Pythian Blog: Technical Track

Step-by-step guide to January 2016 PSU patch apply on 12c drid and RDBMS homes in Linux

Following step by step action plan is for single instance database stored on ASM in 12.1.0.2 on Linux (OEL 6 64 bit in this case.)
Step Description ETA
1 Update the OPATCH utility:   For Database home:   $ unzip p6880880_121010_LINUX.zip -d /u01/app/oracle/product/12.1.0/db_1 $ /u01/app/oracle/product/12.1.0/db_1/OPatch/opatch version   For Grid home:   $ unzip p6880880_121010_LINUX.zip -d /u01/app/oracle/12.1.0.2/grid $ /u01/app/oracle/12.1.0.2/grid/OPatch/opatch version 15 min
2 Create ocm.rsp file:   Note: Press Enter/Return key and don't provide any input and say Yes.   $ export ORACLE_HOME=/u01/app/oracle/12.1.0.2/grid $ $ORACLE_HOME/OPatch/ocm/bin/emocmrsp -no_banner -output /stage/ocm.rsp 5 min
3 Validation of Oracle Inventory   Before beginning patch application, check the consistency of inventory information for GI home and each database home to be patched. Run the following command as respective Oracle home owner to check the consistency.   For database home:   $ /u01/app/oracle/product/12.1.0/db_1/OPatch/opatch lsinventory -detail -oh /u01/app/oracle/product/12.1.0/db_1   For Grid home:   $ /u01/app/oracle/12.1.0.2/grid/OPatch/opatch lsinventory -detail -oh /u01/app/oracle/12.1.0.2/grid   If this command succeeds, it lists the Oracle components that are installed in the home. Save the output so you have the status prior to the patch apply. 5 min
4 Stage the Patch:   $ mkdir /stage/PSUpatch $ cp /stage/p22191349_121020_Linux-x86-64.zip /stage/PSUpatch   Check that the directory is empty. $ cd /stage/PSUpatch $ ls   Unzip the patch as grid home owner.   $ unzip p22191349_121020_<platform>.zip 5 min
5 One-off Patch Conflict Detection and Resolution:   Run it with root user:   /u01/app/oracle/12.1.0.2/grid/OPatch/opatchauto apply /stage/PSUpatch/22191349 -analyze -ocmrf /stage/ocm.rsp   It will ask to rollback identical patches like this:   Analyzing patch(es) on "/u01/app/oracle/12.1.0.2/grid" ... Patch "/stage/PSUpatch/22191349/21436941" is already installed on "/u01/app/oracle/12.1.0.2/grid". Please rollback the existing identical patch first. Patch "/stage/PSUpatch/22191349/21948341" is already installed on "/u01/app/oracle/12.1.0.2/grid". Please rollback the existing identical patch first. Patch "/stage/PSUpatch/22191349/21948344" is already installed on "/u01/app/oracle/12.1.0.2/grid". Please rollback the existing identical patch first. Patch "/stage/PSUpatch/22191349/21948354" is already installed on "/u01/app/oracle/12.1.0.2/grid". Please rollback the existing identical patch first.   So first rollback above 4 patches by going to their directory and issuing with grid owner from grid home:   opatch rollback -id 21948354 -local -oh /u01/app/oracle/12.1.0.2/grid (Repeat for all 4 patches)   Note: In some cases, weirdly, I had to shutdown the has services with root user before patch rollback by using:   /u01/app/oracle/12.1.0.2/grid/bin/crsctl stop has -f   After this again run:   /u01/app/oracle/12.1.0.2/grid/OPatch/opatchauto apply /stage/PSUpatch/22191349 -analyze -ocmrf /stage/ocm.rsp   If analyze command fail then use this with root user:   $ORA_GI_HOME/crs/install/roothas.pl –postpatch   It will start the has services too.   Then again run the analyze command as given above:   It will show something like:   Analyzing patch(es) on "/u01/app/oracle/12.1.0.2/grid" ... Patch "/stage/PSUpatch/22191349/21436941" successfully analyzed on "/u01/app/oracle/12.1.0.2/grid" for apply. Patch "/stage/PSUpatch/22191349/21948341" successfully analyzed on "/u01/app/oracle/12.1.0.2/grid" for apply. Patch "/stage/PSUpatch/22191349/21948344" successfully analyzed on "/u01/app/oracle/12.1.0.2/grid" for apply. Patch "/stage/PSUpatch/22191349/21948354" successfully analyzed on "/u01/app/oracle/12.1.0.2/grid" for apply.   Now you are good to apply the patch. Proceed to next step.         10 min
6 Apply the Patch: (Note: This should apply patch in both GI and RDBMS Home but its unreliable in that sense so after this completes, we need to check opatch lsinventory to make sure that it also applied patches in RDBMS Home)   As root user, execute the following command:   # /u01/app/oracle/12.1.0.2/grid/OPatch/opatchauto apply /stage/PSUpatch/22191349 -ocmrf /stage/ocm.rsp   In case if it doesn’t apply in RDBMS Home, then run:   /u01/app/oracle/product/12.1.0/db_1/OPatch/opatchauto apply /stage/PSUpatch/22191349 -oh /u01/app/oracle/product/12.1.0/db_1 -ocmrf /stage/ocm.rsp   Make sure the above applies both OCW and PSU patches. You can verify that from opatch lsinventory. If only OCW patch is present in output and no PSU (which is likely the case), then issue following from Oracle home with oracle database owner after shutting down database:   /u01/app/oracle/product/12.1.0/db_1/OPatch/opatch apply -oh /u01/app/oracle/product/12.1.0/db_1 -local /stage/PSUpatch/22191349/21948354 60 min
7 Loading Modified SQL Files into the Database:   % sqlplus /nolog SQL> Connect / as sysdba SQL> startup SQL> quit % cd $ORACLE_HOME/OPatch % ./datapatch -verbose 60 min
8 Check for the list of patches applied to the database.   SQL> select action_time, patch_id, patch_uid, version, status, bundle_series, description from dba_registry_sqlpatch; 5 min

No Comments Yet

Let us know what you think

Subscribe by email