Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Usenet -> c.d.o.server -> Re: problems on statistics
Can you narrow it down to one or two slow statements using sql*plus. if you
can then you can work on tuning those statements in particular. You might
consider posting the statement and the explain plan here, together with the
explain plan if you use the RULE hint.
-- Niall Litchfield Oracle DBA Audit Commission UK "Marcelo Parra" <marcelop_at_pidcgroup.com> wrote in message news:abe6ut$ep0$1_at_newsreader.mailgate.org...Received on Fri May 10 2002 - 00:26:37 CDT
> Hi Daniel,
> You are rigth , sorry for this, but I 'd never said Oracle crash, I'd said
> procedures crash, but still it is not the correct term.
>
> The data is:
>
> Operating system platform : Windows NT 4 sp5
> Oracle version and release : Oracle 8.0.5.1.1 with Partitioning , and
in
> a replicated enviroment.
>
> I think that the only problem was the very slow response time speed,
because
> the crashes which I referred to, was
> "ORA-01013: user requested cancel of current operation " on vb
applications
> and oracle reports.
> Later, I found that It was produced by a timeout configuration on ODBC
on
> the client machine.
> Now, supossing it was in that way,
> which is the reason on a CBO running DB to work worst with statistics
than
> without then.
>
> Thanks in advance,
> Marcelo
>
>
> "Daniel Morgan" <dmorgan_at_exesolutions.com> wrote in message
> news:3CDA9824.27440A26_at_exesolutions.com...
> > Marcelo Parra wrote:
> >
> > > Hi , Anyone could tell me wicth is the problem on saving statistics
with
> CBO
> > > that makes crash some procedures or queries?
> > > I have on my production Db this configuration and once I've tried to
> save
> > > statistics and I'had lots of problems .
> > > I don't know any fundament for this to happen, but It did happened.
> > >
> > > Thanks in advance,
> > > Marcelo Parra.
> >
> > I don't know what happened either as you didn't post an operating system
> and
> > platform, a version of Oracle, or the actuall and complete error
message.
> Doing
> > so would, I suspect, be helpful to anyone trying to help you.
> >
> > What I can tell you is that I have never had Oracle crash doing what you
> > describe.
> >
> > Daniel Morgan
> >
>
>
![]() |
![]() |