Re: [HACKERS] Re: [BUGS] General Bug Report: palloc fails with lots of ANDs and ORs
От | Vadim B. Mikheev |
---|---|
Тема | Re: [HACKERS] Re: [BUGS] General Bug Report: palloc fails with lots of ANDs and ORs |
Дата | |
Msg-id | 34BD91D2.8A0985F1@sable.krasnoyarsk.su обсуждение исходный текст |
Ответ на | Re: [BUGS] General Bug Report: palloc fails with lots of ANDs and ORs (Bruce Momjian <maillist@candle.pha.pa.us>) |
Список | pgsql-hackers |
Bruce Momjian wrote: > > > Bruce, > > > > I did some homework. Here is what I have. The default max data segment size on our (AIX 4.1.4) box is around 130000kbytes. > > > > I put together a query which put me just past the threshold of the palloc "out of memory error". It is as follows: > > > > create table outlet ( > > number int, > > name varchar(30), > > ... > > } > > > > create unique index outlet_key on outlet using btree (number); > > > > select count(*) from outlet > > where > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1) or > > (number = 1 and number = 1 and number = 1); > > ... > > > > Wow, looks like a bug. Vadim, why would this happen? I got the same > palloc failure message here, and there is NO data in the table. This is bug in optimizer - try to EXPLAIN query... I have no time to fix it now - could return to this after Feb 1. Vadim
В списке pgsql-hackers по дате отправления: