Re: DBWR pegged out

From: Subodh Deshpande <deshpande.subodh_at_gmail.com>
Date: Thu, 10 Mar 2011 09:52:13 +0530
Message-ID: <AANLkTinbqFpffhaUqbUVJCM0fQdVt_jL9vh-Ja0Mweoi_at_mail.gmail.com>

I think if it is using temp tablespace then its application issue.. check the sorts, usuage of indexes..
random guess is think on re-sizing sga and check utilisation of memory by oracle and at os level..

Quote
oracle has recommended setting _db_fast_obj_ckpt and _db_fast_obj_truncate to false to see if they alleviate UnQuote

after this check whether checkpoint interval becomes consistent and size of archives..other wise in future you may face issues db recovery, if it requires..
thanks..subodh
On 9 March 2011 21:06, Doug Gernaat <dag_at_its.msstate.edu> wrote:

> thanks to all for the info...
>
> this is a sun-fire v890 48G of memory and 6CPU.
> this is a new problem after upgrading to 11gR2.
>
> it is sporadic... i want to say it is encountered mostly
> when there is commit intensive jobs running.
>
> oracle has recommended setting _db_fast_obj_ckpt and
> _db_fast_obj_truncate to false to see if they alleviate
> write complete waits and enq: KO - fast object.
>
> thanks
> -doug-
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

-- 
==============================
DO NOT FORGET TO SMILE TODAY
==============================

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

Received on Wed Mar 09 2011 - 22:22:13 CST