A checkpoint is triggered when the Oracle database service is stopped.

The database on the test machine has been stuck there when it was stopped in the last few days. I used shutdown immediate to stop the database.
I found an article from the InternetArticleBefore shutdown, use the alter system checkpoint command to save the checkpoint. Why is this operation effective?
I went online again to find the role of alter system checkpoint. I learned that this command explicitly allows the Oracle database to execute a checkpoint, ensure that all changes generated by committed transactions are written to the data files on the disk.
And from this article (http://space.itpub.net/94384/viewspace-684442), before the database stops, if the alter system checkpoint is not executed, it will cause the information in the cache not to be written into controlfile in time.
So I guess whether data synchronization will be performed in the background when the database is stopped. The data in the cache will be synchronized to the disk and written to the corresponding data file.
At least this is what the Solaris system does. If an unexpected power failure occurs, you may need to execute fsck to fix file data inconsistency after the system is restarted.