Preinstallation
- Create CRQ for the upgrade
- Update the Os limits
- Copy the required files to one node of the RAC
- Copy PSU and OPatch to all nodes of the RAC
- Unzip the database files
- Create RDBMS response file
- Deploy software
- Copy init.ora and tnsnames.ora to new RDBMS Home
- Upgrade OPatch on each node
- Create ocm response file
- Update config.c with ASM group
- Apply latest PSU on each node
- Update Goldengate Library
- Verify Installation
- Verify a good backup
- Pre-Upgrade check
Upgrade the database
- Turn off cluster database
- Run Pre-Upgrade SQL
- Disable Golden Gate System Trigger
- Shutdown Database
- Update /etc/oratab
- Startup Upgrade
- Upgrade the database
- Check the upgrade for errors
- Run Post-Upgrade SQL
- Enable GoldenGate System Trigger
- Apply PSU to Database and recompile
- Reset Cluster and verify
- Verify correct dump destination
- Shutdown Database
- Update OCR and start
- Compile system triggers
- Verify Oratab file
Update DST
Close CRQ and Release Database
*************************************************************************************************************
Preinstallation
Create CRQ for the Upgrade
Update OS Limits
Calculate the hard and soft limits. These values are a function of memory. The hard limit is total memory and soft limit is 90% of total memory.
As any user
# Get the memlock hard limit(Hardvalue)
grep MemTotal: /proc/meminfo | awk '{print $2}'
# Get the memlock soft limit(Hardvalue)
echo $(( `grep MemTotal: /proc/meminfo | awk '{print $2}'` * 9 / 10 ))
Add the following lines to the /etc/security/limits.conf file
As root user:
#Use the SoftValue and HardValue from above
oracle soft memlock <SoftValue>
oracle hard memlock <HardValue>
grid soft memlock <SoftValue>
grid hard memlock <HardValue>
Copy the required files to one node of the RAC
These instructions assume the following files are copied to /oraworkspace/software location
RDBMS Software Files
- p17694377_121020_Linux_x86-64_1of8.zip(rdbms #1)
- p17694377_121020_Linux_x86-64_2of8.zip(rdbms #2)
Latest PSU
p20299023_121020_Linux-x86-64.zip (PSU 3-copy to all nodes)
Latest OPatch
p6880880_121010_Lunix-x86-64.zip (OPatch -copy to all nodes)
Copy PSU and OPatch to all nodes of the RAC
As Oracle
cd /oraworkspace/software
scp -p p20299023_121020_Linux-x86-64.zip nodeN:/oraworkspace/software
scp -p p6880880_121010_Lunix-x86-64.zip nodeN:/oraworkspace/software
Unzip the database files
As Oracle:
cd /oraworkspace/software
unzip p17694377_121020_Linux_x86-64_1of8.zip
unzip p17694377_121020_Linux_x86-64_2of8.zip
Create RDBMS response files
As Oracle:
cd /oraworkspace/software
grep -i ^[A-Z] database/response/db_install.rsp
Modify the following variables as per RAC environment you are patching
- DECLINE_SECURITY_UPDATES=true
- INVENTORY_LOCATION=/u01/app/oraInventory
- ORACLE_BASE=/u01/app/oracle
- ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
- oracle.install.db.BACKUPDBA_GROUP=oinstall
- oracle.install.db.CLUSTER_NODES=node1,node2
- oracle.install.db.DBA_GROUP=dba
- oracle.install.db.DGDBA_GROUP=oinstall
- oracle.install.db.EEOptionsSelection=true
- oracle.install.db.InstallEdition=EE
- oracle.install.db.KMDBA_GROUP=oinstall
- oracle.install.db.OPER_GROUP=oper
- oracle.install.option=INSTALL_DB_SWONLY
- UNIX_GROUP_NAME=oinstall
Deploy Software
As Oracle
cd /oraworkspace/software/database
./runInstaller -silent -responseFile /oraworkspace/software/dbInstall.rsp
Copy init.ora and tnsnames.ora to new RDBMS Home
The following is done on each node
As Oracle
For each node:
. oraenv
cp - p $ORACLE_HOME/network/admin/tnsnames.ora /u01/app/oracle/product/12.1.0.2/PSU3/network/admin
For Each Database:
.oraenv
cp - p $ORACLE_HOME/dbs/init${ORACLE_SID}.ora ${ORACLE_HOME}/dbs/orapw${ORACLE_SID} \ /u01/app/oracle/product/12.1.0.2/PSU3/dbs
done
Upgrade OPatch on each node
Run the following on each node
As Oracle
cd /u01/app/oracle/product/12.1.0.2/PSU3
unzip -o /oraworkspace/software/p6880880_121010_Linux-x86-65.zip
Create OCM response file:
Check for /var/tmp/ocm.rsp if doesn't exist
As Oracle
The following command has 2 Prompts
1. Email address/User Name: Leave Blank
2. Do you wish to remain uninformed ofsecurity issues(Yes,No)N:Answer 'Y"
export ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
${ORACLE_HOME}/OPatch/ocm/bin/emocmrsp -no_banner -output /var/tmp/ocm.rsp
Example Output:
${ORACLE_HOME}/OPatch/ocm/bin/emocmrsp -no_banner -output /var/tmp/ocm.rsp
Provide your email address to be informed of security issues, install and initiate oracle configuration Manager. Easier for you if you use your My Oracle Support Email address/User Name.
Visit http://www.oracle.com/support/policies.html for details
Email address/Username:
You have not provided an email address for notification of security issues.
Do you Wish to remain uninformed of security issues([Y]es,[N]o [N]:Y
The ocm configuration response file (/var/tmp/ocm.rsp)was successfully created.
Update config.c with ASM group
Correct the value of ASM group in the $ORACLE_HOME/rdbms/lib/config.c file on each node
This will result in ASM group set after the PSU relink in the following steps
As Oracle
export ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
sed -t 's/SS_ASM_GRP ""/SS_ASM_GRP "asmadmin"/' ${ORACLE_HOME}/rdbms/lib/config.c
Apply latest PSU on each Node
As Oracle
. If the grid patch will be used
cd /oraworkspace/software
export ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
${ORACLE_HOME}/OPatch/opatch apply /orworkspace/software/gpsu/20485724/20299023 -oh ${ORACLE_HOME} -ocmrf /var/tmp/ocm.rsp -silent
. Ifthe RDBMS patch will be used
cd /oraworkspace/software
mkdir psu
cd psu
unzip ../p20299023_121020_Linux-x86-64.zip
export ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
${ORACLE_HOME}/OPatch/opatch apply /orworkspace/software/psu/20299023 -oh ${ORACLE_HOME} -ocmrf /var/tmp/ocm.rsp -silent
Verify the output mentions the PSU to all RAC nodes
Example:
All nodes have been patched. You may start oracle instances on the local system and nodes
Composite patch 20299023 successfully applied .
OPatch session completed with warnings.
log file location: /u01/app/oracle/product/12.1.0.2/PSU3/cfgtoollogs/opatch/opaatch2019-12-30_18-56-12PM_1.log
OPatch completed with warnings.
Update Goldengate Library
As Oracle
. oraenv
ln -s ${ORACLE_HOME}/lib/libnnz12.so ${ORACLE_HOME}/lib/libnnz11.so
Verify Installation
As Oracle:
Verify software installed on all nodes
for node in ${nodes} ; do
ssh -q ${node} du -sk /u01/app/oracle/product/12.1.0.2/PSU3
done
Should looks something like the following . Verify the directory size is similar.
5863920 /u01/app/oracle/product/12.1.0.2/PSU3
6260832 /u01/app/oracle/product/12.1.0.2/PSU3
Check the inventory to insure all appears to be in order
for node in ${nodes} ; do
ssh -q ${node} tail /u01/app/oraInventory/ContentsXML/inventory.xml
done
Verify PSU applied on all nodes
for node in ${nodes} ; do
ssh -q ${node} /u01/app/oracle/product/12.1.0.2/PSU3/OPatch/opatch_lsinventory
done
Verify a good backup
Before any change is made to the database , Insure a good backup is in place.
. Check the crontab for the regularly scheduled backup
. Check the logfile to insure the last backup was successful.
. Check each node if RAC if not found on node1 . The backup only needs to run from one node of the RAC.
. If backup is not scheduled or the backups are not successful
Run export backup and create guarantee restore point.
Pre-Upgrade Check
Run the pre-upgrade check on the database to upgrade . Will need to be run individually for each database to upgrade.
As Oracle for each database
# Run for each instance to be upgraded.
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
@/u01/app/oracle/product/12.1.0.2/PSU3/rdbms/admin/preupgrd.sql
exit
EndSQL
Review the output from the above command for each database . Will need to run the required items either before or after the upgrade
./u01/app/oracle/cfgtoollogs/<DBName>/preupgrade/preupgrade.log
./u01/app/oracle/cfgtoollogs/<DBName>/preupgrade/preupgrade_fixups.sql
./u01/app/oracle/cfgtoollogs/<DBName>/preupgrade/postupgrade_fixups.sql
Upgrade the database
Turn Off Cluster Database
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
alter system set cluster_database=false scope=spfile sid='*';
EndSQL
Run Pre-Upgrade SQL
Run the Pre-Upgrade script and the manual items from preupgrd.sql executed above
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
purge dba_recyclebin;
column Name new_value DBName
select name from v\$database;
@/u01/app/oracle/cfgtoollogs/&DBName./preupgrade/preupgrade_fixups.sql
EndSQL
Note that the Preupgrd_fixups.sql script doesn't execute all required items. The "MANUAL ACTION REQUIRED" manually.
Disable Golden Gate System Trigger
sqlplus /nolog <<EndSQL
connect / as sysdba
alter trigger sys.GGS_DDL_TRIGGER_BEFORE disable;
EndSQL
The above will fail if golden gate is not installed. Any error can be ignored.
Shutdown Database
As Oracle
. oraenv
srvctl stop database -d <DBNAME>
Update /etc/oratab
As Oracle, Update /etc/oratab file to the new ORACLE_HOME on all nodes
Startup Upgrade
Startup the database using new ORACLE_HOME
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sydba
Startup Upgrade
EndSQL
Upgrade the Database
As Oracle
. oraenv
cd $ORACLE_HOME/rdbms/admin
$ORACLE_HOME/perl/bin/perl/ catctl.pl -n 8 -l ${ORACLE_HOME}/rdbms/log catupgrd.sql
Example Output
$ORACLE_HOME/perl/bin/perl/ catctl.pl -n 8 -l ${ORACLE_HOME}/rdbms/log catupgrd.sql
...
catctl.pl version 12.1.0.2
Oracle Base =/u01/app/oracle
Analyzing file catupgrd.sql
Log files in /u01/app/oracle/product/12.1.0.2/PSU3/rdbms/admin
catcon: ALL catcon-related output will be written to catupgrd_catcon_30168.lst
catcon: See catupgrd*.log files for output generated by script
catcon: See capupgrd_*.lst files for spool files, if any
Number of Cpus =4
SQL Process Count =8
--------------------------------------------------------------
Phases [0-73]
Serial Phase #: 0 Files: 1 Time: 103s
Serial Phase #: 1 Files: 5 Time: 44s
.....................................
.....................................
.....................................
Serial Phase #: 72 Files: 1 Time: 0s
Serial Phase #: 73 Files: 5 Time: 23
Grand Total Time: 952s
LOG FILES: (Catupgrd*.log)
Upgrade Summary Report Located in :
/u01/app/oracle/product/12.1.0.2/PSU3/cfgtoollogs/DBNAME/upgrade/upg_summary.log
Grand Total Upgrade Time: [0d:0h:15m:52s]
Check Upgrade for errors
As Oracle
cd ${ORACLE_HOME/rdbms/log
grep ^ORA- catupgrd*
Run Post Upgrade SQL
Run the post-upgrade script and manual items from the preupgrd.sql executed above
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
purge dba_recyclebin;
column Name new_value DBName
select name from v\$database;
@/u01/app/oracle/cfgtoollogs/&DBName./preupgrade/postupgrade_fixups.sql
EndSQL
Enable Goldengate System Trigger
sqlplus /nolog <<EndSQL
connect / as sysdba
alter trigger sys.GGS_DDL_TRIGGER_BEFORE enable;
EndSQL
The above will fail if golden gate is not installed. Any error can be ignored.
Apply PSU to Database and Recompile
Check for invalid objects and run the final upgrade verification script
As oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
startup
EndSQL
${ORACLE_HOME}/OPatch/datapatch -apply 20299023
sqlplus /nolog <<EndSQL
connect / as sysdba
@?/rdbms/admin/utlrp.sql
@?/rdbms/admin/utluiobj.sql
@?/rdbms/admin/utlul21s.sql
EndSQL
Reset Cluster and Verify
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
alter system set cluster_database=true scope=spfile and sid='*';
set linesize 100
column banner format a80
column version format a10
column action format a10
column status format a10
column description format a50
select * from v\$version;
select patch_id,version,action,status,description from dba_registry_sqlpatch order by action_time;
EndSQL
Verify Correct Dump Destinations
Be Sure Diagnostic_dest is properly set and *dump_dest default
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
show parameter diagnostic_dest
alter system set diagnostic_dest='/oratrace' scope=spfile sid='*';
show parameter background_dump_dest
alter system reset background_dump_dest scope=spfile sid='*';
show parameter core_dump_dest
alter system reset core_dump_dest scope=spfile sid='*';
show parameter user_dump_dest
alter system reset user_dump_dest scope=spfile sid='*';
EndSQL
Anyone of the above may result in a error that can be ignored.
.ORA-32010: Cannot find entry to delete in SPFILE
Set correct audit_file_dest and insure directories exist
sqlplus /nolog <<EndSQL
connect / as sysdba
column C1 new_name C!
select 'set audit_file_dest=''/oradbaudit/admin//'||name||'/adump''scope=spfilesid=''*''' C1 from v\$database;
alter system &C1.
/
set pagesize 0
set linesize 120
select 'ssh -q ' ||host_name||" mkdir -p /oradbaudit/admin/' ||name||'/adump from v\$database,gv\instance;
EndSQL
Run the above 'ssh -q..." statements from the command line
Shutdown Database
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
shut immediate;
EndSQL
Update OCR and start
As Oracle
. oraenv
srvctl upgrade database -d <DBNAME> -o /u01/app/oracle/product/12.1.0.2/PSU3
srvctl start database d <DBName>
Compile System Triggers
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
select 'alter trigger '||owner||','||object_name||' compile;' from dba_objects where status='INVALID' and object_type='TRIGGER';
EndSQL
Run the above "alter trigger...."statements from the command line
Verify Oratab file
As Oracle
verify /etc/oratab properly updated on all nodes
Update DST
Update DST to version 18 Follow below link to instructions.
- DECLINE_SECURITY_UPDATES=true
- INVENTORY_LOCATION=/u01/app/oraInventory
- ORACLE_BASE=/u01/app/oracle
- ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
- oracle.install.db.BACKUPDBA_GROUP=oinstall
- oracle.install.db.CLUSTER_NODES=node1,node2
- oracle.install.db.DBA_GROUP=dba
- oracle.install.db.DGDBA_GROUP=oinstall
- oracle.install.db.EEOptionsSelection=true
- oracle.install.db.InstallEdition=EE
- oracle.install.db.KMDBA_GROUP=oinstall
- oracle.install.db.OPER_GROUP=oper
- oracle.install.option=INSTALL_DB_SWONLY
- UNIX_GROUP_NAME=oinstall
Update config.c with ASM group
Correct the value of ASM group in the $ORACLE_HOME/rdbms/lib/config.c file on each node
This will result in ASM group set after the PSU relink in the following steps
As Oracle
export ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
sed -t 's/SS_ASM_GRP ""/SS_ASM_GRP "asmadmin"/' ${ORACLE_HOME}/rdbms/lib/config.c
Apply latest PSU on each Node
As Oracle
. If the grid patch will be used
cd /oraworkspace/software
export ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
${ORACLE_HOME}/OPatch/opatch apply /orworkspace/software/gpsu/20485724/20299023 -oh ${ORACLE_HOME} -ocmrf /var/tmp/ocm.rsp -silent
. Ifthe RDBMS patch will be used
cd /oraworkspace/software
mkdir psu
cd psu
unzip ../p20299023_121020_Linux-x86-64.zip
export ORACLE_HOME=/u01/app/oracle/product/12.1.0.2/PSU3
${ORACLE_HOME}/OPatch/opatch apply /orworkspace/software/psu/20299023 -oh ${ORACLE_HOME} -ocmrf /var/tmp/ocm.rsp -silent
Verify the output mentions the PSU to all RAC nodes
Example:
All nodes have been patched. You may start oracle instances on the local system and nodes
Composite patch 20299023 successfully applied .
OPatch session completed with warnings.
log file location: /u01/app/oracle/product/12.1.0.2/PSU3/cfgtoollogs/opatch/opaatch2019-12-30_18-56-12PM_1.log
OPatch completed with warnings.
Update Goldengate Library
As Oracle
. oraenv
ln -s ${ORACLE_HOME}/lib/libnnz12.so ${ORACLE_HOME}/lib/libnnz11.so
Verify Installation
As Oracle:
Verify software installed on all nodes
for node in ${nodes} ; do
ssh -q ${node} du -sk /u01/app/oracle/product/12.1.0.2/PSU3
done
Should looks something like the following . Verify the directory size is similar.
5863920 /u01/app/oracle/product/12.1.0.2/PSU3
6260832 /u01/app/oracle/product/12.1.0.2/PSU3
Check the inventory to insure all appears to be in order
for node in ${nodes} ; do
ssh -q ${node} tail /u01/app/oraInventory/ContentsXML/inventory.xml
done
Verify PSU applied on all nodes
for node in ${nodes} ; do
ssh -q ${node} /u01/app/oracle/product/12.1.0.2/PSU3/OPatch/opatch_lsinventory
done
Verify a good backup
Before any change is made to the database , Insure a good backup is in place.
. Check the crontab for the regularly scheduled backup
. Check the logfile to insure the last backup was successful.
. Check each node if RAC if not found on node1 . The backup only needs to run from one node of the RAC.
. If backup is not scheduled or the backups are not successful
Run export backup and create guarantee restore point.
Pre-Upgrade Check
Run the pre-upgrade check on the database to upgrade . Will need to be run individually for each database to upgrade.
As Oracle for each database
# Run for each instance to be upgraded.
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
@/u01/app/oracle/product/12.1.0.2/PSU3/rdbms/admin/preupgrd.sql
exit
EndSQL
Review the output from the above command for each database . Will need to run the required items either before or after the upgrade
./u01/app/oracle/cfgtoollogs/<DBName>/preupgrade/preupgrade.log
./u01/app/oracle/cfgtoollogs/<DBName>/preupgrade/preupgrade_fixups.sql
./u01/app/oracle/cfgtoollogs/<DBName>/preupgrade/postupgrade_fixups.sql
Upgrade the database
Turn Off Cluster Database
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
alter system set cluster_database=false scope=spfile sid='*';
EndSQL
Run Pre-Upgrade SQL
Run the Pre-Upgrade script and the manual items from preupgrd.sql executed above
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
purge dba_recyclebin;
column Name new_value DBName
select name from v\$database;
@/u01/app/oracle/cfgtoollogs/&DBName./preupgrade/preupgrade_fixups.sql
EndSQL
Note that the Preupgrd_fixups.sql script doesn't execute all required items. The "MANUAL ACTION REQUIRED" manually.
Disable Golden Gate System Trigger
sqlplus /nolog <<EndSQL
connect / as sysdba
alter trigger sys.GGS_DDL_TRIGGER_BEFORE disable;
EndSQL
The above will fail if golden gate is not installed. Any error can be ignored.
Shutdown Database
As Oracle
. oraenv
srvctl stop database -d <DBNAME>
Update /etc/oratab
As Oracle, Update /etc/oratab file to the new ORACLE_HOME on all nodes
Startup Upgrade
Startup the database using new ORACLE_HOME
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sydba
Startup Upgrade
EndSQL
Upgrade the Database
As Oracle
. oraenv
cd $ORACLE_HOME/rdbms/admin
$ORACLE_HOME/perl/bin/perl/ catctl.pl -n 8 -l ${ORACLE_HOME}/rdbms/log catupgrd.sql
Example Output
$ORACLE_HOME/perl/bin/perl/ catctl.pl -n 8 -l ${ORACLE_HOME}/rdbms/log catupgrd.sql
...
catctl.pl version 12.1.0.2
Oracle Base =/u01/app/oracle
Analyzing file catupgrd.sql
Log files in /u01/app/oracle/product/12.1.0.2/PSU3/rdbms/admin
catcon: ALL catcon-related output will be written to catupgrd_catcon_30168.lst
catcon: See catupgrd*.log files for output generated by script
catcon: See capupgrd_*.lst files for spool files, if any
Number of Cpus =4
SQL Process Count =8
--------------------------------------------------------------
Phases [0-73]
Serial Phase #: 0 Files: 1 Time: 103s
Serial Phase #: 1 Files: 5 Time: 44s
.....................................
.....................................
.....................................
Serial Phase #: 72 Files: 1 Time: 0s
Serial Phase #: 73 Files: 5 Time: 23
Grand Total Time: 952s
LOG FILES: (Catupgrd*.log)
Upgrade Summary Report Located in :
/u01/app/oracle/product/12.1.0.2/PSU3/cfgtoollogs/DBNAME/upgrade/upg_summary.log
Grand Total Upgrade Time: [0d:0h:15m:52s]
Check Upgrade for errors
As Oracle
cd ${ORACLE_HOME/rdbms/log
grep ^ORA- catupgrd*
Run Post Upgrade SQL
Run the post-upgrade script and manual items from the preupgrd.sql executed above
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
purge dba_recyclebin;
column Name new_value DBName
select name from v\$database;
@/u01/app/oracle/cfgtoollogs/&DBName./preupgrade/postupgrade_fixups.sql
EndSQL
Enable Goldengate System Trigger
sqlplus /nolog <<EndSQL
connect / as sysdba
alter trigger sys.GGS_DDL_TRIGGER_BEFORE enable;
EndSQL
The above will fail if golden gate is not installed. Any error can be ignored.
Apply PSU to Database and Recompile
Check for invalid objects and run the final upgrade verification script
As oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
startup
EndSQL
${ORACLE_HOME}/OPatch/datapatch -apply 20299023
sqlplus /nolog <<EndSQL
connect / as sysdba
@?/rdbms/admin/utlrp.sql
@?/rdbms/admin/utluiobj.sql
@?/rdbms/admin/utlul21s.sql
EndSQL
Reset Cluster and Verify
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
alter system set cluster_database=true scope=spfile and sid='*';
set linesize 100
column banner format a80
column version format a10
column action format a10
column status format a10
column description format a50
select * from v\$version;
select patch_id,version,action,status,description from dba_registry_sqlpatch order by action_time;
EndSQL
Verify Correct Dump Destinations
Be Sure Diagnostic_dest is properly set and *dump_dest default
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
show parameter diagnostic_dest
alter system set diagnostic_dest='/oratrace' scope=spfile sid='*';
show parameter background_dump_dest
alter system reset background_dump_dest scope=spfile sid='*';
show parameter core_dump_dest
alter system reset core_dump_dest scope=spfile sid='*';
show parameter user_dump_dest
alter system reset user_dump_dest scope=spfile sid='*';
EndSQL
Anyone of the above may result in a error that can be ignored.
.ORA-32010: Cannot find entry to delete in SPFILE
Set correct audit_file_dest and insure directories exist
sqlplus /nolog <<EndSQL
connect / as sysdba
column C1 new_name C!
select 'set audit_file_dest=''/oradbaudit/admin//'||name||'/adump''scope=spfilesid=''*''' C1 from v\$database;
alter system &C1.
/
set pagesize 0
set linesize 120
select 'ssh -q ' ||host_name||" mkdir -p /oradbaudit/admin/' ||name||'/adump from v\$database,gv\instance;
EndSQL
Run the above 'ssh -q..." statements from the command line
Shutdown Database
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
shut immediate;
EndSQL
Update OCR and start
As Oracle
. oraenv
srvctl upgrade database -d <DBNAME> -o /u01/app/oracle/product/12.1.0.2/PSU3
srvctl start database d <DBName>
Compile System Triggers
As Oracle
. oraenv
sqlplus /nolog <<EndSQL
connect / as sysdba
select 'alter trigger '||owner||','||object_name||' compile;' from dba_objects where status='INVALID' and object_type='TRIGGER';
EndSQL
Run the above "alter trigger...."statements from the command line
Verify Oratab file
As Oracle
verify /etc/oratab properly updated on all nodes
Update DST
Update DST to version 18 Follow below link to instructions.
