Re: Multiple SQL version count with cusror_sharing=similar

From: Tanel Poder <tanel_at_poderc.com>
Date: Thu, 2 Jul 2009 11:30:53 +0300
Message-ID: <4602f23c0907020130g3865de28rfc552bf32c9e26d5_at_mail.gmail.com>



Two main reasons, either:
  1. I have hit this problem myself (or a client has hit it), had to troublesheoot it and found this solution or piece of information
  2. I have read about it as someone else has hit such issue and has posted about it in a forum like Oracle-L or Oracle Forums. That's why the lists & forums are great, you spend 5 minutes skimming through the posts every day and when hitting some weird problem, there's a chance you've already heard about it, so will troubleshoot faster.

On Thu, Jul 2, 2009 at 10:52 AM, Neeraj Bhatia <neeraj.dba_at_gmail.com> wrote:

> Hi Tanel,
>
> Thanks for the update. I am on 11.1.0.6, will upgrade and check the
> functionality.
>
> One causual question, how you track version releases and check the
> functionality so quickly. It's really amazing.
>
> Warm Regards,
> Neeraj.
>
> On Thu, Jul 2, 2009 at 3:21 AM, Tanel Poder <tanel_at_poderc.com> wrote:
>
>> In 11.1.0.7 there's a new field HASH_MATCH_FAILED in v$sql_shared_cursor
>> which covers the "unsafe binds due histograms with cursor_sharing=similar"
>> problem case.
>>
>> --
>> Tanel Poder
>> http://blog.tanelpoder.com
>>
>>
>>
>> On Wed, Jul 1, 2009 at 4:39 PM, Rich Jesse <
>> rjoralist_at_society.servebeer.com> wrote:
>>
>>> > BTW, while search in metalink, there are numbers of bug hits related to
>>> the
>>> > issue.
>>> >
>>> > Regards,
>>> > Neeraj Bhatia
>>>
>>>
>>>
>

-- 
Tanel Poder
http://blog.tanelpoder.com

--
http://www.freelists.org/webpage/oracle-l
Received on Thu Jul 02 2009 - 03:30:53 CDT

Original text of this message