mark,
 
ml>WD> For all intents and purposes, I would call that a bug.
 
ml> more of a design defect... but we must also remember that rescans came
ml> later and there was no indication of them having been done... also most
ml> systems were leaf nodes so rescans didn't really matter... the problem
ml> comes when a system feeding others does a rescan and doesn't know about
ml> or take the ^aRESCAN control line into account when tossing... this is
ml> all relatively new, though...
 
Do you have relatives in the Netherlands perhaps?
 
 \%/@rd 

--- DB4 - May 18 2020