KR100837575B1 - Process management system and method - Google Patents

Abstract

According to the present invention, even if a problem occurs in a predetermined process, the administrator can check the schematic diagram of processes dependent on the predetermined process and automatically terminate and restart all related processes instead of terminating and restarting the processes in order. Its purpose is to provide a method for managing processes.

The present invention also provides a program management system including a plurality of servers each of which drives at least one process and a DMP server managing the plurality of servers, the DMP server storing associated process information of each of the plurality of processes. Associated process information storage means; Sensing means for sensing an abnormally operated or abnormally stopped process among the plurality of processes; And extracting, from the associated process storage means, the associated process information of the abnormally operated or abnormally stopped process detected by the sensing means, and at least one of the associated processes of the abnormally operated or abnormally stopped process. Startup means for instructing at least one of said plurality of servers to restart one process.

Linked Process, Monitoring, Restart, Process Management

Description

Process management system and method

1 is a diagram schematically showing the structure of a server management system for processing DMP according to a preferred embodiment of the present invention.

2A is a diagram illustrating an example of a plurality of processes driven according to the above embodiment.

FIG. 2B is a diagram illustrating a configuration of an association process of the Aa process among the processes shown in FIG. 2A.

2C is a diagram illustrating an example of an internal configuration of linked process information stored in the DB server.

FIG. 3 is a diagram illustrating a procedure of processing a process of each level when there is a restart request of a process in the linkage process of the structure shown in FIGS. 2A to 2C.

4 is a flow diagram illustrating a procedure of processing a DMP association process in a DMP server in the server management system according to the present embodiment.

5 is a flow diagram illustrating a procedure for restarting a specific process by the service control server of the server management system according to the present embodiment.

6 is a flow diagram illustrating a procedure for the service control server of the server management system according to the present embodiment to terminate a specific process.

The present invention effectively manages the predetermined process and other processes associated with the process management system, more specifically, when a plurality of servers each need to restart or terminate a predetermined process in a system operating one or more plurality of processes. It relates to a process management system capable of doing, and a process management method using the process management system.

Recently, a process management system that monitors and manages various processes performed in a plurality of servers while operating a plurality of servers under one system has been used. In such a system, a server for monitoring is formed separately to check whether a plurality of other servers and a plurality of processes each of the plurality of servers operate normally through the monitoring server. In addition, the monitoring server may terminate the process and send a restart command to the process in case of malfunction or abrupt operation during the process. By the way, a plurality of processes that proceed in each of the plurality of servers may have a function independent of the process of that server or another server, but at least some processes are associated with at least one of the processes of that server or another server. In some cases. In this case, the monitoring server only needs to terminate and restart the process if the problem process is completely independent of other processes. However, when the problematic process is related to another process, if only the problematic process is restarted, the processes dependent on the problematic process may fail due to abrupt termination and restart of the parent process. .

Therefore, when the problematic process has a predetermined relationship with another process, first, the lower process dependent on the problematic process is identified, and the lower process is first terminated and / or restarted, and then the upper process is executed. Must be terminated and / or restarted sequentially. In this case, when a problem occurs in a predetermined process, the administrator checks the configuration of a process linkage process one by one, and it is common for the administrator to directly proceed with the procedure of sequentially terminating and / or restarting from a lower process to a higher process.

However, this manual shutdown and restart not only puts too much pressure on the administrator, but also often compromises the stability of the system, such as a problem with the entire system if the administrator misses the termination and restart of some processes. Has been. In addition, when the administrator is not able to immediately respond to reasons such as leaving the room for a while or solving other problems, there have been inconveniences such as not using the whole system or some systems.

The present invention solves the problem of the conventional process management system using a server for monitoring, even if a problem occurs in a predetermined process, the administrator checks the configuration of the processes dependent on the predetermined process, and the process in order Instead of shutting down and restarting properly, the goal is to provide a process management method in which all relevant processes can be automatically shut down and restarted.

In addition, an object of the present invention is to provide a process management method that can effectively solve a problem by a means promised in advance when a problem occurs in a predetermined process.

In addition, an object of the present invention is to provide a method that can determine whether a problem occurs in real time and effectively solve the problem in real time, regardless of whether the administrator recognizes a problem in a predetermined process.

In order to achieve the above object, the present invention provides a program management system including a plurality of service control servers each driving at least one process and a data manipulation process (DMP) server for monitoring and managing the plurality of servers. As:

The DMP server comprises: linked process information storage means for storing associated process information of each of the plurality of processes; Sensing means for detecting an abnormally operated or abnormally stopped process among the plurality of processes; And extracting, from the associated process storage means, the associated process information of the abnormally operated or abnormally stopped process detected by the sensing means, and at least one of the associated processes of the abnormally operated or abnormally stopped process. Startup means for providing a message to at least one of said plurality of servers requesting to restart one process.

Further, it is preferable that the start-up means first provide a start-up message of a process located at a lower side of the associated process, and then provide a start-up message of a process located at a higher level.

In addition, it is preferable that the starting means restarts the process located at the lowest when restarting the plurality of processes, and repeats the restarting procedure until the abnormally operating or abnormally stopped process is restarted. Do.

In addition, it is preferable that the extracted associated process includes only information of a process located below the abnormally operated or abnormally stopped process.

In addition, the program management system according to the present invention preferably further comprises termination means for forcibly terminating the process stored in the extracted cooperative process information when the cooperative process information of the process detected by the sensing means is extracted.

In addition, the start means and the end means may be integrally provided so that a start command and an end command may be simultaneously provided.

In addition, the program management system is connected to the DMP server, is provided with information of the abnormally or abnormally stopped process, and the restart command for the abnormally operated or abnormally stopped process is recalled It is preferable to further include at least one management terminal provided to the DMP server.

Process management method comprising a plurality of servers each running at least one process in accordance with the present invention for achieving the above object:

Storing associated process information for each of the plurality of processes;

Identifying an abnormally operated or abnormally stopped process among the plurality of processes;

Extracting linked process information of the abnormally operated or abnormally stopped process among the linked configurations for each of the plurality of processes;

Restarting a process (N step process; N is a natural number of two or more) which is located at the lowest of the associated processes of the abnormally operated or abnormally stopped process;

Restarting the process immediately above the restarted N step process (N-1 step process), and then sequentially restarting the process located again above it;

The sequentially restarting the processes located above the step may be repeated until the process which restarts abnormally or stops abnormally.

In addition, it is preferable that the extracted associated process includes only information of a process located below the abnormally operated or abnormally stopped process.

The method may further include stopping all of a plurality of processes included in the associated process included in the extracted information, after extracting the associated process information of the abnormally operated or abnormally stopped process. desirable. In this case, the step of stopping the plurality of processes included in the associated process included in the extracted information may be performed simultaneously with restarting the plurality of processes, or the process of stopping the process first, and then the process It may be configured to proceed with the restarting step.

Hereinafter, a process management method according to a preferred embodiment of the present invention with reference to the accompanying drawings.

First, FIG. 1 is a diagram schematically showing the structure of a server management system according to a preferred embodiment of the present invention. As shown in FIG. 1, the server management system according to the present embodiment includes a DMP (data manipulation process) server 10, a service control server 21, 22, 23, 24, 25, and a management terminal 31, 32, 33, 34). In the above configuration, the service control servers 21, 22, 23, 24, 25 are each driven at least one process which operates independently or in conjunction with another. In addition, information on whether each process is normally executed in each of the service control servers 21, 22, 23, 24, and 25 is provided to the DMP server 10, and the DMP server 10 is particularly abnormally operated. When the process is detected, it is displayed to the outside through the management terminal (31, 32, 33, 34). Then, the at least one server manager observes whether the service control server 21, 22, 23, 24, 25 is normally operated through the management terminals 31, 32, 33, and 34. In addition, when the DMP server 10 detects an abnormally operating or stopped process, the DMP server 10 operates the processes described below with reference to FIG. This restart procedure may be automatically started by the DMP server 10 or may be initiated by a command of the management terminals 31 to 34. In addition, the DMP server 10 mounts the DB server 15 internally or externally, and stores linkage information on at least one process running in each service control server 21, 22, 23, 24, or 25. do.

Fig. 2A shows a plurality of processes running on each service control server 21, 22, 23, 24, 25 according to the present embodiment. That is, the service control servers 21, 22, 23, 24, and 25 according to FIG. 2A are constituted of five, and are referred to as A, B, C, D, E, and F, respectively. In addition, each of the service control servers 21, 22, 23, 24, and 25 drives six processes in common, and these are called a, b, c, d, e, and f in order. Thus, for example, the fifth process of the C server may be abbreviated as Ce. As will be apparent to those skilled in the art, the order and number of processes of the server and the server may be arbitrarily adopted. For example, the process a of server A and process b of server B may be similar in name but not at all. The process adopted.

In addition, FIG. 2B is a diagram illustrating a linkage configuration of the Aa process (that is, a process of the A server) among the service control server and the process shown in FIG. 2A. Specifically, in the present embodiment, if the process Aa is a one-level process, the one-level process is associated with two-level processes (Ba, Ca, Da, Ea processes), and the Ba process among the two-level processes is again a three-level process (Ab , Bb process). In this way, because the Aa process is associated with a number of other processes (that is, because many processes are dependent on the Aa process), if the Aa process terminates abnormally, it is not enough to restart the Aa process and dependent on the Aa process. All the processes (from the first level process to the N level process) that have been made must be restarted in a predetermined order (the order of the N level process to the first level process). In addition, the linkage process information for each process such as the Aa process is stored in the DB server 15.

FIG. 2C is a diagram illustrating an example of an internal configuration of linked process information stored in the DB server 15. As shown in FIG. 2C, the DB server 15 includes all information necessary for data manipulation processing for all processes such as cooperative process information of the Aa process and cooperative process information of the Ba process. For example, as shown in FIGS. 2B and 2C, in the case of one-level information (Aa process), the two-level information (Ba, Ca, Da, and Ea processes) are directly linked to each other. , The Db process is associated with the Db process, the Db process among the three levels of information is associated with the Bd and Cc processes, which are four levels of information, the Cc process among the four level information is associated with the Ec process, which is five levels of information, It is associated with Ce and Cf processes, which are six levels of information, and the Ce process among six levels of information is associated with the De process, which is seven levels of information, and the De process among seven levels of information, with the Ef process, which is eight levels (N level of FIG. All of this information is stored in the DB server 15, after which the necessary part is extracted by the DMP server 10 and used for the data manipulation process.

FIG. 3 is a diagram illustrating a procedure of processing a process of each level when there is a restart request of an Ec process such as a linkage process of the structure shown in FIGS. 2A to 2C.

As shown in FIG. 3, when a process (Ec process in FIG. 3) abnormally stopped through a predetermined management terminal 33 is detected, the command to stop the Ec process from the management terminal 33 to the DMP server 10. This is provided. When the DMP server 10 is provided with an abort command of the Ec process, the DB server 15 extracts the coordination process information for the Ec process and creates a cooperative process structure for a specific process as shown in FIG. 2B. As mentioned earlier, the associated process structure for a particular process is sufficient to produce a list of lower levels that are affected by a particular process, and does not need to create a higher level list that is not affected by a particular process.

Subsequently, based on the list, the DMP server 10 first instructs the E server 25 to restart the Ef process, which is the lowest level (N level) process, and provides the processing result to the management terminal 33. . Subsequently, the DMP server 10 instructs restart of the De process and the Df process, which are the N-1 level processes, which are the next lower levels, and provides the processing result to the management terminal. In this way, the DMP server 10 instructs the server 21 to 25 which the process is driven sequentially from the lowest process to the highest process in order to restart the process and provides the result to the management terminal 33. The restart of the process and the provision of restart results are completed by proceeding to the Ec process, which was initially deemed necessary to restart.

Although the above has been described with reference to the case of restarting each process, even if it is necessary to terminate each process can proceed by the same procedure and sequence as the restart procedure.

Next, a general procedure of processing the DMP association process in the DMP server 10 in the server management system according to the present embodiment will be described in detail with reference to FIG. 4.

First, when a restart is required for a predetermined process, the DMP server 10 checks the information of the process requiring restart and the linked process information of the process (S101).

Subsequently, the DMP server 10 checks whether a restart response message for at least one of the one or more processes requiring restart is received from the service control servers 21 to 25 (S102). If the restart response message is not received, the DMP server 10 checks whether a message for restarting a specific process is received from the management terminals 31 to 34 (S103). If the restart request message is received, the process proceeds to the next step S104, and if not, the process repeats S102.

After receiving the restart request from the management terminals 31 to 34, the DMP server 10 checks the coordination process information of the process requiring restart (S104), checks the N level information which is the lowest level among the cooperative process information, The service control servers 21 to 25 driving the N level processes are provided with restart messages for processes corresponding to the respective N levels (S105). Processes included in the N level may be composed of one or more than one, and in the case of a plurality of processes, each process may be driven in the same service control server or may be driven in different servers. Subsequently, the DMP server 10 checks whether all restart messages for the process corresponding to the N level are provided (S106), and if not provided again, provides a message not provided again (S107). Or, if all messages have been provided, repeat the previous step S102. That is, as a result of providing a restart message for a process corresponding to the N level, the DMP server 10 checks whether at least one of the processes is restarted (S102), and a signal indicating that at least one process has been restarted is received. If it is input, the process proceeds to the next step S108, and if a signal for restarting is not input, the steps S102 to S107 are repeated.

When a signal indicating that at least one process is restarted is input, the DMP server 102 checks process information corresponding to the N level again, and deletes the process of restarting from the associated process information (S109). Subsequently, after confirming that the restarted process is normally restarted (S110), if the restarted process is not correctly restarted, such as abnormally running, it is assumed that the inspection is necessary, and an error message is sent to a predetermined management terminal. After transmitting (S120) and deleting the association process information (S121), the DMP association process is terminated. On the other hand, if the restarted process is normally driven to notify the management terminal (S111), and checks whether all of the N-level processes are restarted. If all of the N level processes have not been restarted, the step S102 is repeated until all of the N level processes have been restarted.

If all of the N level processes are restarted, it is checked whether the N level is the highest level (1 level) (S113). If the level is 1, all processes are regarded as restarted, and the management terminal is notified that the restart process is completely completed (S118), after deleting the linkage process information (S119), the DMP linkage process ends.

Alternatively, if the restarted N-level process is at least two levels higher, the restart process is restarted by setting the N-1 level process as a new N-level process (S114). In this case, first, the cooperative process information of the new N level is checked, and a restart message for a process corresponding to each N level is provided to the service control servers 21 to 25 driving the new N level process (S115). . Subsequently, the DMP server 10 checks whether all restart messages for the process corresponding to the new N level are provided (S116), and if not provided again, provides the non-provided message (S117). Or, if all messages have been provided, repeat the previous step S102. That is, as a result of providing a restart message for a process corresponding to the N level, the DMP server 10 checks whether at least one of the processes is restarted (S102), and a signal indicating that at least one process has been restarted is received. If it is input, the process proceeds to the next step S108, and if a signal for restarting is not input, the steps S102 to S107 are repeated. When all new N-level processes are completed (S112), as described above, it is checked whether it is at one level, and the above procedure is repeated until the restarted process is at one level.

Next, a procedure of restarting a specific process in the service control servers 21 to 25 provided with the restart message of the specific process from the DMP server 10 will be described with reference to FIG. 5.

First, if the service control server 21 to 25 receives the restart message of the predetermined process from the DMP server 10 (S201), and forms a thread (S202). Subsequently, after executing the formed thread (S203), if there is no response even after a predetermined time elapses, such as when the thread does not run normally or the process does not operate at all due to a system error or the like (S204), the DMP server 10 receives an error. The message is transmitted (S209). Alternatively, if there is a reaction within a predetermined time, it is checked whether the process is executed (S205). If the process is running, the process is forcibly terminated (S206), and the step S204 is repeated. Alternatively, if the process is not running, restart the process (S207), and transmit a success message indicating that the restart process of the specific process is completed to the DMP server 10 (S208).

Next, a procedure of terminating a specific process in the service control servers 21 to 25 that has received the termination message of the specific process from the DMP server 10 will be described with reference to FIG. 6.

First, if the service control server (21 to 25) receives the end message of the predetermined process from the DMP server 10 (S301), and forms a thread (S302). Subsequently, after executing the formed thread (S303), if there is no response even after a certain time elapses, such as the thread does not run normally or the process does not operate at all due to a system error or the like (S204), the DMP server 10 fails. The message is transmitted (S308). Alternatively, if there is a reaction within a predetermined time, it is checked whether the process is executed (S305). If the process is running, the process is forcibly terminated (S306) and the DMP server 10 terminates the specific process. A success message indicating that the procedure is completed is transmitted (S307). Alternatively, when the process is not running, a success message indicating that the termination process of the specific process is completed is immediately transmitted to the DMP server 10 (S307).

The process management system and method according to the present invention has been described in detail above with reference to a preferred embodiment. However, one of ordinary skill in the art will appreciate that various modifications and variations are possible in addition to the systems and methods described in this embodiment. Therefore, the scope of the present invention is limited only by the claims which will be described later.

In the process management system using the DMP server according to the present invention, even if a problem occurs in a predetermined process, the administrator checks the schematic diagram of processes dependent on the predetermined process, and automatically terminates and restarts the process in order, instead of automatically. It can provide a process management method in which all related processes can be terminated and restarted.

In addition, the present invention can provide a process management method that can effectively solve a problem by a means promised in advance when a problem occurs in a predetermined process.

In addition, when a problem occurs in a predetermined process, the present invention can identify a problem in a process in real time regardless of whether the administrator recognizes it and effectively solve it.

Claims (11)

A program management system comprising a plurality of servers each running at least one process and a data manipulation process (DMP) server managing the plurality of servers: The DMP server; Associated process information storage means for storing associated process information of each of the plurality of processes; Sensing means for detecting an abnormally operated or abnormally stopped process among the plurality of processes, and Extract the associated process information of the abnormally operated or abnormally stopped process detected by the sensing means in the associated process storage means, and at least one of the associated processes of the abnormally operated or abnormally stopped process Startup means for providing a message to at least one of said plurality of servers requesting to restart the process of System comprising a. The system according to claim 1, wherein the start-up means first provides a start-up message of a process located below one of the associated processes, and then provides a start-up message of a process located above. The method according to claim 1 or 2, wherein the starting means restarts the lowest-most process when restarting the plurality of processes, and restarts the abnormally operated or abnormally interrupted process. Repeating the restart procedure. The system of claim 1 or 2, wherein the extracted associated process includes only information on processes located below the abnormally operated or abnormally stopped processes. The method of claim 1, wherein the system further comprises terminating means for forcibly terminating the process stored in the extracted associated process information when the associated process information of the process detected by the sensing means is extracted. System. 6. The system according to claim 5, wherein the start-up means and the end means are integrally provided so that a start command and a stop command are provided simultaneously. The system of claim 1 or 2, wherein the system is connected to the DMP server to receive information about the abnormally operated or abnormally stopped process, and wherein the system is connected to the abnormally operated or abnormally stopped process. At least one management terminal for providing a restart command for the DMP server. A process management method comprising a plurality of servers, each of which runs at least one process: Storing associated process information for each of the plurality of processes; Identifying an abnormally operated or abnormally stopped process among the plurality of processes; Extracting linked process information of the abnormally operated or abnormally stopped process among the linked configurations for each of the plurality of processes; Restarting a process (N step process; N is a natural number of two or more) which is located at the lowest of the associated processes of the abnormally operated or abnormally stopped process; And Restarting the process immediately above the restarted N step process (N-1 step process), and then sequentially restarting the process located again above it; And sequentially restarting the processes located above the step until the process which restarts abnormally or stops abnormally stopped. The method of claim 8, wherein the extracted associated process includes only information of processes located below the abnormally operated or abnormally stopped processes. The method of claim 8 or 9, further comprising: extracting the linkage process information of the abnormally operated or abnormally stopped process, and then suspending all the processes included in the linked process included in the extracted information. Process management method further comprising the step of. The method of claim 10, wherein the stopping of the plurality of processes included in the associated process included in the extracted information may be performed simultaneously with restarting the plurality of processes, or stopping the process first. And then restarting the process.

Images