Thread: standby role reversal

I was trying to script a standby role reversal in a 8.3.5 test
environment and something didn't go right.  After the reverse of roles the
new standby said:

 LOG:  invalid resource manager ID in primary checkpoint record

but recovery seemed to pick up.  I think the old standby did not go
down correctly.  How big a mess is this and how might I clean it up if that
has not already happened?