Re: Plan invalidation design

On Feb 18, 9:35 am, t(dot)(dot)(dot)(at)sss(dot)pgh(dot)pa(dot)us (Tom Lane) wrote:
> Russell Smith <mr-r(dot)(dot)(dot)(at)pws(dot)com(dot)au> writes:
>
> > If you replan and immutable function, aren't you possibly messing up a
> > functional index that is using the old function. Hey, if you change an
> > immutable function that has an index, you are in trouble already.
>
> True.

While I agree that if you change an immutable function used by an
index, your index will break, I do not understand how re-planning it
will cause problems. Is the worry that the index will not pick up on
the new plan?

Andrew