What is configuration item testing in software testing? What is the definition and what does it do?

CI testingI think we have to be clear about two concepts first:

Software configuration items: I think that a software configuration item is a product that has been developed, has entered the configuration management, and is ready to be provided to the customer. It can be executable or product documentation.

Software Requirements SpecificationThe software requirements specification is a document obtained during the requirements analysis in the early stage of the project, which describes the user's requirements and is the common understanding of the project in the initial stage, such as some interface design and process description, which is the basis of the entire development work.

SoCI testing, it can be understood as a kind of check of the software configuration item, to check whether it is consistent with the software requirements specification. For example, perform a functional test on the executable ** to see if its functionality is consistent with the requirements in the software requirements specification. Alternatively, a document review of a product document has been conducted to see if the installation steps have been described as required in the requirements specification, or if the interface described in the document is the same as in the requirements specification.

Therefore, the configuration item test needs to be in theUnit testswithIntegration testingAfter that.

I understand that the order of testing should be: unit tests - > integration tests - > CI tests - > system tests - > confirmation tests, and if there are changes to the project, regression tests are also required. Of course, this is only to help to understand, and in practice it will certainly not be done in order.

Personal point of view: The configuration item test here can be simply regarded as the next level of unit testing, that is, the test of individual configuration items, here if you don't understand the test, just change the word, check In some systems, the scope of work given to the test role includes document checking, static checking, which is the so-called configuration item test. After this part of the work is completed, a series of follow-up work such as unit testing, integration testing, system testing, etc. will be carried out.

Summary. The basic principles of software testing require 1Testing should be based on user needs All testing standards should be based on meeting customer needs, and from the user's point of view, the most serious errors are those that cause the program to fail to meet the requirements.

The environment should be configured according to the needs of the user, and the test and evaluation results should be carried out according to the user's usage habits. If the system does not meet the needs and expectations of the customer, then the development of the system is a failure. At the same time, there is no point in finding and fixing defects in the system.

Early involvement and exposure of the user to the prototype system during the development process is a precautionary measure to avoid such problems. Sometimes, the test results of the product may be perfect, but the end customer does not buy it. Because, this perfectly developed product may not be the product that the customer really wants.

2.Doing a good job in software testing is the key to doing a good job in software testing Software testing is an organized, planned, and step-by-step activity, so testing must be organized and planned, and the test plan must be strictly implemented to avoid the arbitrariness of testing.

Can you add, I don't quite understand it.

The basic old principle of software testing requires 1Testing should be based on user needs All testing standards should be based on meeting customer needs, and the most serious errors in terms of user roundness are those that cause the program to fail to meet the requirements. The environment should be configured according to the needs of the user, and the test and evaluation results should be carried out according to the user's usage habits.

If the system does not meet the needs and expectations of the customer, then the development of the system is a failure. At the same time, there is no point in finding and fixing defects in the system. Early involvement and exposure of the user to the prototype system during the development process is a precautionary measure to avoid such problems.

Sometimes, the test results of the product may be perfect, but the end customer does not buy it. Because, this perfect product balance from a development perspective may not be the product that the customer really wants. 2.

Doing a good job in software testing is the key to doing a good job in software testing Software testing is an organized, planned, and step-by-step activity, so testing must be organized and planned, and the test plan must be strictly implemented to avoid the arbitrariness of testing.

The definition of software testing is: the use of professional testing tools, test methods to find software vulnerabilities, and then submit to the development department for modification, software testing engineers are the gatekeepers of software quality, occupy a very important position in the enterprise.

The recognition of software testing in China is relatively late compared with foreign countries, and there are still few universities that offer software testing disciplines, so talents in this area are still very scarce in China. According to the statistics of China Talent Network, the current shortage of software testing talents is more than 300,000 people, and the annual rate is increasing by 20%, and domestic and foreign IT industry giants are stepping up competition for software testing talents.

It is an indisputable fact that the salary of a good test engineer is several times that of an average employee in a traditional industry. Therefore, there are many people who switch careers to software testing every year. Everyone hopes to change their fate by learning a skill, and their ideas and courage are commendable.

The Eight Basic Principles of Software Testing:

All software testing should be traced back to user requirements. Conduct early and continuous software testing. When designing test cases, you should include reasonable and unreasonable inputs and the corresponding expected outputs.

Pay full attention to the clustering phenomenon in the test. Programmers should avoid checking their own programs. Try to avoid the arbitrariness of the test.

Each test result should be thoroughly examined. Retain test documentation, including test plans, use cases, error statistics, and final analysis reports.

We've discussed the issue of software testing many times in previous issues, especially with regard to the structure and composition of the requirements document. Today, let's take a look at what the test basis in software testing really means. I hope that through the reading of this article, you will have a better understanding of software testing.

In the Fundamentals of Software Testing, we have come up with a concept called 'testbasis'. In a literal sense, a test basis is a document or information that we can use to test and analyze the test and write the use case. He is used to guide our tests, from which we can extract answers to fundamental questions such as 'what to test' and 'how to test'. Without a basis for testing, there is no way to start testing.

When it comes to the basis of testing, we will directly think of the requirements document, which may be presented in different formats depending on the characteristics of the project: such as the requirements specification format, or the prototype schema, etc. Depending on the difference between the content decomposition and the description of the requirements document, it may be presented as a user story or a product requiring a document (PRD).

It's important to understand that testing is not just about requirements documents. Essentially, he should include all the information that can guide us in the test. Let's take a look at them one by one, what information we can use as a basis for testing, and how to use them.

One is the design documentation of the development department, including the architecture design that we mentioned in the software lifecycle, the output of the detailed design phase.

When the development department carries out the above design work, it is possible to produce more abundant documents than in the requirements stage, such as architecture design drawings, algorithm design drawings, detailed design specifications of modules, interface definition documents, database design specifications, interface design line drawings, etc.

In practice, you will find that the design documents produced by the development department often contain more detailed and rich definition information about the product. Based on the information provided in these documents, we can go one step further and determine the scope and content of our tests at a deeper level.

Of course, theoretically, these design documents are only analytical outputs made by the development department in order to implement the requirements, and they may not necessarily match the initial product requirements or even the user needs. The premise of our use of these design documents is that there is sufficient communication between the development department and the requirements department, and their product design is in line with the requirements.

Changping Computer Training suggests that in order to confirm this fact, we testers may need to go back and forth between the development team and the requirements team (or users) to seek a positive answer (a typical situation is that the requirements team's interpretation of the user's requirements is not detailed enough, in fact, he may not have thought about it to a corresponding depth at all, and he may not have any opinion on the design given by the development through subjective judgment, but can only express approval).

Software crack testing includes functional testing, performance testing, and automated testing. However, the performance test basically has the following points:

1.Stress test.

2.Load testing.

3.Capacity testing.

4.Concurrent testing.

5.Durability testing.

6.Strength test.

In layman's terms, it is to simulate the real operation of the user and verify whether the software can meet the needs of the user.

This includes load testing, strength testing, database capacity testing, benchmarking, and competitive testing.

1. Load test.

Load testing, a type of performance test, refers to whether the program can afford to run data in an overloaded environment. In this type of testing, test subjects are subjected to different workloads to evaluate and evaluate the performance behavior of the test subjects under different workload conditions, as well as their ability to continue to operate normally.

2. Strength test.

Strength test, a performance test, he is the software system operation in the case of particularly low system resources. This type of testing can often be written about the level of hardware and software required by the system.

The purpose of implementing and performing such tests is to find out errors caused by insufficient resources or resource contention. If there is not enough memory or disk space, the test subject may exhibit defects that would not be obvious under normal conditions. Other defects can be caused by contention for shared resources, such as database locks or network bandwidth.

Strength testing can also be used to determine the maximum amount of work that a test subject can handle.

3. Database capacity test.

Database capacity testing refers to inserting a certain amount of data into a database table through a stored procedure to see if the relevant pages can display the data in a timely manner. Database capacity testing has test subjects process large amounts of data to determine if they have reached the limits that will cause the software to fail. The capacity test will also determine the maximum load or amount of work that the test object can sustainably handle in a given amount of time.

4. Benchmarking.

A benchmark is a test that compares known existing systems to see if they are competitive with similar products.

5. Competitive test.

Software competes for the use of various resources (data records, memory, etc.) and competes with other related systems for resources.

The performance test indicators of software testing are mainly as follows:

1.response time; The DAO response time generally includes the network time and answer interval + application processing time. The time the user sends a request to the time the user receives the response data returned by the server, which is the response time.

Generally within 3-5 seconds, the performance is good.

2.Throughput; It refers to the number of requests processed by the system per unit of time, which indicates the load capacity of the system to a certain extent.

4.resource utilization; One of the key test indicators is resource utilization, which analyzes CPU usage and disk usage when the system is running, which is also an important basis for improving performance in subsequent work.

Answer]: bThis question tests the understanding of software test configuration management. The software sells Zhonghui test configuration management including 4 basic activities: configuration item identification, configuration item control (change control), configuration status report, and configuration audit (see the "Limb Training Software Evaluator Tutorial").