Creating a numpy arrays of all possible combinations of elements

I would like to creare a Numpy array, which consists of all possible combinations.
I have a number of samples, which may have a set of possible values.
Currently I use recursion, but am looking for faster methods to work with a high number of combinations.
A possible number of combinations is (n of values)^samples

import numpy as np
import itertools 

samples=3
range1=range(1) 
range2=range(20,101,10)
a=np.zeros((samples))
total=np.empty((0,samples))
limit=1000
def nested(samples,limit):
    if samples>0:
        for j in itertools.chain(range1, range2):
            a[samples-1]=j
            global total
            total=np.vstack((total,a))
            nested(samples-1,limit)

nested(samples, limit)
total=np.unique(total, axis=0)
print(total.shape)

With currently applied recursion method, it takes a lot of time to build an array with 1M+ combinations.

>Solution :

You can use itertools.product() for this:

samples = 3
values = list(itertools.chain(range1, range2)) # all iterables
total = [list(item) for item in itertools.product(*[values for _ in range(samples)])]