1. Assess the scope of impact of the bug
1. Analyze the number of users
affected by the bug 2. Analyze the severity of the bug's impact
2. Solve online problems
When the scope of the bug is relatively small, it is usually solved by fixing the bug. The methods are as follows:
1. Understand the scene of the bug, business operations, and strive to reproduce the bug
. 2. The developer combines various logs (system logs, database) when the bug appears. Log, operation log, debug log), locate the cause of the bug
3. After the developer has modified the bug, the tester will verify it to ensure that the bug has been fixed
. 4. Fix the bug according to the project planning release/upgrade time node The code is posted online, and the bugs are resolved. Bugs with a
relatively large impact
When the scope of the bug is relatively large, if it is still solved by fixing the bug, the impact on the user or the company's loss cannot be controlled. At this time, the most important thing is to minimize the scope of the problem. The method is as follows:
1. When it is impossible to clarify the reason for the introduction of the problem, you can avoid it by rolling back the version
2. Some user functions can be downgraded or closed through the background configuration
3. If it is a performance problem such as insufficient resources, you can pass Restart the system or expand the capacity to solve the problem, and then observe the
above several ways to avoid the problem only to help us gain time. After avoiding the problem, we still need to locate the problem according to the previous bug fix method, fix the problem, and release the fixed code. On, the bug will be completely resolved.
Three. Backtracking online issues
Check whether other businesses have the same type of problems.
Analyze the reasons for
the bugs. Supplement the test cases that caused the bugs.