Josh Berkus <josh@agliodbs.com> writes:
> (c) we *think* that other columns of the table, including other indexed
> columns, are changed much more frequently than the intarray column is.
> Currently doing analysis on that.
Yeah, I noticed that your statistics for the table showed far more
updates than insertions or deletions. If the intarray itself didn't
change often, that would result in lots of duplicate entries being made
in the index. They'd get cleaned by vacuum eventually, but maybe not
fast enough to avoid the one-live-tuple-per-page syndrome that Alexander
was speculating about.
regards, tom lane