Re: Recreating unique indices without having dups & downtime. - Mailing list pgsql-general

From Charles Tassell
Subject Re: Recreating unique indices without having dups & downtime.
Date
Msg-id 4.3.2.7.2.20000709192830.00d7b800@mailer.isn.net
Whole thread Raw
In response to Recreating unique indices without having dups & downtime.  (Denis Perchine <dyp@perchine.com>)
List pgsql-general
You should just be able to create the same index with a different name,
then drop the original.  I don't think there are any conflicts with indexes
with the same structure, as long as they have different names.  I've never
actually tried this myself  though.

If the DTD problem within transactions was fixed in 7.0x, you might even be
able to do this within a transaction.  I don't remember hearing that DTD
operations were now reliable within transactions with the 7.0 release
announcement though.

At 11:34 AM 7/9/00, Denis Perchine wrote:
>Hello,
>
>Is it possible to recreate unique indices without having dups (in the case
>of drop/create),
>or downtime (in the case of reindex)? This is real problem.
>
>--
>Sincerely Yours,
>Denis Perchine
>
>----------------------------------
>E-Mail: dyp@perchine.com
>HomePage: http://www.perchine.com/dyp/
>FidoNet: 2:5000/120.5
>----------------------------------


pgsql-general by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: Is level a reserved keyword? (FYI)
Next
From: Philip Warner
Date:
Subject: Announce: Testers needed for pg_dump with BLOB support.