Six modes of fault handling for telecom billing software

I. Mechanism analysis based on the code logic, analyze the various responsibility chains that arise from the problem. Rigorously analyze and troubleshoot each step of the chain of responsibility. Issue confirmation by collecting code, logs, and other information for suspicious steps. Two. The black box reproduces the collection of configuration and phenomena related to the problem business, compares the problem data collection with the configuration and phenomena of the data collection without problems, sums up the law of the problem and speculates the factors that may cause the problem. Three. Add Feedback
1. Increase the log and add feedback
2. Add new impact to the system and observe the change of the problem

Four. Historical analysis
1. According to the time of the problem, and the operation before and after the relevant time, determine whether there is a certain operation (such as upgrade, configuration data refresh, etc.) cause the problem 2. If the problem changes over time, then the trend of inductive change. Learn about the changes in the environment during this time, and see if there are any mutations before or after the change.
Five. Environmental inspection
A thorough analysis of all information in the environment, including individual process logs, may reveal scenarios that were missed during the analysis. Other problems may also be found, which can be eliminated to reduce the current problem. Six. Avoidance of problems 1. Change the structure of the problem-prone programme to avoid problems. 2. Add scripts to quickly correct data after problems occur. 3. In the code, directly determine the output results, if there is a problem, modify and then output.