Covariance under a triangular joint density

Let the random vector $(X,Y)$ have the joint probability density function $$f(x,y)=\begin{cases}\dfrac1x,&\text{if }0<y<x<1,\ 0,&\text{otherwise.}\end{cases}$$ Then $\mathrm{Cov}(X,Y)$ is equal to