Re: [GENERAL] Bug in postgres 9.6.2?

2017-07-21 Thread Tom Lane
greigwise writes: > If I can provide a pg_dump backup with a db where I can reproduce the error > and then also my postgresql.conf along with the query, would that be what > you need for a test case? Sounds like enough. regards, tom lane -- Sent

Re: [GENERAL] Bug in postgres 9.6.2?

2017-07-21 Thread greigwise
If I can provide a pg_dump backup with a db where I can reproduce the error and then also my postgresql.conf along with the query, would that be what you need for a test case? Thanks, Greig -- View this message in context:

Re: [GENERAL] Bug in postgres 9.6.2?

2017-07-20 Thread Tom Lane
greigwise writes: > I do still have the 9.6.2 instance. > Here is the explain analyze results from the 9.6.2 instance: Oh ... you have parallel query enabled, eh? In that case I'm less surprised, there were multiple parallel-query fixes since 9.6.2. Although those still

Re: [GENERAL] Bug in postgres 9.6.2?

2017-07-20 Thread greigwise
I do still have the 9.6.2 instance. Here is the explain analyze results from the 9.6.2 instance: Aggregate (cost=764612.56..764612.57 rows=1 width=8) (actual time=695.166..695.166 rows=1 loops=1) CTE test -> Nested Loop (cost=3345.90..757594.63 rows=311908 width=4) (actual

Re: [GENERAL] Bug in postgres 9.6.2?

2017-07-20 Thread Tom Lane
greigwise writes: > So, I have this query with nothing non-deterministic in it, yet I can run it > multiple times and get different results in postgres 9.6.2: > ... > Is it possible that this is a bug in 9.6.2 that was fixed in 9.6.3? Hard to tell without more information

[GENERAL] Bug in postgres 9.6.2?

2017-07-20 Thread greigwise
So, I have this query with nothing non-deterministic in it, yet I can run it multiple times and get different results in postgres 9.6.2: with test as ( select g.id2 from staging a join dim_1 e on e.id1 = a.id1 and a.created_at >= e.effective_at and a.created_at < e.expired_at join dim_2 g on