I encountered some weird behavior with the FullbodyIK node in 5.4. For some reasons in 5.4 the FullbodyIK solver fails to pin effector targets to their desired location/rotation. The attached video illustrates the problem and is quite self explaining I guess. The first part of the video shows the FullbodyIK in 5.4 (bugged) where the head does not reach its desired target. The second part of the video shows the exact same setup in 5.3 where I get my desired and expected result of pinning the head to its target.
I hope I am not mistaken and understood the design of the FullbodyIK correctly, but I would assume what I get in 5.3 is/was intended and by design. In any case, the same setup should deliver the same result so it’s safe to say that something is off I guess.