KR19990058887A - Process control method that acts as a parent or child process by configuration of different processes - Google Patents

Abstract

The present invention relates to a program method in a UNIX system that supports multi-process. When the processes that should be configured in a relationship between a parent and a child are configured as different processes for various reasons, they serve as child processes. It is about how to control the process.

Process control method consisting of different processes of the present invention to act as a parent process, in the case that the parent-child process structure is not used, the step of checking whether the process of performing the child role is first read by reading the shared memory If the process is running, storing in each variable, checking if a message has arrived from the process acting as a child, if the message has arrived, storing it in that variable, otherwise checking if there is work to do Performing a work instruction by calculating a maximum work execution time, and destroying and reinitializing a child process if a work completion result is not delivered within the maximum work time.

Process control method consisting of different processes according to the present invention to perform the role of a child process, in the case of not using a parent-child process structure, the step of registering its PID in the shared memory, attempt to initialize Step, when the work order command is received from the parent process, the step of reporting the result.

The present invention can improve the characteristics of using resources efficiently by configuring different processes without using a parent-child process structure when the time for generating a child process or overload is large.

Description

Process control method that acts as a parent or child process by configuration of different processes

The present invention relates to a program method in a UNIX system that supports multi-process. When the processes that should be configured in a relationship between a parent and a child are configured as different processes for various reasons, they serve as child processes. It is about how to control the process.

In a UNIX system, there is usually only one live process waiting for a particular purpose. When a particular operation is performed, a child process is created and executed.

The biggest advantage of constructing a process structure between parent and child is that the parent process can save the child process only when it needs to work, and if the child's behavior is abnormal, it can kill the child process and revive it to increase the work efficiency. To reduce unnecessary memory waste.

The child process is automatically terminated after the specific task is finished, and the parent process that created the child process receives the termination message and then resumes waiting.

The reason for organizing processes in this way is to allow for multiple tasks at the same time. However, depending on the situation, it can not be configured. For example, there is a problem in that it takes a lot of time to initialize as a child process is newly created, or if there is an overload, it is necessary to consider how to construct a parent-child process structure.

Other objects and advantages of the present invention will become more apparent upon reading the following detailed description and with reference to the drawings below:

The present invention is to solve the problems as described above, but in order to configure the process in a parent-child relationship, but in reality difficult, it is intended to enable the efficient use of resources by configuring and controlling with different processes.

1 is a block diagram illustrating a process control method serving as a parent process by configuration of different processes according to an embodiment of the present invention.

2 is a block diagram illustrating a process control method that acts as a child process by configuration of different processes according to an embodiment of the present invention.

One preferred embodiment of the process control method is composed of different processes of the present invention created to achieve the above object to serve as a parent process,

In the case of not using the parent-child process structure,

Checking whether a process serving as a child by reading the shared memory for the first time is running;

Storing in each variable if the process is running;

Checking whether a message has arrived from a process playing a child role;

Storing the message in a corresponding variable if the message arrives, and checking whether there is content to be worked on;

Calculating a maximum work execution time and performing a work instruction; And

If the work completion result is not delivered within the maximum work time, the process includes destroying and reinitializing the child process.

In the present invention, in the storing in the respective variables, each variable is preferably an initialization success variable and a task success variable.

One preferred embodiment of the process control method composed of different processes according to the present invention for achieving the above object to play the role of a child process, in the case of not using a parent-child process structure,

Registering its PID in shared memory;

Attempting an initialization operation;

Receiving a work order command from the parent process includes performing this and reporting the result.

In the present invention, in the PID registration step, the PID registration is preferably registered in a unique memory so that the process of performing the child role can be eliminated, if necessary,

In the attempting step, it is preferable to continue to attempt the initialization even if the initialization fails.

It is desirable to write the state for all of the above steps to the shared memory for control of the parent processor.

While the invention is susceptible to various modifications and alternative forms, only specific embodiments thereof are shown in the drawings and will be described in detail. It is to be understood, however, that the present invention is not limited to the specific forms referred to in the specification, but rather that the invention is intended to cover all modifications, equivalents, and substitutions within the spirit and scope of the invention as defined by the appended claims. It should be understood to include.

Hereinafter, a method of controlling a process serving as a parent or child process by a configuration of different processes of the present invention will be described with reference to FIGS. 1 and 2.

The present invention provides a method of controlling a process by imitating it to be almost similar, although not a relationship between parent and child. When a child acting process is first created, it writes its own process identification (PID) to shared memory, and when a parent acting process needs to remove a child role, it can read and remove the PID in its native memory.

The child role process writes all the information that the parent process needs for control into its own memory and also informs the parent process with a message.

The reason for notifying the message is to inform the user of the change as soon as possible. The reason for writing the intrinsic memory is to know the operation state of the child process when the process of the parent role dies and restarts.

Referring to FIG. 1, a process acting as a parent checks whether a process of a child role is running after first reading native memory. If it is running, it is saved in initialization success variable and task success variable respectively. It then checks to see if a message from the child role process has arrived and if so, stores it in that variable, otherwise checks whether there is something to work on. If the child role process is alive and ready to work, the maximum task execution time is calculated and ordered. If the task completion result is not delivered within the maximum task execution time, the child process is removed and newly initialized.

Referring to FIG. 2, the child role process registers its PID in shared memory so that the parent process can be removed as needed and registers a signal so that it can first send a message to the parent process before it is removed. do. After that, the initialization operation is performed, and even if it fails, it continues to attempt initialization. When a work order command comes from a parent process, it does it and reports the result. All of this progress is recorded in shared memory so that the parenting process knows what the childing process is doing at any time.

According to the present invention operating as described above, if the time for generating a child process, or the overload is large, it is possible to configure a different process without using a parent-child process structure, thereby improving the characteristics of using resources efficiently. have.

Claims (6)

In the case of not using the parent-child process structure, Checking whether a process serving as a child by reading the shared memory for the first time is running; Storing in each variable if the process is running; Checking whether a message has arrived from a process playing a child role; Storing the message in a corresponding variable if the message arrives, and checking whether there is content to be worked on; Calculating a maximum work execution time and performing a work instruction; And And destroying and reinitializing the child process if the work completion result is not delivered within the maximum work time. The method of claim 1, wherein in the storing of the variables in the variables, each variable is composed of different processes that are initialization success variables and task success variables. In the case of not using the parent-child process structure, Registering its PID in shared memory; Attempting an initialization operation; A process control method that is composed of different processes that act as child processes, including the step of receiving work order instructions from the parent process and reporting the results. 4. The process of claim 3, wherein in the PID registration step, the PID registration is a process that is configured as a different process to perform a child process role, registering in a unique memory so that a process for performing a child role can be eliminated as necessary. Control method. 4. The process control method according to claim 3, wherein in the attempting step, the initial process continues to be attempted even if the initialization fails. 6. The process control method according to any one of claims 3 to 5, wherein the processes for all the steps are written in a shared memory for control of a parent processor, and are configured as different processes.