Re: Order by and timestamp

[Björn Lundin <b.f.lundin@xxxxxxxxx>]

16 mars 2020 kl. 01:41 skrev Tom Lane <tgl@xxxxxxxxxxxxx>:

Adrian Klaver <adrian.klaver@xxxxxxxxxxx> writes:

On 3/15/20 2:33 PM, Björn Lundin wrote:

I then did ’select * from AMARKETS order by STARTTS’

Is amarkets in more then one schema?

Yeah, it's hard to think of any explanation other than "the query used a
corrupt index on startts to produce the ordering".  But your \d doesn't
show any index on startts.  So maybe there's more than one amarkets
table?

Yes - in other schemas - described in reply to Adrain

But the schema_path does not point to them

And those two other tables are empty

Another possibly-useful bit of evidence is to see what EXPLAIN shows as
the query plan for this query.

bnl=> explain select * from amarkets order by startts;

                              QUERY PLAN                               

-----------------------------------------------------------------------

 Sort  (cost=10702.57..10939.29 rows=94691 width=106)

   Sort Key: startts

   ->  Seq Scan on amarkets  (cost=0.00..2875.91 rows=94691 width=106)

(3 rader)

--

Björn Lundin

b.f.lundin@xxxxxxxxx