Re: Import bug

Altering a users quota to zero after they've created tables is a perfectly valid (though slightly unusual) thing to do, so I think you've got a point, sort of.

I'd agree with you that you might consider it a problem, except that it's dead easy to fix it: you just re-set the user's quota in the new database, and run import again for that particular schema. You then re-re-set the quota when all is finished. Fiddly, perhaps. But not terminal.

I think, fundamentally, it's an issue of perception as to what import does. All it does, in fact, is to read a bunch of 'create table' and 'insert into' statements from the dump file. It's therefore dumb, and can't be expected to finesse things by inventing commands which aren't in the dump file -that really would be a bug if it did!

I find that once people open up a dump file in the text editor of their choice and have a look at what it contains (look only!! it's a binary file, so don't save or edit!!!), then it becomes much more obvious what import is capable of: it's only reading a script, after all.

Regards
HJR "Chuck" <chuckh_at_softhome.net> wrote in message news:afidij$erb69$1_at_ID-85580.news.dfncis.de...
>
> "Daniel Morgan" <dmorgan_at_exesolutions.com> wrote in message
> news:3D1C8ACA.BFB2FB16_at_exesolutions.com...
> > I don't think that is what Tom said at all. And I find your original
> posting
> > logically inconsistent. How exactly does a schema come to have tables in
a
> > tablespace in which its quota is zero?
>
> Easy. After tables are created the DBA sets the quota on the tablespace to
> 0? How is that logically inconsistent?
>
> >
> > And what developer, or DBA, could possibly think that they are going to
> violate
> > the security model by building objects in a tablespace for which they
have
> no
> > privilege to create anything.
> >
> > The only bug I see here is in not doing reasonable preparation before
> trying the
> > import.
>
> The problem here is an export can be taken of a database, that when
imported
> into an empty database will not restore the database to it's original
> condition. I call that a *serious* problem with the import utility.
>
>
>
Received on Fri Jun 28 2002 - 15:46:53 CDT