Re: how to dump and comprehend a rollback block

From: NetComrade <andreyNSPAM_at_bookexchange.net>
Date: Fri, 11 Jul 2003 18:16:42 GMT
Message-ID: <3f0efe03.343001229@nyc.news.speakeasy.net>

Actually, wrapping the SQL with a savepoint and rollback is much better than doing a commit:

http://groups.google.com/groups?q=author:netcomrade&start=30&hl=en&lr=&ie=UTF-8&oe=UTF-8&selm=3e96eed6.3973444080%40nyc.news.speakeasy.net&rnum=35

This may be due to the A5200 disk arrays that we have :) But there are a lot of positive things by doing rollbacks, mainly minimal redo generated. Since we have 10k's of distributed queries per min, it all adds up.

On Sat, 05 Jul 2003 10:18:50 +0200, Jaap W. van Dijk <j.w.vandijk.removethis_at_hetnet.nl> wrote:

>There are issues involving distributed SELECT's causing TX locks on
>rollback. Do you use these? Then maybe the following link can put you
>on track:
>
>http://www.jlcomp.demon.co.uk/faq/dblink_commit.html
>
>HTH, Jaap.

.......
We use Oracle 8.1.7.4 on Solaris 2.7 boxes remove NSPAM to email Received on Fri Jul 11 2003 - 13:16:42 CDT