xes. Obviously, the optimizer
grossly miscalculates the costs, how can we help it?
On Fri, Feb 4, 2022 at 9:01 AM Tomas Vondra
wrote:
> On 2/4/22 05:21, A Shaposhnikov wrote:
> > Tomas,
> >
> > thank you! The query:
> >
> > select 1 from data as d, da
imizer would be able to infer it itself.
Thank you
On Thu, Feb 3, 2022 at 4:24 PM Tomas Vondra
wrote:
>
>
>
> On 2/3/22 20:32, A Shaposhnikov wrote:
> > I made a mistake yesterday claiming that the created statistics
> > changed the row counts in the estimates - it did no
y low cost regardless of size. Another guess is that it almost
ignores the "limit" parameter or uses it in some incorrect way.
On Thu, Feb 3, 2022 at 12:31 PM Vijaykumar Jain
wrote:
>
>
> On Fri, 4 Feb 2022 at 01:03, A Shaposhnikov wrote:
>>
>> I made a mistake yeste
I made a mistake yesterday claiming that the created statistics
changed the row counts in the estimates - it did not - I looked at
the wrong query yesterday. In the correct query plan the row estimate
still differs from the actual by many orders of magnitude:
Nested Loop (cost=1.01..27584834.53
