Re: awr history
From: Nuno Souto <dbvision_at_iinet.net.au>
Date: Fri, 14 Jan 2011 20:54:36 +1100
Message-ID: <4D301D5C.4000204_at_iinet.net.au>
Daniel W. Fink wrote,on my timestamp of 14/01/2011 6:09 AM:
> I will add one more caveat - remember that values reset after instance restarts.
> So if you are depending on calculating the deltas (the only option available in
> statspack), you need to test for an instance restart and adjust your logic
> appropriately. I have a script
> (http://www.optimaldba.com/scripts/CustomStatspack/optstat_statistic_matrix_weekly.sql)
> that has that logic in the factored subqueries at the beginning.
>
Date: Fri, 14 Jan 2011 20:54:36 +1100
Message-ID: <4D301D5C.4000204_at_iinet.net.au>
Daniel W. Fink wrote,on my timestamp of 14/01/2011 6:09 AM:
> I will add one more caveat - remember that values reset after instance restarts.
> So if you are depending on calculating the deltas (the only option available in
> statspack), you need to test for an instance restart and adjust your logic
> appropriately. I have a script
> (http://www.optimaldba.com/scripts/CustomStatspack/optstat_statistic_matrix_weekly.sql)
> that has that logic in the factored subqueries at the beginning.
>
One way to work around that problem is to test each delta. If it is negative, it is a direct indicator of a startup in the middle. Works fine for me, but of course with the caveat that for long deltas it is not reliable.
-- Cheers Nuno Souto in sunny Sydney, Australia dbvision_at_iinet.net.au -- http://www.freelists.org/webpage/oracle-lReceived on Fri Jan 14 2011 - 03:54:36 CST