Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: Single-statement 'write consistency' on read committed. Oh, really?
Hello,
Please see my comments below.
"APC" <andrew.clarke_at_logicacmg.com <mailto:andrew.clarke_at_logicacmg.com>>
wrote in message
<news:58428d02.0312030425.528c46a8_at_posting.google.com>...
> Following from previous posting: I was using a slightly different set
> of SQL than Tengiz used, which explains my different results on the
> 9.2 setup. Having re-done the tests using Tengiz's actual code I have
> duplicated his results.
I am glad that Oracle _is_ behaving consistently at least with respect to executing scenario 2 on different installations. Thank you for re-running the test.
>
> However, I now understand that this is the Oracle is supposed to work,
> and the results are *not* inconsistent.
>
Well, I am not so sure.
Firstly, regarding your test, it's easy enough to predict its outcome iff one knows about the Oracle re-start pecularity:
Now why I am not happy about the observed behaviour. Briefly, the reasons are as follows:
<...skipped...>
Rgds. Received on Wed Dec 03 2003 - 16:48:44 CST