About questions related to the Riemann's Hypothesis.

Question

Let n be an integer , if n satisfies Robin's Inequality ($\sigma(n)$ /n < $e^{\gamma}$ lnlnn) say n is 'regular'. If n doesn't satisfy Robin's Inequality say n is a 'counter' or a counter-example. One can prove if a and b are coprime and are both counterexamples of Robin's Inequality then so is (a b). Also if m is a regular number where m is not a prime power them m has a proper divisor h such that 1 < h < m , where h is also regular. Let A be the minimum counter that has no proper divisor that is regular; if A has at least two distinct prime divisors p and q so A = (p Q) where q divides Q, this would imply p is a counter but I don't think a single prime is a counterexample to Robin's Inequality. Is it? So A would be a prime power, but similarly a prime power is not a counterexample is this right? These arguments seem to imply every counter example has a proper divisor that is regular.