After much drama, I managed to patch the GI and RDBMS homes to PSU 12.1.0.2.5. With that out of the way, I can now create a database. Instead of using manual commands, I will use dbca
.
None.
Ran the following commands as the oracle user to set up the environment prior to using dbca
:
export ORACLE_HOME=/opt/app/oracle/product/12.1.0/dbhome_1 export PERL5LIB=${ORACLE_HOME}/perl/lib export PATH=${ORACLE_HOME}/bin:${ORACLE_HOME}/perl/bin:${PATH}
dbca
was invoked as follows:
dbca
On the following screen, chose database creation:
Clicked Next.
Create a database with default configuration
Clicked Next.
No problems were found.
No action was required on the following screen:
Clicked Finish.
The following progress page appeared:
On the Creating Pluggable Databases step, the following message appeared—there were no other details:
Clicked Ignore.
This was followed by the following message:
Clicked Ignore.
Then another message (not shown) before failing altogether.
According to the screen shots, there were two (2) error messages:
The first error returned far too many hits (even when qualified with DBCA), while the second only returned one (1) hit:
This is not an exact match.
The alert log for EMREPOS shows:
Wed Dec 30 08:52:01 2015 alter pluggable database open ORA-65000 signalled during: alter pluggable database open... Wed Dec 30 08:54:08 2015 Shutting down instance (abort) License high water mark = 10 Wed Dec 30 08:54:08 2015 USER (ospid: 13453): terminating the instance Wed Dec 30 08:54:09 2015 Instance terminated by USER, pid = 13453 Wed Dec 30 08:54:09 2015 Instance shutdown complete
MOS Doc ID 2046208.1 is not a match at all. There was really a missing PDB name.
Assuming that the database creation is repairable, I went on.
I added the appropiate entry to /etc/oratab as follows:
cat >>/etc/oratab <<DONE emrepos:/opt/app/oracle/product/12.1.0/dbhome_1:N # line added by Douglas 2015/12/31 DONE
Used . oraenv
to set up the Oracle environment for emrepos.
I reviewed the alert log for the EMREPOS database to find the name of the SPFILE used on the last startup.
Used srvctl
under the database home to define the database resource for EMREPOS:
srvctl add database \ -db emrepos \ -oraclehome /opt/app/oracle/product/12.1.0/dbhome_1 \ -domain yaocm.id.au \ -spfile "+DATA/EMREPOS/PARAMETERFILE/spfile.269.899763205" \ -role PRIMARY \ -startoption "READ WRITE" \ -stopoption "IMMEDIATE" \ -instance emrepos \ -policy AUTOMATIC \ -diskgroup "DATA,FRA"
Used the orapwd
command to create a password file for the EMREPOS database as follows:
orapwd file="+DATA" dbuniquename=EMREPOS sysbackup=y sysdg=y syskm=y
The output was:
Enter password for SYS: Enter password for SYSBACKUP: Enter password for SYSDG: Enter password for SYSKM:
Note: Using orapwd
after srvctl add database
updates the CRS resource automatically. Using the commands in the reverse order means that you will have to specify the password file name in srvctl add database
command.
Used the following command to see the current configuration of the CRS resource for the EMREPOS database:
srvctl config database -db emrepos -all
The output was:
Database unique name: emrepos Database name: Oracle home: /opt/app/oracle/product/12.1.0/dbhome_1 Oracle user: oracle Spfile: +DATA/EMREPOS/PARAMETERFILE/spfile.269.899763205 Password file: +DATA/EMREPOS/PASSWORD/pwdemrepos.271.899898713 Domain: yaocm.id.au Start options: read write Stop options: immediate Database role: PRIMARY Management policy: AUTOMATIC Disk Groups: DATA,FRA Services: Database is enabled OSDBA group: OSOPER group: Database instance: emrepos
Used the following command to start up the EMREPOS database instance:
srvctl start database -db emrepos
No error messages were produced, nor were there any in the alert log.
select pdb_id,pdb_name,status from cdb_pdbs;
The output was:
PDB_ID PDB_NAME STATUS ---------- -------------------- --------- 2 PDB$SEED NORMAL