Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: suspicious standard Oracle Linux start/stop script

Re: suspicious standard Oracle Linux start/stop script

From: Niall Litchfield <n-litchfield_at_audit-commission.gov.uk>
Date: Fri, 30 May 2003 14:16:39 +0100
Message-ID: <3ed759b7$0$29719$ed9e5944@reading.news.pipex.net>


"Sybrand Bakker" <gooiditweg_at_sybrandb.demon.nl> wrote in message news:peqcdvce8adlqminseu6j3e16u512cbc2l_at_4ax.com...
> On Thu, 29 May 2003 16:42:21 +0100, "Niall Litchfield"
> <n-litchfield_at_audit-commission.gov.uk> wrote:
>
> >My advice
> >
> >modify the dbshut sql script so it does a shutdown abort. Oracle will
> >recover on startup.
> >
>
> AAAAAAAAAAAAAAAAAAAARRRRRRRRRRRRRRRRRRRRGGGGGGGGGGGGGGGGGGGGGGGGGGGGG
>
> Seems like we need a competition for the most stupid advice offered.
> There have been quite many the last few days.

The basis for my advice. dbshut is being called at system shutdown. The system will in any case kill off the Oracle processes (kill -9) if shutdown normal/immediate is taking too long. So shutdown immediate/normal has its dangers of an entirely uncontrolled shutdown occuring - this is what the OP was concerned about.

In contrast shutdown abort is very quick, and Oracle *will* recover from the online redo logs on startup. As howard mentions there is of course the danger of losing the relevant logfile, but that also must apply to shutdown immediate.

-- 
Niall Litchfield
Oracle DBA
Audit Commission UK
Received on Fri May 30 2003 - 08:16:39 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US