Re: Interesting Bugs in 12cR1

  • From: Niall Litchfield <niall.litchfield@xxxxxxxxx>
  • To: rjoralist2@xxxxxxxxxxxxxxxxxxxxx
  • Date: Thu, 11 Jul 2013 19:00:38 +0100

The first one seems reasonable to me (though a bit odd who writes select *
from subselect first up). The feature works, sometimes and there's code
available as a workaround for when it doesn't, what would you have done? I
see 4 options
a) pull the new feature
b) pull the legacy code which it conflicts with
c) put in place a workaround
d) delay the release further

The interesting thing to me is the language of "legacy code constraints".
It wouldn't entirely surprise me if the constraints were there for good
reason and the workaround could encounter new versions of old issues.

I've been looking at threaded execution and came across the sys auth issue,
to date I've yet to find any great benefit for this process model, which
isn't the default. I haven't done anywhere near enough testing (probably
need to write a java program to test it out properly) but at the moment its
looking like one to file in the 'good in theory' bin.
On Jul 11, 2013 5:46 PM, "Rich Jesse" <rjoralist2@xxxxxxxxxxxxxxxxxxxxx>
wrote:

> Hemant writes:
>
> > Some bugs that I found interesting :
> >
> http://hemantoracledba.blogspot.com/2013/07/interesting-bugs-in-12cr1.html
>
> Bug 16047938 gets me because it's a new feature that's broken.  If it was a
> regression, I could see how that might be missed -- the product is huge --
> but shouldn't QA be verifying that a NEW feature actually works?
>
> Like anything, I'm sure there's more to the story that we'll never know,
> but
> it seems embarrassing.
>
> Bug 13877504 -- eww.  The better answer to me would be to just turn off
> threading.  Or change every cronjob that logs in via external
> authentication.
>
> Like I've said before: I'll wait for a few point releases before even
> trying
> -- it seems pointless to do so before the ironing is done.
>
> Rich
>
> --
> //www.freelists.org/webpage/oracle-l
>
>
>


--
//www.freelists.org/webpage/oracle-l


Other related posts: