Re: High "in memory undo latch" waits and cpu consumption
From: Harish Kalra <harish.kumar.kalra_at_gmail.com>
Date: Wed, 19 Mar 2008 17:13:04 +1100
Message-ID: <11ac22130803182313w441796f8p392534251c39fba2@mail.gmail.com>
Date: Wed, 19 Mar 2008 17:13:04 +1100
Message-ID: <11ac22130803182313w441796f8p392534251c39fba2@mail.gmail.com>
Josh,
Oracle acquire "In memory undo latch" to access IMU structures in the memory (Shared Pool) if you are you are using "In memory Undo". Quick way to fix this issue is to turn off the In Memory Undo by setting in_memory_undo to false.
Please share the AWR or statspack report incase you want to tune this.
Thanks & Regards
-Harish Kalra
On 3/19/08, Josh Collier <Josh.Collier_at_banfield.net> wrote:
>
> An update statement that is being run with bind variables by many
> concurrent sessions is spending all its time sucking down cpu and waiting on
> "in memory undo latch".
>
> Do you guys know anything about this latch? I cannot find much about it.
>
>
-- http://www.freelists.org/webpage/oracle-lReceived on Wed Mar 19 2008 - 01:13:04 CDT