Setting up a Physical Standby database using OEM 12c

Though I am not a big fan of settting up things through OEM, I wanted to give it a try on setting up a Physical database instance using OEM 12c.

Steps and the explanations are provided below:

Earlier before starting the setup, I cloned the oracle home to the dataguard server and also pushed the Oracle 12c Agent to the server.

Now coming to the setup.....

From the database home in OEM 12c, Click "Availability" --> "Add Standby Database" to start the dataguard setup process.

In our case, the Primary database was registered with a listener using a non-standard port (1521) and the OEM throwed out an error message that LOCAL_LISTENER parameter must be set.

Choose the type of standby database to be created. In our case, we were creating a Physical Standby database.


 

OEM provides several ways to create the standby database. We used the online backup method which in turn uses RMAN duplicate from active database.

In the Backup options screen, we need to provide the RMAN degree of parallelism, the host credentials used and the standby redo log names. We already had the Host Credentials saved as Named Credentials in OEM so we used the "Existing" credentials.

In the Database location screen, we need to provide the Standby database name, hostname and the host credentials for the standby database server.

In the "File location" screen, I missed out the small "Customize" button which might have allowed me to specify the new file location for the dataguard instance. But I chose to keep the file names and locations same as that of the primary database.

In this screen, we also need to chose the listener that will be used to connect to the Dataguard instance.

In the "Configuration" screen, we chose to Use the Dataguard Broker so that OEM will do the DG Broker related configurations.

Review the setup and click the Finish button. A new OEM job will be submitted to complete the Dataguard setup.

We can monitor the job progress just by clicking the "view job" link in the previous screen which will take us to the job activity screen.

In our case, the setup was failing with "Timeout" errors while performing the RMAN duplicate setup. This was due to the Windows 2008 firewall which was running in the Dataguard server. We had to create few rules for the Oracle Database in order to resolve the issue.

One more error we encountered is that the primary database had the LOG_ARCHIVE_DEST parameter configured which we need to reconfigure to use LOG_ARCHIVE_DEST_1 parameter.

Once these issues were fixed, the dataguard setup completed without any errors.

  

Cool...Isn't it!!!

We could drill down to each step and see what action were done in the background.  Posting all those information will make the blog more bloated.

We built this standby instance mainly for reporting purpose. The standby database will be in recovery mode for an hour in the morning and evening and rest of the time will be opened in read-only mode so that it can be used for reporting purposes.

I will post more on how we accomplished through DGMGRL (Dataguard Broker) commands and OEM 12c.