ERR ORR <rd0002(at)gmail(dot)com> writes:
> My understanding from your replies is that this behavior with *b-tree
> indices* is not considered a bug but rather a case of "works as designed",
> yet still and apart from the solution of my particular problem, the fact
> that in 9.2.4 it is so bad that a simple query sometimes does not finish
> after 10+ minutes with incessant disk activity tends to look like a problem
> to me.
It is no better or worse in 9.2.4 than in any prior release.
regards, tom lane