not able to do RMAN clone

[sivakumarocp]

Hi Tim,

Actually I am trying do RMAN clone from production database to one of our recovery host. at the end of cloning stage it got failed with below error.

Code: Select all

channel ORA_AUX_DISK_2: restored backup piece 1
piece handle=/BACKUPS17/rep_to_b36/orprod42/budump/ORPROD42_20120907_54620_1 tag=COORAPROD42_F_201209070705
channel ORA_AUX_DISK_2: restore complete, elapsed time: 00:24:58
channel ORA_AUX_DISK_1: restored backup piece 1
piece handle=/BACKUPS17/rep_to_b36/orprod42/budump/ORPROD42_20120907_54628_1 tag=COORAPROD42_F_201209070705
channel ORA_AUX_DISK_1: restore complete, elapsed time: 00:16:31
Finished restore at 08-SEP-12
sql statement: CREATE CONTROLFILE REUSE SET DATABASE "ORRECO42" RESETLOGS ARCHIVELOG
  MAXLOGFILES     64
  MAXLOGMEMBERS      2
  MAXDATAFILES      500
  MAXINSTANCES     8
  MAXLOGHISTORY    10517
 LOGFILE
  GROUP  1 ( '/ORA/LOG1/orreco42/log1A.rdo', '/ORA/LOG2/orreco42/log1B.rdo' ) SIZE 50 M  REUSE,
  GROUP  2 ( '/ORA/LOG1/orreco42/log2A.rdo', '/ORA/LOG2/orreco42/log2B.rdo' ) SIZE 50 M  REUSE,
  GROUP  3 ( '/ORA/LOG1/orreco42/log3A.rdo', '/ORA/LOG2/orreco42/log3B.rdo' ) SIZE 50 M  REUSE,
  GROUP  4 ( '/ORA/LOG1/orreco42/log4A.rdo', '/ORA/LOG2/orreco42/log4B.rdo' ) SIZE 50 M  REUSE,
  GROUP  5 ( '/ORA/LOG1/orreco42/log5A.rdo', '/ORA/LOG2/orreco42/log5B.rdo' ) SIZE 50 M  REUSE,
  GROUP  6 ( '/ORA/LOG1/orreco42/log6A.rdo', '/ORA/LOG2/orreco42/log6B.rdo' ) SIZE 50 M  REUSE
 DATAFILE
  '/ORA/DATA_TEMP/orreco42/system01.dbf'
 CHARACTER SET UTF8


contents of Memory Script:
{
   switch clone datafile all;
}
executing Memory Script

released channel: ORA_AUX_DISK_1
released channel: ORA_AUX_DISK_2
released channel: ORA_AUX_DISK_3
released channel: ORA_AUX_DISK_4
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of Duplicate Db command at 09/08/2012 03:43:23
RMAN-03015: error occurred in stored script Memory Script
RMAN-06136: ORACLE error from auxiliary database: ORA-19563:  header validation failed for file

RMAN>


before to do this clone operation I ran Full backup script and it got complete successfully but eventhen cloning operation is getting fail with above error.
I am not sure how to proceed further, the production database size is around 830G.

DB version : 10.2.0.4
O/S : solaris 10

pls let me now how to overcome this error.

[Tim...]

Hi.

RMAN definitely thinks the file has a bogus header. Are you sure it is using the most recent files to perform a duplicate? RMAN can pick any files that are available. It may for instance not use the latest file for a recovery, but pick an earlier file and wind it forward with archived redo logs.

If you can confirm the file being used is from the latest backup, then I think you need to contact Oracle support and get them ti diagnose the issue.

Cheers

Tim...

[sivakumarocp]

Hi Tim,

Thanks for the reply.
I found the above error caused due to identical filename was created in different directory on production database and i was trying to restore all datafiles into single directory on target host thats the reason why it got failed.

to resolve this issue i used SET NEW NAME COMMAND along with RMAN DUPLICATE command but again it got failed with below error.

Code: Select all

cataloged datafile copy
datafile copy filename=/ORA/DATA_TEMP/orreco42/stage_arch03.dbf recid=25 stamp=793461637

input datafile copy recid=17 stamp=793461629 filename=/ORA/DATA_TEMP/orreco42/staged04.dbf
datafile 19 switched to datafile copy
input datafile copy recid=18 stamp=793461630 filename=/ORA/DATA_TEMP/orreco42/undo02.dbf
datafile 20 switched to datafile copy
input datafile copy recid=19 stamp=793461631 filename=/ORA/DATA_TEMP/orreco42/staged05.dbf
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of Duplicate Db command at 09/08/2012 14:01:51
RMAN-03015: error occurred in stored script Memory Script
RMAN-03009: failure of switch command on clone_default channel at 09/08/2012 14:01:51
ORA-19625: error identifying file
ORA-27036: translation error, unable to expand file name
Additional information: 7202
Additional information: 2

RMAN>

I am not able to trace where the issue is! though all datafiles has been restored still it got failed.

I am not sure about these two errors.

ORA-19625: error identifying file
ORA-27036: translation error, unable to expand file name

what does it mean.



Below is the RMAN script which I am using for this duplicate DB.

Code: Select all

RUN
{
  SET NEWNAME FOR DATAFILE 28 TO '/ORA/DATA_TEMP/staged06.dbf';
  DUPLICATE TARGET DATABASE TO orreco42
  PFILE = /oracle/init_reco42.ora
  NOFILENAMECHECK
  LOGFILE GROUP 1 ('/ORA/LOG1/orreco42/log1A.rdo',
                   '/ORA/LOG2/orreco42/log1B.rdo') SIZE 50M REUSE,
          GROUP 2 ('/ORA/LOG1/orreco42/log2A.rdo',
                   '/ORA/LOG2/orreco42/log2B.rdo') SIZE 50M REUSE,
          GROUP 3 ('/ORA/LOG1/orreco42/log3A.rdo',
                   '/ORA/LOG2/orreco42/log3B.rdo') SIZE 50M REUSE,
          GROUP 4 ('/ORA/LOG1/orreco42/log4A.rdo',
                   '/ORA/LOG2/orreco42/log4B.rdo') SIZE 50M REUSE,
          GROUP 5 ('/ORA/LOG1/orreco42/log5A.rdo',
                   '/ORA/LOG2/orreco42/log5B.rdo') SIZE 50M REUSE,
          GROUP 6 ('/ORA/LOG1/orreco42/log6A.rdo',
                   '/ORA/LOG2/orreco42/log6B.rdo') SIZE 50M REUSE;
}


I am storing staged06.dbf file under these two directories /ORA/DATA_TEMP/ and /ORA/DATA_TEMP/orreco42 in target host.


I am bale to see below entries in alert log.

Code: Select all

Sun Sep  9 10:34:07 2012
CREATE CONTROLFILE REUSE SET DATABASE "ORRECO42" RESETLOGS ARCHIVELOG
  MAXLOGFILES     64
  MAXLOGMEMBERS      2
  MAXDATAFILES      500
  MAXINSTANCES     8
  MAXLOGHISTORY    10517
 LOGFILE
  GROUP  1 ( '/ORA/LOG1/orreco42/log1A.rdo', '/ORA/LOG2/orreco42/log1B.rdo' ) SIZE 50 M  REUSE,
  GROUP  2 ( '/ORA/LOG1/orreco42/log2A.rdo', '/ORA/LOG2/orreco42/log2B.rdo' ) SIZE 50 M  REUSE,
  GROUP  3 ( '/ORA/LOG1/orreco42/log3A.rdo', '/ORA/LOG2/orreco42/log3B.rdo' ) SIZE 50 M  REUSE,
  GROUP  4 ( '/ORA/LOG1/orreco42/log4A.rdo', '/ORA/LOG2/orreco42/log4B.rdo' ) SIZE 50 M  REUSE,
  GROUP  5 ( '/ORA/LOG1/orreco42/log5A.rdo', '/ORA/LOG2/orreco42/log5B.rdo' ) SIZE 50 M  REUSE,
  GROUP  6 ( '/ORA/LOG1/orreco42/log6A.rdo', '/ORA/LOG2/orreco42/log6B.rdo' ) SIZE 50 M  REUSE
 DATAFILE
  '/ORA/DATA_TEMP/orreco42/system01.dbf'
 CHARACTER SET UTF8
Sun Sep  9 10:34:07 2012
WARNING: Default Temporary Tablespace not specified in CREATE DATABASE command
Default Temporary Tablespace will be necessary for a locally managed database in future release
Setting recovery target incarnation to 1
Sun Sep  9 10:34:07 2012
Successful mount of redo thread 1, with mount id 4078318879
Sun Sep  9 10:34:07 2012
Completed: CREATE CONTROLFILE REUSE SET DATABASE "ORRECO42" RESETLOGS ARCHIVELOG
  MAXLOGFILES     64
  MAXLOGMEMBERS      2
  MAXDATAFILES      500
  MAXINSTANCES     8
  MAXLOGHISTORY    10517
 LOGFILE
  GROUP  1 ( '/ORA/LOG1/orreco42/log1A.rdo', '/ORA/LOG2/orreco42/log1B.rdo' ) SIZE 50 M  REUSE,
  GROUP  2 ( '/ORA/LOG1/orreco42/log2A.rdo', '/ORA/LOG2/orreco42/log2B.rdo' ) SIZE 50 M  REUSE,
  GROUP  3 ( '/ORA/LOG1/orreco42/log3A.rdo', '/ORA/LOG2/orreco42/log3B.rdo' ) SIZE 50 M  REUSE,
  GROUP  4 ( '/ORA/LOG1/orreco42/log4A.rdo', '/ORA/LOG2/orreco42/log4B.rdo' ) SIZE 50 M  REUSE,
  GROUP  5 ( '/ORA/LOG1/orreco42/log5A.rdo', '/ORA/LOG2/orreco42/log5B.rdo' ) SIZE 50 M  REUSE,
  GROUP  6 ( '/ORA/LOG1/orreco42/log6A.rdo', '/ORA/LOG2/orreco42/log6B.rdo' ) SIZE 50 M  REUSE
 DATAFILE
  '/ORA/DATA_TEMP/orreco42/system01.dbf'
 CHARACTER SET UTF8
Sun Sep  9 10:36:01 2012
Switch of datafile 28 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 2 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 3 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 4 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 5 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 6 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 7 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 8 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 9 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 10 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 11 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 12 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 13 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 14 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 15 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 16 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 17 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 18 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 19 complete to datafile copy
  checkpoint is 9329582429334
Switch of datafile 20 complete to datafile copy
  checkpoint is 9329582429334
oracle:orreco42@bosora01 /oracle/admin/orreco42/bdump >


while restoring file id 21 the problem occurs, but i compared file size with source and target it looks like same to me.

Code: Select all


On Source DB:
============
  FILE_ID FILE_NAME                                  Size_GB
--------- --------------------------------------- ----------
       20 /ORA/DATA02/orprd42/staged05.dbf         6.4453125
       21 /ORA/DATA02/orprd42/staged_2.dbf                 8
       22 /ORA/DATA01/orprd42/stage_arch02.dbf     .93359375
       23 /ORA/DATA02/orprd42/staged_3.dbf         2.9296875

On Target DB:
===========
oracle:orreco42@bsora01 /ORA/DATA_TEMP/orreco42 > du -sh staged05.dbf staged_2.dbf stage_arch02.dbf staged_3.dbf
 6.4G   staged05.dbf
 8.0G   staged_2.dbf
 956M   stage_arch02.dbf
 2.9G   staged_3.dbf

staged_2.dbf has 8G in both source and target. eventhen why its getting fail!
i posted all the info which i know, please let me know if you need any more info.

[Tim...]

Hi.

The short answer is I don't know. I think you need to raise an SR with Oracle support. Maybe it's a bug?

Cheers

Tim...

[sivakumarocp]

I opened ticked with Oracle, lets see what they are saying.

[Tim...]

[sivakumarocp]

Hi Tim,

Got solution from Oracle for my RMAN duplicate issue. Oracle given below suggestion to overcome this issue.

"
the bug is caused by datafiles using a underscore in their names and are of the same length (the filename has the same number of characters).

In your case, the following 'set newname ...' commands are being used:

set newname for datafile 16 to
"/ORA/DATA_TEMP/orreco42/staged02.dbf";

....

set newname for datafile 21 to
"/ORA/DATA_TEMP/orreco42/staged_2.dbf";

Here you can see that both filenames are the same in size (number of characters), the only difference is the underscore being used for file 21 instead of the 0 in file 16.

The workaround is to explicitly use the 'SET NEWNAME FOR DATAFILE ...' for all datafiles that have another datafile with the same name except for the underscore as explained above.
For those datafiles, you either replace the underscore by another character or add a character to the filename so the number of characters are different. "


This is because of the "_" in datafile name convention. while doing rman clone I stored datafiles whichever having underscore "_" into separate directory then duplicate went successfully

[Tim...]

OK. Good to know.

Cheers

Tim...