Different definitions of percentiles
It was mentioned earlier that there are several competing definitions of the upper and lower quartile. All such definitions split the data approximately into quarters but there is not a unique way to do this.
There is even less agreement about the precise definition of other percentiles, and different computer software finds them in different ways. The definitions are usually based on a smoothed version of the cumulative distribution function.
The differences between the different definitions are small if the data set is large.
If your conclusion about the data would change with a different definition of the percentiles, you are over-interpreting the data.