Oracle Apps DBA

My photo
More than 5+ Years of IT Experience in administering Enterprise wide Multi Vendor UNIX Servers, Oracle Databases (8i to 11gR2), Middle-tiers, Applications and Clusters. I am a Sun Certified System Administrator (SCSA) for Solaris 10, oracle 10g database certified associate(OCA), Oracle 10g Database certified professional (OCP) and Oracle E-Business suite R12 certified professionl(OCP)

Oracle Database and Applications

Sunday, 14 November 2010

Apps 11i clone ORA-27101: shared memory realm does not exist

The below issue is encountered while cloning from one of our production Instance.
Error stack from the ApplydbTier logfile:


ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
SVR4 Error: 2: No such file or directory
Connected to an idle instance.
ORA-27102: out of memory
SVR4 Error: 22: Invalid argument
BEGIN dbms_backup_restore.zeroDbid(0); END;

*
ERROR at line 1:
ORA-01034: ORACLE not available
ERROR at line 1:
ORA-01034: ORACLE not available


declare lib_ver varchar2(3);
*
ERROR at line 1:
ORA-01034: ORACLE not available


ERROR:
ORA-01012: not logged on


Disconnected
exit_code=1

Checking for errors ...

ORA-27102: out of memory
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-27101: shared memory realm does not exist
ORA-27102: out of memory
ORA-01034: ORACLE not available
ORA-27102: out of memory
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
ORA-01034: ORACLE not available
The database has not been successfully created. Shutting down the instance ...

SQL*Plus: Release 10.2.0.4.0 - Production on Sun Nov 14 13:09:10 2010


The error code itself indicates that there is an insufficient memory required for starting up an oracle instance.

The issue is with the larger SGA allocated to an new instance. While running adcfgclone. pl, it created new context file with SGA, PGA, Shared_Pool as per the PRODUCTION parameters. But on clone Server the resource were not available as per these parameters.

Context file entries created with the parameters as per the PRODUCTION parameter file:


SGA_TARGET oa_var="s_db_sga_target" 18G SGA_TARGET
SHARED_POOL_SIZE oa_var="s_db_shared_pool_size" 3G SHARED_POOL_SIZE
PGA_AGGREGATE_TARGET oa_var="s_db_pga_aggregate_target" 3G PGA_AGGREGATE_TARGET


Just we need to modify these parameters in context file as per the resources of clone server.

Modified entries:


SGA_TARGET oa_var="s_db_sga_target" 4G SGA_TARGET
SHARED_POOL_SIZE oa_var="s_db_shared_pool_size" 1G SHARED_POOL_SIZE
PGA_AGGREGATE_TARGET oa_var="s_db_pga_aggregate_target" 1G PGA_AGGREGATE_TARGET


After changing these parameters in _.xml run adcfgclone.pl again by specifying the complete path of the context file.


bash-3.00$ perl adcfgclone.pl appsTier /db/clonedat/dbase/10.2.0/appsutil/clone/bin/clonenew_testdb.xml


After this adcfgclone.pl completed successfully without any issues.

No comments: