RE: unique constraint violation on 8i import but not on 9i
- From: "Mark W. Farnham" <mwf@xxxxxxxx>
- To: <ag@xxxxxxxxxxxx>, <oracle-l@xxxxxxxxxxxxx>
- Date: Wed, 4 Jul 2007 11:07:10 -0400
Your guess seems the most likely cause of the problem to me as well and that
is what I would check first before I did full research to see exactly what
was causing the error in 8i, which I guess would be to load a version of the
table in 8i without the primary key enabled and then check for duplicates.
If that came up empty (no duplicates), I'd probably see if the primary key
could be enabled. If that found duplicates, I'd be looking for a bug in 9i
or a timing issue about when indexes and constraints were enabled or a
difference in value interpretation between 8i and 9i including nls,
characterset, and anything else I could think of, unless finding out what
was wrong in 9i was not important. And I'd sure be wondering how the
duplicate ever got into 8i in the first place, but that is a different
matter.
Yep, I think your guess that there was existing data is a very good guess
and worth a quick check before embarking on detailed exploration of the
exact cause of the problem.
My guess is that we differ a wee bit about what each of us thinks is a
guess, but that we both endorse applying logic and science to
troubleshooting. Maybe a better way to say that is "good guessing prunes the
research tree; silly guessing just wastes time." Or perhaps "x seems so much
more likely than all other possibilities combined and it is so cheap to
check that I'm going to rule that out first, before any other research."
Probably that is what you consider "proper troubleshooting." Me too.
Regards,
mwf
-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx]
On Behalf Of Alex Gorbachev
Sent: Wednesday, July 04, 2007 8:57 AM
To: oracle-l@xxxxxxxxxxxxx
Subject: Re: unique constraint violation on 8i import but not on 9i
Why on earth would someone think that the problem would be resloved
faster with guess than with proper trobleshooting?
How much time would it take to check content of the table and see that
there were any rows already before the import?
<snip>
--
//www.freelists.org/webpage/oracle-l
Other related posts:
