10 Replies Latest reply on Sep 8, 2011 12:19 PM by David Heisler-Oracle

    MW 2.19 multi node upgrade patching fails on second, third, etc node.

    David Heisler-Oracle
      When using Maintenance Wizard 2.19 with multi-node configuration patching will fail after the first node with an error in the adpatch log that states invalid options as seen below.


      "AutoPatch warning:
      Ignoring unrecognized option: "options".

      Connecting to APPS......Connected successfully.


      AutoPatch error:
      You must be in Maintenance Mode to apply patches.
      You can use the AD Administration Utility to set Maintenance Mode."

      Please update this forum if you hit this issue and I can send you the fix until Maintenance Wizard 2.20 is released.

      Thanks
        • 1. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
          user12010879
          I can't see any logs for the patching or adadmin on the second node. How can I tell if MW is even trying to run it on this node?

          Many Thanks,
          • 2. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
            David Heisler-Oracle
            If you click on the execute button you should see the log of the execution of the step. Please cut and paste that content here for review.
            • 3. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
              user12010879
              As requested here's a paste of the command log, I can just see entries for the admin node and not the forms/web node.

              Many Thanks

              ###############################################################################
              INFO - Help on monitoring when multiple drivers and nodes are applied.
              ###############################################################################
              If the web window times out, click on the x and rerun it to view the updated log
              file. It will change as the patch continues to run in background. Just keep marking it
              success until it is complete or check the active patch node manually:
              $ tail -f $APPL_TOP/admin/${TWO_TASK}/log/patches_PRE_1703_1_patchit11i*.log

              Each node, if multi-node, will be determined and the patch drivers will be applied
              on the right nodes in the right order. Look for the ACTION: statement.
              ###############################################################################
              ACTION: execute on node: apps , patch=patches_PRE_1703_1 driver=u_merged.drv
              ###############################################################################
              patchit11i.sh REMOTE SIZE: 47535 LOCAL SIZE: 47535
              OK - no update needed for patchit11i.sh, same size as on EOF: 47535 bytes.

              Found release REL=11.5.10.2 from fnd_product_groups.
              Starting adpatch for: patch=patches_PRE_1703_1 force= driver=u_merged.drv
              Using defaults file /u10/orafun/appl/admin/FUN/adalldefaults.txt.

              Starting patch with the following command:

              /u10/orafun/appl/ad/11.5.0/bin/adpatch defaultsfile=/u10/orafun/appl/admin/FUN/adalldefaults.txt logfile=patches_PRE_1703_1_patchit11i.log patchtop=/u10/orafun/appl/patch/patches_PRE_1703_1 driver=u_merged.drv workers=6 interactive=no options=hotpatch,novalidate,nocompilejsp,nocompiledb,nomaintainmrc,nogenerateportion restart=n

              ###############################################################################
              Confirms adpatch is not still active, and views Check patch log times.
              ###############################################################################
              ... PROCESS LISTING of adpatch
              ... ls -l of patches_PRE_1703_1_patchit11i.log to see date/time stamp
              -rw-r--r-- 1 orafun dba 53761 Aug 30 14:41 /u10/orafun/appl/admin/FUN/log/patches_PRE_1703_1_patchit11i.log
              ... last 20 lines of log file: /u10/orafun/appl/admin/FUN/log/patches_PRE_1703_1_patchit11i.log

              for details.


              Purging timing information for prior sessions.

              sqlplus -s APPS/***** @/u10/orafun/appl/ad/11.5.0/admin/sql/adtpurge.sql 10 1000

              Done purging timing information for prior sessions.

              AutoPatch is complete.

              AutoPatch may have written informational messages to the file
              /u10/orafun/appl/admin/FUN/log/patches_PRE_1703_1_patchit11i.lgi

              Errors and warnings are listed in the log file
              /u10/orafun/appl/admin/FUN/log/patches_PRE_1703_1_patchit11i.log

              and in other log files in the same directory.


              ###############################################################################
              SUMMARY - patchit11i.sh - Completed application of patch: patches_PRE_1703_1, driver: driver=u_merged.drv
              ###############################################################################
              OK - adpatch completed successful.

              INFO: Not starting patch alert monitor, to start: Login Engineer, User Admin, Update Profile


              Redo the Action
              Completed Successfully Post Hour Minute
              • 4. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
                David Heisler-Oracle
                It appears you may have a misconfiguration in your project setup.

                Please login to sysadmin - click on the roles tab and from the project administrator drop down choose your current project.

                Please check under the Define Nodes and verify that you have defined a different host name / login id for each tier you have.

                After making your changes you will need to run the configuration again.

                Let me know
                • 5. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
                  user12010879
                  Yes I can confirm that I have both 11.5.10 nodes registered. The login account and host name is correct. In addition I used the autoconfig file to register both nodes and verified that the MW node could ssh to them.
                  • 6. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
                    David Heisler-Oracle
                    Please provide the output of this sql command:

                    select instance_id, tier_id, host, platform from eof_instance_tiers order by instance_id;

                    Also - how many steps do you have in this Task - I have only four listed. There should be a step for each tier apps , conc, forms, web

                    Task: 11i Pre-Upgrade Patching      
                    Comments      Step      Audit      Setup      Execute      Validate
                    User Notepad      1--2-Create Merged patch for Rel 11i - Part 1 (Required)      Step Audit           Do Execute      
                    User Notepad      1--3-Apply Merged patch to the APP node - Part 1 (Required)      Step Audit           Execute did NOT COMPLETE successfully. Please CHECK THE LOG.      
                    User Notepad      1--4-Compile and generate using adadmin (Required)      Step Audit           Execute did NOT COMPLETE successfully. Please CHECK THE LOG.      
                    User Notepad      1--5-Verify that all 11i patches were installed (Required)      Step Audit           Do Execute
                    • 7. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
                      user12010879
                      INSTANCE_ID TIER_ID HOST PLATFORM
                      ----------- ---------- ---------------------------------------- ----------
                      1703 115_WEB lassell.npl.co.uk SunOS
                      1703 12_FRM appstest.npl.co.uk Linux
                      1703 115_FRM lassell.npl.co.uk SunOS
                      1703 12_CM appstest.npl.co.uk Linux
                      1703 115_DB verrier.npl.co.uk SunOS
                      1703 115_ADM verrier.npl.co.uk SunOS
                      1703 115_CM verrier.npl.co.uk SunOS
                      1703 12_WEB appstest.npl.co.uk Linux
                      1703 12_ADM appstest.npl.co.uk Linux

                      9 rows selected.
                      • 8. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
                        David Heisler-Oracle
                        Ok please execute the following two commands and return the output. I believe I have found the issue based on your last update. The output from these two sql commands will verify the problem.

                        You also did not answer the second part of my previous post concerning # of steps in the Task 11i Pre-Upgrade Patching


                        select 'PATCH_NODES~' || Eof_Tools.get_node_list_pre(1703) from dual;


                        select 'PATCH_NODES~' || Eof_Tools.get_node_list_post(1703) from dual;
                        • 9. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
                          user12010879
                          select 'PATCH_NODES~' || Eof_Tools.get_node_list_pre(1703) from dual;

                          'PATCH_NODES~'||EOF_TOOLS.GET_NODE_LIST_PRE(1703)
                          --------------------------------------------------------------------------------
                          PATCH_NODES~apps forms


                          select 'PATCH_NODES~' || Eof_Tools.get_node_list_post(1703) from dual;

                          'PATCH_NODES~'||EOF_TOOLS.GET_NODE_LIST_POST(1703)
                          --------------------------------------------------------------------------------
                          PATCH_NODES~apps

                          I'll have a look at the 2nd part and get back to you.

                          Many Thanks.
                          • 10. Re: MW 2.19 multi node upgrade patching fails on second, third, etc node.
                            David Heisler-Oracle
                            Ok, this is a bug with code that creates the steps to apply the merged patches. The code is looking at the POST node list instead of the PRE node list. In your case you are going from a multi-node 11i instance to a single node R12 instance.

                            Let me see if I can get you the fix.

                            No need to check the steps anymore.

                            Thanks