Re: CREATE INDEX and HOT - revised design - Mailing list pgsql-hackers

From Simon Riggs
Subject Re: CREATE INDEX and HOT - revised design
Date
Msg-id 1174502620.6069.43.camel@silverbirch.site
Whole thread Raw
In response to Re: CREATE INDEX and HOT - revised design  (Bruce Momjian <bruce@momjian.us>)
List pgsql-hackers
On Wed, 2007-03-21 at 13:29 -0400, Bruce Momjian wrote:
> Pavan Deolasee wrote:
> > On 3/21/07, Bruce Momjian <bruce@momjian.us> wrote:
> > >
> > >
> > >
> > > I am worried that will require CREATE INDEX to wait for a long time.
> > 
> > 
> > 
> > Not unless there are long running transactions. We are not waiting
> > for the lock, but only for the current transactions to finish.
> 
> Waiting for all transactions might take a while, no?

It would be nice if we could force CREATE INDEX to only run outside a
transaction block, but that just isn't possible if you care about
existing applications.

Any PL function running a CREATE INDEX would fail and I'm pretty sure
there's lots of them. Many of them use dynamic SQL, so you wouldn't even
be able to scan them prior to run-time to know that they would fail.

--  Simon Riggs              EnterpriseDB   http://www.enterprisedb.com




pgsql-hackers by date:

Previous
From: Grzegorz Jaskiewicz
Date:
Subject: Re: [PATCHES] Bitmapscan changes
Next
From: Heikki Linnakangas
Date:
Subject: Re: CREATE INDEX and HOT - revised design