The "fixed, root-caused, verified, closed, and then it comes back wearing a different face" framing is the sharpest part of this already -- that pattern almost always means the original fix patched the symptom at the layer where it was visible, while the actual root cause lived one layer down (toolchain, instruction set, whatever ARM64 changed underneath everyone). Curious whether the second occurrence traced back to literally the same root cause as the first one, or whether it took an unrelated bug sharing the same blast radius to make the connection.
The "fixed, root-caused, verified, closed, and then it comes back wearing a different face" framing is the sharpest part of this already -- that pattern almost always means the original fix patched the symptom at the layer where it was visible, while the actual root cause lived one layer down (toolchain, instruction set, whatever ARM64 changed underneath everyone). Curious whether the second occurrence traced back to literally the same root cause as the first one, or whether it took an unrelated bug sharing the same blast radius to make the connection.