Universal system method of reappearing-free quick problem positioning and system diagnosis

Abstract

The invention relates to nesting bi-characteristics fast problem locating and information self filtering general systematic method. It diagnoses and locates various abnormal problems in system running by systematic building one set of context environment description behavior and information self synthesizing mechanism which can greatly save locating problem required time and various resources demand, is convenient for various personnel to adopt proper matching measure to treat the problem. The invention uses multilayer characteristic sign, nesting type synthesizing mechanism, and general assemble output service mechanism to finish no recurrence fast locating problem and system diagnosis.

Description

technical field [0001] The present invention is applied to software systems and related systems that complete system information description and diagnosis through software, and relates to a general system method for fast problem location and system information description and diagnosis without recurrence, so that various personnel can take appropriate matching measures to deal with it in time Problem or diagnose system status. Background technique [0002] With the all-round popularization of information technology and the continuous growth and changes of various application requirements, hardware equipment and software systems at all levels are developing and evolving in the direction of complexity and refinement. Due to the characteristics of the software system itself and the existing development model, As a result, the software system has become larger and larger, with more and more layers, and gradually formed a structural relationship of layer-by-layer dependence, whic...

AI Technical Summary

Problems solved by technology

[0005] Analyze the main bottlenecks in the entire development process. In addition to the necessary upfront time investment in the requirements and design process, a major and long time consumption exists in various types of tests. This step has a significant impact on the quality and performance of the system. Decisive role, but from the actual situation, a lot of time in this process is spent on other processing that has nothing to do with the test case: such as problem reproduction, problem location (including preliminary location and precise location), problem owner search and problem In terms of finding the cause; another problem in this process is: how to find all the system errors that should be found in a large amount of chain output information that may be generated when a certain type of abnormality occurs in the system, so as to avoid the neglect of problems caused by human factors, so as to Prevent neglected problems from appearing in the officially released system to the greatest extent

[0009] There are various unsatisfactory aspects in such a traditional problem-handling mode: First, from the perspective of the whole process, this kind of processing method is similar to the serial processing mode, because it is impossible to establish a link between the person who discovers the problem and the person in charge of the problem module. The direct correspondence between the two leads to a lengthy process and takes a very long time to deal with and solve the problem. Once the person who finds the problem is inexperienced or does not understand the system composition, he has to use layer-by-layer reporting to complete the final location of the problem. During this period In addition to replacing the equipment, the problem finder can hardly do any remedy. In practice, it is impossible to replace the equipment in many cases. If there are multiple problems across multiple modules and levels, the time will be multiplied; the other is the processing mode. The bottleneck is that developers are limited and restricted by relevant tools and servers in version acquisition. In most cases, it is worthwhile to always put version issues that have little to do with the problem itself in the list of tasks to be dealt with. Consider an improvement

Not only does it cause all kinds of people to spend a lot of time on contacting the person in charge at all stages, but it also introduces huge risks and chain losses to the system

[0010] Due to the complexity of the system itself and the naming standards adopted by most systems, the file name and code line number provided in the language standard to calibrate the output position in the corresponding version consume a lot of limited output buffer resources, followed by The side effect of this method is that this method seriously affects the integrity of the presentation of the problem-related context information. On this basis, the further problem is that developers often have to repeat the problem generation process through reproduction. On the one hand, the problem reproduction itself produces Huge delay, on the other hand, the response ability of this processing method to accidental events may be ignored due to human reasons, and if the problem is not found in other testing processes, its potential harm may be related to the success or failure of the entire process, especially In the critical state of multi-task concurrent system interaction, due to the uncertainty caused by multiple physical factors, some reproduction processes are extremely difficult or even difficult to reproduce manually. At this time, we can only rely on the luck and experience of engineers, which constitute potential system quality risk

[0011] Another significant disadvantage of this processing mode is that due to the layer-by-layer nesting of the system, an abnormal error often leads to a series of abnormal output, and the final output information may be various. Since the output itself consumes system resources seriously, it will be greatly reduced. Affects the performance of real-time systems that are extremely demanding on time, and in the case where multitasking is common today, the time resources consumed by a single task in output may cause system task switching and the resulting possible output loss, all of which are Influencing factors on problem solving

Images