Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Mailing Lists -> Oracle-L -> RE: how to force the DB to SKIP tx recovery ?

RE: how to force the DB to SKIP tx recovery ?

From: Vikas Khanna <vkhanna_at_quark.co.in>
Date: Tue, 23 Jul 2002 02:23:33 -0800
Message-ID: <F001.0049F271.20020723022333@fatcity.com>


Rahul,
Instance recovery requires Rollforward and Rollbackwards. If the transaction is not committed it has to be rollbacked once the instance gets started. Let us assume that there are no committed transactions which were left in the buffer cache to be flushed back on to datafiles. This means I am left with Uncommitted txns to get them rollbackwards. This is done by SMON in the background or Server process which needs those blocks but the Instance and the db are opened and ready for use.

The db is opened ie. what u wanted and that's the way Oracle does work.

-----Original Message-----
Sent: Tuesday, July 23, 2002 2:10 PM
To: 'vkhanna_at_quark.co.in'
Cc: 'Oracle List I'

Vikas, even if the parameters were set for the checkpoint to happen often, all the
transactions *not* commited would still be rolled back !! for example, when i shut down the PC
in the middle of doing an update of 1/2 a million rows !!! (how would these parameters help?)

> ----------
> From: Vikas Khanna[SMTP:vkhanna_at_quark.co.in]
>
> No, Instance recovery has to take place internally, SMON has to recover
> the
> committed transactions which were left hanging in the cache to datafiles
> so
> that the files become synchronized.
>
> After that the db gets opened and roll backwards takes its own time. It
> depends SMON does this job in the background and any other Server process
> if
> needs that block can rollback the uncommitted transaction and proceed with
> the block.
>
> If you wish to recover the instance recovery very fast then set the
> LOG_CHECKPOINT_INTERVAL,LOG_CHECKPOINT_TIMEOUT,FAST_START_IO_TARGET
> parameters to force the checkpoints quite often, this would open the
> instance under crash quite fast as it has to perform less work. But more
> the
> checkpoints initiated more the performance degradations as B'ground
> processes would be under tremendous work.
>
> Hope it clarifies.
> Vikas Khanna
>

-- 
Please see the official ORACLE-L FAQ: http://www.orafaq.com
-- 
Author: Vikas Khanna
  INET: vkhanna_at_quark.co.in

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).
Received on Tue Jul 23 2002 - 05:23:33 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US