cluster, and determines a health rate corresponding to the application server cluster according to a response message fed back by the application server cluster. The monitoring platform in the monitoring system determines one application server in the application server cluster as a proxy server, and triggers the proxy server to send a health check message to other application servers in the application server cluster. The monitoring platform determines the health rate corresponding to the application server cluster according to the response message fed back by the other application servers and obtained and sent by the proxy server.

4. The failure handling method of a cluster of application servers according to claim 1, characterized by, Further comprising: The monitoring system sends fault alarm information to a terminal device, and forwards the fade-out processing result generated and sent by the link load balancer to the terminal device.

5. A monitoring system, characterized by Comprise: A proxy server configured to send a health check message to an application server in an application server cluster, and determine a health rate corresponding to the application server cluster according to a response message fed back by the application server cluster. A monitoring platform configured to, when judging that the health rate corresponding to the application server cluster is less than a preset health rate, determine whether the application server cluster satisfies a preset fade-out condition according to a number of virtual servers in a fade-in state fed back by a link load balancer. The monitoring platform is further configured to, when judging that the application server cluster satisfies the preset fade-out condition, determine that the application server cluster is faulty, and send fade-out control information to the link load balancer, so that the link load balancer performs fade-out processing on the application server cluster according to the fade-out control information. The monitoring platform is specifically configured to: When judging that the health rate corresponding to the application server cluster is less than a preset health rate, send query information to the link load balancer, so that the link load balancer determines a number of virtual servers in a fade-in state in a link load balancing resource pool according to the query information, and feeds back the number of virtual servers in the fade-in state to the monitoring platform; wherein the link load balancing resource pool contains all virtual servers corresponding to the application server cluster in an application service system. Determine whether the number of virtual servers in the fade-in state is greater than a preset fade-in threshold according to the number of virtual servers in the fade-in state fed back by the link load balancer. When judging that the number of virtual servers in the fade-in state is greater than the preset fade-in threshold, determine that the application server cluster satisfies the preset fade-out condition.

6. The monitoring system of claim 5, wherein, The proxy server is specifically configured to: sending a health check message to each application server in the application server cluster respectively, and determining the health rate corresponding to the application server cluster according to whether each application server in the application server cluster feeds back a response message within a preset time period; or sending a health check message to each application server in the application server cluster respectively, and determining the health rate corresponding to the application server cluster according to a status code in the response message fed back by each application server in the application server cluster and a preset status code; or sending a health check message to each application server in the application server cluster respectively, and determining the health rate corresponding to the application server cluster according to whether each application server in the application server cluster feeds back a response message within the preset time period and a status code in the response message fed back and the preset status code; obtaining a request message sent by a monitoring platform in the monitoring system, and sending the health rate corresponding to the application server cluster to the monitoring platform.

7. A monitoring platform, characterized in that comprise: a transceiver module, configured to determine an application server in an application server cluster as a proxy server, and trigger the proxy server to send a health check message to other application servers in the application server cluster; a processing module, configured to determine a health rate corresponding to the application server cluster according to a response message fed back by the other application servers and obtained and sent by the proxy server; the processing module is further configured to, when judging that the health rate corresponding to the application server cluster is less than a preset health rate, determine whether the application server cluster satisfies a preset fade-out condition according to a number of virtual servers in a fade-in state fed back by a link load balancer; the transceiver module is further configured to, when judging that the application server cluster satisfies the preset fade-out condition, determine that the application server cluster is faulty, and send fade-out control information to the link load balancer, so that the link load balancer performs fade-out processing on the application server cluster according to the fade-out control information; the processing module is configured to, when judging that the health rate corresponding to the application server cluster is less than a preset health rate, determine whether the application server cluster satisfies a preset fade-out condition according to a number of virtual servers in a fade-in state fed back by a link load balancer, and specifically configured to: when judging that the health rate corresponding to the application server cluster is less than a preset health rate, send query information to the link load balancer, so that the link load balancer determines the number of virtual servers in a fade-in state in a link load balancing resource pool according to the query information, and feeds back the number of virtual servers in a fade-in state to a monitoring system; wherein the link load balancing resource pool contains all virtual servers corresponding to the application server cluster in an application service system; determine whether the number of virtual servers in a fade-in state is greater than a preset fade-in threshold according to the number of virtual servers in a fade-in state fed back by the link load balancer; When it is determined that the number of the virtual servers gradually entering is greater than the preset threshold of gradual entering, it is determined that the application server cluster satisfies the preset condition of gradual exiting.

8. A failure handling system for an application server cluster, characterized by Comprising: a link load balancer, an application server cluster, and the monitoring system according to any one of claims 5-6; or, the link load balancer, the application server cluster, and the monitoring platform according to claim 7.

Images