Re: re SHUTDOWN ABORT -- was RE: Debate on rc commands Solaris and

Alter system checkpoint... You don't say...

Hey, this is the first time this thread has concluded without the usual "you guys better watch out b/c yer gonna break your database!" post.

I'd say this universal support for ABORT over IMMEDIATE represents a dramatic change in the prevailing DBA attitude over, say, two years ago.

How do you suppose that happened?

:-)

The only dissenter was Dan. Dan, what's the difference between a kernel transaction and a regular transaction? Are you talking about the O/S kernel or Oracle? Can you explain in more detail what the kernel transaction does to make Oracle unrecoverable after ABORT?

I'm still mulling over that 'alter system checkpoint.' Sounds familiar.

--
Jeremiah Wilton
http://www.speakeasy.net/~jwilton

On Sat, 1 Feb 2003, [iso-8859-1] Connor McDonald wrote:



> Agreed. All that matters is the redo logs.  If Oracle

> had named the shutdown options more accurately, that

> is

> 

> shutdown abort => shutdown fast

> shutdown immediate => shutdown hopefully

> shutdown transactional => shutdown when hell freezes

> over

> shutdown normal => shutdown never

> 

> then I'm pretty sure I know what everyone would be

> using...An 'alter system checkpoint' just before the

> abort also helps startup times.

> 

>  --- Hemant K Chitale <hkchital_at_singnet.com.sg> wrote:

> > 

> > Comparing the shutdown scenarios with what would

> > happen if someone

> > were to "pull the plug" on the server is the best

> > way to explain the behaviour

> > of Oracle.

> > What if

> >    a. Someone switched-off the server [there's a

> > power failure and the UPS

> > doesn't kick in] ?

> >    b.  The server crashes with a Unix Panic

> > well, the Oracle instance dies and yet, [as long as

> > the disks are still good,

> > even if you have to do an "fsck"] there's Instance

> > Recovery which goes through

> > successfully.

> > So what's the problem with SHUTDOWN ABORT ?  Its

> > about the same thing,

> > but not as bad.

> > 

> > I consistently use SHUTDOWN ABORT, STARTUP RESTRICT,

> > SHUTDOWN NORMAL

> > on Production, Mission-Critical databases.  I've

> > never had a problem with a

> > SHUTDOWN ABORT.

> > Even a clone of an ABORTed instance can work if the

> > online redo logs and 

> > current

> > control-file are also cloned !

> > 

> > I have seen SHUTDOWN IMMEDIATE hang very many times

> > [particularly when

> > DBMS_JOB jobs are running or the un-intelligent

> > Intelligent Agent is 

> > connected].

> > 

> > At 05:29 PM 31-01-03 -0800, you wrote:

> > >Dan,

> > >

> > >If you look at Rama Velpuri's most famous treatise

> > on B & R - namely

> > >'Oracle8 Backup & Recovery Handbook' (Oracle

> > Press), you will see the path

> > >that the various shutdown options take. [Page 43,

> > Chapter 2: The Oracle

> > >Architecture and Configuration]. I belive this is

> > the closest that we could

> > >come to a published look under the covers. Some one

> > mentioned a switch

> > >logfile prior to the shutdown abort. In any case, a

> > shutdown abort would not

> > >be as bad as someone pulling the plug on the juice

> > (as in Cleaning lady: 'I

> > >need to plug in my hoover... aaah - here's a socket

> > I can use). The instance

> > >goes down not-so-gracefully, but the disk is still

> > safe as long as the

> > >server stays up.

> > >

> > > > -----Original Message-----

> > > > From: Fink, Dan [mailto:Dan.Fink_at_mdx.com]

> > > >

> > > > I know I am going to catch major grief for this,

> > but I avoid

> > > > shutdown abort

> > > > when possible. Shutdown immediate can take some

> > time, as it will do

> > > > transaction rollback in serial (where recovery

> > at startup can do it in

> > > > parallel and as needed). The main reason I avoid

> > it is that

> > > > there are known

> > > > bugs that will cause the database to be

> > unrecoverable when a

> > > > shutdown abort

> > > > is done while a kernel transaction is being

> > performed.

> > > > Granted, I have not

> > > > personally encountered this, and the chances of

> > encountering

> > > > it are slight,

> > > > but why take the risk?

> > > >

> > > > <Ducking and running for cover>

> > > > Dan Fink

> > > >

> > > > -----Original Message-----

> > > > Sent: Thursday, January 30, 2003 6:15 AM

> > > > To: Multiple recipients of list ORACLE-L

> > > >

> > > >

> > > >

> > > > I'll echo that sentiment.

> > > >

> > > > 'shutdown abort', 'startup restrict' was a

> > regular part of my

> > > > shutdown scripts beginning in 1994 with 7.0.16,

> > as

> > > > 'shutdown immediate' wasn't all that reliable,

> > even in situations

> > > > where it should have worked.

> > > >

> > > > Jared

> > > >

> > > > On Wednesday 29 January 2003 15:53, John

> > Kanagaraj wrote:

> > > > > Rao,

> > > > >

> > > > > And where did you read that 'shutdown abort'

> > is not

> > > > recommended? This is

> > > > > another myth that has been busted a while ago.

> > A shutdown

> > > > abort followed

> > > > by

> > > > > a startup restrict and a normal shutdown is

> > the way to go

> > > > when dealing

> > > > with

> > > > > rogue sessions that open a connection and

> > never shutdown.

> > > > In such cases, a

> > > > > shutdown immediate will _never_ return

> > (certainly not

> > > > within your 5 to 10

> > > > > minutes). I have been using this method for

> > more than 8 years now -

> > > > > starting at 7.0.16 fyi. The trick in this case

> > is to script

> > > > it into the rc

> > > > > commands.

> > > > >

> > > > > John Kanagaraj

> > > > > Oracle Applications DBA

> > > > > DBSoft Inc

> > > > > (W): 408-970-7002

> > > > >

> > > > > I don't know what the future holds for me, but

> > I do know

> > > > who holds my

> > > > > future!

> > > > >

> > > > > ** The opinions and statements above are

> > entirely my own

> > > > and not those of

> > > > > my employer or clients **

> > > > >

> > > > >

> > > > > -----Original Message-----

> > > > > Sent: Wednesday, January 29, 2003 11:49 AM

> > > > > To: Multiple recipients of list ORACLE-L

> > > > >

> > > > >

> > > > > Paula,

> > > > >

> > > > > Shutdown abort is not recommended as the file

> > checkpointing

> > > > is not done

> > > > > during shutdown abort.  If you need to perform

> > shutdown

> > > > abort, then, it is

> > > > > preferred to bring up the db with startup

> > restrict (so that

> > > > the users

> > > > > wouldn't connect) and then, cleanly shutdown

> > the db and

> > > > bring it up again.

> > > > >

> > > > > Tell to your sys admins. that shutdown

> > immediate would take

> > > > some time

> > > > > (about 5 to 10 minutes) depending on the

> > activity on your

> > > > db.  They would

> > > > > have to wait for that much time before calling

> > a DBA during

> > > > system boots.

> > > > >

> > > > > Rao

> > > > >

> > > > > -----Original Message-----

> > > > > Sent: Wednesday, January 29, 2003 11:30 AM

> > > > > To: Multiple recipients of list ORACLE-L

> > > > >

> > > > >

> > > > >

> > > > > System Administrator says he doesn't trust

> > that the rc

> > > > commands will stop

> > > > > if the database doesn't want to shutdown and

> > even if it

> > > > does would want to

> > > > > shutdown with scripts beforehand so that a DBA

> > could

> > > > connect and resolve

> > > > > the issue.  Other DBA says this is all wrong

> > and rc commands should

> > > > include

> > > > > shutdown immediate of database.  In the past I

> > had setup 2

> > > > processes in

> > > > the

> > > > > system scripts for the sys admin - shutdown

> > immediate -

> > > > wait .... shutdown

> > > > > abort - on a read-only DSS system which of

> > course allows

> > > > some room for

> > > > this

> > > > > type of activity.  I kind of would want to

> > know if a

> > > > database was going to

> > > > > be shutdown with an abort esp. in OLTP system

> > and do it myself.

> > > > >

> > > > >

> > > > >

> > > > > - any ideas

> > > > --

> > > > Please see the official ORACLE-L FAQ:

> > http://www.orafaq.net

> > > > --

> > > > Author: Jared Still

> > > >   INET: jkstill_at_cybcon.com

> > > >

> > > > Fat City Network Services    -- 858-538-5051

> > http://www.fatcity.com

> > > > San Diego, California        -- Mailing list and

> > web hosting services

> > > >

> >

> ---------------------------------------------------------------------

> > > > 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).

> > > > --

> > > > Please see the official ORACLE-L FAQ:

> > http://www.orafaq.net

> > > > --

> > > > Author: Fink, Dan

> > > >   INET: Dan.Fink_at_mdx.com

> > > >

> > > > Fat City Network Services    -- 858-538-5051

> > http://www.fatcity.com

> > > > San Diego, California        -- Mailing list and

> > web hosting services

> > > >

> >

> ---------------------------------------------------------------------

> > > > 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).

> > > >

> > >--

> > >Please see the official ORACLE-L FAQ:

> > http://www.orafaq.net

> > >--

> > >Author: John Kanagaraj

> > >   INET: john.kanagaraj_at_hds.com

> > >

> > >Fat City Network Services    -- 858-538-5051

> > http://www.fatcity.com

> > >San Diego, California        -- Mailing list and

> > web hosting services

> >

> >---------------------------------------------------------------------

> > >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).

> > 

> > Hemant K Chitale

> > My web site page is :  http://hkchital.tripod.com

> > 

> > 

> > -- 

> > Please see the official ORACLE-L FAQ:

> > http://www.orafaq.net

> > -- 

> > Author: Hemant K Chitale

> >   INET: hkchital_at_singnet.com.sg

> > 

> > Fat City Network Services    -- 858-538-5051

> > http://www.fatcity.com

> > San Diego, California        -- Mailing list and web

> > hosting services

> >

> ---------------------------------------------------------------------

> > 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).

> >  

> 

> =====

> Connor McDonald

> web: http://www.oracledba.co.uk

> web: http://www.oaktable.net

> email: connor_mcdonald_at_yahoo.com

> 

> "GIVE a man a fish and he will eat for a day. But TEACH him how to fish, and...he will sit in a boat and drink beer all day"

> 

> __________________________________________________

> Do You Yahoo!?

> Everything you'll ever need on one web page

> from News and Sport to Email and Music Charts

> http://uk.my.yahoo.com

> -- 

> Please see the official ORACLE-L FAQ: http://www.orafaq.net

> -- 

> Author: =?iso-8859-1?q?Connor=20McDonald?=

>   INET: hamcdc_at_yahoo.co.uk

> 

> Fat City Network Services    -- 858-538-5051 http://www.fatcity.com

> San Diego, California        -- Mailing list and web hosting services

> ---------------------------------------------------------------------

> 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).

> 

> 


-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.net
-- 
Author: Jeremiah Wilton
  INET: jwilton_at_speakeasy.net

Fat City Network Services    -- 858-538-5051 http://www.fatcity.com
San Diego, California        -- Mailing list and web hosting services
---------------------------------------------------------------------
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).