Work Essay - After Making Mistakes

In the new year, whether it is work, schoolwork, or other personal goals, we will continue to move forward. First of all, I wish you a safe and prosperous new year.

From the end of last year to today, my department has been implementing a major update of the business model. It took nearly two months, right? As it involves the establishment of the charging method, the rigor of the cash flow is different. In addition, the cash flow must be connected with the functions of the financial department, combined with the payment method and the bank behind it.

In the process of collaborative operation, there are many troubles that were not encountered in previous development, such as operation time delay, the judgment of correct/incorrect return value is different from what you imagined, etc., you cannot directly see the other party's program, and the waiting process is really It's a painful thing. After all, if you encounter errors in the development of your own product, you can easily understand the part and reason of the error through the internal specifications. Even if you don’t understand the original design, you can ask the person who designed it.

One thing to add, I am responsible for the bias display, and the key financial flow chain and financial statement calculations are in charge of the seniors. Although there will definitely be a lot of things waiting to start in the next year (after all, they are still receiving salary), but you can breathe a sigh of relief after finishing the launch today.

At least that's what we probably thought this morning.

In the case of major updates, in order to prevent users from accidentally touching and generating dirty data during the update process, the external connection will not be opened until after the update is re-accepted by the plan. Therefore, the closing time of today's station was announced last week from 10:00 to 14:00. , as a result, many unexpected situations occurred during the acceptance.

API (Application Programming Interface) is a kind of thing, which will access the database or other data sources for combination and calculation according to HTTP requests with different paths and parameters during operation, and then return the processed data to the requesting object.

It is absolutely abnormal for the API to stop working in the middle of the execution, but it has encountered several times after going online today. Before the problem is found and the acceptance is completed, the time of opening the website can only be gradually delayed.

In the afternoon, the supervisor found the problem, called several of our back-end engineers, showed us a piece of code, and made sure we understood its meaning.

 catch (AthenaException e) {
    e.printStackTrace();
    System.exit(1);
}

Well, to put it simply, the predecessors misused the official sample code during development, resulting in the entire program being stopped once an error occurred.

After the supervisor confirmed that we understood, he asked the seniors to make corrections, and the others continued to carry out their original work. This is not to blame, just to avoid the same problem from happening again.

Thinking back to last year, this happened to me a few times.

One of them couldn't work normally after going online on Thursday, and many colleagues couldn't handle it immediately after working overtime, so they had to restore it back to before the update, and go online again next Monday after Friday's correction.
However, I encountered other problems next Monday, which were finally solved after 10:30 in the evening. The above are all caused by a part of the functions I developed.

At that time, my colleagues and supervisors who worked overtime together did not blame me. They just showed a little helpless feeling when they found out that it was not completed/wrong again. The supervisor reminded me to apply for overtime the next day. "You are really working when you stay, so there is no problem in applying." At the stand-up meeting, the project manager only said, "Everyone worked hard yesterday, and the problem is solved." I am very grateful for everyone's tolerance.

When encountering an error, first find the cause, then find the solution, and finally synchronize the information as a whole to ensure that others will not step on the same pit a second time in the future. Everything is for the sake of overall efficiency and quality.

I believe this experience will have a long-term impact on me.