TWI585687B - The method, device, and system of the plug - Google Patents

Description

Method, device and system for running plug-in program

The invention relates to the field of computers, and in particular to a method, device and system for operating a plug-in program.

With the development of network technology, some network tools have been used more and more widely. For example, instant messaging tools have gradually become an indispensable network tool in user life. At the same time, with the widespread use of these network jobs, users have also created more demand for them. In this context, the software developed by the developers of these network tools is often unable to meet the needs of users, that is, the services provided by the core business programs of these network tools can no longer meet the needs of customers. Therefore, by formulating specifications, the program modules that conform to the specifications, that is, the plug-ins, are allowed to access the software of the corresponding network tools, so that the plug-in program and the core business program together provide more services for the users. This has become an unstoppable trend.

Currently, there are several ways to run the plugin:

1. The plug-in program and the core business program run in the same process. Thus, the support system of the plug-in program is simple, and the plug-in program and the core business program have high interaction efficiency, but because in a process, once the plug-in has a problem, it will directly lead to The core business program is abnormal or crashes, and because of the large number of modules in the core business program, it is difficult to specifically locate which module or plug-in has a problem;

2. The plug-in program runs outside the running process of the core business program, and each plug-in program has a process. This prevents the core business program from being affected by the plug-in program, but when the number of plug-ins is large, it will be occupied a lot. System resources of the client to reduce the user experience;

3. The plugin runs outside the running process of the core business program, and all the plugins run in a specific process, thus saving the system resources of the client, but when an external program is abnormal, it may cause all the plugins. Not available or abnormal, but also not conducive to the specific positioning of which plug-in problem.

It can be seen that the existing plug-in running mode has higher stability requirements for plug-ins, or more user resources are occupied. However, with the increase of plug-in developers, the technical level is not good, and the stability of the plug-in is difficult to guarantee. Usually it is not like the core business program. The plug-in will often crash or malfunction, so use the above method 1 or 3. , because the individual plug-ins run unstable, resulting in the core business program or all plug-ins running abnormally. At the same time, the resources of the client are limited. The above method 2 will occupy the resources of the client too much, which may cause the client to run slowly, and may even cause the client system to crash.

In view of this, the embodiment of the present invention provides a method for operating a plug-in program to solve the problem of unstable system operation caused by the operation mode of the plug-in in the prior art.

A method for operating a plug-in program according to an embodiment of the present invention includes: obtaining a stability level of a plug-in program to be run; determining a plug-in running mode corresponding to the obtained stable level according to a correspondence between the saved stable level and the plug-in running mode. The running plugin is run according to the determined plugin running mode.

The user terminal of the plug-in running according to the embodiment of the present invention includes: a first acquiring unit, configured to obtain a stable level of the plug-in program to be run; and a storage unit, configured to save a correspondence between the stable level and the running mode; a unit, configured to determine, according to a correspondence between a stable level and a running mode saved by the storage unit, a plug-in running mode corresponding to the obtained stable level; and an operating unit configured to run the to-be-running plug-in according to the determined plug-in running mode .

A system for operating a plug-in program according to an embodiment of the present invention includes: a user end, configured to obtain a stable level of a plug-in to be run from a stable level of a plug-in saved by a server, according to a saved stable level and an operating mode. Corresponding relationship, determining a plug-in running mode corresponding to the obtained stable level, running the to-be-operated plug-in according to the determined plug-in running mode; and the server is configured to save the stable level of the plug-in.

In the embodiment of the present invention, the stability level of the plug-in to be run is obtained, and according to the corresponding relationship between the saved stable level and the running mode, the plug-in running mode corresponding to the obtained stable level is determined, and the operating mode is determined according to the determined plug-in running mode. The running plug-in program is mentioned, so that different plug-in running modes are selected according to different stability levels of the plug-in, thereby improving the stability of the system operation and effectively controlling the resource occupation of the plug-in.

In the embodiment of the present invention, when the service provided by the core business program of the network tool running on the user end cannot meet the requirements of the user, the plug-in program of the network tool needs to be run to meet the user's needs, and the plug-in program runs. The process of referring to FIG. 1 specifically includes: Step 101: Obtain a stability level of the plug-in to be run.

Here, the stability level of the plug-in to be run can be obtained from the stable level of the plug-in saved by the server. It is also possible to obtain the stability level of the plug-in to be run from the stable level of the locally saved plug-in. The stability level of the plug-in can be divided according to the running status of the plug-in in the set time. For example, the stability level is divided into three levels, which are stable, normal, and unstable. Here, an upper threshold is set, and a lower threshold is set. When the number of occurrences of the abnormality of the plugin is greater than or equal to the upper threshold, the stability level of the plugin is unstable; when the set time, the plugin runs. When the number of occurrences of the abnormality is less than or equal to the lower threshold, the stability level of the plugin is stable. When the number of times the abnormality of the plugin is abnormal is between the lower threshold and the upper threshold, the stability level of the plugin is For ordinary. Of course, the stability level can also be divided into only two levels, stable and normal. Then, only one threshold is set. When the number of abnormalities of the plug-in running is greater than the threshold within the set time, the stability level of the plug-in is Ordinary, otherwise stable. It is also possible to divide the stability level into four levels, stable, relatively stable, normal and unstable, then it is necessary to set three thresholds, a first threshold, a second threshold, and a third threshold, whose size is abnormally decremented, then, when setting The stability level of the plugin is unstable when the number of times the exception occurs in the plug-in program is greater than or equal to the first threshold; when the set time is exceeded, the number of times the plug-in program runs abnormally is greater than or equal to the second threshold, and is less than When the first threshold is used, the stability level of the plugin is normal; when the number of times the abnormality of the plugin operation is greater than or equal to the third threshold and less than the second threshold, the stability level of the plugin is compared. Stable; when the number of times the exception occurs during the set time is less than the third threshold, the stability level of the plugin is stable.

Here, the server or the local client can establish a corresponding relationship between the plug-in and the stability level. After determining the plug-in information of the plug-in to be run, for example, identifying the information, the stability level of the plug-in to be run can be obtained.

Step 102: Determine a plug-in running mode corresponding to the obtained stable level according to the correspondence between the saved stable level and the running mode.

Here, the correspondence between the stability level and the operation mode may be set in advance and saved in the storage unit of the client. The plug-in running mode includes: starting different running processes for different plug-in programs to run the plug-in running program, and the correspondence between the stable level and the plug-in running mode includes: when the stable level is stable, the plug-in The program running mode is to run the plugin in the shared loader process; when the stability level is normal, the plugin running mode is to start the independent loader process to run the plugin. The loader process is an executable program that can be created by the core service program of the network tool running on the client, or by other programs in the client, including the shared loader process and the independent load. Into the process. A shared loader process can run multiple plugins, and a separate loader process runs only one plugin. In this way, multiple stable levels of stable plug-ins running in a common loader process can reduce the occupation of client resources; the stability level is normal plug-in running in a separate loader process, which can be reduced The plug-in increases the stability of the system by affecting the core business process and the stability of the plug-in.

Here, when the stability level includes stability, normality, and instability, the correspondence between the stability level and the plug-in operating mode further includes: when the stability level is unstable, the plug-in running mode is to abandon the running plug-in, or notify the user to select whether to run. The plugin. When the user chooses to run, start the independent loader to run the plugin; when the user chooses not to run, give up running the plugin.

When the stability level includes stability, stability, normality and instability, the correspondence between the stability level and the plug-in operating mode includes: when the stability level is relatively stable, the plug-in running mode is put into the shared loader process. The plugin, here, the shared loader can be a shared loader that is stable with a stable level, or it can be another shared loader.

Thus, the stability level of the to-be-operated plug-in obtained in step 101 is different, and the operation mode is different.

Step 103: Run the plug-in to be run according to the plug-in running mode determined in step 102.

In this embodiment, an exception may occur during the running process of the plug-in program. Therefore, the method for running the exception program in the present embodiment further includes: when the plug-in program runs abnormally, the client acquires the abnormal information of the plug-in running, and saves the abnormality. The information is sent to the server for saving, and the abnormal information includes: the cause of the abnormality, and the plug-in information of the running plug-in. The cause of the exception includes: the loader process crashes, the plug-in execution thread is dead, or the memory or processor that the plug-in program takes too long. The plugin information specifically includes: the identification information, name, version, and the like of the plugin.

In this way, the client or the server can determine the stability level of the plug-in according to the abnormal information of the plug-in running, and can periodically adjust, that is, according to the saved abnormal information, to determine the abnormal number of times the plug-in appears within the set time. And adjusting the stability level of the plugin according to the number of abnormalities. The specific adjustment process includes: after setting the time trigger, that is, after the set time arrives, acquiring the plug-in information of the plug-in program whose usage times exceed the threshold in the set time, and determining the acquired plug-in according to the saved abnormal information. The number of abnormalities of the information within the set time, and the stability level of the plug-in is automatically adjusted according to the number of abnormalities. Of course, for the next regular adjustment, the number of abnormalities and the number of uses need to be reset. Here, it can also be manually adjusted. The customer service personnel can count the running status of the plug-in program within a certain period of time. When the number of times the plug-in program runs abnormally exceeds the set value within the set time, the stability level of the plug-in program is modified.

The plug-in running process of running the network tool on the client side can be managed by the core business program of the network tool, or can be managed by other programs in the client side. In this embodiment, the core business program not only provides the core business of the network tool, but also includes a program for managing the plug-in program. The loader process is a stand-alone executable in the client, and the core business program can be a load process.

The embodiments of the present invention are further described in detail below with reference to the accompanying drawings. Referring to FIG. 2, the specific process of the plug-in running is as follows: Step 201: The core service program obtains the stability level of the plug-in to be run.

Here, when the core service program requests the server to obtain the stability level of the to-be-runned plug-in, the server obtains the plug-in information of the to-be-runned plug-in according to the request, for example, the identification information, the saved plug-in and the plug-in program. In the correspondence between the stable levels, the stability level of the corresponding to-be-running plug-in that queries the identification information is returned, the queried stable level is returned to the core business program, and the number of times the recorded plug-in plug-in is used is incremented. Among them, the level of stability includes stability, generality and instability.

When the core service program obtains the stable level of the to-be-runned plug-in from the local client, the client queries the stability level of the to-be-run plug-in in the stable level of the saved plug-in, and returns the queried stable level to the core service. The program, and increments the number of times the recorded plug-in is used. Similarly, the level of stability includes stability, generality, and instability.

Step 202: The core service program determines that the stability level of the to-be-running plug-in is stable, normal, or unstable. When the stability level is stable, step 203 is performed; when the stability level is normal, step 206 is performed; when the stability level is If it is unstable, go to step 207.

Step 203: The core service program determines whether the shared loader process exists. When it exists, step 205 is performed, and when it does not exist, step 204 is performed.

Step 204: The core business program creates a shared loader process.

Step 205: The shared loader process runs a plugin to be run.

Step 206: The core business program establishes a separate loading process for the normal to-be-running plug-in, and the loading process runs the plug-in to be run.

Step 207: The plugin running process ends.

In the above embodiment, when there is no shared loader process, step 204 creates a shared loader process. Here, the core business process can also be directly used as a shared loader in the core business process. A stable plugin that runs at a stable level.

In the embodiment of the present invention, when the plug-in program runs abnormally, the user terminal may also obtain the abnormal information of the plug-in program, and save the abnormal information or send the information to the server. The plugin is running in the loader process. Therefore, if the plugin runs abnormally, that is, when the loader process is abnormal, the client obtains the exception information of the loader process and saves the exception information, or Send to the server, see Figure 3, the process is as follows: Step 301: Get the exception information of the loader process, the exception information includes: the cause of the exception of the loader process, and the plug-in running in the loader process Plugin information.

Here, when the loader process is a separate loader process, the exception information includes: the cause of the exception, and the plugin information of the plugin running in the independent loader process, when the loader process is shared When the loader process, the exception information includes: the cause of the exception, and the plugin information of all the plugins running in the shared loader process. The exception information can also include: the exception context of the loader process. Here, the exception context includes: memory image of the loader process, function stacking context, plug-in program loaded by the loader, and other technical information. Through the analysis of this information, you can know the time corresponding to the context, the plug-in running in the loader process, the instructions currently being executed, and so on.

The client can obtain the cause of the abnormality of the loader process through detection. The cause of the exception may be that the loader process has crashed, the plugin execution thread has been deadlocked, or the plugin has taken up too much memory or processor for a long time. Here, the loader process can self-test to find the cause of the exception, or the other process of the client can detect the loader process to find the cause of the exception.

The client can also obtain the abnormal information of the loader process through the form of a periodic heartbeat report, that is, the loader process periodically sends a message to the core business program, which is called a heartbeat, when the client core business program is If the heartbeat report of the loader process is not received within the set time, it indicates that the loader process has an exception.

Step 302: Send the obtained abnormal information to the server.

Here, the UE can directly send the obtained abnormal information to the server, and can also process the abnormal information within the set time, for example, statistics the abnormal number of occurrences of each plug-in in the set time, and then send it to the server.

In this step, of course, only the acquired abnormal information is saved, and is not sent to the server, and the abnormal information in the set time is directly processed.

After the client or the server collects the abnormal information, it can periodically adjust the stability level of the plug-in according to the collected abnormal information. Referring to Figure 4, the process of periodically adjusting the stability level of any plug-in is as follows: Step 401: Arrive After the set time, the number of times the plug-in is used within the set time is obtained.

During the set time, when the client accesses the server once and obtains the stable level of the plug-in to be run, the server increments the number of times the plug-in is used, so that the server saves each plug-in. The number of times the program is used during the set time. Alternatively, each time the client obtains the stable level of the to-be-runned plug-in from the local, the number of times the plug-in is to be incremented is incremented, so that the client saves the number of times each plug-in is used within the set time. .

Here, the plug-in to be adjusted in this embodiment runs 120 times in a set month.

Step 402: Compare the obtained usage times with the set threshold. When the acquired usage times are greater than or equal to the set threshold, step 403 is performed. Otherwise, the process ends, the plug-in usage rate is not high, and the stability level does not need to be adjusted.

Step 403: According to the saved abnormal information, the abnormal number of occurrences of the plug-in is obtained.

The exception information obtained by the client or the server each time includes the plug-in information. Therefore, the number of abnormalities in the abnormal information of the plug-in program in the set time can be queried, for example, within one month. 30 times.

Step 404: Adjust the stability level of the plugin according to the number of times the plugin is used and the number of abnormalities.

Here, the stability level is divided into three levels, which are stable, normal, and unstable. Set an upper threshold, a lower threshold, for example, the upper threshold is 20%, and the lower threshold is 5%. When the number of abnormal times divided by the number of uses is greater than or equal to the upper threshold, the stability level of the plugin is unstable; When the number of abnormal times divided by the number of uses is less than or equal to the lower threshold, the stability level of the plugin is stable. When the number of abnormal times divided by the number of uses is between the lower threshold and the upper threshold, the plugin is The stability level is normal. For example, the initial stability level of the plug-in program is normal, and in this adjustment period, that is, within one month set, the number of uses is 120 times, and the number of abnormal times is 30 times, then, 30/120=25%, 25 %>20%, the stability level of the plugin is adjusted to be unstable.

Of course, the stability level can also be divided into only two levels, stable and normal, then only one threshold is set. When the number of abnormal times divided by the number of uses is greater than the threshold, the stability level of the plugin is normal, otherwise For stability. It is also possible to divide the stability level into four levels, stable, relatively stable, normal and unstable, then it is necessary to set three thresholds, the first threshold, the second threshold, and the third threshold, whose size is abnormally decremented, then, when When the number of abnormal times divided by the number of uses is greater than or equal to the first threshold, the stability level of the plugin is unstable; when the number of abnormal times divided by the number of uses is greater than or equal to the second threshold and less than the first threshold, The stability level of the plugin is normal; when the number of abnormal times divided by the number of uses is greater than or equal to the third threshold and less than the second threshold, the stability level of the plugin is relatively stable; when the number of abnormal times is divided by When the value obtained by the number of times is less than the third threshold, the stability level of the plugin is stable.

Step 405: The number of times of use and the number of abnormal times are reset, that is, the number of times of use and the number of abnormal times are cleared, and the counting in the next adjustment period is restarted.

In the embodiment of the present invention, the stability level of the plug-in program can be divided into multiple types, and the user can use multiple shared loader processes in the process of running the plug-in program. For example, the stability level is divided into: stable and stable. Normal, and unstable, then all stable plugins with stable levels can be put into a common loader process, and all stable plugins with stable levels can be put into another shared loader. Running in the process. Then, the correspondence between the stability level saved on the client and the running mode of the plugin includes: when the stable level of the acquisition is stable, the running mode of the plugin is to run the plugin in the first shared loader process; When the stability level is relatively stable, the plug-in running mode is to run the plug-in in the second shared loader process; when the stable level is obtained, the plug-in running mode is to start the independent loader process. Plug-in; when the stable level obtained is unstable, the plug-in running mode is to abandon the running plug-in.

According to the foregoing method of the embodiment of the present invention, a client running the plug-in program can be constructed. Referring to FIG. 5, the first acquiring unit 510, the storage unit 520, the first determining unit 530, and the running unit 540 are included. The first obtaining unit 510 is configured to obtain a stable level of the plug-in to be run.

The storage unit 520 is configured to save a correspondence between the stability level and the operation mode.

The first determining unit 530 is configured to determine, according to the correspondence between the stability level and the operating mode saved by the storage unit 520, the plug-in operating mode corresponding to the acquired stable level.

The running unit 540 is configured to run the to-be-running plug-in according to the determined plug-in running mode.

Each of the above units may be a separate program in the user terminal, or may execute the functions of the first obtaining unit 510, the storage unit 520, and the first determining unit 530 by a core business program, and the loader process performs the function of the running unit 540. .

Further, the user terminal further includes a second obtaining unit, configured to obtain abnormal information of the running of the plugin, wherein the abnormal information includes: an abnormal reason, and plugin information of the plugin. The storage unit 520 is further configured to store the abnormal information.

In this way, after the set time is triggered, the user determines the number of abnormal times that the plug-in program appears within the set time according to the abnormal information stored in the storage unit, and adjusts the stability level of the plug-in according to the abnormal number of times. Specifically, the plug-in program information of the plug-in program whose usage times exceed the threshold in the set time period is obtained, and the abnormality information of the acquired plug-in program information is determined according to the saved abnormal information, and is adjusted according to the abnormal number of times. The stability level of the plugin, and the number of abnormalities and the number of uses are reset.

Of course, the UE may further include a sending unit, configured to send the abnormal information acquired by the second acquiring unit.

In the embodiment of the present invention, the stability level of the plug-in to be run can be obtained from the stable level of the plug-in saved by the server. Therefore, a system running the plug-in can be constructed, including FIG. 6, including the client 100 and Server 200.

The client 100 is configured to obtain a stable level of the plug-in to be run from the stable level of the plug-in saved by the server 200, and determine a plug-in corresponding to the obtained stable level according to the corresponding relationship between the saved stable level and the running mode. In the running mode, the to-be-running plug-in is run according to the determined plug-in running mode.

The server 200 is configured to save the stability level of the plug-in.

Therefore, the client 100 includes: a first obtaining unit 110, a storage unit 120, a first determining unit 130, and an operating unit 140. The first obtaining unit 110 is configured to obtain a stable level of the plug-in to be run.

The storage unit 120 is configured to save a correspondence between a stability level and an operation mode.

The first determining unit 130 is configured to determine, according to the correspondence between the stable level and the operating mode saved by the storage unit 120, the plug-in operating mode corresponding to the acquired stable level.

The running unit 140 is configured to run the to-be-running plug-in according to the determined plug-in running mode.

The second obtaining unit 150 is further configured to obtain the abnormal information of the operation of the plugin. The storage unit 120 is further configured to store the abnormal information.

When the client 100 runs the plug-in, the client 100 further includes a sending unit 160, which is used to obtain the abnormal information of the plug-in, and sends the abnormal information to the server 200, where the abnormal information includes: the cause of the abnormality, and the The plugin information about the plugin.

Further, the server 200 includes a storage unit 210 for not only storing the stability level of the plug-in but also receiving and storing the abnormal information.

The server 200 further includes an adjusting unit 220, configured to determine, according to the abnormal information saved by the storage unit 210, the number of abnormalities that occur in the set time within the set time, and adjust the stability of the plugin according to the quantity. level.

For the convenience of description, the above-mentioned parts of the client and the server are separately described by functions into various units. Of course, the functions of each unit can be implemented in the same or multiple software or hardware in the practice of the present invention.

In the embodiment of the present invention, the stability level of the plug-in to be run is obtained, and according to the correspondence between the saved stable level and the running mode, the plug-in running mode corresponding to the obtained stable level is determined, according to the determined plug-in. The program running mode is used to run the plug-in to be run, so that different stable level plug-ins are put into different loader processes to avoid the failure of the core business program due to the instability of the plug-in program. Improve the stability of web tools. Putting a plug-in program with higher stability and belonging to the same stable level into a shared loader can save system resources of the client and improve the stability of the client system. And by collecting the abnormal information of the plug-in program and dynamically adjusting the stability level of the plug-in, the unstable plug-in can be eliminated, and the user experience is improved.

It will be apparent that those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention. Thus, it is intended that the present invention cover the modifications and modifications of the inventions

100. . . user terminal

110. . . First acquisition unit

120. . . Storage unit

130. . . First determining unit

140. . . Operating unit

150. . . Second acquisition unit

160. . . Sending unit

200. . . server

210. . . Storage unit

220. . . Adjustment unit

510. . . First acquisition unit

520. . . Storage unit

530. . . First determining unit

540. . . Operating unit

1 is a flow chart of the operation of an exception program according to the present invention;

2 is a flow chart of a specific method for implementing an exception program according to the present invention;

3 is a flow chart of obtaining an abnormality information of a plug-in running according to an embodiment of the present invention;

4 is a flowchart of adjusting a stability level of a plug-in according to an embodiment of the present invention;

FIG. 5 is a structural diagram of a user end of a running plug-in according to an embodiment of the present invention; FIG.

FIG. 6 is a structural diagram of a system for running a plug-in according to an embodiment of the present invention.

Claims (16)

A method for operating a plug-in program, comprising: obtaining a stability level of a plug-in program to be run before the current running of the plug-in program, wherein the plug-in program has a stable level based at least in part on the saved plug-in and Determining the correspondence between one of the plurality of possible stability levels; determining the operating mode of the plug-in corresponding to the obtained stable level according to the correspondence between the saved stability level and the plug-in operating mode; The plugin run mode runs the plugin to be run. The method of claim 1, wherein the obtaining a stable level of the plug-in to be run comprises: obtaining a stable level of the plug-in to be run from a stable level of the plug-in saved by the server; or Get the stability level of the plug-in to be run from the stable level of the locally saved plug-in. The method of claim 1, wherein the plugin running mode comprises: starting different running processes for the plugin to run the plugin to be run for different stability levels of the plugin. The method of claim 1 or 3, wherein the plugin running mode comprises: when the obtained stable level is stable, the plugin running mode is to run the plugin in the shared loader process; When the stable level obtained is normal, the plugin running mode is to start the independent loader process to run the plugin. The method of claim 1, wherein the method further comprises: when the plugin runs abnormally, the method further comprises: acquiring and saving the abnormal information of the plugin running. The method of claim 5, wherein the abnormal information includes: an abnormal cause, and plugin information of the plugin. The method of claim 5, wherein, when obtaining a stable level of the to-be-running program from a stable level of the plug-in saved by the server, the method further comprises: sending the abnormal information to the servo Save it. The method of claim 2, wherein the method further comprises: periodically adjusting the stability level of the saved plugin. The method of claim 5, wherein the periodic adjustment comprises: determining, according to the saved abnormal information, an abnormal number of occurrences of the plug-in within the set time; and adjusting the stability of the plug-in according to the abnormal number of times level. An apparatus for operating a plug-in program, comprising: a first obtaining unit, configured to obtain a stable level of a plug-in program to be run before the current running of the plug-in program, wherein the plug-in program has at least a stable level According to the saved plugin and multiple possible Determining a correspondence between one of the stability levels; a storage unit for storing a correspondence between the stability level and the operation mode; and a first determining unit configured to correspond to the stability level and the operation mode according to the storage unit Determining an acquired plug-in operating mode corresponding to the stable level; and running a unit for running the to-be-running plug-in according to the determined plug-in operating mode. The device of claim 10, wherein the device further comprises: a second obtaining unit, configured to acquire abnormal information of the operation of the plug-in; and the storage unit is further configured to store the abnormal information. The device of claim 11, wherein the device further comprises: a second determining unit, configured to determine, according to the abnormal information stored by the storage unit, an abnormal number of times that the plug-in program occurs within the set time; Used to adjust the stability level of the plugin according to the number of abnormalities. The device of claim 10, wherein the device further comprises: a sending unit, configured to send the abnormal information. A system running a plug-in program, comprising: a client, a stable level of a plug-in saved from a server Before the current running of the plug-in program, the stability level of the plug-in program to be run is obtained, and the plug-in running mode corresponding to the obtained stable level is determined according to the corresponding relationship between the saved stable level and the running mode, according to the determined plug-in. a program running mode to run the to-be-running plug-in, wherein the level of stability of the plug-in is determined based at least in part on a correspondence between the saved plug-in and one of a plurality of possible stability levels; The server is used to save the stability level of the plugin. The system of claim 14, wherein the user terminal is further configured to obtain abnormal information about the operation of the plugin, and send the abnormal information to the server. The system of claim 15 , wherein the server is further configured to receive and save abnormal information, determine an abnormal number of occurrences of the plug-in according to the saved abnormal information, and adjust the plug-in according to the abnormal number of times. Stability level.