The likelihood's truncation collapses the posterior to a bounded uniform

Let $f(x|\theta)=e^{-(x-\theta)} x\ge\theta$ and let the prior for $\theta$ be given by $\pi(\theta)=e^{-\theta} 0<\theta<\infty$. Then: