Re: [SQL] query doesn't always follow 'correct' path..

2013-02-18 Thread Bert
Hello, there were 3 hours in between the 2 queries. so I guess new data was loaded already. new data is being loaded with that etl_run_id. wkr, Bert On Mon, Feb 18, 2013 at 4:20 PM, Виктор Егоров wrote: > 2013/2/18 Bert > >> When I don't touch the indexscan setting I get the following output

Re: [SQL] query doesn't always follow 'correct' path..

2013-02-18 Thread Julien Cigar
On 02/18/2013 16:20, Julien Cigar wrote: On 02/18/2013 15:39, Bert wrote: Hello, Thanks the nice people on irc my problem is fixed. I changed the following settings in the postgres.conf file: default_statistics_target = 5000 -> and I analyzed the tables after the change of course -> now I only

Re: [SQL] query doesn't always follow 'correct' path..

2013-02-18 Thread Julien Cigar
On 02/18/2013 15:39, Bert wrote: Hello, Thanks the nice people on irc my problem is fixed. I changed the following settings in the postgres.conf file: default_statistics_target = 5000 -> and I analyzed the tables after the change of course -> now I only got 2 plans anymore, in stead of 3 defa

Re: [SQL] query doesn't always follow 'correct' path..

2013-02-18 Thread Виктор Егоров
2013/2/18 Bert > When I don't touch the indexscan setting I get the following output: > Total query runtime: 611484 ms. > 20359 rows retrieved. > and the following plan: http://explain.depesz.com/s/sDy > > However, when I put set enable_indexscan=off; in fron of the same query I > get the followi

Re: [SQL] query doesn't always follow 'correct' path..

2013-02-18 Thread Bert
Hello, Thanks the nice people on irc my problem is fixed. I changed the following settings in the postgres.conf file: default_statistics_target = 5000 -> and I analyzed the tables after the change of course -> now I only got 2 plans anymore, in stead of 3 cpu_tuple_cost = 0.1 -> by setting this va

Re: [SQL] query doesn't always follow 'correct' path..

2013-02-18 Thread Bert
Hello, yes, the tables are vacuumed every day with the following command: vacuum analyze schema.table. The last statistics were collected yesterday evening. I collected statistics about the statistics, and I found the following: table_name; starttime; runtime "st_itemseat";"2013-02-17 23:48:42";"0

Re: [SQL] query doesn't always follow 'correct' path..

2013-02-18 Thread Frank Lanitz
Am 18.02.2013 10:43, schrieb Bert: > Does anyone has an idea what triggers this bad plan, and how I can fix it? Looks a bit like wrong statistics. Are the statistiks for your tables correct? Cheers, Frank -- Sent via pgsql-sql mailing list (pgsql-sql@postgresql.org) To make changes to your sub