S/N
|
Steps
|
Command
|
|
1.
|
Download Latest CPU Patch from Metalink after
every three months. Also download the opatch utility with version the latest
version
|
|
|
2.
|
SCP CPU Patch to the destination, Where it has to
be applied.
|
scp
–rp
|
|
3.
|
Shutdown the oracle database and listener on both
nodes
|
SQL>shutdown
immediate
ps
–ef |greptns
Kill
-9
|
|
4.
|
Take backup of $ORACLE_HOME and Ora Inventory on
both nodes
|
cp
–rp $ORACLE_HOME /backup/OraHome_bkp_
|
|
5.
|
Run /usr/sbin/slibclean if on AIX Based Systems
before applying the patch.
|
/usr/sbin/slibclean
|
|
6.
|
Unzip the patch file.
|
Use
unzip or gunzip to uncompress the files.
|
|
7.
|
After unzip, Go to the patch directory on 1st
node e.g. /oracle10g/cpu_apr_2007/5901891
|
cd
/oracle10g/cpu_apr_2007/5901891
|
|
8.
|
Set ORACLE_HOME env value
|
On
Unix Systems use export ORACLE_HOME=
|
|
9.
|
Add Opatch utility to PATH env
|
Export
PATH=$PATH:$ORACLE_HOME/OPatch
|
For
running opatch from patch number directory.
|
10.
|
Apply the CPU Patch on 1st node
|
$opatchnapply
–local
|
Make
sure that you are using opatch version 10.2.0.4 napply only works with this
version. Check opatch version with $opatch version cmd
|
11.
|
If you want to apply patch on both nodes in a
single attempts then use global
|
$opatchnapply
–global
|
This
command will apply the patch on both nodes simultaneously. Please make sure
that database and listener are down on both nodes
|
12.
|
If applying with –local
|
If
you are applying the patch one by one on each node then follow the steps 2nd
to 10thon each node.
|
|
Post installation steps
S/N
|
Steps
|
Command
|
Remarks
|
1.
|
After installing the patch, go to
$ORACLE_HOME/cpu directory on any of the node
|
cd
$ORACLE_HOME/cpu
|
This
activity is required on only one node.
|
2.
|
Run catcpu.sql script as SYSDBA
|
sqlplus /nolog
SQL> CONNECT /
AS SYSDBA
SQL>
@catcpu.sql
|
This
script should be run on one node only. It is not required on other nodes of
RAC
|
3.
|
Check for invalid objects of SYS and SYSTEM
Schema
|
SQL>select
count(*),status,owner from dba_objects where owner in (‘SYS’,’SYSTEM’) and
status <>’VALID’ group by status,owner
|
This
script should be run on one node only. It is not required on other nodes of
RAC
|
4.
|
If above SQL return any rows then, Run utlrp.sql
|
SQL>
@?/rdbms/admin/utlrp.sql
|
This
script should be run on one node only. It is not required on other nodes of
RAC
|
5.
|
Re-Check invalid objects of SYS and SYSTEM
Schema.
|
|
|
Output
Check with
opatch utility that patch has been installed.
1.
$ opatchlsinventory
2.
Repeat this command on all nodes and check the
patch number and date.
3.
SQL> select owner,status,count(*) from
dba_objects where owner in (‘SYS’,’SYSTEM’) and status<>’VALID’ group by
owner,status
4.
Release the system to application team and ask them
to check the functionality.
This comment has been removed by a blog administrator.
ReplyDeleteThis comment has been removed by a blog administrator.
ReplyDelete