KR102109536B1 - Method for diagnosing and handling obstacle of server based on obstacle type - Google Patents

Abstract

According to an embodiment of the present invention, in a method for an integrated management server to diagnose and respond to a server failure based on a type of failure, (a) monitoring for the web server based on real-time status information received from the web server Performing, if a failure is detected, diagnosing the type of the detected failure; (b) performing a response to the detected failure by transmitting a control command corresponding to the diagnosed failure type to the web server on the basis of a rule for diagnosis and response to the failure; And (c) generating and storing the result information on the diagnosis and response to the failure, and transmitting the generated result information to the manager terminal.

Description

METHOD FOR DIAGNOSING AND HANDLING OBSTACLE OF SERVER BASED ON OBSTACLE TYPE}

The present invention relates to a method for diagnosing and responding to a server failure based on a failure type, and more specifically, a complex failure occurring in a server is diagnosed according to a type, and response control is automatically performed based on a preset rule for the diagnosed failure. By implementing it, it is possible to automatically pre- respond or immediately respond to a diagnosis failure, thereby preventing and preventing the failure, and thus to a method of enabling the web server to effectively overcome the failure situation.

In general, a web server or a web application server should be able to provide web services without failure even in the case of a rapid increase in server users.

However, it is impossible to prevent a failure during server operation. Accordingly, by monitoring the server in real time, it detects a failure that occurs and promptly responds to it, thereby minimizing the problem of server operation due to failure. Development efforts are continuing.

In the case of a web server or a web application server, even if it is a simple failure occurring in the server, if these failures are connected in a complex manner, a serious problem may arise that the entire server cannot provide a web service.

However, in the case of the existing server failure detection and resolution process, only a small number of failure types are diagnosed, and only the basic response functions such as providing a failure alarm and restarting the server are provided for the response according to the diagnosis.

In other words, the existing server failure diagnosis and response process only handles basic response to an already occurring failure such as providing an alarm, making it difficult to respond to various server failure types, and proactively diagnosing and resolving problems with the server. Since it is not a method to prevent the occurrence of a server failure by responding in advance, there is a disadvantage in that it is impossible to fundamentally respond to a serious server failure such as inability to service the entire server.

Therefore, it is possible to immediately respond to various major types of failures that can occur in the server, and to automatically diagnose and respond to failures before they are propagated to larger and more complex failure situations, so that the failure of the web server can be effectively overcome. In addition, there is an increasing demand for technology development to provide a web service with minimal obstacles to users, and an urgent need is to solve the above-mentioned problems.

The present invention is to solve the above-mentioned problems of the prior art, and prevents and prevents various server failures caused by various failure diagnosis and response control based on various failure types rather than processing some failures on the server. Its purpose is to make it possible.

The present invention aims at diagnosing and responding to a failure so that the entire server can be prevented from becoming out of service by simply connecting the types of failures and spreading them to a larger and more complicated failure situation.

The objects of the present invention are not limited to the objects mentioned above, and other objects not mentioned will be clearly understood from the following description.

According to an embodiment of the present invention for achieving the above object, the integrated management server, in the method of diagnosing and responding to a server failure based on the type of failure, (a) real-time status information received from the web server Performing monitoring on the web server as a basis, and if a failure is detected, diagnosing the type of the detected failure; (b) performing a response to the detected failure by transmitting a control command corresponding to the diagnosed failure type to the web server on the basis of a rule for diagnosis and response to the failure; And (c) generating and storing the result information on the diagnosis and response to the failure, and transmitting the generated result information to the manager terminal.

The real-time status information may include server resource information and node resource information for the web server.

The step (a) may include diagnosing the type of the detected failure by performing an analysis based on the real-time status information and a failure diagnosis and response rule stored in advance in the integrated management server.

The step (a) is to diagnose the type of the detected failure by performing an analysis based on the history information in which the real-time status information and the result information on the diagnosis and response to the failure in the integrated management server are accumulated and stored. It can contain.

The step (b) may include performing a response to the detected failure by sending alarm information for the diagnosed failure type to the manager terminal.

In step (a), the type of the detected failure is a request exceeding failure, a large database data inquiry failure, a database connection request exceeding failure, a service delay failure, an insufficient memory failure, a request exceeding failure and a stopping failure of a specific service. And diagnosing with any one.

In the steps (a) and (b), when it is determined through the monitoring that the available thread usage of the web server exceeds a threshold, the request is diagnosed as a failure, and the diagnosis and response are set in advance. On the basis of the rules for, it may include transmitting a control command to bypass the user request to a temporary page to the web server.

Control to bypass the user request to a temporary page, if it is determined that the available thread usage in the web server has exceeded the threshold, and the request exceeded fault is diagnosed, the web server itself automatically performs the request exceeded fault diagnosis It can be characterized as possible.

In the steps (a) and (b), when it is determined that a large database data query for a service is requested from the user through the monitoring, the large database data query is diagnosed as a failure, and a preset failure diagnosis and response is determined. Based on the rule, it may include the step of forcibly terminating the corresponding service requested from the user or transmitting a control command to limit the service by a predetermined condition to the web server.

The control for forcibly terminating the corresponding service requested from the user or restricting the service by a predetermined condition is determined as a request for a large database data query for the service from the user in the web server, so that the large database data search is performed. When a failure is diagnosed, the web server itself can automatically perform a large database data inquiry failure diagnosis immediately.

In the steps (a) and (b), if it is determined through the monitoring that the database connection group (Pool) in the web server exceeds a threshold, database connection of a specific data source is excessively used, database connection Diagnose as a request exceeded failure, and send a control command to bypass the user request to a temporary page or forcibly reduce the waiting time for acquiring the corresponding database connection to the web server, based on the rules for diagnosing and responding to the failure. It may include the steps.

Steps (a) and (b) may cause a service delay failure if it is determined that a response to a service requested from the user is delayed through the monitoring or a request for a service provided after a predetermined time is delayed from the user. Diagnosing, and based on a predetermined rule for diagnosis and response to the failure, may include transmitting a control command to the web server to limit the thread usage rate for the service diagnosed with the failure.

In the above steps (a) and (b), out of memory (OOM) occurs due to an error in service logic or excessive use of the system through the monitoring, or an automatic memory collection phenomenon (Full Garbage) Collection, Full GC) If the frequency of occurrence increases or it is determined that a situation in which the memory cannot be released is continuously generated, it is diagnosed as an out-of-memory failure, and a dump generation command is based on the rules for diagnosing and responding to the failure. And transmitting a server restart command to the web server and transmitting alarm information to the manager terminal.

In the steps (a) and (b), if it is determined that a user request for a specific service received through the monitoring to the web server exceeds a limit that can be processed by the web server, a request exceeding failure of the specific service may result. Diagnosis, and transmits a request control command to the web server to sequentially enter the service for a plurality of users based on the rules for diagnosis and response to a preset failure, and alarm information for the diagnosis of the failure. It may include the step of transmitting to the administrator terminal.

In the steps (a) and (b), if it is determined through the monitoring that the web server is in a running state that does not perform any function even though it is in a running state, it is diagnosed as a stopping fault, and a preset fault The method may include transmitting a dump generation command and a server restart command to the web server based on the rules for diagnosis and response, and transmitting alarm information for the corresponding diagnosis failure to the manager terminal.

The step (c) may include generating the result information on the diagnosis and response to the failure in the form of statistical information arranged in a time series and transmitting the result information to the manager terminal.

The step (c) may include generating result information on the diagnosis and response to the failure at a predetermined period set in advance or at the request of the administrator received from the administrator terminal.

In step (c), when an information inquiry request including a predetermined search condition is received from the administrator terminal, the result of the diagnosis and response to the failure corresponding to the received predetermined search condition is generated and transmitted to the administrator terminal. It may include the steps.

On the other hand, according to another embodiment of the present invention, by performing a monitoring for the web server based on the real-time status information received from the web server, if a failure is detected, a failure diagnosis unit for diagnosing the type of the detected failure; A failure processing unit that performs a response to the detected failure by transmitting a control command corresponding to the diagnosed failure type to the web server based on a preset rule for failure diagnosis and response; A result providing unit generating and transmitting result information on the diagnosis and response to the manager terminal; And a storage unit for storing result information on the generated fault diagnosis and response.

The real-time status information may include server resource information and node resource information for the web server.

The failure diagnosis unit may diagnose the type of the detected failure by performing analysis based on the real-time status information and a failure diagnosis and response rule stored in advance in the integrated management server.

The failure diagnosis unit may diagnose the type of the detected failure by performing an analysis based on the real-time status information and history information in which result information for diagnosis and response of the failure in the integrated management server is accumulated and stored.

The failure processing unit may respond to the detected failure by transmitting alarm information for the diagnosed failure type to the manager terminal.

The failure diagnosis unit diagnoses the type of the detected failure as one of a request exceeded failure, a large database data inquiry failure, a database connection request exceeded failure, a service delay failure, an insufficient memory failure, a request exceeded failure and a stop failure of a specific service. can do.

If it is determined through the monitoring that the available thread (Thead) usage of the web server has exceeded the threshold value, the failure diagnosis unit diagnoses a request exceeded failure, and the failure processing unit detects the request exceeded failure in advance. Based on the established fault diagnosis and response rules, a control command for bypassing a user request to a temporary page may be transmitted to the web server.

Control to bypass the user request to a temporary page, if it is determined that the available thread usage in the web server has exceeded the threshold, and the request exceeded fault is diagnosed, the web server itself automatically performs the request exceeded fault diagnosis It can be characterized as possible.

When the failure diagnosis unit determines that a large database data query for a service is requested from the user through the monitoring, the failure diagnosis unit diagnoses a large database data query failure, and the failure processing unit determines that the large database data query failure occurs, Based on a preset rule for diagnosis and response to a failure, a control command for forcibly terminating the corresponding service requested by the user or restricting the service by a predetermined condition may be transmitted to the web server.

The control for forcibly terminating the corresponding service requested from the user or restricting the service by a predetermined condition is determined as a request for a large database data query for the service from the user in the web server, so that the large database data search is performed. When a failure is diagnosed, the web server itself can automatically perform a large database data inquiry failure diagnosis immediately.

If it is determined through the monitoring that the database connection group (Pool) in the web server exceeds a threshold and database connection of a specific data source (Datasource) is excessively used, diagnoses a database connection request excess failure. , When the failure processing unit is diagnosed as a failure exceeding the database connection request, the user request is bypassed to a temporary page or the waiting time for obtaining the corresponding database connection is enforced based on a preset rule for diagnosis and response to the failure. The control command reduced to can be transmitted to the web server.

The failure diagnosis unit diagnoses a service delay failure when it is determined that a response to a service requested from a user is delayed through the monitoring or a request for a service provided after a predetermined time is delayed from the user, and the failure processing unit In the case of being diagnosed as the service delay failure, a control command for limiting the thread utilization rate for a service diagnosed with a failure may be transmitted to the web server based on a rule for diagnosis and response to a failure.

The fault diagnosis unit may cause an out of memory (OOM) or an automatic memory collection phenomenon (Full Garbage Collection, Full GC) due to an error in service logic or excessive memory use in the system through the monitoring. When it is determined that the frequency is increased or a situation such as the inability to release the memory is continuously generated, the diagnosis is made of an insufficient memory failure, and when the failure processing unit is diagnosed as the insufficient memory failure, preset failure diagnosis and response Based on the rules, a dump generation command and a server restart command may be transmitted to the web server and alarm information may be transmitted to the manager terminal.

When the user's request for a specific service received by the web server exceeds the limit that can be processed by the web server through the monitoring, the failure diagnosis unit diagnoses a request exceeding failure of the specific service, and the failure processing unit In the case of being diagnosed as a failure exceeding the request of the specific service, the web server issues a request control command to sequentially enter a service for a plurality of users based on a preset rule for diagnosis and response of the failure. And transmit the alarm information for the diagnosis of the failure to the manager terminal.

If it is determined through the monitoring that the web server is in a stopped state (Hang) that does not perform any function even though the web server is in a running state, the fault diagnosis unit diagnoses a stop fault, and the fault processing unit diagnoses the stop fault In this case, based on a preset rule for diagnosis and response to a failure, a dump generation command and a server restart command may be transmitted to the web server, and alarm information for the diagnosis failure may be transmitted to the manager terminal.

The result providing unit may generate result information on the diagnosis and response to the failure in the form of statistical information arranged in a time series and transmit the result information to the manager terminal.

The result providing unit may generate result information on the diagnosis and response to the failure by a request of an administrator received from the administrator terminal or a predetermined period.

When the information inquiry request including the predetermined search condition is received from the administrator terminal, the result providing unit may generate result information on the diagnosis and response to the failure corresponding to the received predetermined search condition and transmit the result information to the administrator terminal. .

According to an embodiment of the present invention, it is possible to prevent and prevent various failures that may occur in a server by performing a complex server failure diagnosis and response based on various failure types, rather than processing for some failure phenomenon. In operation, reliability and stability can be ensured.

According to another embodiment of the present invention, by performing a rule-based failure diagnosis and response based on a preset rule, when a server failure occurs, a response to a failure is automatically performed based on a preset rule, the server The obstacle can be overcome immediately.

According to another embodiment of the present invention, the server performs integrated management and real-time control of the server based on the resource information of the web server, and thus has an advantage that server uninterrupted control is possible.

It should be understood that the effects of the present invention are not limited to the above-described effects, and include all effects that can be deduced from the configuration of the invention described in the detailed description or claims of the present invention.

1 is a diagram schematically showing a configuration of a system for diagnosing and responding to server failure based on a type of failure according to an embodiment of the present invention.
2 is a block diagram showing the configuration of an integrated management server according to an embodiment of the present invention.
3 is a flowchart illustrating a process in which a failure of a web server is diagnosed according to a type according to an embodiment of the present invention, and a failure is handled accordingly.
4 is a diagram illustrating a screen displayed on a manager terminal when providing real-time monitoring information according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating a screen displayed on a manager terminal when providing result information on diagnosis and response to a failure according to an embodiment of the present invention.
6 is a diagram illustrating a screen displayed on a manager terminal when providing result information according to a request exceeded fault diagnosis according to an embodiment of the present invention.
7 is a diagram illustrating a screen displayed on a manager terminal when providing result information according to diagnosis of a large database data inquiry failure according to an embodiment of the present invention.
8 is a diagram illustrating a screen displayed on a manager terminal when providing result information according to a diagnosis of a failure exceeding a database connection request according to an embodiment of the present invention.
9 is a diagram illustrating a screen displayed on a manager terminal when providing result information according to a service delay failure diagnosis according to an embodiment of the present invention.
10 is a diagram illustrating a screen displayed on a manager terminal when providing result information according to a diagnosis of a memory shortage failure according to an embodiment of the present invention.
11 is a diagram illustrating a screen displayed on a manager terminal when providing result information according to diagnosis of a request exceeded failure of a specific service according to an embodiment of the present invention.
12 is a diagram illustrating a screen displayed on a manager terminal when providing result information according to diagnosis of a stop disorder according to an embodiment of the present invention.
13 is a diagram illustrating a screen displayed on a manager terminal when providing result information on diagnosis and response to a failure in the form of statistics arranged in a time series according to an embodiment of the present invention.
14 is a diagram illustrating a screen displayed on a manager terminal when providing diagnosis history search result information according to an embodiment of the present invention.

Hereinafter, the present invention will be described with reference to the accompanying drawings. However, the present invention may be implemented in various different forms, and thus is not limited to the embodiments described herein. In addition, in order to clearly describe the present invention in the drawings, parts irrelevant to the description are omitted, and like reference numerals are assigned to similar parts throughout the specification.

Throughout the specification, when a part is "connected" to another part, this includes not only "directly connected" but also "indirectly connected" with another member in between. . Also, when a part “includes” a certain component, this means that other components may be further provided instead of excluding other components, unless otherwise stated.

In this specification, the web server may be a web server (Web server) or a web application server (Web Application Server, WAS), and has been described in the form of a single server, but may be configured as a plurality of server types. to be. Accordingly, in the present specification, the integrated management server can simultaneously monitor a plurality of web servers.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram schematically showing a configuration of a system for diagnosing and responding to server failure based on a type of failure according to an embodiment of the present invention.

Referring to Figure 1, the failure type based server failure diagnosis and response system according to an embodiment of the present invention, the user terminal 100, the web server 200, the integrated management server 300 and the administrator terminal 400 It can be composed of.

Each of the user terminal 100, the web server 200, the integrated management server 300, and the manager terminal 400 may be connected through a communication network. Here, the communication network may be configured regardless of its communication mode such as wired or wireless, a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), etc. It can be composed of various communication networks.

The user terminal 100 and the administrator terminal 400 are based on all types of handhelds equipped with a touch screen panel such as a mobile phone, a smart phone, a personal digital assistant (PDA), a portable multimedia player (PMP), a tablet PC, and the like. It may include a wireless communication device of, in addition to a desktop PC, tablet PC, laptop PC, IPTV including a set-top box, it may also include a device provided with a basis for installing and running applications.

The manager terminal 400 may install an application providing a server integrated management service and store it in memory. For example, the administrator terminal 400 accesses an app store server (not shown) in which various applications are uploaded, and then downloads a server integrated management service providing application that provides various services such as monitoring information inquiry and failure history information provision. Can be installed.

According to an embodiment of the present invention, the administrator terminal 400 may be the same device as the user terminal 100 accessing a webpage provided with a server integrated management service.

The user terminal 100 may be provided with an interface and a screen for a web service provided from the web server 200, through which the user can select a service to be used.

When any one of the web services is selected from the user through the provided interface, the user terminal 100 may transmit a corresponding service request signal to the web server 200, and the web server 200 may display a corresponding web service screen Alternatively, the interface may be transmitted to the user terminal 100 and provided to the user.

The web server 200 may transmit resource information to the integrated management server 300. At this time, the resource information includes server resource information and node resource information, and through the resource information, the integrated management server 300 may perform real-time monitoring of the web server 200. The node may refer to a system such as a physical device or a container in which a web server is installed.

According to an embodiment of the present invention, the web server 200 transmits resource information to the integrated management server 300 when a service provision request signal is received from the user terminal 100, or a separate request signal Even if it is not received, it is possible to automatically transmit resource information every predetermined period to perform real-time monitoring.

The integrated management server 300 may perform real-time monitoring based on resource information received from the web server 200, and diagnose a failure through analysis.

The integrated management server 300 may diagnose a failure of the web server 200 based on a preset rule, and may perform response control for the diagnosed failure.

At this time, the preset rule is for a diagnosis type response method for an occurrence disorder, and a response method for each type of failure may be matched in advance.

The types of diagnosis include request over-diagnosis, large database data inquiry diagnosis, database connection request over-diagnosis, service delay diagnosis, memory shortage diagnosis, request over-diagnosis and stop diagnosis of a specific service. This may include moving to the bypass page, forcibly shutting down the service, controlling service restrictions, restricting service access, providing alarms, generating dumps, and restarting services.

The integrated management server 300 may cause the web server 200 to overcome the failure by transmitting a corresponding control command matching the diagnosed failure type to the web server 200.

According to an embodiment of the present invention, in the case of a request exceeding failure or a large database inquiry failure, it is determined whether the threshold of the available thread usage in the web server 200 is exceeded, or a large database inquiry request for a service from a user As it is possible to determine whether or not it is possible, the web server 200 itself can perform the response control as soon as it is determined that the corresponding failure has occurred, without the intervention of the integrated management server 300. Accordingly, a problem that misses an appropriate time to respond to the failure may not occur due to the time required as the integrated management server 300 diagnoses the failure and transmits the corresponding control control command.

When the failure of the web server 200 is diagnosed and corresponding control is performed, the integrated management server 300 may match and store the diagnosed failure type information and corresponding corresponding information accordingly.

In addition, the integrated management server 300 transmits real-time monitoring information to the manager terminal 400 so that server monitoring information can be provided through the manager terminal 400. The corresponding response history may be transmitted to the manager terminal 400 to be provided to the manager.

In addition, when there are multiple web servers 200, the integrated management server 300 may selectively provide real-time monitoring information for each server to the administrator terminal 400 for the plurality of web servers 200.

The integrated management server 300 may generate a report on the result of diagnosis and response to the failure and transmit it to the manager terminal 400. At this time, the integrated management server 300 may provide the statistical information listed in time series for the diagnosis and response results of the failure, so that the administrator can grasp the time-based failure pattern.

The manager terminal 400 may display real-time monitoring information received from the integrated management server 300, failure diagnosis notification information, response history information according to the occurrence of the failure, and a report on the failure diagnosis and response results to the administrator.

In addition, the administrator terminal 400, when a request for information inquiry for a predetermined failure diagnosis and response is input from the user through the interface related to provision of the integrated management service provided from the integrated management server 300, the request information is integrated management server ( 300) to receive corresponding information.

2 is a block diagram showing the configuration of the integrated management server 300 according to an embodiment of the present invention.

Referring to FIG. 2, the integrated management server 300 according to an embodiment of the present invention may include a fault diagnosis unit 310, a fault processing unit 320, a result providing unit 330, and a storage unit 340. have.

The failure diagnosis unit 310 may perform real-time monitoring of the web server 200 based on resource information received from the web server 200.

Specifically, the failure diagnosis unit 310 may perform real-time monitoring by performing analysis based on the received resource information and the history information data for the failure diagnosis and response results.

When a failure is detected through the monitoring, the failure diagnosis unit 310 may diagnose the failure according to a predetermined type of failure. At this time, the predetermined type of failure that can be diagnosed may include request over-diagnosis, large database data inquiry diagnosis, database connection request over-diagnosis, service delay diagnosis, memory shortage diagnosis, request over-diagnosis and stop diagnosis of a specific service.

According to an embodiment of the present invention, the failure diagnosis unit 310 performs monitoring based on resource information received from the web server 200, and an available thread of the web server 200 is requested by an excessively generated user service request ( Thread) If it is determined that the usage exceeds the threshold, it is possible to diagnose a request exceeding failure with respect to the web server 200.

According to another embodiment of the present invention, the failure diagnosis unit 310 performs monitoring based on resource information received from the web server 200, and when it is determined that a large database data query for the service is requested from the user, The web server 200 may diagnose a large database data inquiry failure.

According to another embodiment of the present invention, the failure diagnosis unit 310 performs monitoring based on resource information received from the web server 200, delays processing time of the database in the web server 200, and locks the database. If it is determined that the database connection is excessively occupied by, for example, the database connection group (Pool) exceeds the threshold, and it is determined that the database connection of a specific data source (Datasource) is excessively used, the web server 200 The database connection request exceeded fault can be diagnosed.

According to another embodiment of the present invention, the fault diagnosis unit 310 performs monitoring based on resource information received from the web server 200 to respond to a service requested from a user due to a temporary network or linkage system problem. If it is determined that a request for a service provided is received after the delay or a predetermined time is delayed from the user, the service delay failure may be diagnosed with respect to the web server 200.

According to another embodiment of the present invention, the fault diagnosis unit 310 performs monitoring based on resource information received from the web server 200, due to an error in service logic or excessive memory use of the system. If it is determined that an out-of-memory (OOM) problem has occurred, the out-of-memory failure may be diagnosed with respect to the web server 200.

In addition, the fault diagnosis unit 310 performs monitoring based on the resource information received from the web server 200, such that the frequency of occurrence of a full garbage collection (full GC) increases or memory release is impossible If it is determined to occur continuously, the web server 200 may be diagnosed in advance that a memory shortage failure will occur.

According to another embodiment of the present invention, the fault diagnosis unit 310 performs monitoring based on resource information received from the web server 200, and requests a user for a specific service received by the web server 200 If it is determined that the limit that can be processed by the web server 200 is exceeded, it is possible to diagnose a failure exceeding a request for a specific service for the web server 200.

According to another embodiment of the present invention, if it is determined that the failure diagnosis unit 310 is in a hang state in which no function is performed even when the web server 200 is in the active state, the web server 200 is checked. Stop failure can be diagnosed.

The failure processing unit 320 transmits a corresponding control command matched for each type to the web server 200 with respect to the type of failure for the web server 200 diagnosed through the failure diagnosis unit 310, so that the web server ( 200) can be overcome.

In other words, when a failure is diagnosed through the failure diagnosis unit 310, the failure processing unit 320 automatically generates a response control command corresponding to the diagnosis failure type based on a preset failure diagnosis and response rule. 200), the web server 200 can automatically and immediately overcome the generated failure.

According to an embodiment of the present invention, when a request exceeded failure is generated or diagnosed to occur through the failure diagnosis unit 310, the failure processing unit 320 issues a control command to bypass the excessively generated user request to a temporary page. By transmitting to the web server 200, so that the web server 200 is maintained in a stable state, it is possible to perform a preliminary response or an immediate response to a failure.

As the failure processing unit 320 transmits a bypass control command to the temporary page, when it is determined that the available thread usage of the web server 200 does not exceed a threshold value through the failure diagnosis unit 310, a user generated thereafter The request may no longer send control commands to bypass the temporary page.

According to another embodiment of the present invention, when a large database data inquiry failure has occurred or is diagnosed to occur through the failure diagnosis unit 310, the failure processing unit 320 forcibly terminates the corresponding service requested by the user or is predetermined. By transmitting a control command to limit the service by the condition to the web server 200, the web server 200 is maintained in a stable state, so that it is possible to perform a preliminary response or an immediate response to a failure.

According to another embodiment of the present invention, when the database connection request exceeded failure is diagnosed through the failure diagnosis unit 310, the failure processing unit 320 bypasses the generated user request to a temporary page or waits to acquire the database connection By transmitting a control command to forcibly reduce the time to the web server 200, it is isolated from the failure so that the web server 200 is maintained in a stable state, and an immediate response to the failure can be implemented.

The failure processing unit 320 transmits a control command of the temporary page bypass, and when it is determined that the database connection group does not exceed a threshold value through the failure diagnosis unit 310, the service request generated thereafter is no longer temporary. You may not send a control command to bypass the page or forcibly reduce the waiting time for obtaining the connection to the database.

According to another embodiment of the present invention, when a service delay failure occurs or is diagnosed to occur through the failure diagnosis unit 310, the failure processing unit 320 issues a control command to limit the thread usage rate for the service in which the failure is diagnosed. By transmitting to the web server 200, communication quality of service (Quality of Service, QoS) is ensured, so it is possible to perform a preliminary response or an immediate response to a failure.

When the failure processing unit 320 transmits the thread usage limit control command, if it is determined that the service delay problem no longer occurs through the failure diagnosis unit 310, the thread usage limit control release command to the web server 200 It is possible to prevent the usage rate limiting control from being carried out.

According to another embodiment of the present invention, when it is diagnosed that a memory shortage failure has occurred through the failure diagnosis unit 310, the failure processing unit 320 transmits a dump generation command and a server restart command to the web server 200, It is possible to perform an immediate response to a failure by transmitting the alarm information on the occurrence of the failure to the manager terminal 400.

In addition, if it is previously diagnosed that a memory shortage failure will occur through the failure diagnosis unit 310, the failure processing unit 320 may transmit a dump generation command to the web server 200 and may cause a memory shortage failure to the administrator terminal 400. By providing an alarm for, it is possible to perform a preliminary response to an insufficient memory failure.

According to another embodiment of the present invention, when a request exceeding failure of a specific service is generated or diagnosed to be generated through the failure diagnosis unit 310, the failure processing unit 320 sequentially enters services for a plurality of users By sending a request control command to the web server 200, and by transmitting the alarm information for the diagnosis of the failure to the administrator terminal 400, so as to prevent the departure of the user's service use bar, preemptive response to the failure Or you can respond immediately.

According to another embodiment of the present invention, when it is diagnosed that a stop failure has occurred through the failure diagnosis unit 310, the failure processing unit 320 transmits a dump generation command and a server restart command to the web server 200, and an administrator By transmitting the alarm information for the corresponding diagnostic failure to the terminal 400, the web server 200 can be quickly recovered from the failure, and the entire system can be prevented from occurring, so that an immediate response to the failure can be performed. have.

The result providing unit 330 may transmit real-time monitoring information performed on the basis of the resource information received from the web server 200 to the manager terminal 400 to provide it to the manager.

In addition, the result providing unit 330 may generate result information on the diagnosis and response to the failure and transmit it to the manager terminal 400.

According to an embodiment of the present invention, the result providing unit 330 generates the result information on the diagnosis and response of the failure as time-series statistical information of the date or time of diagnosis of the failure and provides it to the administrator terminal 400, thereby allowing the administrator Time-based failure patterns can be identified.

The result information on the diagnosis and response to the disorder may include analysis information on the cause of occurrence of each diagnosis type, information on the result of the diagnosis of the disorder over time, and information on the diagnosis and response on the specific date.

The result providing unit 330 may generate and provide fault diagnosis and response result information according to an administrator request received from the manager terminal 400, or automatically generate and provide it every predetermined predetermined period.

The storage unit 340 stores the fault diagnosis and response result information generated by the result providing unit 330, so that the generated fault diagnosis and response result information can be used for analysis for diagnosis of a disorder according to real-time monitoring in the future. Through this, it is possible to enable the disorder diagnosis unit 310 to perform organic or comprehensive analysis.

In addition, the storage unit 340 may store history information corresponding to the requested condition when requesting the inquiry of the history information on the diagnosis and response to the failure from the administrator terminal 400 by storing the result of the diagnosis and response to the failure. have.

3 is a flowchart illustrating a process in which a failure of the web server 200 is diagnosed according to a type according to an embodiment of the present invention, and a failure is handled accordingly.

First, the web server 200 may transmit real-time status information for the web server 200 by transmitting resource information including server resource information and node resource information to the integrated management server 300 (S301).

The integrated management server 300 may perform real-time monitoring based on real-time status information received from the web server 200, that is, resource information (S302).

When a failure is detected during monitoring, the integrated management server 300 may perform a failure diagnosis for each type based on rules for preset failure diagnosis and response (S303). At this time, the diagnosis of each type of failure may be performed according to an organic or comprehensive analysis based on information on the result of the diagnosis and response stored in the server.

When the fault type is diagnosed, the integrated management server 300 transmits a control command corresponding to the diagnosed fault type to the web server 200 based on the preset rules for fault diagnosis and response, thereby diagnosing the fault type. A response process for a failure may be performed (S304).

According to an embodiment of the present invention, the integrated management server 300 diagnoses a request exceeded failure when it is determined that the available thread usage of the web server 200 exceeds a threshold due to excessive service requests through real-time monitoring. , In response to a preset diagnosis and response rule, a control command for bypassing an excessively generated user request to a temporary page is transmitted to the web server 200 to respond to the diagnosed failure.

For example, when a mass service request is received from a user to the web server 200, as the integrated management server 300 transmits a temporary page bypass command to the web server 200, the web server 200 is sent to the user. By providing a temporary page, it is possible to prevent a request exceeding failure from occurring or to quickly overcome a request exceeding failure that has occurred.

After the response to the diagnosis failure, if it is determined that the available thread usage of the web server 200 is lower than the threshold, the integrated management server 300 transmits the temporary page bypass control command to the web server 200 You may not.

In the case of the request exceeding failure, the web server 200 may be determined in the integrated management server 300 as well as the web server 200, as it is possible to determine whether a mass service request from a user exceeds a preset available thread usage threshold. ), If it is determined that the mass service request from the user exceeds a preset available thread usage threshold, the web server 200 does not receive a temporary page bypass command from the integrated management server 300, the web server 200 itself By providing a temporary page to the corresponding user, it is possible to prevent a request excess failure from occurring or to more quickly overcome the generated request excess failure.

According to another embodiment of the present invention, the integrated management server 300 diagnoses a large database data inquiry failure when a large database data inquiry for a service is requested from a user through real-time monitoring, and a preset diagnosis And a response to the diagnosed failure by forcibly terminating the corresponding service requested from the user according to the rule for the response or by transmitting a control command to the web server 200 to limit the service by a predetermined condition. have.

For example, if it is determined that a large amount of memory of the web server 200 is used as a specific user requests a large database data query for the service, the integrated management server 300 commands the service to be terminated to the web server 200 By transmitting the data, the web server 200 may forcibly terminate the service requested from the user, so that a large database data inquiry failure does not occur or the generated large database data inquiry failure can be quickly overcome.

In the case of the large-capacity database data inquiry failure, as well as the integrated management server 300, the web server 200 can determine whether a large database data inquiry request from a user exceeds a threshold of a preset number of inquiry requests. When the web server 200 determines that a request for querying a large amount of database data from a user exceeds a threshold of a preset number of query requests, the web server 200 does not receive a control command from the integrated management server 300, The web server 200 itself controls the service to be limited by a predetermined condition by forcibly shutting down the corresponding service requested from the user or controlling the service to be restricted according to a predetermined condition for diagnosis and response. It can be overcome quickly.

According to another embodiment of the present invention, if the integrated management server 300 determines that the database connection is excessively occupied due to a processing time delay of a database in the web server 200 through a real-time monitoring, a database lock state, etc., that is, When the database connection group exceeds the threshold and it is determined that the database connection of a specific data source is excessively used, the database connection request exceeded fault is diagnosed, and the user request generated according to the preset rules for diagnosis and response By sending a control command to the web server 200 to bypass the temporary page or forcibly reduce the waiting time for acquiring the database connection, a response to the diagnosed failure may be performed.

After performing the response to the diagnosis failure, if it is determined that the database connection group does not exceed the threshold value, the integrated management server 300 sends the temporary page bypass control command or the corresponding database connection acquisition waiting time forced shortening control command to the web server. It may not be transmitted to (200).

According to another embodiment of the present invention, the integrated management server 300 is provided after the response of the service requested from the user is delayed or a predetermined time is delayed from the user due to a temporary network, a connection system problem, etc. through real-time monitoring. When it is determined that the request for the service is received, the service delay diagnosis is performed, and a control command for limiting the thread usage rate for the service diagnosed with the failure according to the preset rules for diagnosis and response is sent to the web server 200. By transmitting, it is possible to respond to the diagnosed disorder.

After the response to the diagnosis failure, if it is determined that the service delay problem does not occur, the integrated management server 300 may transmit the thread utilization limit control release command to the web server 200 so that the utilization limit control is not performed. have.

According to another embodiment of the present invention, if the integrated management server 300 determines that a memory shortage problem is caused by an error in service logic or excessive memory use of the system through real-time monitoring, the memory shortage fault is diagnosed, and Respond to the diagnosed failure by sending a dump generation command and a server restart command to the web server 200 according to the established rules for diagnosis and response, and sending alarm information for a failure occurrence to the manager terminal 400 Can be done.

In addition, if it is determined that the frequency of occurrence of an automatic memory collection phenomenon increases or a memory release cannot be continuously performed through real-time monitoring, the integrated management server 300 may cause an insufficient memory failure to the web server 200 in advance. Diagnosing, by sending a dump generation command to the web server 200 according to the rules for pre-set diagnosis and response, and provides an alarm on whether or not an out-of-memory failure may occur to the administrator terminal 400, thereby detecting a diagnosed failure. It is possible to conduct a preliminary response to Korea.

According to another embodiment of the present invention, the integrated management server 300 is that the user's request for a specific service received by the web server 200 through real-time monitoring exceeds the limit that can be processed by the web server 200 If it is determined, diagnoses a request exceeded failure of a specific service, transmits a request control command to the web server 200 to sequentially enter a service for a plurality of users according to a preset rule for diagnosis and response, and , By transmitting the alarm information for the diagnosis of the failure to the manager terminal 400, it is possible to respond to the diagnosed failure.

According to another embodiment of the present invention, if the integrated management server 300 determines that the web server 200 is in a running state that does not perform any function through real-time monitoring, it stops the failure. Diagnosis, and transmits a dump generation command and a server restart command to the web server 200 according to the preset rules for diagnosis and response, and transmits alarm information for the diagnosis failure to the manager terminal 400, thereby diagnosing Responds to disabled disorders.

The integrated management server 300 may store fault diagnosis and response result information in the server, and generate result information according to a request received from a manager or a predetermined cycle, and provide the result to the manager terminal 400. (S305).

At this time, the diagnosis and response result information of the failure may be generated and provided in the form of information arranged in time series according to time or date units.

In addition, the generated and stored fault diagnosis and response result information may be used as real-time monitoring or analysis data when fault diagnosis is performed later.

4 is a diagram illustrating a screen displayed on the manager terminal 400 when providing real-time monitoring information according to an embodiment of the present invention.

First, as shown in FIG. 4, the integrated management server 300 may provide a screen or interface that enables real-time monitoring information confirmation and integrated management of the web server 200 to the administrator terminal 400.

The integrated management server 300 may provide the alarm 401 to the manager terminal 400 when the result information on the diagnosis and response to the failure is generated, and when the alarm 401 is selected by the manager, the manager terminal 400 ) May display and provide a list 402 of result information on failure diagnosis and response generated by the integrated management server 300.

If any one of the generated result information for the diagnosis and response to the failure 403 is selected by the administrator, the administrator terminal 400 displays and provides the result information for the corresponding diagnosis and response, which is detailed in FIG. 5. I will explain.

When the execution window 404 for providing alarm history information is selected and input from the administrator, the administrator terminal 400 may display and provide the alarm history information 405 provided from the integrated management server 300. At this time, the alarm history information may be provided only for the history information within the last month.

The manager terminal 400 may display summary information 406 for a plurality of servers and nodes through a real-time monitoring screen provided from the integrated management server 300 and provide it to the manager.

In addition, the administrator terminal 400 may provide detailed information 407 for each of the plurality of web servers 200 through a real-time monitoring screen.

5 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information on diagnosis and response to a failure according to an embodiment of the present invention.

As illustrated in FIG. 5, the result information for diagnosis and response to a failure may be provided by being divided into common information 501, diagnosis result information 502, and detailed information 503.

In particular, the diagnosis result information can be checked for each time zone through the diagnosis result information 502, and detailed information for each diagnosis can be checked as detailed information 503 for a specific diagnosis time point is provided.

Specifically, the common information 501 includes information on the number of times the diagnosis was started (Start Time), the last time the fault was detected, the end time of the diagnosis (End Time), and the total number of times the fault was detected (Total Diagnostics) Repeat Count), and the result information generation count information (Total Report Creation Count), which is the number of times result information for fault diagnosis and response is generated, may be included.

According to an embodiment of the present invention, if it is determined that the diagnosed failure type is the same type, the integrated management server 300 may generate up to 3 result information, thereby preventing a plurality of identical result information from being generated. On the other hand, if it is determined that the diagnosed failure type is not the same type, the integrated management server 300 may generate up to 100 result information.

Therefore, through the common information 501, when the same diagnosis repetition number information is greater than the result information generation number information, it can be determined from the administrator's point of view that a plurality of the same diagnosis has occurred.

Meanwhile, the diagnosis result information 502 may be provided in the order in which the disorders are diagnosed, and the types of the faults generated according to the diagnosis time may be separately provided in color.

If any one of the plurality of provided diagnostic result information 502 is selected, detailed information 503 for the selected diagnostic result may be provided.

The detailed information 503 may be provided by commonly including the fault diagnosis time information 504, the dump generation time information 505, the generated dump file 506, and the corresponding information 507 for the diagnosed fault type. .

6 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information according to a request exceeded fault diagnosis according to an embodiment of the present invention.

As illustrated in FIG. 6, when providing the result information according to the request exceeded fault diagnosis, the detailed information 503 includes threshold information 601 used at the time of diagnosis and actual utilization information for each connector exceeding the threshold value ( 602) and service information 603 for a plurality of requested 5 cases when diagnosing a disorder may be further included and provided.

At this time, the service information 603 for a number of requests for diagnosis of a failure may include service name and average number of calls per service, and average number of calls is measured and calculated three times at 5 second intervals. It may be the average number of times.

7 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information according to diagnosis of a large database data inquiry failure according to an embodiment of the present invention.

As shown in FIG. 7, when providing the result information according to the diagnosis of a large database data query failure, the detailed information 503 includes threshold information 701 used during diagnosis and a list of services requesting the query of large database data 702 This may be included further.

8 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information according to a diagnosis of a failure exceeding a database connection request according to an embodiment of the present invention.

As illustrated in FIG. 8, when providing the result information according to the diagnosis of a database connection request exceeded failure, the detailed information 503 includes a threshold information 801 used at the time of diagnosis and a data source in which database connection group usage exceeds a threshold value And usage rate information 802 for each data source.

9 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information according to a service delay failure diagnosis according to an embodiment of the present invention.

As illustrated in FIG. 9, when providing the result information according to the service delay failure diagnosis, the detailed information 503 may further include the threshold information 901 and the service delay list 902 used in the diagnosis. .

At this time, the service delay list 902 includes the service name, the total number of services being performed at the time of failure diagnosis, the number of services determined to be service delay compared to the total number of services, and service execution time information based on the time of failure diagnosis. Can be.

10 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information according to a diagnosis of a memory shortage failure according to an embodiment of the present invention.

As illustrated in FIG. 10, when providing the result information according to the diagnosis of insufficient memory failure, the detailed information 503 may further include and provide detailed item information 101 about the insufficient memory failure.

The detailed item information 101 about the memory shortage failure may include and be provided with a heap utilization rate, a frequency of automatic memory collection, a leak, a request for querying a large database data, and file name information when creating a heap dump.

11 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information according to diagnosis of a request exceeded failure of a specific service according to an embodiment of the present invention.

As illustrated in FIG. 11, when providing result information according to a request exceeding fault diagnosis of a specific service, the detailed information 503 includes threshold information 111 used during diagnosis, and connector information occupied by the target service ( 112) and a list 113 of five performed services may be further included and provided.

At this time, the list 113 of the five services performed may include the service name and the number of service requests.

12 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information according to diagnosis of a stop disorder according to an embodiment of the present invention.

As illustrated in FIG. 12, when providing the result information according to the diagnosis of the stop disorder, the detailed information 503 may further include connection timeout time information and retry count information 121.

13 is a diagram illustrating a screen displayed on the manager terminal 400 when providing result information on diagnosis and response to a failure in the form of statistics arranged in a time series according to an embodiment of the present invention.

As shown in FIG. 13, the manager terminal 400 may display the annual failure diagnosis result 131 in the form of a time series based on a date provided from the integrated management server 300 and provide it to the manager, by the manager When a specific day is selected (132), the diagnosis result generated on the selected date may be displayed and provided in the form of result information 133 listed based on time.

14 is a diagram illustrating a screen displayed on the manager terminal 400 when providing diagnosis history search result information according to an embodiment of the present invention.

The administrator may search the summary information of the failure diagnosis of the web server 200 to be searched through the administrator terminal 400 and the result information of the diagnosis and response to the failure (141).

At this time, the result information for the diagnosis and response to the failure is stored in the integrated management server 300 for 7 days, and the result information generated after 7 days may be automatically deleted.

When any one of the dates in which the failure is diagnosed by the administrator is selected (142), the administrator terminal 400 may display and provide a result information list 143 for diagnosis and response to the failure generated on the date.

As described above, according to an embodiment of the present invention, it is possible to prevent and prevent various failures that may occur in a server by performing a complex server failure diagnosis and response based on various failure types, rather than processing for some failure phenomenon. Bar and reliability can be secured in server operation.

According to another embodiment of the present invention, by performing a rule-based failure diagnosis and response based on a preset rule, when a server failure occurs, a response to a failure is automatically performed based on a preset rule, the server The obstacle can be overcome immediately.

According to another embodiment of the present invention, the server performs integrated management and real-time control of the server based on the resource information of the web server, and thus has an advantage that server uninterrupted control is possible.

It should be understood that the effects of the present invention are not limited to the above-described effects, and include all effects that can be deduced from the configuration of the invention described in the detailed description or claims of the present invention.

The above description of the present invention is for illustration only, and those skilled in the art to which the present invention pertains can understand that the present invention can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined form.

The scope of the present invention is indicated by the following claims, and all modifications or variations derived from the meaning and scope of the claims and their equivalent concepts should be interpreted to be included in the scope of the present invention.

100: user terminal
200: web server
300: integrated management server
310: disability diagnosis unit
320: fault handling unit
330: result providing unit
340: storage unit
400: administrator terminal

Claims (36)

A method for the integrated management server to diagnose and respond to a server failure based on the type of failure,
(a) performing monitoring on the web server based on real-time status information received from the web server, and if a failure is detected, diagnosing the type of the detected failure;
(b) the web server transmits a control command corresponding to the diagnosed fault type to the web server based on a preset rule for diagnosis and response of the fault, so that the web server responds to the detected fault according to the control command. Controlling to automatically perform a line response or an immediate response; And
(c) generating and storing result information for diagnosis and response to the failure, and transmitting the generated result information to the manager terminal,
Step (a) is,
Analysis based on the real-time status information including the server resource information and the node resource information for the web server and the rules for diagnosing and responding to the failure, requesting the detected type of failure exceeded failure, large-capacity database data A method of diagnosing and responding to a server failure based on a type of failure, comprising diagnosing one of a query failure, a database connection request exceeded failure, a service delay failure, an out-of-memory failure, and a request for a particular service.
According to claim 1,
Step (c) is,
A method of diagnosing and responding to a server failure based on a failure type, further comprising the step of limiting the number of generations of result information on the failure diagnosis and response to a failure diagnosed with the same type to a preset number or less.
According to claim 1,
Result information on the diagnosis and response to the disorder
Diagnosis execution time indicating the first time the failure was detected, diagnosis end time indicating the last time the failure was detected, the same diagnosis repetition number information indicating the total number of occurrences of the same type of fault as the diagnosed fault type, and fault diagnosis and response A method for diagnosing and responding to a server failure based on a type of failure, including information on the number of result information generated indicating the total number of times the result information has been generated.
According to claim 1,
Step (a) is,
And analyzing a diagnosis based on the real-time status information and history information in which result information for diagnosis and response of the failure in the integrated management server is accumulated, and diagnosing the detected type of the failure. How to diagnose and respond to server failures.
According to claim 1,
Step (b) is,
And performing a response to the detected failure by sending the alarm information for the diagnosed failure type to the manager terminal.
delete According to claim 1,
Steps (a) and (b) are:
When it is determined through the monitoring that the available thread usage of the web server exceeds a threshold, the request is diagnosed as a failure, and the user request is temporarily based on a preset failure diagnosis and response rule. And transmitting a control command to bypass the page to the web server.
The method of claim 7,
Control to bypass the user request to a temporary page,
When it is determined that the usage of available threads in the web server exceeds a threshold, and a request exceeded fault is diagnosed, the web server itself can automatically perform a request exceeded fault diagnosis, which is based on a fault type based server fault How to diagnose and respond.
According to claim 1,
Steps (a) and (b) are:
When it is determined through the monitoring that a large database data query for the service is requested from the user, the large database data query is diagnosed as a failure, and the request requested by the user is based on a preset rule for diagnosis and response to the failure. A method of diagnosing and responding to a server failure based on a type of failure, comprising the step of forcibly terminating the service or transmitting a control command to limit the service by a predetermined condition to the web server.
The method of claim 9,
The control for forcibly terminating the corresponding service requested from the user or for restricting the service by a predetermined condition,
When it is determined that a large database data inquiry for a service is requested from a user in the web server, and a large database data inquiry failure is diagnosed, the web server itself can automatically perform a large database data inquiry failure diagnosis immediately Type-based server failure diagnosis and response method.
According to claim 1,
Steps (a) and (b) are:
When it is determined through the monitoring that the database connection group (Pool) in the web server exceeds the threshold and database connection of a specific data source is excessively used, it is diagnosed as a failure exceeding the database connection request, and is set in advance. Based on the rules for diagnosis and response to a failure, the method comprising a type of failure-based server, comprising: sending a control command to the web server to bypass a user request to a temporary page or forcibly reduce a waiting time for obtaining a corresponding database connection. Disability Diagnosis and Response Methods.
According to claim 1,
Steps (a) and (b) are:
When it is determined through the monitoring that a response to a service requested from a user is delayed or a request for a service provided after a predetermined time is delayed from the user, it is diagnosed as a service delay failure, and a preset failure diagnosis and response Based on the rule for, comprising the step of transmitting a control command for limiting the thread usage rate for the service diagnosed with a failure to the web server, the failure type based server failure diagnosis and response method.
According to claim 1,
Steps (a) and (b) are:
Through the monitoring, out of memory (OOM) occurs due to errors in service logic or excessive memory use in the system, or the frequency of occurrence of full garbage collection (full GC) increases or memory If it is determined that the situation of non-cancellation occurs continuously, it diagnoses as an out-of-memory failure, and transmits a dump generation command and a server restart command to the web server based on a preset rule for diagnosis and response to a failure. And transmitting the alarm information to the manager terminal.
According to claim 1,
Steps (a) and (b) are:
When it is determined through the monitoring that a user request for a specific service received by the web server exceeds a limit that can be processed by the web server, it is diagnosed as a request exceeding failure of a specific service, and diagnoses and responds to a preset failure. And transmitting a request control command for sequentially entering service for a plurality of users to the web server based on a rule for, and transmitting alarm information for the diagnosis of the failure to the manager terminal. , Failure type based server failure diagnosis and response method.
According to claim 1,
Steps (a) and (b) are:
If it is determined through the monitoring that the web server is in a stopped state that does not perform any function even though it is in the running state, it is diagnosed as a stopped fault and based on the rules for preset fault diagnosis and response, And sending a dump generation command and a server restart command to the web server, and transmitting alarm information for a corresponding diagnosis fault to the manager terminal.
According to claim 1,
Step (c) is,
And generating the result information on the diagnosis and response in the form of statistical information arranged in a time series and transmitting the result to the manager terminal.
According to claim 1,
Step (c) is,
And generating result information for the diagnosis and response to the failure at the request of the administrator received from the administrator terminal or a predetermined period of time.
According to claim 1,
Step (c) is,
When the request for information inquiry including a predetermined search condition is received from the manager terminal, generating and transmitting result information on the diagnosis and response to the failure corresponding to the received predetermined search condition, and transmitting the result information to the manager terminal. Type-based server failure diagnosis and response method.
A failure diagnosis unit that performs monitoring on the web server based on real-time status information received from the web server, and detects a type of the detected failure when a failure is detected;
The web server responds in advance to the detected failure according to the control command by transmitting a control command corresponding to the diagnosed fault type to the web server based on a preset rule for fault diagnosis and response. A failure processing unit that controls to perform an immediate response automatically;
A result providing unit generating and transmitting result information on the diagnosis and response to the manager terminal; And
It includes a storage unit for storing the result information for the generated diagnosis and response to the failure,
The disorder diagnosis unit,
Analysis based on the real-time status information including the server resource information and the node resource information for the web server and the rules for diagnosing and responding to the failure, requesting the detected type of failure exceeded failure, large-capacity database data An integrated management server that diagnoses any one of inquiry failure, database connection request exceeded failure, service delay failure, insufficient memory failure, specific service request exceeded failure and stop failure.
The method of claim 19,
The result providing unit,
An integrated management server that limits the number of generations of result information for the diagnosis and response to the diagnosis of a failure diagnosed with the same type to a preset number or less.
The method of claim 19,
Result information on the diagnosis and response to the disorder
Diagnosis execution time indicating the first time the failure was detected, diagnosis end time indicating the last time the failure was detected, the same diagnosis repetition number information indicating the total number of occurrences of the same type of fault as the diagnosed fault type, and fault diagnosis and response Integrated management server including result information generation number information indicating the total number of times the result information is generated.
The method of claim 19,
The disorder diagnosis unit,
An integrated management server that diagnoses the type of the detected failure by performing an analysis based on the real-time status information and history information in which result information for diagnosis and response of failures in the integrated management server is accumulated.
The method of claim 19,
The failure processing unit,
An integrated management server that responds to the detected failure by sending alarm information for the diagnosed failure type to the manager terminal.
delete The method of claim 19,
The failure diagnosis unit diagnoses as a request exceeded failure when it is determined that the usage amount of the available thread (Thead) of the web server exceeds a threshold through the monitoring,
The failure processing unit, when diagnosed as the request exceeded failure, based on the rules for the preset failure diagnosis and response, transmits a control command to bypass the user request to a temporary page, the integrated management server.
The method of claim 25,
Control to bypass the user request to a temporary page,
If it is determined that the available thread usage in the web server exceeds a threshold, and a request exceeded fault is diagnosed, the web server itself can automatically perform a request exceeded fault diagnosis immediately, integrated management server.
The method of claim 19,
When it is determined that a query for a large amount of database data for a service is requested from a user through the monitoring, the failure diagnosis unit diagnoses a failure in a large amount of database data inquiry,
When it is determined that the large-capacity database data inquiry failure is determined, the failure processing unit forcibly terminates the corresponding service requested from the user or performs a service under a predetermined condition based on a preset rule for diagnosis and response to the failure. An integrated management server that transmits control commands to be restricted to the web server.
The method of claim 27,
The control for forcibly terminating the corresponding service requested from the user or for restricting the service by a predetermined condition,
When it is determined that a large database data inquiry for a service is requested from a user within the web server, and a large database data inquiry failure is diagnosed, the web server itself can automatically perform a large database data inquiry failure diagnosis immediately and is integrated. Management server.
The method of claim 19,
If it is determined through the monitoring that the database connection group (Pool) in the web server exceeds a threshold and database connection of a specific data source (Datasource) is excessively used, diagnoses a database connection request excess failure. ,
When the failure processing unit is diagnosed as a failure exceeding the database connection request, the user request is bypassed to a temporary page or the waiting time for obtaining the corresponding database connection is forcibly based on a preset rule for diagnosis and response to the failure. An integrated management server that transmits a control command to reduce to the web server.
The method of claim 19,
The failure diagnosis unit diagnoses a service delay failure when it is determined that a response to a service requested from a user is delayed through the monitoring or a request for a service provided after a predetermined time is delayed from the user,
The failure processing unit, when diagnosed as the service delay failure, transmits a control command for limiting the thread usage rate for the service diagnosed with the failure to the web server, based on a preset rule for diagnosing and responding to the failure. , Integrated management server.
The method of claim 19,
The fault diagnosis unit may cause an out of memory (OOM) or an automatic memory collection phenomenon (Full Garbage Collection, Full GC) due to an error in service logic or excessive memory use in the system through the monitoring. If it is judged that the frequency increases or the situation such as the inability to release the memory occurs continuously, it is diagnosed as insufficient memory,
The failure processing unit, when diagnosed as the memory shortage failure, transmits a dump generation command and a server restart command to the web server based on a preset rule for fault diagnosis and response, and sends alarm information to the manager terminal. To the integrated management server.
The method of claim 19,
The failure diagnosis unit diagnoses a request exceeded failure of a specific service when it is determined that a user request for a specific service received through the monitoring exceeds a limit that can be processed by the web server,
The failure processing unit, when diagnosed as a failure exceeding the request of the specific service, based on a preset rule for failure diagnosis and response, requests control commands for sequentially entering services for a plurality of users Integrated management server that transmits to the web server and transmits the alarm information for the diagnosis of the failure to the manager terminal.
The method of claim 19,
If it is determined through the monitoring that the web server is in a stopped state that cannot perform any function even though the web server is in a running state, the fault diagnosis unit diagnoses a stop fault,
The failure processing unit, when diagnosed as the stop failure, transmits a dump generation command and a server restart command to the web server based on a preset rule for diagnosis and response to a failure, and the diagnosis failure to the manager terminal Integrated management server, to send alarm information about.
The method of claim 19,
The result providing unit,
An integrated management server that generates result information for the diagnosis and response to the failure in the form of statistical information arranged in a time series and transmits it to the manager terminal.
The method of claim 19,
The result providing unit,
Integrated management server that generates the result information for the diagnosis and response to the failure by a predetermined period set in advance or the request of the administrator received from the manager terminal.
The method of claim 19,
The result providing unit,
When an information inquiry request including a predetermined search condition is received from the manager terminal, the integrated management server generates and transmits result information for diagnosis and response to a failure corresponding to the received predetermined search condition.

Images