Not sure Kolmogorov Smirnov Test is working as it should

So I have some data, Bequerels from Cs-137 decay from some samples. I have to check if the chosen data belongs to normal distributed sample values. Q-Q-plots suggest they do, Anderson Darling test too. Kolmogorov gives strange values. The Dataframe looks something like this:

Z79V0001 Z79V0003_1 Z790003_2

0 5.3738e+09 2.5241e+09 3.2927e+09

1 . . .

2 . . .

Code is like this:

for col in columns:
  print([col])
  print(stats.kstest(df[[col]].dropna().values, 'norm', args=(min(df[col]),max(df[col]))))
  print(min(df[col]))#just to check

The output is something like

['Z79V0001']

KstestResult(statistic=0.62697, pvalue=1.23716e-17)

As I understood the p-value should be large for normal distributed data.

Do I have to normalize my data first? I am afraid I am comparing the Gauss CMD with the wrong scale?

Every help is appreciated.