Someone just pointed out to me that *our* tables were at 8704 also!
  It turns out that I had filed RSCS DIRECT on MAINT's 191 instead of
  of RSCS's 191 when I did the update.  I'd just assumed the past
  couple of weeks that everything was fine.
 
I guess it is a good thing SFAUSTIN was directly connected to us or
  we could have been the culprits!
 
The moral of the story is that we shouldn't assume that nodes that
  have backlevel routing tables are 'slackers'.  Perhaps they made
  a similar mistake.
 
Regards,
         David