Re: New 8.1.6 bug of the day

buckeye714_at_my-deja.com wrote:
>
> This is a known Oracle Bug and a patch is available. We ran into the
> problem a couple of months ago. We immediately called Oracle Support
> and sent them the tables, some data and the query that returned the bad
> results. They were able to reproduce it in their labs. Turns out it is
> an old Bug with the Rule Based Optimizer that was once fixed but has
> been re-introduced in 8.1 Following is a snippit from the TAR we filed:
>
> <Bug:1176886>
> This bug is effectively a duplicate of <Bug:385681> re-introduced in
> 8.1 If an SMJ and unique index scan and the index scan is known
> to return only a single row then a query may either dump or return
> incorrect results.
> Workaround: Use CBO
>
> As I said, they supplied a patch, but I can no longer find it. Any one
> with concerns should call Oracle Support and reference the above Bug
> numbers.

I searched through the Oracle Metalinks site and did find the 1176886 bug, but the last time that any work was done on (at least according to the log in the bug report) was in late April. I was unable to find any patches which have any reference to the bug. The bug report did state that the workaround was to use the Cost Based Optimizer with statistics and then include the ALL_ROWS hint in the SELECT statement. The problem with the workaround it is the code review itself could be an enormous undertaking. Received on Mon May 15 2000 - 00:00:00 CDT