…
Forgive me if this is insulting anyone's intelligence or education but just to verbalize what I know about this stuff (and maybe someone can correct me if I'm wrong - my one strengths-of-materials class was very long ago): If you have a long piece of uniform material (like a virgin length of square aluminum tubing) and you apply a bending force to it, the material will start to bend into a mostly-uniform curve from end to end. The outside of the bend will all stretch and the inside of the bend will all compress and the forces will be spread over the entire length, and the material won't fail (crack, etc.) until the whole piece is very close to failure at every point along its length. However, if you modify the uniform piece in some way (drill a hole in it, weld or unevenly heat it enough to affect the metallurgy, change its shape somehow, thicken part of it, etc.), then it is no longer uniform and any bending forces will tend to concentrate somewhere. Meaning a certain point will absorb more of the overall force. Since this is generally occurring at a structurally weak point, it may end up being stressed beyond the point of failure even though the rest of the material is still well within its elastic deformation range (meaning a bend from which it will completely return to original shape).
…