At three o'clock this afternoon, I received a piece of data from the next hour.
Today is at 3 pm on February 1st, and the file creation time is at 4 pm on February 1st.
Look at the picture below: Today's time is less than 3:30, the time of failure
Due to the different time zones, I understand that DATETIME is inconsistent in different countries and different time zones-but the software I use does not go abroad, and time is a very important parameter. If DATETIME is wrong, it will malfunction.
The query found that: datetime is stored in the data as a floating point number, which is roughly like this:
21433.2342432
may be because the floating point number is too small, and it is possible to cut the too small number directly when intercepting.
It may be because the system has a bug when converting int to datetime, or it involves instability.
It may be because datetime has multiple rules, and the matching rules are misplaced.
The date is less wrong, the time is more...
The probability of occurrence is about one in ten million.
The reason is not clear.
———————————————————————————————————————————————— -
update: when system get datetime there are two possibilities: get time from the client, obtaining datetime from the server. When the client has a BUG, there may be time problems.
The datetime should be added from the server and the datetime of the server shall prevail.
What I said before was wrong when I was uneducated.