I was able to add the suggested indexes like stage.event__00075000((body->>'SID'::text)); and indeed these helped the QA environment use those indexes instead of sequence scanning.
I'm still perplexed by my original question, why production uses the partition index and qa does not?
Index Scan using ix_event__00014695_landing_id on event__00014695 e_3 (cost=0.56..39137.89 rows=37697 width=564) │
│ Index Cond: (landing_id = t_sap.landing_id)
Ultimately I think this is just highlighting the need in my environment to set random_page_cost lower (we are on an SSD SAN anyway..), but I dont think I have a satisfactory reason by the row estimates are so bad in the QA planner and why it doesnt use that partition index there.
On Fri, Sep 15, 2017 at 3:59 PM, Mike Broers <mbroers@xxxxxxxxx> wrote:
That makes a lot of sense, thanks for taking a look. An index like you suggest would probably further improve the query. Is that suggestion sidestepping the original problem that production is evaluating the landing_id bit with the partition index and qa is sequence scanning instead?AND exists (select 1 from t_sap where e.landing_id = t_sap.landing_id)) as rankings;Based on the difference in row estimate I am attempting an analyze with a higher default_statistic_target (currently 100) to see if that helps.On Fri, Sep 15, 2017 at 3:42 PM, Tom Lane <tgl@xxxxxxxxxxxxx> wrote:Mike Broers <mbroers@xxxxxxxxx> writes:
> If Im reading this correctly postgres thinks the partition will return 6.5
> million matching rows but actually comes back with 162k. Is this a case
> where something is wrong with the analyze job?
You've got a lot of scans there that're using conditions like
> │ -> Seq Scan on event__99999999 e_1 (cost=0.00..2527828.05 rows=11383021 width=778) (actual time=25522.389..747238.885 rows=42 loops=1)
> │ Filter: (((body ->> 'SID'::text) IS NOT NULL) AND (validation_status_code = 'P'::bpchar))
> │ Rows Removed by Filter: 12172186
While I'd expect the planner to be pretty solid on estimating the
validation_status_code condition, it's not going to have any idea about
that JSON field test. That's apparently very selective, but you're just
getting a default estimate, which is not going to think that a NOT NULL
test will exclude lots of rows.
One thing you could consider doing about this is creating an index
on (body ->> 'SID'::text), which would prompt ANALYZE to gather statistics
about that _expression_. Even if the index weren't actually used in the
plan, this might improve the estimates and the resulting planning choices
enough to make it worth maintaining such an index.
Or you could think about pulling that field out and storing it on its own.
JSON columns are great for storing random unstructured data, but they are
less great when you want to do relational-ish things on subfields.
regards, tom lane
