Perform the post patching work:
1. Apply the catbundle in all the database.
2. Run Rman catalog upgraede for PROD databases .
2. Verify patch is applied on all nodes for user ORACLE the 2 patches, and for GRID 3 patches. Verify patch is applied. Check inventory.
3. Verify cluster.
4. Verify databases.
ps -ef | grep pmon make sure all DBs comes back up.
5. Relocate services back to original location.
6. Start agents. Sudo to omsagent user.
9. Make sure file system comes back on all nodes.
df -h and compare with root home directory saved log for prior patching df -h file system. IF needed and missing any mount it.
10. If we stopped GG. Then start goldengate.
Start manager and all extract and replicat.
11. Start the jobs
b. Enable disabled jobs :
c. Change the job schedules on crontab which were changed for patching week:
11. Bring up Apps now.
Ask the Apps team to start the apps and notify the teams patching has been done.
That is it. you are done with patching for the server. Good job.
1. Apply the catbundle in all the database.
cd $ORACLE_HOME/rdbms/admin
menu to source oracle environment
sqlplus / as sysdba
@catbundle.sql psu apply
@utlrp
|
connect and run the upgrade command twice.
rman catalog rman_db_name/linuxpassword@OMSA
upgrade catalog
upgrade catalog
do the same for all other database.
|
2. Verify patch is applied on all nodes for user ORACLE the 2 patches, and for GRID 3 patches. Verify patch is applied. Check inventory.
login as user oracle: set environement for oracle
export PATH=$PATH:$ORACLE_HOME/OPatch
opatch lsinventory
Oracle should have 25869727 25920335
Login as Grid and then forset grid environement for ASM.
export PATH=$PATH:$ORACLE_HOME/OPatch
opatch lsinventory
GRID should 22502505 25869727 25920335
|
3. Verify cluster.
login as root
. oraenv
+ASM1
crsctl check cluster
crsctl stat res -t
|
ps -ef | grep pmon make sure all DBs comes back up.
5. Relocate services back to original location.
srvctl relocate service -d test -s service1 -i test2 -t test1 -f
|
6. Start agents. Sudo to omsagent user.
cd /tools/agents/oracle/occ/agent12c/agent_inst/bin
./emctl start agent
./emctl status agent
|
9. Make sure file system comes back on all nodes.
df -h and compare with root home directory saved log for prior patching df -h file system. IF needed and missing any mount it.
10. If we stopped GG. Then start goldengate.
Start manager and all extract and replicat.
11. Start the jobs
b. Enable disabled jobs :
exec DBMS_SCHEDULER.DISABLE('OWNER.JOB_NAME');
|
| There are three cronjobs which ran at 1PM GMT changed to 5AM for patching week. Change the crontab for those three job from 00 13 * * 0 ===> 00 5 * * 0 |
11. Bring up Apps now.
Ask the Apps team to start the apps and notify the teams patching has been done.
That is it. you are done with patching for the server. Good job.
No comments:
Post a Comment