RE: Rollback Segment Shrinkage Question

Steve,

It's definitely possible that this is one of the "myths" about Oracle. I remember Walter Lindsay and his papers on rollback segments, they were like lightbulbs going off for me, I finally "got it" when I saw him present.

As Gaja said, Cary may have been part of the team that came up with this number and as he became more visible to the user community, many things became "his".

Thanks for the walk down memory lane :)

Rachel

>From: "Steve Adams" <steve.adams_at_ixora.com.au>
>To: "Rachel Carmichael" <carmichr_at_hotmail.com>, "Multiple recipients
>of list ORACLE-L" <ORACLE-L_at_fatcity.com>
>Subject: RE: Rollback Segment Shrinkage Question
>Date: Mon, 23 Oct 2000 09:41:12 +1000
>
>Hi Rachel,
>
>I think that the ascription of this rule of thumb to Cary Millsap is
>mistaken.
>The first bulletin that I saw containing this recommendation and the
>probabilistic model behind it was published by Walter Lindsay in Support
>Notes
>early in 1990 just a few month after Cary had joined Oracle.
>
>@ Regards,
>@ Steve Adams
>@ http://www.ixora.com.au/
>@ http://www.christianity.net.au/
>
>
>-----Original Message-----
>From: Rachel Carmichael [mailto:carmichr_at_hotmail.com]
>Sent: Sunday, 22 October 2000 10:46
>To: Multiple recipients of list ORACLE-L
>Subject: RE: Rollback Segment Shrinkage Question
>
>
>The minimum value for minextents doesn't necessarily have to be 20, it all
>depends on how your system is used. Oracle "suggests" 20 as the minimum
>because Cary Millsap once said it should be 20.
>
>there are no absolutes in this.
>
>My $.02
>
>Rachel
>
>
> >From: Gillian M <zmkj04_at_yahoo.com>
> >Reply-To: ORACLE-L_at_fatcity.com
> >To: Multiple recipients of list ORACLE-L <ORACLE-L_at_fatcity.com>
> >Subject: RE: Rollback Segment Shrinkage Question
> >Date: Sat, 21 Oct 2000 11:20:20 -0800
> >
> >Note that to determine the value for optimal:
> >value of initial extent * value for minextents
> >
> >The minimum value for minextents should be 20. By
> >default it is 2.
> >
> >G!
> >--- Gaja Krishna Vaidyanatha <gajav_at_yahoo.com> wrote:
> > > Hi Rachel,
> > >
> > > Just to add another point to what you have mentioned
> > > here, and
> > > that is the increased probability of 'ORA-1555
> > > snapshot too old'
> > > occuring when OPTIMAL clause is set too low in the
> > > rollback
> > > segments. This is especially true when your
> > > environment
> > > supports a large number of small transactions and
> > > some long
> > > running reports.
> > >
> > > Cheers,
> > >
> > > Gaja
> > >
> > > --- Rachel Carmichael <carmichr_at_hotmail.com> wrote:
> > > > nope, it doesn't. Oracle only checks the size of
> > > the rollback
> > > > against
> > > > optimal when you need to move from one extent
> > > within the
> > > > rollback segment to
> > > > another.
> > > >
> > > > You are right that when it shrinks automatically
> > > it can have a
> > > > performance
> > > > hit but that's usually minimal. What's important
> > > to check is
> > > > to see if your
> > > > rollback segments are extending and shrinking
> > > constantly. If
> > > > they are, you
> > > > have sized them and/or the optimal parameter
> > > incorrectly.
> > > >
> > > >
> > > > >From: "Cale, Rick T (Richard)"
> > > <RICHARD.T.CALE_at_saic.com>
> > > > >Reply-To: ORACLE-L_at_fatcity.com
> > > > >To: Multiple recipients of list ORACLE-L
> > > > <ORACLE-L_at_fatcity.com>
> > > > >Subject: RE: Rollback Segment Shrinkage Question
> > > > >Date: Tue, 17 Oct 2000 08:25:54 -0800
> > > > >
> > > > >I think after the transaction completes it will
> > > shrink back
> > > > to optimal.
> > > > >Keep
> > > > >in mind their is a performance
> > > > >hit and you should monitor the frequency of
> > > dynamic
> > > > extensions.
> > > > >
> > > > >Rick
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Carle, William T (Bill), NBSO
> > > [SMTP:wcarle_at_att.com]
> > > > > > Sent: Tuesday, October 17, 2000 10:21 AM
> > > > > > To: Multiple recipients of list ORACLE-L
> > > > > > Subject: Rollback Segment Shrinkage Question
> > > > > >
> > > > > > Hi,
> > > > > >
> > > > > > At what point in time does a rollback
> > > segment shrink
> > > > back to its
> > > > > > optimal
> > > > > > value?
> > > > > >
> > > > > >
> > > > > > Bill Carle
> > > > > > AT&T
> > > > > > Database Administrator
> > > > > > 816-995-3922
> > > > > > wcarle_at_att.com
> > > > > >
> >
> >
> >__________________________________________________
> >Do You Yahoo!?
> >Yahoo! Messenger - Talk while you surf! It's FREE.
> >http://im.yahoo.com/
> >--
> >Please see the official ORACLE-L FAQ: http://www.orafaq.com
> >--
> >Author: =?iso-8859-1?q?Gillian=20M?=
> > INET: zmkj04_at_yahoo.com
> >
> >Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
> >San Diego, California -- Public Internet access / Mailing Lists
> >--------------------------------------------------------------------
> >To REMOVE yourself from this mailing list, send an E-Mail message
> >to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
> >the message BODY, include a line containing: UNSUB ORACLE-L
> >(or the name of mailing list you want to be removed from). You may
> >also send the HELP command for other information (like subscribing).
>
>_________________________________________________________________________
>Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
>
>Share information about yourself, create your own public profile at
>http://profiles.msn.com.
>
>--
>Please see the official ORACLE-L FAQ: http://www.orafaq.com
>--
>Author: Rachel Carmichael
> INET: carmichr_at_hotmail.com
>
>Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051
>San Diego, California -- Public Internet access / Mailing Lists
>--------------------------------------------------------------------
>To REMOVE yourself from this mailing list, send an E-Mail message
>to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in
>the message BODY, include a line containing: UNSUB ORACLE-L
>(or the name of mailing list you want to be removed from). You may
>also send the HELP command for other information (like subscribing).
>