SEO

In Oracle 8.1.7.4 (on Solaris 8), archived redo log files were the
same size as the original redo log file (i.e., archiving a 10 MB redo
log file resulted in a 10 MB archived redo log file in the archive
destination directories). In Oracle 10g release 2 (10.2.0.1.0 on
Solaris 9), the archived redo log files are all significantly smaller
than the originals. With 10 MB redo logs, the largest archived redo
log file is about 8.8 MB and most of them are more like 7.5 MB (out of
150 total archived redo logs).

We're archiving to 3 separate destinations and the archived redo
logs are identical to each other in all 3 places (but different
from the originals). I compared a few originals with their
archived copies and other than a few bytes that differ
in the first 1 KB, the files are identical until the archived
copy ends (1-2 MB before the original file ends). This indicates
to me that the reason for the size difference is NOT due to
compression of the originals.

Is there some change to the way redo log archiving works between 8i
and 10g? Could there be some reason why log file switches would be
happening well before the end of the redo log file?

I have a TAR open with Oracle about this, but haven't gotten any
useful information so far.

Dave
--

David Hollenberg dhollen@NOmosisSPAM.org http://www.mosis.org/
MOSIS Service Voice 310-448-8704
Information Sciences Institute FAX 310-823-5624

I think there are some parameters backup & recover related

Oracle force a checkpoint at it's calculated time, before fill up the
current online log.

You can check these parameter

AnySQL from China
http://www.anysql.net/en/
http://www.anysql.net/en/mydmp.php

David Hollenberg wrote:
> In Oracle 8.1.7.4 (on Solaris 8), archived redo log files were the
> same size as the original redo log file (i.e., archiving a 10 MB redo
> log file resulted in a 10 MB archived redo log file in the archive
> destination directories). In Oracle 10g release 2 (10.2.0.1.0 on
> Solaris 9), the archived redo log files are all significantly smaller
> than the originals. With 10 MB redo logs, the largest archived redo
> log file is about 8.8 MB and most of them are more like 7.5 MB (out of
> 150 total archived redo logs).
>
> We're archiving to 3 separate destinations and the archived redo
> logs are identical to each other in all 3 places (but different
> from the originals). I compared a few originals with their
> archived copies and other than a few bytes that differ
> in the first 1 KB, the files are identical until the archived
> copy ends (1-2 MB before the original file ends). This indicates
> to me that the reason for the size difference is NOT due to
> compression of the originals.
>
> Is there some change to the way redo log archiving works between 8i
> and 10g? Could there be some reason why log file switches would be
> happening well before the end of the redo log file?

redo has been redone in 10g (I think R1). Its possible that archiving
might have changed as well. That said the most common reason for logs
switching before they are full is because the log switch is scheduled
(or coded in the app).

--
Niall Litchfield
Oracle DBA
http://www.niall.litchfield.dial.pipex.com