Problem: Define for $\alpha > 0$ the subset of $\mathbb{R}^2$ given as $$ D_{\alpha} = \left\{(x,y) \in \mathbb{R}^2 \mid 0 < y, 0 < x < y^{\alpha} < 1 \right\}. $$ Define the function $$f : (0, \infty) \times (0, \infty) \to [0, \infty): (x,y) \mapsto \frac{1}{(x+y)^2}. $$ For which values of $\alpha > 0$ is $$ \int_{D_{\alpha}} f \ d\lambda < + \infty $$ ?
Attempt: I first wanted to change coordinates to polar coordinates. But I don't know how to redefine the subset $D_{\alpha}$ in terms of polar coordinates. Since $f$ is a positive function, we can switch order of integration (Fubini). So I wrote (I'm not sure if the integration bounds are correct) $$ \int_{D_{\alpha}} f \ d \lambda = \int_{0}^1 \bigg( \int_{0}^{y^{\alpha}} \frac{1}{(x+y)^{2}} dx \bigg) dy = \int_{0}^{y^{\alpha}} \bigg( \int_0^1 \frac{1}{(x+y)^2} dy \bigg) dx = \int_{0}^{y^{\alpha}} \bigg( \frac{1}{x} - \frac{1}{x+1} \bigg) dx. $$ I'm not sure how to proceed. I can only use linearity if I know the two terms inside the brackets are integrable. But $$ \int_{0}^{y^{\alpha}} \frac{dx}{x} = \lim_{\epsilon \to 0} \ln \bigg( \frac{y^{\alpha}}{\epsilon} \bigg) $$ and I don't think I can get this finite for any $\alpha$.
Help/advice with this problem is appreciated.