Re: Oracle EE 12.1 standardize on bigfile tablespaces?

From: Chris Stephens <cstephens16_at_gmail.com>
Date: Thu, 23 Feb 2017 14:25:52 +0000
Message-ID: <CAEFL0swNjfDbr1h_CQYJUgmjJtosF+xDY6y1bMsnbWnLhjwpPA_at_mail.gmail.com>



in my opinion, we've failed as DBA's if we ever allow a diskgroup to approach full.

so far, the only disadvantage that seems significant to to me s an even where a datafile restore is required.

thanks for all the input!

On Wed, Feb 22, 2017 at 8:54 AM Yong Huang <dmarc-noreply_at_freelists.org> wrote:

> > standardizing on bigfile tablespaces for all application data.
>
> In addition to what others (Mladen Gogala etc.) have said, there's one
> other annoyance with a bigfile tablespace. Suppose you have diskgroups DATA
> and FRA and this bigfile tablespace is in DATA. One night you run out of
> space in DATA due to fast growth of this tablespace. You can't use the free
> space in FRA to temporarily solve the problem because you can't add a
> datafile to a bigfile tablespace. Although it's not a good idea to scatter
> some data in FRA, a smallfile tablespace at least gives you this
> flexibility in an emergency solution.
>
> Speaking of restoring a huge datafile versus a small one. Don't forget to
> try RMAN block media recovery first. The time it takes is not related to
> the size of the datafile where the corruption occurs. If this doesn't work,
> then try data file media recovery, and a small datafile is obviously better.
>
> Yong Huang
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

--
http://www.freelists.org/webpage/oracle-l
Received on Thu Feb 23 2017 - 15:25:52 CET

Original text of this message