Re: Re: [COMMITTERS] pgsql: Fix free space map to correctly track the total amount of FSM
От | Tatsuo Ishii |
---|---|
Тема | Re: Re: [COMMITTERS] pgsql: Fix free space map to correctly track the total amount of FSM |
Дата | |
Msg-id | 20071004.152349.81959457.t-ishii@sraoss.co.jp обсуждение исходный текст |
Ответ на | Re: Re: [COMMITTERS] pgsql: Fix free space map to correctly track the total amount of FSM (Decibel! <decibel@decibel.org>) |
Список | pgsql-hackers |
> On Tue, Oct 02, 2007 at 09:07:54PM -0400, Alvaro Herrera wrote: > > Decibel! wrote: > > > > > Hrm... what about adding output to vacuum verbose that indicates how many > > > pages in a relation have free space? That would allow something like > > > pgfouine to see how many FSM pages were needed. It would also make it > > > easier to identify relations that could stand a vacuum full/reindex/cluster > > > (though you'd also want to know something like average free space per > > > page). > > > > Rather than wasting time fixing minor FSM issues, I would favor > > rewriting the stuff so that the FSM is disk-spillable. > > Sure, but this would also likely be a 20 line change to vacuum... These proposals would not help, at least me at all. Since I was talking about the pre 8.2 versions. There's 0 chance these changes are backported to previous versions. I'm thinkg about writing a small function which will do something 8.2 or later's vacuum does(telling the right FSM pages needed). -- Tatsuo Ishii SRA OSS, Inc. Japan
В списке pgsql-hackers по дате отправления: