Re: [Q] what differtent between logical standby database and physical standby DB?
- From: Tanel Põder <tanel.poder.003@xxxxxxx>
- To: <oracle-l@xxxxxxxxxxxxx>
- Date: Fri, 9 Apr 2004 18:45:16 +0300
Hi!
By construction I mean, that there is much more in a redolog than just data
block numbers, byte offsets and changes to the blocks - meaning that Oracle
still has to read and interpret a lot of supporting information for
constructing the block level changes for the physical recovery, so in case
of a major corruption (e.g. the whole redo block is blanked), Oracle will
likely detect it (by giving an ORA-600 for example). On the other hand,
minor corruptions (e.g. one bit flipped) could get pass through "logical"
recovery equally as well, since the SQL statements applied to database are
constructed partially from the same redo vectors anyway.
By parsing I meant that the actual changes to datablocks have to be read out
from complex redo structures like redo blocks -> redo records -> redo
vectors -> various information such is opcode, absolute file#, RDBA, SCN,
SEQ etc.. before we actually get to the real changed data itself..
Basically we are talking about the same thing, but from slightly different
perspective.
AFAIK, KCR is the component taking care of redo construction and KCV is used
for recovery. KCB is just taking care of the general buffer management like
buffer hashes and LRU maintenance.
Tanel.
----- Original Message -----
From: "Riyaj Shamsudeen" <rshamsud@xxxxxxxxxxxx>
To: <oracle-l@xxxxxxxxxxxxx>
Sent: Thursday, April 08, 2004 8:15 PM
Subject: RE: [Q] what differtent between logical standby database and
physical standby DB?
> Tanel
> Guess, I am not understanding the word "construction" applied towards a
> physical standby. Change vectors describe the changes at a lowest
> granular level and KCB layer( I think..) has to apply the change vectors
> directly to the data block buffer (of course, after reading the buffer
> if not present in the cache etc..etc). I don't understand what need to
> be constructed for the physical standby application. Are you refering to
> redo block checksum verification or something along those lines ? You
> also said parsing has to be done for the change vectors. Can you please
> help me understand that too ?
> I agree completely that lot more need to be done to reconstruct the SQL
> in the case of logical standby.
>
> Thanks a lot, Tanel!
>
> Riyaj "Re-yas" Shamsudeen
> Certified Oracle DBA
>
>
> -----Original Message-----
> From: oracle-l-bounce@xxxxxxxxxxxxx
> [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Tanel Põder
> Sent: Thursday, April 08, 2004 11:41 AM
> To: oracle-l@xxxxxxxxxxxxx
> Subject: Re: [Q] what differtent between logical standby database and
> physical standby DB?
>
>
> > Um, in the physical standby case, aren't the archivelogs simply =
> > transferred to the standby server in their entirety? To the best of
> > my = knowledge, physical standby doesn't do logmining, right?
>
> It doesn't do logmining, but however, in both cases the changes have to
> be constructed from change vectors (and supplemental information) in
> redo records, the difference is that in logical apply scenario sql
> statements are constructed and passed to higher layer for sql
> processing, but in physical apply just datablocks are read and changes
> are applied to them directly (using kernel cache layer?).
>
> The point I was trying to make was, that in both cases you have to parse
> the redo records to get the changes, that way you'd detect corruptions
> equally in both cases as well... yes SQL apply ignores some of the redo
> vectors in logs but on the other hand, it requires additional,
> supplemental records for its operations...
>
> Tanel.
>
>
> ----------------------------------------------------------------
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> ----------------------------------------------------------------
> To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx put
> 'unsubscribe' in the subject line.
> --
> Archives are at //www.freelists.org/archives/oracle-l/
> FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
> -----------------------------------------------------------------
>
>
> -- Attached file included as plaintext by Ecartis --
> -- Desc: Signature
>
> The information transmitted is intended only for the person or entity to
> which it is addressed and may contain confidential and/or privileged
> material. If the reader of this message is not the intended recipient,
> you are hereby notified that your access is unauthorized, and any review,
> dissemination, distribution or copying of this message including any
> attachments is strictly prohibited. If you are not the intended
> recipient, please contact the sender and delete the material from any
> computer.
>
>
> ----------------------------------------------------------------
> Please see the official ORACLE-L FAQ: http://www.orafaq.com
> ----------------------------------------------------------------
> To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx
> put 'unsubscribe' in the subject line.
> --
> Archives are at //www.freelists.org/archives/oracle-l/
> FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
> -----------------------------------------------------------------
>
----------------------------------------------------------------
Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------
To unsubscribe send email to: oracle-l-request@xxxxxxxxxxxxx
put 'unsubscribe' in the subject line.
--
Archives are at //www.freelists.org/archives/oracle-l/
FAQ is at //www.freelists.org/help/fom-serve/cache/1.html
-----------------------------------------------------------------
Other related posts:
