SLAC ESD Software Engineering Group
Stanford Linear Accelerator Center

Oracle  -MCCO

Shutdown and Startup

SLAC Computing
Software Engineering
Detailed
Unix
 

 


The MCCO database is running in a DataGuard environment.   The components of this environment are a DataGuard broker which will not be described herein,  a primary database,  a physical standby database, two listeners for user connections, and a listener to handle the data being transferred from the primary database machine to the secondary.   The broker is not being considered here because this is describing normal shutdown and startup procedures which do not lead to failover.

 

The shutdown order is as follows:

 

  • Shutdown mcc-observer
  • Shutdown the oracle databases on mccora1, mccora2
  • Shutdown the servers mccora1, mccora2

 

 

The startup order is as follows:

 

  • Startup the servers mccora1, mccora2
  • Startup the oracle databases on mccora1, mccora2
  • Startup mcc-observer, and manually startup the Observer process
 

 

1.   Logging into  mccora1 and Mccora2.

 

You must log in with an account that can issue ' su oracle'

 

2.   Setting the Environment

 

      su oracle

      source ~oracle/cora10.2.0

      setenv ORACLE_SID MCCO

 

 

Shutting the databases down

 

1.  Determining which database is primary and  which is the physical standby

 

       sqlplus /nolog

               connect / as sysdba

               select database_role from v$database;

                exit

 

The query will return either PRIMARY or   PHYSICAL STANDBY

 

On the machine which hosts the physical standby database issue:

 

lsnrctl stop

 

sqlplus /nolog

           connect / as sysdba

 

On the machine which hosts the primary database issue:

 

lsnrctl stop

 

sqlplus  /nolog

     connect / as sysdba

     shutdown immediate

     exit

 

Wait for the shutdown to complete

DO NOT ISSUE  shutdown abort AS IT WILL CAUSE  AN ATTEMPTED FAILOVER!!

 

Returning to the physical standby machine issue

 

         alter database recover managed standby database cancel;

        shutdown immediate

        exit

 

At the OS command prompt issue

 

lsnrctl stop log_listener

 

 

Starting the databases up

 

On the physical standby  machine issue

 

lsnrctl start log_listener

sqlplus /nolog       

connect /  as sysdba

         startup mount

 

 

wait  for the startup to complete

 

On the primary database machine issue

 

sqlplus /nolog

        connect / as sydba       

startup

         exit

 

Wait for the startup to complete

 

Return to the  physical standby machine and  issue

 

              alter database recover managed standby database using current logfile disconnect;

              exit

 

Wait for the command to complete

 

 

Return to the  primary database  machine and issue

 

lsnrctl start

 

Return to the physical standby database machine and isssue

 

lsnrctl start

 

 


 

 

 

 
 

 


[SLAC ESD Software Engineering Group][ SLAC Home Page]


Author: Ken Brobeck

Modified: 01-Mar-2011