Re: Expdp a subset of fields from a table

From: Robert Freeman <>
Date: Mon, 29 Jun 2009 12:54:17 -0700 (PDT)
Message-ID: <>

Don't suppose you are doing RMAN backups? If you are, in most cases it should detect any corruption.


 Robert G. Freeman
Oracle ACE
Oracle Database 11g RMAN Backup and Recovery (Oracle Press) - ON IT'S WAY SOON!
OCP: Oracle Database 11g Administrator Certified Professional Study Guide (Sybex)
Oracle Database 11g New Features (Oracle Press)
Portable DBA: Oracle  (Oracle Press)
Oracle Database 10g New Features (Oracle Press)
Oracle9i RMAN Backup and Recovery (Oracle Press)
Oracle9i New Features (Oracle Press)
Other various titles out of print now...
The LDS Church is looking for DBA's. You do have to be a Church member in
good standing. A lot of kind people write me, concerned I may be breaking
the law by saying you have to be a Church member. It's legal I promise! :-)

From: Rich <>
To: Michael McMullen <>
Cc: Oracle-L Freelists <>
Sent: Monday, June 29, 2009 1:04:59 PM
Subject: Re: Expdp a subset of fields from a table

Hi William:
"Your best option is to create a view and
export the view."
Hey, now that's creative.  However, Oracle throws:
ORA-39166: Object TEST_EXPDP_HIST_T2 was not found.
ORA-31655: no data or metadata objects selected for job

Michael et al,
This table is only insert - no delete, no update.
We also control when the inserts happen; the table is a sort of quasi partition using a view and more inexpensive disks.  It's just part of our ILM.
We have looked at delayed block cleanout and are working with Oracle to try prove whether that is or is not the issue.

We have tried setting the event (1555 at errorstack level 3:10200), however, we don't get any trace from doing so - Oracle on Linux x86_64.
I did trace the session and it does show increasing values of: 
cleanouts only - consistent read gets        546440
cleanout - number of ktugct calls            546440
in v$sessstat during the export which confirms they are happening.

We have tried analyze and it completed, however, the error (1555) persists.
Using Oracle Notes 452341.1 and 787004.1 (faster) we have found no corruption in the LOB area.
I think this is why Oracle is pressing for potential corruption in the table which is why I would like to try an expdp without the LOB.

Thanks for your help,

On Mon, Jun 29, 2009 at 11:32 AM, Michael McMullen <> wrote:

I’ve always thought that performance
and ora-1555 go hand in hand, slow query on changing data, odds are you see
ora-1555 but I also get these on my lob tables even when there have been no
data changes for hours. Look at delayed block cleanout. You could always trace
the session or set an event on ora-1555 to make sure. It could even be an
underlying index causing the errors. Most of our data is bulk loaded so we
usually do a combination of undo_retention,fts on the data, analyze all data
and analyze all indexes to avoid the error. Delayed block cleanouts are
supposed to be rare but I run into it all the time.

________________________________ [] On Behalf Of Rich
Sent: Monday, June 29, 2009 1:33

To: Blanchard, William
Cc: Oracle-L Freelists
Subject: Re: Expdp a subset of
fields from a table 
Hi William,
I tried that:
$ORACLE_HOME/bin/expdp `/home/oracle/dbserver_dba` TRACE=480300
logfile=HSR_TRANSLATION_HISTORY_TIER2_NO_LOB_`date +%d%b%Y`_expdp.log

I get ORA-00933: SQL command not properly ended

Hi Michael,
It's not a performance issue, 1555 issue.
We checked for corruption in the LOB area to no avail and Oracle is saying they
think it might be the table segment, not the LOB segment.
I asked in the SR, however, thought I'd ask here, also.

On Mon, Jun 29, 2009 at 10:10 AM, Blanchard, William <>
Just include a select in your expdp command/parfile.
QUERY = <schema>.<table>: SELECT …

Received on Mon Jun 29 2009 - 14:54:17 CDT

Original text of this message