For those who propose drilling halfway through to reduce the liklihood of cracking, I wouldn't assume that without some analysis. It wouldn't surprise me at all if drilling a hole partway through created a higher stress concentration in the arm than if the hole passed completely through. This is especially true if you just use a drill bit or an end mill to make the hole. The sharp point or edge at the bottom of the hole would concern me. A ball end mill would be better. But again, if the diameter is too small, you can still create a stress concentration at the bottom.

If it were me, I'd model the thing in CAD and do some FEA analysis before I started making chips. Hmm....that might be fun (yeah, I know, I need to get out more).