On Tue, Nov 17, 2020 at 04:58:45PM +0100, Frits Jalvingh wrote:
> Hi Justin, thanks for your help!
> I have attached both plans, both made with set enable_nestloop = false in
> the attachments.
> On the Postgresql 13 server work_mem is 64MB. It cannot really be higher
> there because Postgresql does not control its use of memory, setting it
> higher on this VM will cause the OOM killer to kill Postgresql for some
> queries.
Can you try to get an explain just for this query with either increased
work_mem or hash_mem_multiplier ?
Or possibly by messing with the cost parameters, including seq_page_cost.
Making all cost_* params 1000x smaller might allow the disable cost to be
effective.
--
Justin
