1 2 Previous Next 16 Replies Latest reply: Sep 21, 2012 11:49 AM by r_g_financials_dba RSS

    Startup migrate hanging

    765776
      I am upgrading database from 10.2.0.4 to 11.2.0.1 . While starting the instance in new 11g home in Migrate mode its took 1 hour .. I have checked there is no instance recovery is going on that time. Its in Sun cluster Active - passive database. I don't face this issue in 10g...

      Edited by: Mano on Feb 11, 2011 2:11 PM
        • 1. Re: Startup migrate hanging
          Srini Chavali-Oracle
          Pl indicate which document you are following for the upgrade and which step you are in.

          The official document is here - http://download.oracle.com/docs/cd/E11882_01/server.112/e17222/toc.htm

          What messages are there in the alert log file ? What is the OS version ?

          HTH
          Srini
          • 2. Re: Startup migrate hanging
            765776
            As Part for R12 upgrade , I am upgrading the database from 10.2.0.4 to 11.2.0.1 following below note

            Interoperability Notes EBS R12 with Database 11gR2 [ID 1058763.1]

            Steps done so far,

            Install 11.2.0.1 software
            Install Example CD - 11.2.0.1
            Create nls directory - 11.2.0.1
            Apply the patch 8730312 - 11.2.0.1
            Apply interop patch 9218789 - 11.2.0.1
            Empty recyclebin in 10.2.0.4 database
            Apply DST patches in 10.2.0.4 database -- (7695070 (prereq for 8524113),8524113)
            Gather system statistics in 10.2.0.4 database -- EXEC DBMS_STATS.GATHER_DICTIONARY_STATS;
            Ran utlu112i.sql in 10g database
            started the database in 11g home in migrate mode which took 1 hour

            Currently catupgrd.sql is running on

            SunOS 5.10 Generic_139555-08 sun4u sparc SUNW,SPARC-Enterprise

            there is no message in Alert logfile

            ALTER DATABASE MOUNT
            Control file expanded from 2386 to 2548 blocks for upgrade.
            Changing di2dbun from to proa081
            Expanded controlfile section 32 from 10 to 31 records
            The number of logical blocks in section 32 remains the same
            Successful mount of redo thread 1, with mount id 2731133030
            Database mounted in Exclusive Mode
            Lost write protection disabled
            Create Relation IPS_PACKAGE_UNPACK_HISTORY
            Completed: ALTER DATABASE MOUNT
            Fri Feb 11 20:46:35 2011
            ALTER DATABASE OPEN MIGRATE
            LGWR: STARTING ARCH PROCESSES
            Fri Feb 11 20:46:35 2011
            ARC0 started with pid=20, OS id=28409
            ARC0: Archival started
            LGWR: STARTING ARCH PROCESSES COMPLETE
            ARC0: STARTING ARCH PROCESSES
            Fri Feb 11 20:46:36 2011
            ARC1 started with pid=21, OS id=28411
            Fri Feb 11 20:46:36 2011
            ARC2 started with pid=22, OS id=28413
            Fri Feb 11 20:46:36 2011
            ARC3 started with pid=23, OS id=28415
            ARC1: Archival started
            ARC2: Archival started
            ARC1: Becoming the 'no FAL' ARCH
            ARC1: Becoming the 'no SRL' ARCH
            ARC2: Becoming the heartbeat ARCH
            Thread 1 advanced to log sequence 11995 (thread open)
            Thread 1 opened at log sequence 11995
            Current log# 4 seq# 11995 mem# 0: /u22/app/oraredo/proa081/log4a.dbf
            Current log# 4 seq# 11995 mem# 1: /u24/app/oraredo/proa081/log4b.dbf
            Successful open of redo thread 1
            MTTR advisory is disabled because FAST_START_MTTR_TARGET is not set
            SMON: enabling cache recovery
            Fri Feb 11 20:46:36 2011
            NSA2 started with pid=24, OS id=28419
            Successfully onlined Undo Tablespace 1.
            SMON: enabling tx recovery
            ARC3: Archival started
            ARC0: STARTING ARCH PROCESSES COMPLETE
            Database Characterset is UTF8
            Updating 10.2.0.4.0 NLS parameters in sys.props$
            -- adding 11.2.0.1.0 NLS parameters.
            Archived Log entry 23850 added for thread 1 sequence 11994 ID 0xffffffff9fa9d749 dest 1:
            Incremental checkpoint up to RBA [0x2edb.5.0], current log tail at RBA [0x2edb.2c.0]
            Fri Feb 11 20:46:39 2011
            Beginning log switch checkpoint up to RBA [0x2edc.2.10], SCN: 11725979644285
            Thread 1 advanced to log sequence 11996 (LGWR switch)
            Current log# 5 seq# 11996 mem# 0: /u21/app/oraredo/proa081/log5a.dbf
            Current log# 5 seq# 11996 mem# 1: /u23/app/oraredo/proa081/log5b.dbf
            Archived Log entry 23851 added for thread 1 sequence 11995 ID 0xffffffff9fa9d749 dest 1:
            Fri Feb 11 20:46:48 2011
            Stopping background process MMNL
            Stopping background process MMON
            Starting background process MMON
            Fri Feb 11 20:46:50 2011
            MMON started with pid=17, OS id=28446
            Starting background process MMNL
            Fri Feb 11 20:46:50 2011
            MMNL started with pid=18, OS id=28448
            ALTER SYSTEM enable restricted session;
            ALTER SYSTEM SET systemtrig_enabled=FALSE SCOPE=MEMORY;
            Autotune of undo retention is turned off.
            ALTER SYSTEM SET undoautotune=FALSE SCOPE=MEMORY;
            ALTER SYSTEM SET undo_retention=900 SCOPE=MEMORY;
            ALTER SYSTEM SET aq_tm_processes=0 SCOPE=MEMORY;
            ALTER SYSTEM SET enable_ddl_logging=FALSE SCOPE=MEMORY;
            Resource Manager disabled during database migration: plan '' not set
            ALTER SYSTEM SET resource_manager_plan='' SCOPE=MEMORY;
            ALTER SYSTEM SET recyclebin='OFF' DEFERRED SCOPE=MEMORY;
            Resource Manager disabled during database migration
            Fri Feb 11 20:47:11 2011
            replication_dependency_tracking turned off (no async multimaster replication found)
            Fri Feb 11 20:48:36 2011
            Incremental checkpoint up to RBA [0x2edc.2540.0], current log tail at RBA [0x2edc.268f.0]
            Fri Feb 11 21:08:38 2011
            Incremental checkpoint up to RBA [0x2edc.2b3f.0], current log tail at RBA [0x2edc.2b3f.0]
            Fri Feb 11 21:42:20 2011
            XDB UNINITIALIZED: XDB$SCHEMA not accessible
            Completed: ALTER DATABASE OPEN MIGRATE
            Fri Feb 11 21:47:15 2011
            Incremental checkpoint up to RBA [0x2edc.24b4d.0], current log tail at RBA [0x2edc.3cf9d.0]
            Fri Feb 11 21:54:20 2011

            Edited by: Mano on Feb 11, 2011 2:26 PM
            • 3. Re: Startup migrate hanging
              Srini Chavali-Oracle
              Is there a reason you are not using DBUA ? Pl see these MOS Docs for manual upgrades

              Complete Checklist for Manual Upgrades to 11gR2          (Doc ID 837570.1)
              Oracle 11gR2 Upgrade Companion          (Doc ID 785351.1)

              Also, which R12 version are you upgrading to ?

              I am not sure why you are using the MIGRATE command - you should be using the UPGRADE command

              http://download.oracle.com/docs/cd/E11882_01/server.112/e17222/upgrade.htm#CACGGHJC

              HTH
              Srini
              • 4. Re: Startup migrate hanging
                765776
                I have followed all the noted properly . In order to get more control on upgrade we have chosen Manual upgrade mode. Same upgrade process went fine on standalone server (No os cluster) . But in cluster env. its taking 1 hour to starting the db.

                Upgrade process

                Database

                10.2.0.4 ---> 11.2.0.1

                Application

                12.0.6 ----> 12.1.1 ---> 12.1.2
                • 5. Re: Startup migrate hanging
                  Srini Chavali-Oracle
                  Pl elaborate on what you mean by "more control over the upgrade". The DBUA method is recommended - it has worked without issues for me for 11i database upgrades from 10gR2 to 11gR1.

                  HTH
                  Srini
                  • 6. Re: Startup migrate hanging
                    765776
                    I dont see any issue for using catupgrd.sql you had any issue in past using catupgrd.sql especially for performance issue like this..
                    • 7. Re: Startup migrate hanging
                      Srini Chavali-Oracle
                      It has been years since I have performed manual upgrades. They are error prone since you have to manually execute all of the steps.

                      Srini
                      • 8. Re: Startup migrate hanging
                        ora_tech
                        Hi,
                        I dont see any issue for using catupgrd.sql you had any issue in past using catupgrd.sql especially for performance issue like this..
                        As updated by srini Oracle Always recommends to use DBUA for all your database upgrades. I also had been involved in so many version of database upgrades in apps 11i and R12 upgrades from 9i to 10g to 11gr1 to 11gr2 . I also used DBUA as my upgrade procedure and it completed without any issues.

                        thanks,
                        X A H E E R
                        • 9. Re: Startup migrate hanging
                          765776
                          I dont have any issue on running catupgrd.sql or DBUA. The issue is database startup migrate its took long time after upgrade also it took 40 min to open . There is no recovery is going on..
                          • 10. Re: Startup migrate hanging
                            ora_tech
                            Hi,
                            I dont have any issue on running catupgrd.sql or DBUA. The issue is database startup migrate its took long time after upgrade also it took 40 min to open . There is no recovery is going on..
                            What is the size of your database and number of invalid objects in db?

                            thanks,
                            X A H E E R
                            • 11. Re: Startup migrate hanging
                              765776
                              180 GB . Invalid objects - 76 - belongs to apps ...
                              • 12. Re: Startup migrate hanging
                                ora_tech
                                Hi,
                                180 GB . Invalid objects - 76 - belongs to apps ...
                                Its normal.

                                Why you are using startup migrate. AFAIK this was used prior 10.2.0.x releases. Please refer below MOS tech note.

                                *What is 'STARTUP MIGRATE'? [ID 252273.1]*

                                thanks,
                                X A H E E R
                                • 13. Re: Startup migrate hanging
                                  Lubiez Jean-Valentin
                                  Hello,
                                  Completed: ALTER DATABASE MOUNT
                                  Fri Feb 11 20:46:35 2011
                                  ...
                                  Completed: ALTER DATABASE OPEN MIGRATE
                                  Fri Feb 11 21:47:15 2011
                                  Yes it's rather strange.

                                  What about the Server ?
                                  Do you have enough memory on it ?
                                  Do you have any log at the OS level ?

                                  If there's no error at the Oracle level, the problem may come from the OS or the Hardware.


                                  Best regards,
                                  Jean-Valentin
                                  • 14. Re: Startup migrate hanging
                                    765776
                                    There is no error message return on OS level log , have enough CPU (95% IDLE) and 20GB free memory . I have raised SR for the same with oracle finally they concluded its BUG in R12

                                    BUG 8284764 - RCA: R12:CREATE ACCOUNTING PROCESS DOES NOT DROP QUEUE TABLE

                                    select queue from xla.AQ$XLA_ACCOUNTING_QTAB;

                                    no rows selected

                                    select QUEUE_TABLE,count(*) from dba_queues group by QUEUE_TABLE order by 2 desc;

                                    QUEUE_TABLE COUNT(*)
                                    ------------------------------ ----------
                                    XLA_ACCOUNTING_QTAB 836418
                                    ASO_ORDER_FEEDBACK_T 5
                                    TBLMIG_MSG_QTAB 4


                                    Resolution

                                    1. Apply patch 9131790:R12.XLA.B
                                    This will ensure that message queues are cleanedup for future runs of Create Accounting.

                                    2. For the current message queues need to apply the datafix
                                    1 2 Previous Next