On Tue, Nov 8, 2011 at 7:19 PM, Greg Smith <greg@2ndquadrant.com> wrote:
> On 11/08/2011 05:07 PM, Robert Treat wrote:
>>
>> It's already easy to get "good enough" numbers based on user space
>> tools with very little overhead, so I think it's more important that
>> the server side tool be accurate rather than fast.
>
> What user space method do you consider good enough here? I haven't found
> any approximation that I was really happy with; wouldn't have bothered with
> this otherwise.
>
check_postgres and the pg_bloat_report both use a method of comparing
on disk size vs estimated size based on table structure (or index
info). Run regularly, it's certainly possible to keep bloat under
control. That said, I'd still like to see something more accurate.
Robert Treat
conjecture: xzilla.net
consulting: omniti.com