Re: gcs log flush sync

We were in the midst of a huge performance problem today on a 6 way RAC and this occurred enough times to be both unusual and interesting. Even more interesting is that this wait looks like it's related to I/O some way. Even more interesting, we had identified writes as a real issue, especially redo writes.

Thanks for the explanation!! Always good to hear from the folks that wrote the book on waits!

Thanks again
John

• Original Message ---- From: K Gopalakrishnan <kaygopal_at_gmail.com> To: orclwzrd_at_yahoo.com Cc: oracle-l <oracle-l_at_freelists.org> Sent: Thursday, October 9, 2008 5:18:12 PM Subject: Re: gcs log flush sync

John,

I don't discount this as an idle wait. Basically you (LMS) are/is waiting for LGWR to flush the pending redo during CR fabrication. Without going too much in to details, you can turn off the behaviour by setting _cr_server_log_flush to false.

Are you looking for the details on this wait event or having a performance problem and suspecting the wait?

On Thu, Oct 9, 2008 at 12:22 PM, John D Parker <orclwzrd_at_yahoo.com> wrote:
> Anyone have any solid information on this wait event? The 10.2 docs indicate
> it's an idle event, but it's name is far too ominous to just be an idle
> event of the same pedigree as null event...
>
> Thanks,
> John
>
> PS. I'm seeing it on a 6 way 10.2.0.4 RAC on RHEL ES 4
>
>

-- 
Best Regards,
K Gopalakrishnan
Co-Author: Oracle Wait Interface, Oracle Press 2004
http://www.amazon.com/exec/obidos/tg/detail/-/007222729X/

Author: Oracle Database 10g RAC Handbook, Oracle Press 2006
http://www.amazon.com/gp/product/007146509X/



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