Re: awr history
- From: Kellyn Pedersen <kjped1313@xxxxxxxxx>
- To: daniel.fink@xxxxxxxxxxxxxx, Brandon.Allen@xxxxxxxxxxx
- Date: Thu, 13 Jan 2011 14:41:24 -0800 (PST)
Now aren't you guys smart! :P
I did actually keep track of instance restarts and had a flag in the repository
table, but it was a quick way to fix that problem and wasn't really the best
way.
Skipping numbers in the snap_id's? I'm trying to remember if I've actually
seen
that, (and I may have and just was oblivious to it!) Something to look into...
Thank you gentlemen, I do not plan on giving up the scripts and code I wrote to
wrap around Karl's scripts- I'm kind of partial to it... :)
Kellyn Pedersen
Mulit-Platform Database Administrator
http://www.linkedin.com/in/kellynpedersen
www.dbakevlar.com
________________________________
From: Daniel W. Fink <daniel.fink@xxxxxxxxxxxxxx>
To: Brandon.Allen@xxxxxxxxxxx
Cc: "oracle-l@xxxxxxxxxxxxx" <oracle-l@xxxxxxxxxxxxx>
Sent: Thu, January 13, 2011 12:09:05 PM
Subject: Re: awr history
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.
On 1/13/2011 12:00 PM, Allen, Brandon wrote:
>These are nice scripts, but just one caution on a problem I just
>noticed with them – if your snap_ids skip numbers in their
>sequence
>then the stats for those intervals will be omitted because the
>query
>relies on logic like this: “AND s1.snap_id = s0.snap_id
>+
>1”
>
>Another way of doing this without the limitation above is to use
>the
>LAG function – here is an example:
>
>select a.snap_id, a.snap_time, round((b.value-lag(b.value) over
>(order by a.snap_time)) / (86400*(a.snap_time-lag(a.snap_time)
>over
>(order by a.snap_id)))) as "Rds/sec" from stats$snapshot a,
>
>stats$sysstat b where a.snap_id = b.snap_id and b.name = 'physical
>reads' and a.snap_time > sysdate-1 order by 1;
>
>Regards,
>Brandon
>
>From:oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On
>Behalf Of Kellyn Pedersen
>
>
>
>Karl must have missed this email, but I'll promote his scripts
>for
>him... I've had way too much fun with them not to! :)
>
>The AWR gen script, (awr_genwl.sql) was a source of great fun,
>that only after a few, minor tweaks and a table creation, I
>was retaining high level AWR data by dbid in a repository. How
>nice it was to create many pretty, colorful graphs for
>non-technical managers when they were quick to say, "It was the
>
>databases fault!" :D
>http://karlarao.wordpress.com/scripts-resources/
>
>
>
>
>________________________________
Privileged/Confidential Information may be contained in this message
or
attachments hereto. Please advise immediately if you or your employer
do
not consent to Internet email for messages of this kind. Opinions,
conclusions and other information in this message that do not relate to
the official business of this company shall be understood as neither
given nor endorsed by it.
>
> No virus found in this incoming message. Checked by AVG - www.avg.com
>Version: 9.0.872 / Virus Database: 271.1.1/3376 - Release Date: 01/12/11
>12:34:00
>
>
Other related posts:
