Horton

Re: [HACKERS] Re: [QUESTIONS] MySQL benchmark page\

Bruce Momjian <maillist@candle.pha.pa.us>

View thread
From: Bruce Momjian <maillist@candle.pha.pa.us>
To: vadim@sable.krasnoyarsk.su (Vadim B. Mikheev)
Cc: scrappy@hub.org, herouth@oumail.openu.ac.il, pgsql-hackers@postgreSQL.org
Date: 1998-02-04T02:37:51Z
Lists: pgsql-hackers 
> 
> The Hermit Hacker wrote:
> > 
> >         So, why is pg_class locked for the duration of a vacuum when the
> > vacuum is being performed for the whole database when it should (I think)
> > only need to be locked when updates are happening to it?
> 
> Yes, this is how it should be. This long lock comes from very old time.
> When I visited vacuum last time (year ago) I hadn't time to decide
> why this lock is here and could it be just removed or not. 
> Still no time :)
> 
> Vadim
> 
> 

Added to TODO:

	* Make VACUUM on database not lock pg_class
	* Make VACUUM ANALYZE only use a readlock




-- 
Bruce Momjian
maillist@candle.pha.pa.us