Re: Significant discrepancy in V$OSSTAT since upgrading to 10.2.0.5
From: joel garry <joel-garry_at_home.com>
Date: Wed, 21 Jul 2010 11:22:49 -0700 (PDT)
Message-ID: <0e7e2df5-e2dc-46db-b795-29c6fa3cbbdf_at_l25g2000prn.googlegroups.com>
On Jul 21, 8:51 am, Mladen Gogala <n..._at_email.here.invalid> wrote:
> I used V$OSSTAT for monitoring in 10.2.0.4 and it was much more accurate
> than is the case with 10.2.0.5. Look at this:
>
> SQL> column stat_name format a20
> SQL> select stat_name,value from v$osstat
> 2 where stat_name like '%TIME';
>
> STAT_NAME VALUE
> -------------------- ----------
> IDLE_TIME 80163480
> BUSY_TIME 42444869
> USER_TIME 32543474
> SYS_TIME 7838644
> IOWAIT_TIME 8463912
> NICE_TIME 1
> RSRC_MGR_CPU_WAIT_TI 0
> ME
>
> SQL> select 42444869/80163480 from dual;
>
> 42444869/80163480
> -----------------
> .529478872
>
> SQL> !sar -u 3 5
> Linux 2.6.9-22.ELsmp (oracle13) 07/21/2010
>
> 11:45:27 AM CPU %user %nice %system %iowait %idle
> 11:45:30 AM all 17.83 0.00 8.08 6.00 68.08
> 11:45:33 AM all 19.23 0.00 8.83 2.66 69.28
> 11:45:36 AM all 14.51 0.00 6.67 0.83 77.98
> 11:45:39 AM all 19.67 0.00 8.25 1.58 70.50
> 11:45:42 AM all 13.33 0.00 7.00 1.25 78.42
> Average: all 16.92 0.00 7.77 2.47 72.85
>
> SQL>
>
> So, V$OSSTAT tells me that my CPU resources are 52.9% busy while sar
> tells me that those very same resources are 70% idle. The "top" monitor
> confirms that "sar" is right and not Oracle.
> --http://mgogala.byethost5.com
Date: Wed, 21 Jul 2010 11:22:49 -0700 (PDT)
Message-ID: <0e7e2df5-e2dc-46db-b795-29c6fa3cbbdf_at_l25g2000prn.googlegroups.com>
On Jul 21, 8:51 am, Mladen Gogala <n..._at_email.here.invalid> wrote:
> I used V$OSSTAT for monitoring in 10.2.0.4 and it was much more accurate
> than is the case with 10.2.0.5. Look at this:
>
> SQL> column stat_name format a20
> SQL> select stat_name,value from v$osstat
> 2 where stat_name like '%TIME';
>
> STAT_NAME VALUE
> -------------------- ----------
> IDLE_TIME 80163480
> BUSY_TIME 42444869
> USER_TIME 32543474
> SYS_TIME 7838644
> IOWAIT_TIME 8463912
> NICE_TIME 1
> RSRC_MGR_CPU_WAIT_TI 0
> ME
>
> SQL> select 42444869/80163480 from dual;
>
> 42444869/80163480
> -----------------
> .529478872
>
> SQL> !sar -u 3 5
> Linux 2.6.9-22.ELsmp (oracle13) 07/21/2010
>
> 11:45:27 AM CPU %user %nice %system %iowait %idle
> 11:45:30 AM all 17.83 0.00 8.08 6.00 68.08
> 11:45:33 AM all 19.23 0.00 8.83 2.66 69.28
> 11:45:36 AM all 14.51 0.00 6.67 0.83 77.98
> 11:45:39 AM all 19.67 0.00 8.25 1.58 70.50
> 11:45:42 AM all 13.33 0.00 7.00 1.25 78.42
> Average: all 16.92 0.00 7.77 2.47 72.85
>
> SQL>
>
> So, V$OSSTAT tells me that my CPU resources are 52.9% busy while sar
> tells me that those very same resources are 70% idle. The "top" monitor
> confirms that "sar" is right and not Oracle.
> --http://mgogala.byethost5.com
Should it be busy_time / ( busy_time + idle_time )?
select 42444869/(42444869+80163480) from dual;
42444869/(42444869+80163480)
.346182534
(from http://blog.tanelpoder.com/seminar/seminar-files/#comments , I could be totally misunderstanding, of course.)
jg
-- _at_home.com is bogus. So California requires big electricity providers to get 20% of the power from renewable resources. SDGE finances a wind farm in Montana that sells the power to Canada, and gets 45% of its investment as a write-off in tax credits and accelerated depreciation, and can count it as part of the 20%. Meanwhile, I have to worry about rolling blackouts power-spiking all my devices with computers in them, including prophylactic blackouts because the power grid sparks fires. Because they don't keep their database of trees that need to be trimmed up to date.Received on Wed Jul 21 2010 - 13:22:49 CDT