Am in the process of upgrading our system from R12.0.4 64bit to 12.1.1 which is running on RHEL 5.3 single node with database version 126.96.36.199.
During the upgrade i got an error:- FAILED: file MSDODPCODE.sql on worker 2
On drilling down in adwork002.log it mentioned that "not connected to oracle".
So checking in the alert log file i found an error which led me to the trace file which is listed below. I know for a fact its something to do with OLAP as it wasnt migrated when i did the migration from 32bit to 64bit.
Trace file /data/db/tech_st/11.2.0/admin/bkp_erpbkp/diag/rdbms/bkp/bkp/trace/bkp_ora_18967.trc
Oracle Database 11g Enterprise Edition Release 188.8.131.52.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
ORACLE_HOME = /data/db/tech_st/11.2.0
System name: Linux
Node name: erpbkp.tcaa.go.tz
Version: #1 SMP Wed Dec 17 11:41:38 EST 2008
Instance name: bkp
Redo thread mounted by this instance: 1
Oracle process number: 42
Unix process pid: 18967, image: firstname.lastname@example.org
*** 2013-05-31 17:02:05.441
*** SESSION ID:(886.6019) 2013-05-31 17:02:05.441
*** CLIENT ID:() 2013-05-31 17:02:05.441
*** SERVICE NAME:(bkp) 2013-05-31 17:02:05.441
*** MODULE NAME:(SQL*Plus) 2013-05-31 17:02:05.441
*** ACTION NAME:() 2013-05-31 17:02:05.441
Exception [type: SIGSEGV, Address not mapped to object] [ADDR:0x2B6348806ED7] [PC:0x47F8EB4, __intel_new_memcpy()+2164] [flags: 0x0, count: 1]
Incident 20339 created, dump file: /data/db/tech_st/11.2.0/admin/bkp_erpbkp/diag/rdbms/bkp/bkp/incident/incdir_20339/bkp_ora_18967_i20339.trc
ORA-07445: exception encountered: core dump [__intel_new_memcpy()+2164] [SIGSEGV] [ADDR:0x2B6348806ED7] [PC:0x47F8EB4] [Address not mapped to object] 
ssexhd: crashing the process...
Shadow_Core_Dump = PARTIAL
ksdbgcra: writing core file to directory '/data/db/tech_st/11.2.0/admin/bkp_erpbkp/diag/rdbms/bkp/bkp/cdump'
Ensure you have no invalid objects and if any exist, then compile them and restart the failed worker. Please also make sure the number of workers you are using to run this upgrade is not more than (1-1.5 * CPUs).
ORA-600/ORA-7445/ORA-700 Error Look-up Tool [ID 153788.1]
ORA-7445 [memcpy] [ID 310531.1]
Full Database Export Fails With ORA-03113 and ORA-01041 After Migrated To 64bit [ID 1497414.1]
Please confirm that you have the patches in the following docs applied.
Database Preparation Guidelines for an E-Business Suite Release 12.1 Upgrade [ID 761570.1]
Interoperability Notes EBS 12.0 and 12.1 with Database 11gR2 [ID 1058763.1]
If you still have the same issue, then please log a SR and upload the database log/trace files generated to Oracle support.
As promised am here to update you on how we solved the issue through Oracle Support.
EPB AW's Cannot Be Attached Or Do Not Exist [ID 795247.1].
Now this applies only for the AW's belonging to EPB as for the rest i deleted them manually.
Restarting the failed worker brought about this error:-
sqlplus -s APPS/***** @/apps/apps/apps_st/appl/msd/12.0.0/patch/115/sql/MSDODPCODE.sql
ERROR at line 1:
ORA-33292: Insufficient permissions to access analytic workspace APPS.ODPCODE
using the specified access mode.
ORA-06512: at "SYS.DBMS_AW", line 67
ORA-06512: at "SYS.DBMS_AW", line 131
ORA-06512: at "SYS.DBMS_AW", line 977
ORA-06512: at "SYS.DBMS_AW", line 930
ORA-33262: Analytic workspace APPS.ODPCODE does not exist.
ORA-06512: at "SYS.DBMS_AW", line 901
ORA-06512: at "APPS.MSD_AW_LOADER", line 10
ORA-06512: at line 3
So i followed the below doc and all is well now. The actual problem is 32bit AW's in a 64bit Database.
Thanks alot and good evening.
ORA-33292 Insufficient permissions to access analytic workspace APPS.ODPCODE Error When Applying Patch 9058835 [ID 977196.1]