Hi ALL,
11gR2
OL6
We have the following RMAN backup scripts
1. Level 0 > run on Sundays
2. Incremental > run on Mondays to Saturdays
======================
Level 0 Backup Script:
======================
# Identify the backup location
export BACKUP_DIR=/oracle/RMAN_Backup
# Source the environment file, if any
. /oracle/PROD/db/tech_st/11.2.0/PROD_ebizdb.env
# Clean out older backups. Delete all backups older than 10 days
#find $BACKUP_DIR/* -mtime +10 ! -wholename $BACKUP_DIR/ -delete
# Clean out older backups. Delete all backups older than 10 days
find $BACKUP_DIR/* -mtime +7 ! -wholename $BACKUP_DIR/ -delete
# Create Directory for the new day's backup
date
mkdir $BACKUP_DIR/`date +%F`
export BACKUP_DIR=$BACKUP_DIR/`date +%F`
export LOGFILE=${BACKUP_DIR}/LogRMAN
# Run the RMAN backup script and spool the output
rman target / nocatalog log=$LOGFILE << EOF
run {
allocate channel ch1 type disk format='$BACKUP_DIR/FPPC_PROD_full_%d_%t_s%s_s%p_CtlFile.bkp';
set limit channel ch1 kbytes=2000000;
backup current controlfile;
release channel ch1;
allocate channel ch1 type disk format='$BACKUP_DIR/FPPC_PROD_full_%t_s%s_s%p_DB.bkp';
allocate channel ch2 type disk format='$BACKUP_DIR/FPPC_PROD_full_%t_s%s_s%p_DB.bkp';
allocate channel ch3 type disk format='$BACKUP_DIR/FPPC_PROD_full_%t_s%s_s%p_DB.bkp';
allocate channel ch4 type disk format='$BACKUP_DIR/FPPC_PROD_full_%t_s%s_s%p_DB.bkp';
set limit channel ch1 kbytes=2000000;
set limit channel ch2 kbytes=2000000;
set limit channel ch3 kbytes=2000000;
set limit channel ch4 kbytes=2000000;
backup full filesperset 4 (database include current controlfile);
release channel ch1;
release channel ch2;
release channel ch3;
release channel ch4;
allocate channel ch1 type disk format='$BACKUP_DIR/FPPC_PROD_full_%d_%t_s%s_s%p_ArcLog.bkp';
allocate channel ch2 type disk format='$BACKUP_DIR/FPPC_PROD_full_%d_%t_s%s_s%p_ArcLog.bkp';
set limit channel ch1 kbytes=4000000;
set limit channel ch2 kbytes=4000000;
backup archivelog all filesperset 2;
release channel ch1;
release channel ch2;
crosscheck archivelog all;
delete noprompt archivelog until time 'sysdate-1';
}
exit;
EOF
# Compress the file
echo "Full RMAN Backup Completed Successfully ..."
echo "Backup successful: " `date` >> $LOGFILE
# Compress the file
echo "Now zipping files ... " `date`
echo "Zipping Files : " `date` >> $LOGFILE
# gzip -r $BACKUP_DIR
# echo "Zipping Files completed: " `date` >> $LOGFILE
# echo "Zipping files completed ..."
# echo ===================================
==========================================
INCREMENTAL BACKUP
==========================================
# Identify the backup location
export BACKUP_DIR=/oracle/RMAN_Backup
# Source the environment file, if any
. /oracle/PROD/db/tech_st/11.2.0/PROD_ebizdb.env
# Clean out older backups. Delete all backups older than 10 days
#find $BACKUP_DIR/* -mtime +10 ! -wholename BACKUP_DIR -delete
# Clean out older backups. Delete all backups older than 7 days
find $BACKUP_DIR/* -mtime +7 ! -wholename BACKUP_DIR -delete
# Create Directory for the new day's backup
date
mkdir $BACKUP_DIR/`date +%F`
export BACKUP_DIR=$BACKUP_DIR/`date +%F`
export LOGFILE=${BACKUP_DIR}/LogRMAN
# Run the RMAN backup script and spool the output
rman target / nocatalog log=$LOGFILE << EOF
run {
allocate channel ch1 type disk format='$BACKUP_DIR/FPPC_PROD_incr_%d_%t_s%s_s%p_CtlFile.bkp';
set limit channel ch1 kbytes=2000000;
backup current controlfile;
release channel ch1;
allocate channel ch1 type disk format='$BACKUP_DIR/FPPC_PROD_incr_%d_%t_s%s_s%p_ArcLog.bkp';
allocate channel ch2 type disk format='$BACKUP_DIR/FPPC_PROD_incr_%d_%t_s%s_s%p_ArcLog.bkp';
set limit channel ch1 kbytes=2000000;
set limit channel ch2 kbytes=2000000;
backup archivelog all filesperset 2;
release channel ch1;
release channel ch2;
crosscheck archivelog all;
delete noprompt archivelog until time 'sysdate-1';
}
exit;
EOF
# Compress the file
echo "Incremental RMAN Backup Completed Successfully ..."
echo "Backup successful: " `date` >> $LOGFILE
# Compress the file
echo "Now zipping files ... " `date`
echo "Zipping Files : " `date` >> $LOGFILE
# gzip -r $BACKUP_DIR
# echo "Zipping Files completed: " `date` >> $LOGFILE
# echo "Zipping files completed ..."
# echo ===================================
Then I run the Restore Command:
1. $ORACLE_HOME/bin/rman target /
2. startup nomount pfile='/oracle/PROD/db/tech_st/11.2.0/dbs/initPROD.ora';
3. restore controlfile from '/oracle/RMAN_Backup/2017-01-05/FPPC_PROD_full_PROD_932489637_s31_s1_CtlFile.bkp';
4. alter database mount;
5. catalog start with '/oracle/RMAN_Backup';
6. crosscheck backup;
7. delete noprompt expired backup;
8. restore database;
But I got error in #8
Error validating file dummy (399) in piece /oracle/RMAN_Backup/2017-04-26_full/FPPC_PROD_full_942323108_s1485_s2_DB.bkp: missing header
Error validating file dummy (4) in piece /oracle/RMAN_Backup/2017-04-26_full/FPPC_PROD_full_942323108_s1485_s2_DB.bkp: missing header
Error validating file dummy (352) in piece /oracle/RMAN_Backup/2017-04-26_full/FPPC_PROD_full_942323108_s1485_s2_DB.bkp: missing header
Error validating file dummy (3) in piece /oracle/RMAN_Backup/2017-04-26_full/FPPC_PROD_full_942323108_s1486_s2_DB.bkp: missing header
Error validating file dummy (5) in piece /oracle/RMAN_Backup/2017-04-26_full/FPPC_PROD_full_942323108_s1486_s2_DB.bkp: missing header
Exception [type: SIGSEGV, Address not mapped to object] [ADDR:0x7F684AF82B84] [PC:0x7EAD87E, kgcdm()+4604] [flags: 0x0, count: 1]
Errors in file /oracle/PROD/db/tech_st/11.2.0/admin/PROD_ebizdb/diag/rdbms/prod/PROD/trace/PROD_ora_3842.trc (incident=915678):
ORA-07445: exception encountered: core dump [kgcdm()+4604] [SIGSEGV] [ADDR:0x7F684AF82B84] [PC:0x7EAD87E] [Address not mapped to object] []
Incident details in: /oracle/PROD/db/tech_st/11.2.0/admin/PROD_ebizdb/diag/rdbms/prod/PROD/incident/incdir_915678/PROD_ora_3842_i915678.trc
Use ADRCI or Support Workbench to package the incident.
See Note 411.1 at My Oracle Support for error and packaging details.
Wed Apr 26 15:44:53 2017
Dumping diagnostic data in directory=[cdmp_20170426154453], requested by (instance=1, osid=3842), summary=[incident=915678].
Wed Apr 26 15:44:56 2017
=====================================

Please help....
Kind regards,
jc