Re: Incremental Commits and Disappearing Sessions

From: Dereck L. Dietz <dietzdl_at_ameritech.net>
Date: Tue, 26 Feb 2008 21:37:56 GMT
Message-ID: <UE%wj.12635$J41.5332@newssvr14.news.prodigy.net>

"Anurag Varma" <avoracle_at_gmail.com> wrote in message news:35ca7a77-2d6b-4fe0-9f6b-39ea813f585b_at_k2g2000hse.googlegroups.com...
> On Feb 25, 7:54 pm, "Dereck L. Dietz" <diet..._at_ameritech.net> wrote:
>> Oracle 10.2.0.3.0
>> Windows 2003 Server
>> 16GB Memory (93% SGA( 93% Buffer Cache,<3% Shared Pool, <3% Large Pool),
>> 7%
>> PGA)
>>
>> In my continuing saga of disappearing sessions our off-site DBA has told
>> us
>> to do incremental commits.
>>
>> It's always been my understanding that you only want to commit at the end
>> of
>> your logical transaction. After reading Tom Kyte's book that only
>> reinforced that understanding.
>>
>> Could anybody shed any light on why he would suggest this?
>
>
> No clue. Ask your DBA why he suggests that?
>
> I recall you started this discussion way back stating that there was a
> ORA-7445 error associated with it (right?). An ORA-7445 should leave a
> trace on the server side.
> Other than that, have your tried tracing this session?
>
> Anurag

He said he applied a patch for the ORA-7445 errors but our processes are still disappearing. Received on Tue Feb 26 2008 - 15:37:56 CST

Original text of this message