Well, that's why they call them random numbers. Assuming the random number generator works right, you have a roughly 10% chance of drawing the same number twice if you're sampling a population of 30 numbers 10 times with replacement.
If you think of x as a specific number and n as the cardinality of the set,
for the first draw: P(x) = (1/n). Then the prob of not getting the same number again is P(x<>the first number)=(1/(n-1)). Assuming the draws are independent, the likelihood of not getting a repeat over a number of draws diminishes rapidly.
That's why Russian Roulette isn't a game you want to play often.
If you want to check a solution over a range of values, you can uniformly generate the values for the variables you're interested in, solve, and check the results.
JS