> Also, somebody made a real good point about rolled-back insertions.
> Even if the only command you ever apply to the table is INSERT, you
> could still have dead rows in the table if some of those transactions
> occasionally roll back.
hmm... That's true. I don't think autovacuum doesn't anything to account
for the concept of rolledback inserts. I suppose in most real world
situations that number is going to be small enough to be ignored, but not
in all cases. Is there anyway for the stats system to report the
information about rolledback inserts? In fact autovacuum probably has a
similar deficiency for rolled back deletes but not a rolled back update.
Anyone think this is enough of an issue that it needs more attention?
Matt