Re: [PERFORM] How to get explain plan to prefer Hash Join

Sorry about that, excel clipped off the dates. -- View this message in context: http://postgresql.nabble.com/How-to-get-explain-plan-to-prefer-Hash-Join-tp5841450p5841633.html Sent from the PostgreSQL - performance mailing li

Re: [PERFORM] How to get explain plan to prefer Hash Join

On Thu, Mar 12, 2015 at 8:59 AM, atxcanadian wrote: > Here is the output: > > > > This is after I've manually ran analyze. > The "last_*" columns are only showing times, and not full timestamps. Does your reporting tool drop th

Re: [PERFORM] How to get explain plan to prefer Hash Join

Here is the output: This is after I've manually ran analyze. -- View this message in context: http://postgresql.nabble.com/How-to-get-explain-plan-to-prefer-Hash-Join-tp5841450p5841610.html Sent from the PostgreSQL - perform

Re: [PERFORM] How to get explain plan to prefer Hash Join

On Wed, Mar 11, 2015 at 5:35 PM, atxcanadian wrote: > > I'm a little perplexed why the autovacuum wasn't keeping up. Any > recommendations for those settings to push it to do a bit more analyzing of > the tables?? > What does pg_stat_user_tables show for that table?

Re: [PERFORM] How to get explain plan to prefer Hash Join

Isn't a random_page_cost of 1 a little aggressive? We are currently setup on Amazon SSD with software RAID 5. -- View this message in context: http://postgresql.nabble.com/How-to-get-explain-plan-to-prefer-Hash-Join-tp5841450p5841605.html Sent from the PostgreSQL - performance mailing list ar

Re: [PERFORM] EXPLAIN (no ANALYZE) taking an hour for INSERT FROM SELECT

Gunnlaugur Thor Briem writes: > Yes, I think that's it: I've just realized that immediately prior to the > INSERT, in the same transaction, an unfiltered DELETE has been issued; i.e. > the whole table is being rewritten. Then the INSERT is issued ... with a > WHERE clause on non-existence in the (