As the title states, how would I go about finding the positive integer solutions of
$$\frac{1}{x_1}+\frac{1}{x_2}+\cdots+\frac{1}{x_n}+\frac{1}{x_1 x_2 \cdots x_n}=1$$?
Thank you for your help.
Edit: I think I've made some progress. I conjecture that you can expand the set {a, a+1, a*(a+1)+1, a*(a+1)(a(a+1)+1)+1, ...} (i.e. multiply all previous solutions and add 1), where a=2, to size n, and that will always be a solution in the case of n variables. This is far from finding all positive integer solutions, though.
Edit 2: For example {2}, {2,3}, {2,3,7}, {2,3,7,43}, {2,3,7,1807}, {2,3,7,43,1807,3263443} are all solutions in the case where $n=$ size of the solution set respectively.
This is known as the improper Znám problem.
All of the solutions for $n\le 7$, as well as an algorithm to compute them are given in Brenton and Hill's 1998 paper. All of the 119 solutions for $n=8$ are reported to have been found in Brenton and Vasiliu's 2002 paper, but the link containing the solutions appears to have decayed. Regenerating the first hundred or so of these solutions is easy, but finishing the last few will probably take a couple of weeks of computer time. Finding all the solutions for $n=9$ looks intractable without a breakthrough in the search techniques.
Brenton, Lawrence; Hill, Richard (1988), "On the Diophantine equation 1=Σ1/ni + 1/Πni and a class of homologically trivial complex surface singularities", Pacific Journal of Mathematics 133 (1): 41–67
Brenton, Lawrence; Vasiliu, Ana (2002), "Znám's problem", Mathematics Magazine 75 (1): 3–11