On Thu, 2005-12-22 at 10:40 -0500, Tom Lane wrote:
> Simon Riggs <simon@2ndquadrant.com> writes:
> > Sorry, I missed that.
>
> And you evidently still didn't understand it. Locking both pages does
> not fix the problem, because it doesn't guarantee that there's not a
> concurrent indexscan in flight from one to the other. If you move items
> from one page to the other in the opposite direction from the way the
> scan is going, then it will miss those items. If we try to fix this by
> making scans lock one page before releasing the previous, then we'll
> create a bunch of deadlock cases.
Thank you for explaining things; I'm sorry you had to do it twice.
I'm ever optimistic, so I try to resist the temptation to stay quiet in
case I make a mistake.
Best Regards, Simon Riggs