Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: DB Buffer Cache Size
Connor McDonald <connor_mcdonald_at_yahoo.com> wrote in message news:<412B468D.72D_at_yahoo.com>...
> Howard J. Rogers wrote:
> >
> [snip]
>
> > I simply don't understand the basis of the question. I offer you a bodge-job
> > that fixes nothing, but makes things appear OK on the surface for $10,000;
> > and a comprehensive, thorough and "correct" fix of the underlying issues
> > for $15,000; and the gist of your question implies you'd pick the $10,000
> > option?
> >
> [snip]
>
> > Even the lowest end of town must known that 10K today, plus 10K next year,
> > plus 10K the year after is not a 'quick-and-cheap' alternative to paying
> > 15K-right-now-and-that's-all.
> >
> > Regards
> > HJR
>
> Depends where you sit in the hierarchy...My experience is:
>
> Consultant: recommends $15k fix
> DBA/Developer: prepares case/spec for $15k fix
> Project Mgr: asks what can be done for only $10k
> DBA/Developer: presents $10k and $15k option - stresses importance of
> latter
> Project Mgr: presents $10k and $15k option to senior mgt - stresses
> importance of saving $5k
> Senior Mgt: opts for $10k
>
> Its like a place I'm at, where little or no referential integrity exists
> on one of the sub-systems. We presented a case for fixing the current
> data and adding RI, to which the response was:
>
> "What percentage data integrity problems do we currently have?"
>
> Hmmmmm....I always thought data integrity was a boolean.
>
> Connor McDonald
Connor,
It used to be a boolean, before the capability existed to enable a constraint NOVALIDATE. I can see why one would want in the case of exchanging partitions, but other than that, its too easy for someone or some piece of code to simply enable/re-enable a constraint NOVALIDATE and move on, leaving the garbage right where it lay. I see it on a daily basis. Job security, perhaps.
Principle of uninintended consequences in action.
I'm not saying that the EXCEPTIONS INTO <exceptions_table> clause isn't helpful.
I wonder if there will be an analogous feature for FORCED LOGGING by tablespace, with constraints whereby the dba can specify that a constraint should not be allowed to be enabled NOVALIDATE. I'd settle for such an action being able to be audited, I guess.
-bdbafh Received on Tue Aug 24 2004 - 13:15:03 CDT