Soft audit Checklist・No15: Test Management Technology (2/2)
Introducing development technology and test management items in the software development audit checklist
Software development auditing is an auditing method that Gutara’s father arbitrarily named to determine the capabilities of software development contractors. Overview of software development audit and pre-preparation of the audit at the time and the work of the audit on the day for, have been introduced in other articles. Also, please refer to another article for the points when using the audit checklist used in software development audits .
Continuing from the previous article, this article introduces the remaining check items for test management in the development technology checklist.
[Item number: TM-04]
After checking the quantity of the test, the next step is to check the quality of the test . However, the quality of the test is not a simple numerical value, and it is annoying that there is no reliable way to judge whether the quality is good or bad. There is no point in worrying about it, so first check the quality of the test from the perspective of the ratio of abnormal system tests .
In fact, if you set the purpose of the test to be that the function works , you only need to test the normal system . In fact do the software development overseas, the test requirements specification to write in Aru function is to make sure that the work for the test of the normal system Thus not, There is also a good case that.
If the purpose of the test is set to allow the function to operate stably under various operating conditions, it is assumed that it can occur in an environment where the device is actually used, such as a heavy load or a lot of noise. Anomalous tests are also included to confirm that it works even in various situations . In addition, since communication protocols often specify the processing when an error occurs, there is also a quasi-normal system test to confirm that the error processing designed and implemented is executed correctly . The ratio of these abnormal and quasi-normal tests is used to determine whether the quality of the tests is good.
[Item number: TM-05]
By looking at the ratio of abnormal tests in a little more detail, that is , the ratio of tests for each test category , you can judge the quality of the test from another perspective. As for the ratio of tests for each test category, this is not the correct answer, but by comparing it with our existing products, we can see the difference between the test of the contractor and the test of our company, so the test that is insufficient It will be easier to make specific test quality improvement requests, such as having more tests in the category.
[Item number: TM-06]
Father Gutara thinks that the upgrade function of the farm needs to be of higher quality than other functions . Unfortunately, software may be released with potential bugs in the current software, which is becoming larger in scale. Even if there is a software bug in a product running on the market, if you can upgrade to a bug-fixed version of the farm via the net, you can eliminate the bug. If the version upgrade function of such a farm shouldn’t work due to a bug, I don’t want to imagine that. Therefore, when testing the firmware version upgrade function, pay special attention to make sure that the quality of the test is sufficient.
In addition, firmware and software are used interchangeably in the article by Father Gutara. From the meaning of the word, it seems good to refer to the software that can be rewritten in the market, while the farm is closer to the hardware and cannot be rewritten by version upgrade etc. after shipping. However, since the work related to quality such as planning, design, implementation, and testing is the same for both farm and software, Father Gutara regards farm and software as similar things. Therefore, please read the article, thinking that the names farm and software both refer to the same thing.
[Item number: TM-07]
When testing the upgrade function of a farm, the quantity of tests is important as well as the quality of the tests . Especially in the case of terminal devices used in the homes of end users, there are tens of thousands to hundreds of thousands of products, and the fur version upgrade function is used in various environments and timings. It is not possible to prepare all cases at the test stage, but we have conducted a total of hundreds to thousands of version upgrade tests to confirm the normal operation of the version upgrade function at various timings. I want to keep it.
[Item number: TM-08]
Wireless LAN has spread to every home in the last few years. As the number of products that support wireless LAN is increasing in various home appliances, it is becoming more common for our products to have wireless LAN functions. Of course, if the product has a wireless LAN function, we will also test the wireless LAN , but since the wireless LAN has a wide range of preparations for the test environment and how to proceed with the test, carefully consider it in order to perform a good quality test. You need to proceed. Carefully check that the wireless LAN test is made with such points in mind.
[Item number: TM-09]
As the scale of software has increased, the number of man-hours and days required for testing has also increased. When it comes to tests for hundreds of man-days or tens of days, if you do not properly arrange the test environment, equipment, test personnel, etc. and proceed with the test, test waiting will occur here and there, or test equipment you are playing. Will occur, which will reduce the efficiency of the test . Also, once you start testing, it’s important to make sure you’re on track.
You need to be especially careful when things are n’t going according to plan , and the quality of the software you’re testing may be so bad that the tests are stuck here and there. The test can measure the quality of the software from the result, but you can also know the quality of the software from the middle of the test . To that end, we will focus on whether there is a system in which tests are properly planned and the progress of tests is properly traced.
[Item number: TM-10]
The biggest deliverable of the test is the bug detected in the test. Bugs are fed back to the development department together with information such as how to reproduce them, and the bugs are fixed to improve the quality of the software. However, there are many other test artifacts besides the bugs we found. For example, suppose you test similar quality and quantity on similar software A and B and find 30 bugs for software A and 60 for software B.
If all the bugs found in both software are fixed, the final software quality can be judged to be the same for both A and B. However, not all latent bugs can be found by testing, so there is also the idea that software B is more likely to have latent bugs , and software A has better quality. Since such information is also a product of the test, we will check from the perspective of whether the test results are being used effectively, including such information.
Next to test management are check items for test content
Next to the confirmation of test management, the items to be confirmed about the test contents will be introduced in the next article.