RE: RMAN testing

  • From: "Sweetser, Joe" <JSweetser@xxxxxxxx>
  • To: <veeeraman@xxxxxxxxx>, "Andy Rivenes" <arivenes@xxxxxxxx>
  • Date: Thu, 14 Feb 2008 11:30:34 -0700

One workaround might be to use a link, if that's feasible in your
environment - though I would be careful if we're talking about mucking
about in OH.
 
-joe

________________________________

From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Ram Raman
Sent: Thursday, February 14, 2008 11:13 AM
To: Andy Rivenes
Cc: oracle-l
Subject: Re: RMAN testing


 
The problem is that the first run of the RMAN happens successfully and
creates backup files in "/psoft/u02/...", but the second run looks for
file in $OH/dbs (/psoft/u01). It is trying to recover files in OH/dbs.
In the beginning (yesterday when I started testing) /psoft/u01 was where
RMAN was doing the backup. I delete the old backups several times and
rerun the RMAN. Somehow RMAN seems to want to restore in the original
location of OH/dbs only in the second run. 
 
Is there any command to tell RMAN not to look in OH/dbs while doing the
restore in the second run? I run crosscheck after deleting backups. 
 
 
RMAN *second* run looking for files in OH/dbs (/psoft/u01...) :
________________________________________________________________________
___________________________
 
channel oem_disk_backup: starting incremental datafile backupset restore
channel oem_disk_backup: specifying datafile copies to recover
recovering datafile copy fno=00001
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-SYSTEM_FNO-1_0aj8k74m
recovering datafile copy fno=00002
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-SYSTEM_FNO-2_0ej8k83a
recovering datafile copy fno=00003
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-SYSTEM_FNO-3_0fj8k84d
recovering datafile copy fno=00004
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-SYSTEM_FNO-4_02j8k597
recovering datafile copy fno=00005
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-UNDOTBS1_FNO-5_0gj8k8
5g
recovering datafile copy fno=00006
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-SYSAUX_FNO-6_03j8k5bj
recovering datafile copy fno=00007
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-TSBIG_FNO-7_0sj8kcsh
recovering datafile copy fno=00008
name=/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691
_TS-TSBIG_FNO-8_16j8kt9o
channel oem_disk_backup: reading from backup piece
/psoft/u02/oracle10g/cds/ATES/bkup/1jj8mftv_1_1
released channel: oem_disk_backup
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of recover command at 02/14/2008 11:51:08
ORA-19870: error reading backup piece
/psoft/u02/oracle10g/cds/ATES/bkup/1jj8mftv_1_1
ORA-19625: error identifying file
/psoft/u01/oracle10/app/product/10.2.0/dbs/data_D-ATES_I-2542000691_TS-S
YSTEM_FNO-4_02j8k597
ORA-27037: unable to obtain file status
HP-UX Error: 2: No such file or directory
Additional information: 3
________________________________________________________________________
________________________
   
   
My run part of the RMAN script now looks like this:
_____________________________________________________________________
run {
allocate channel oem_disk_backup device type disk;
configure CHANNEL device type 'SBT_TAPE' clear;
sql 'alter system switch logfile';
CONFIGURE CHANNEL DEVICE TYPE DISK CLEAR;
configure channel 1 device type disk clear;
configure CHANNEL device type disk format
'/psoft/u02/oracle10g/cds/ATES/bkup/%U';
recover copy of database with tag 'incr_ates';
CONFIGURE CHANNEL DEVICE TYPE DISK CLEAR;
configure channel 1 device type disk clear;
backup as compressed backupset
     incremental level 1 cumulative  copies=1 tag 'incr_ates'
     for recover of copy with tag 'incr_ates'
     format '/psoft/u02/oracle10g/cds/ATES/bkup/%U'
     database plus archivelog delete input;
sql 'alter system switch logfile';
    }
________________________________________________________________________
_____________
   
   
 


 
On 2/14/08, Andy Rivenes <arivenes@xxxxxxxx> wrote: 

        It might have to do with your saved configuration. Try a "show
all"
        to see what is set, and then you can run a "CLEAR", like you did
for
        the SBT_TAPE setting, for the setting(s) that you want to reset.
        
        
        At 08:03 PM 2/13/2008, Ram Raman wrote:
        >Hi all,
        >
        >I am trying to simulate what is in production in dev box and
test
        >out the existing setup. We have Oracle
<http://10.2.0.2>10.2.0.2.
        >ATES is a small database that I created in dev. Here is the
script
        >that I take backup with:
        >
        
>______________________________________________________________________
        >
        >/psoft/u02/oracle10g/cds/ATES $ cat backup_ATES.sh
        >export ORACLE_HOME=/psoft/u01/oracle10/app/product/10.2.0
        >export ORACLE_SID=ATES
        >export LD_LIBRARY_PATH=$ORACLE_HOME/lib
        >
        >$ORACLE_HOME/bin/rman <<EOF
        >
        >#connect catalog <mailto:rman/xxxx@rmanprd>rman/xxxx@rmanprd;
        >connect target;
        >
        >run {
        >allocate channel oem_disk_backup device type disk;
        >configure CHANNEL device type 'SBT_TAPE' clear;
        >sql 'alter system switch logfile';
        >configure CHANNEL device type disk format
        >'/psoft/u02/oracle10g/cds/ATES/bkup/%U';
        >recover copy of database with tag 'incr_ates';
        >backup as compressed backupset
        >      incremental level 1 cumulative  copies=1 tag 'incr_ates'
        >      for recover of copy with tag 'incr_ates'
        >      format '/psoft/u02/oracle10g/cds/ATES/bkup/%U'
        >      database plus archivelog delete input;
        >sql 'alter system switch logfile';
        >     }
        >exit
        >EOF
        
        
        

Other related posts: