6 Scenarios Using the DGMGRL Command-Line Interface

6.1 Prerequisites for Getting Started

One of the prerequisites for using DGMGRL is that a primary database and any standby databases must already exist. The DG_BROKER_START initialization parameter must be set to TRUE for all databases in the configuration. You must use a server parameter file with the broker (see Section 2.1 and Section 7.1.3).

Convert the initialization parameter files (PFILE) on both primary and standby databases into server parameter files (SPFILE), if necessary. Use the following SQL*Plus command:

CREATE SPFILE='spfilename' FROM PFILE='pfilename';

If an instance was not started with a server parameter file, then you must shut down the instance and restart it using the server parameter file.

After starting the Oracle instance, set the DG_BROKER_START=TRUE initialization parameter using the SQL ALTER SYSTEM statement. The parameter value will be saved in the server parameter file. The next time you start the Oracle instance, the broker is started automatically, and you do not need to issue the SQL ALTER SYSTEM statement again.

The following assumptions are made for the scenarios in this chapter:

• TCP/IP is used to connect to primary and standby databases.

• The standby database has been created from backups of the primary database control files and datafiles as described in the Oracle Data Guard Concepts and Administration.

• The scenarios in this chapter assume the following broker configuration:

  • The configuration name is DRSolution.

  • The database unique name (DB_UNIQUE_NAME) of the primary database is North_Sales.

  • The database unique name (DB_UNIQUE_NAME) of the remote standby database is South_Sales.

  • The protection mode is maximum performance mode.

  • There are standby redo log files configured for both the primary and standby database. The transport mode for both databases is ASYNC.

  • The standby database is a physical standby database.

6.2 Scenario 1: Creating a Configuration

This section provides examples that create a broker configuration named
DRSolution that includes a primary and standby database named North_Sales and South_Sales.

The following steps show how to create a configuration and add one physical standby database.

Step 1   Invoke DGMGRL.

To start DGMGRL, enter dgmgrl at the command-line prompt on a system where Oracle Data Guard is installed:

$ dgmgrl

The DGMGRL prompt is displayed:

DGMGRL>

Step 2   Connect to the primary database.

Before you specify any command (other than the HELP, EXIT, or QUIT), you must first connect to the primary database using the DGMGRL CONNECT command.

The account from which you connect to the database (SYS in this example) must have SYSDG or SYSDBA privileges on the primary and standby databases.

Note:

You do not have to include AS SYSDBA on the CONNECT command; it is the default setting.

The following examples show two variations of the CONNECT command. Example 6-1 shows how to connect to the default database on the local system, and Example 6-2 includes the Oracle Net Services connect identifier
(North_Sales.example.com) to make a connection to a database located on a remote system. In both examples, you are prompted for a password.

Example 6-1 Connecting to the Primary Database on the Local System

DGMGRL> CONNECT sysdg;
Password: password
Connected as SYSDG.

Example 6-2 Connecting to the Primary Database on a Remote System

DGMGRL> CONNECT sysdg@North_Sales.example.com;
Password: password
Connected as SYSDG.

Step 3   Clear existing remote redo transport destinations.

You must clear any remote redo transport destinations on the primary database that do not have the NOREGISTER attribute, before a configuration can be created. Otherwise, the following error message is returned when you attempt to create the configuration:

ORA-16698: LOG_ARCHIVE_DEST_n parameter set for object to be added
 
Failed.

To clear LOG_ARCHIVE_DEST_n settings, use the ALTER SYSTEM SET LOG_ARCHIVE_DEST_n=" " SQL*Plus command.

Step 4   Create the broker configuration.

A broker configuration is initially created with just a primary database, which in this case is called North_Sales. In a later command, you will add the standby database, South_Sales.

Note:

The names for the primary and standby databases must match their database unique names. Fetch these from their DB_UNIQUE_NAME initialization parameter as follows:

SQL> SHOW PARAMETER DB_UNIQUE_NAME;

Use the CREATE CONFIGURATION command to create the DRSolution configuration and define the primary database, North_Sales:

DGMGRL> CREATE CONFIGURATION 'DRSolution' AS
>  PRIMARY DATABASE IS 'North_Sales'
>  CONNECT IDENTIFIER IS North_Sales.example.com;
  

DGMGRL returns the following information:

Configuration "DRSolution" created with primary database "North_Sales"

The name North_Sales is the value of this database's DB_UNIQUE_NAME initialization parameter.

Step 5   Show the configuration information.

Use the SHOW CONFIGURATION command to display a brief summary of the configuration:

DGMGRL> SHOW CONFIGURATION;

DGMGRL returns the following information:

Configuration - DRSolution
 
  Protection Mode: MaxPerformance
  Databases:
    North_Sales - Primary database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
DISABLED

Step 6   Add a standby database to the configuration.

To add a standby database to the DRSolution configuration, use the ADD DATABASE command.

The following command defines South_Sales as a standby database, which is the standby database associated with the primary database called North_Sales:

DGMGRL> ADD DATABASE 'South_Sales' AS
>  CONNECT IDENTIFIER IS South_Sales.example.com;

DGMGRL returns the following information:

Database "South_Sales" added

The name South_Sales is the value of the database's DB_UNIQUE_NAME initialization parameter.

Use the SHOW CONFIGURATION command to verify that the South_Sales database was added to the DRSolution configuration:

DGMGRL> SHOW CONFIGURATION;

DGMGRL returns the following information:

Configuration - DRSolution
 
  Protection Mode: MaxPerformance
  Databases:
    North_Sales  - Primary database
      South_Sales  - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
DISABLED
 
DGMGRL>

6.3 Scenario 2: Setting Database Properties

After you create the configuration with DGMGRL, you can set database properties at any time. For example, the following statements set the LogArchiveFormat and StandbyArchiveLocation database properties for the South_Sales standby database:

DGMGRL> EDIT DATABASE 'South_Sales' SET PROPERTY 'LogArchiveFormat'='log_%t_%s_%r_%d.arc';
Property "LogArchiveFormat" updated.

DGMGRL> EDIT DATABASE 'South_Sales' SET PROPERTY 'StandbyArchiveLocation'='/archfs/arch/';
Property "StandbyArchiveLocation" updated.

Use the SHOW DATABASE VERBOSE command to view all properties and their values for a database. The following example shows the properties for the South_Sales database.

DGMGRL> SHOW DATABASE VERBOSE 'South_Sales';
 
Database - South_Sales
 
  Role:            PHYSICAL STANDBY
  Intended State:  OFFLINE
  Transport Lag:   (unknown)
  Apply Lag:       (unknown)
  Apply Rate:      (unknown)
  Real Time Query: OFF
  Instance(s):
    south_sales1
 
  Properties:
    DGConnectIdentifier             = 'South_Sales.example.com'
    ObserverConnectIdentifier       = ''
    LogXptMode                      = 'ASYNC'
    RedoRoutes                      = ''
    DelayMins                       = '0'
    Binding                         = 'optional'
    MaxFailure                      = '0'
    MaxConnections                  = '1'
    ReopenSecs                      = '300'
    NetTimeout                      = '30'
    RedoCompression                 = 'DISABLE'
    LogShipping                     = 'ON'
    PreferredApplyInstance          = ''
    ApplyInstanceTimeout            = '0'
    ApplyLagThreshold               = '0'
    TransportLagThreshold           = '0'
    TransportDisconnectedThreshold  = '0'
    ApplyParallel                   = 'AUTO'
    StandbyFileManagement           = 'AUTO'
    ArchiveLagTarget                = '0'
    LogArchiveMaxProcesses          = '5'
    LogArchiveMinSucceedDest        = '1'
    DbFileNameConvert               = 'dbs/t, dbs/ct, dbs/bt, dbs/ct'
    LogFileNameConvert              = 'dbs/t, dbs/ct, dbs/bt, dbs/ct'
    FastStartFailoverTarget         = ''
    InconsistentProperties          = '(monitor)'
    InconsistentLogXptProps         = '(monitor)'
    SendQEntries                    = '(monitor)'
    LogXptStatus                    = '(monitor)'
    RecvQEntries                    = '(monitor)'
    StaticConnectIdentifier         = '(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=south_sales1.example.com)(PORT=2840))
(CONNECT_DATA=(SERVICE_NAME=South_Sales_DGMGRL.example.com)(INSTANCE_NAME=south_sales1)(SERVER=DEDICATED)))'
    StandbyArchiveLocation          = 'USE_DB_RECOVERY_FILE_DEST'
    AlternateLocation               = ''
    LogArchiveTrace                 = '255'
    LogArchiveFormat                = 'db3r_%d_%t_%s_%R.arc'
    TopWaitEvents                   = '(monitor)'
 
Database Status:
DISABLED

If broker management of the database is enabled, setting a database property value causes the underlying parameter value to be changed in the corresponding database, and the value for the changed parameter is reflected in the server parameter file. Thus, if the database is shut down and restarted outside of Oracle Enterprise Manager Cloud Control (Cloud Control) and DGMGRL (such as from the SQL*Plus interface), the database uses the new parameter values from the updated server parameter file when it starts. However, you should not make changes to the redo transport services initialization parameters through SQL statements. Doing so will cause an inconsistency between the database and the broker.

You can change a property if the database is enabled or disabled. However, if the database is disabled when you change a property, the change does not take effect until the database is enabled.

6.4 Scenario 3: Enabling the Configuration and Databases

So far, the DRSolution configuration is disabled, which means it is not under the control of the Data Guard broker. When you finish configuring the databases into a broker configuration and setting any necessary database properties, you must enable the configuration to allow the Data Guard broker to manage it.

You can enable:

• The entire configuration, including all of its databases

• A standby database

Enable the entire configuration.

You can enable the entire configuration, including all of the databases, with the following command:

DGMGRL> ENABLE CONFIGURATION;
Enabled.

Show the configuration.

Use the SHOW command to verify that the configuration and its databases were successfully enabled:

DGMGRL> SHOW CONFIGURATION;

DGMGRL returns the following information:

Configuration - DRSolution
 
  Protection Mode: MaxPerformance
  Databases:
    North_Sales  - Primary database
      South_Sales  - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

Enable the database.

This step is unnecessary except if the standby database was previously disabled with the DISABLE DATABASE command. Normally, enabling the configuration also enables the standby database.

DGMGRL> ENABLE DATABASE 'South_Sales';
Enabled.

Show the database.

DGMGRL> SHOW DATABASE 'South_Sales';

Database - South_Sales
 
  Role: PHYSICAL STANDBY
  Intended State: APPLY-ON
  Transport Lag: 0 seconds (computed 1 second ago)
  Apply Lag: 0 seconds (computed 1 second ago)
  Apply Rate: 1.54 MByte/s
  Real Time Query: OFF
  Instance(s):
    south_sales1
 
Database Status:
SUCCESS

6.5 Scenario 4: Setting the Configuration Protection Mode

You can change the protection mode of the configuration at any time. However, it is best if you do this when there is no activity occurring in the configuration if you are moving to the maximum protection or maximum availability modes.

This scenario sets the protection mode of the configuration to the MAXAVAILABILITY mode. Note that this protection mode requires that there be at least one standby configured to use standby redo log files and it must receive redo via SYNC or FASTSYNC mode if it receives redo directly from the primary database. If the standby receives redo via a far sync instance, then the far sync instance must receive redo via SYNC or FASTYSYNC mode and the standby must receive redo from the far sync instance via ASYNC mode.

Step 1   Configure standby redo log files, if necessary

Because you will be setting the protection mode to the MAXAVAILABILITY mode, it is important to ensure that sufficient standby redo log files are configured on the standby database.

Step 2   Configure redo transport mode appropriately.

Configure the standby to receive redo via SYNC or FASTSYNC mode, if the standby receives redo directly from the primary database. If the standby receives primary redo via a far sync instance, then configure the far sync instance to receive redo via SYNC or FASTSYNC mode and configure the standby to receive redo via ASYNC mode. For example:

DGMGRL> EDIT DATABASE 'South_Sales' SET PROPERTY 'LogXptMode'='SYNC';
Property "LogXptMode" updated

The broker will not allow this command to succeed unless the standby database is configured with standby redo log files in the configuration.

Step 3   Change the overall protection mode for the configuration.

Use the EDIT CONFIGURATION command to upgrade the broker configuration to the MAXAVAILABILITY protection mode:

DGMGRL> EDIT CONFIGURATION SET PROTECTION MODE AS MAXAVAILABILITY;
Succeeded.

If the configuration is disabled when you enter this command, the actual protection mode change is not applied until you enable the configuration with the ENABLE CONFIGURATION command. The broker will not allow you to enable the configuration if it does not find a standby database in the configuration that can support the requirements of the protection mode.

Step 4   Verify the protection mode has changed.

Use the SHOW CONFIGURATION command to display the current protection mode for the configuration:

DGMGRL> SHOW CONFIGURATION;
 
Configuration - DRSolution
 
  Protection Mode: MaxAvailability
  Databases:
    North_Sales  - Primary database
      South_Sales  - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

See Also:

Section 4.6, "Managing Data Protection Modes"

6.6 Scenario 5: Setting up Maximum Availability Mode with a Far Sync Instance

A far sync instance can be used with maximum availability protection mode if the primary and standby database are geographically far enough apart to make the use of synchronous transport mode impractical. The example in this section shows how to add a far sync instance to the configuration and then set up the RedoRoutes property for all members of the configuration. Setting of RedoRoutes property for the far sync instance enables it to send redo data based on either the North_Sales or South_Sales database being the primary.

Step 1   Add the far sync instance to the configuration

Issue the following commands to add the far sync instance named FS to the broker configuration:

DGMGRL> ADD FAR_SYNC 'FS' AS CONNECT IDENTIFIER IS FS.EXAMPLE.COM;
DGMGRL> ENABLE FAR_SYNC 'FS';
DGMGRL> SHOW CONFIGURATION;
  
Configuration - DRSolution
 
  Protection Mode: MaxPerformance
  Databases:
  North_Sales - Primary database
    South_Sales - Physical standby database
    FS - Far Sync
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

In this output, South_Sales and FS are indented under North_Sales. This indicates that the primary is sending redo data to both South_Sales and FS.

Step 2   Set up the RedoRoutes property for North_Sales and FS

Issue the following commands so that the current primary (North_Sales) sends redo data to the far sync instance only, and the far sync instance sends redo data to South_Sales when North_Sales is a primary:

DGMGRL> EDIT DATABASE 'North_Sales' SET PROPERTY RedoRoutes='(LOCAL : FS SYNC)';
DGMGRL> EDIT FAR_SYNC 'FS' SET PROPERTY RedoRoutes='(North_Sales : South_Sales');
DGMGRL> SHOW CONFIGURATION;
 
DGMGRL> SHOW CONFIGURATION;
 
Configuration - DRSolution
 
  Protection Mode: MaxPerformance
  Databases:
  North_Sales - Primary database
    FS - Far Sync
      South_Sales - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

The indentation scheme in the output above indicates that North_Sales sends redo data to FS and FS sends redo data to South_Sales.

Step 3   Upgrade to maximum availability protection mode

Issue the following commands to upgrade the protection mode to maximum availability:

DGMGRL> EDIT CONFIGURATION SET PROTECTION MODE AS MaxAvailability;
DGMGRL> SHOW CONFIGURATION;
 
Configuration - DRSolution
 
  Protection Mode: MaxAvailability
  Databases:
  North_Sales - Primary database
    FS - Far Sync
      South_Sales - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

Actions Required When Roles Change

If South_Sales ever becomes the primary and North_Sales becomes a standby database that receives redo data from the far sync instance, FS, then the RedoRoutes property must be set for South_Sales. Additionally, a new rule must be included in the RedoRoutes property for FS to allow it to send redo data to North_Sales when South_Sales is the primary database.

If these rules are not created, then North_Sales will not be able to receive redo data. To check whether the proper rules have been set, use the SHOW CONFIGURATION WHEN PRIMARY IS command to see what the redo transport configuration would be if South_Sales were the primary database. For example:

DGMGRL> SHOW CONFIGURATION WHEN PRIMARY IS 'South_Sales';
 
Configuration when South_Sales is primary - DRSolution
 
  Members:
  South_Sales  - Primary database
    FS     - Far Sync 
      North_Sales  - Physical standby database 
 
  Members Not Receiving Redo:
  North_FS - Physical standby database
  Warning: ORA-16685: database does not receive redo data

To correct this error, set the RedoRoutes property for South_Sales and FS as follows:

DGMGRL> EDIT DATABASE 'South_Sales' SET PROPERTY RedoRoutes='(LOCAL : FS SYNC)';
DGMGRL> EDIT FAR_SYNC 'FS' SET PROPERTY RedoRoutes=('North_Sales : South_Sales)(South_Sales : North_Sales)';

After the change to the RedoRoutes property for South_Sales and FS is complete, use the SHOW CONFIGURATION WHEN PRIMARY IS command to confirm that the error has been cleared:

DGMGRL> SHOW CONFIGURATION WHEN PRIMARY IS 'South_Sales';
 
Configuration when South_Sales is primary - DRSolution
 
  Members:
  South_Sales  - Primary database
    FS     - Far Sync 
      North_Sales  - Physical standby database 

6.7 Scenario 6: Enabling Fast-Start Failover and Starting the Observer

You can enable fast-start failover from any site, including the observer site, while connected to any database in the broker configuration. Enabling fast-start failover does not trigger a failover. Instead, it allows the observer that is monitoring the configuration to initiate a fast-start failover if conditions warrant a failover.

Fast-start failover can be enabled in configurations operating in either maximum performance or maximum availability protection mode. This section describes the steps to enable fast-start failover and start the observer where the configuration protection mode is set to maximum availability mode.

Step 1   Ensure standby redo logs are configured on the primary and target standby databases.

Standby redo logs must be configured on the primary and standby databases. You must stop log apply services prior to configuring standby redo logs.

See Also:

Oracle Data Guard Concepts and Administration for instructions on configuring standby redo log files

Step 2   Configure redo transport mode appropriately.

Configure the primary and standby databases to receive redo via SYNC or FASTSYNC mode, if they receive redo directly rather than via a far sync instance. If either database receives redo via a far sync instance, then configure the far sync instance to receive redo via SYNC or FASTSYNC mode and configure the database to receive redo via ASYNC mode. For example:

DGMGRL> EDIT DATABASE 'North_Sales' SET PROPERTY 'LogXptMode'='SYNC';
Property "LogXptMode" updated

DGMGRL> EDIT DATABASE 'South_Sales' SET PROPERTY 'LogXptMode'='SYNC';
Property "LogXptMode" updated

The broker does not allow these commands to succeed unless the databases are configured with standby redo log files.

Step 3   Set the FastStartFailoverTarget configuration property.

If you have two or more standby databases, set up the FastStartFailoverTarget configuration property on the primary database to indicate the desired target standby database. The broker reciprocally sets this property for the target standby database to indicate the primary database as its future target standby database when fast-start failover is actually enabled. There is no need for you set this property on the target standby as this is done for you automatically.For example:

DGMGRL> EDIT DATABASE 'North_Sales' SET PROPERTY FastStartFailoverTarget='South_Sales';
Property "FastStartFailoverTarget" updated

Step 4   Upgrade the protection mode to MAXAVAILABILITY, if necessary.

If it is necessary to upgrade the protection mode, use the following DGMGRL EDIT CONFIGURATION command. For example:

DGMGRL> EDIT CONFIGURATION SET PROTECTION MODE AS MAXAVAILABILITY;

Step 5   Enable Flashback Database on the primary and target standby databases, if necessary.

If it is not already enabled on the primary and standby databases, enable Flashback Database by issuing the following statements on each database:

ALTER SYSTEM SET UNDO_RETENTION=3600 SCOPE=SPFILE;
ALTER SYSTEM SET UNDO_MANAGEMENT='AUTO' SCOPE=SPFILE;
SHUTDOWN IMMEDIATE;
STARTUP MOUNT;
SHOW PARAMETER UNDO;
ALTER SYSTEM SET DB_FLASHBACK_RETENTION_TARGET=4320 SCOPE=BOTH;
ALTER DATABASE ARCHIVELOG;
ALTER SYSTEM SET db_recovery_file_dest_size=<size>;
ALTER SYSTEM SET db_recovery_file_dest=<directory-specification>;
ALTER DATABASE FLASHBACK ON;
ALTER DATABASE OPEN;

Ensure the UNDO_RETENTION and DB_FLASHBACK_RETENTION_TARGET initialization parameters are set to sufficiently large values so that reinstatement is still possible after a prolonged outage.

Step 6   Start the observer.

Start the observer by logging into the observer computer and running DGMGRL. Connect to the configuration as a user who has the SYSDG or SYSDBA privilege and then issue the START OBSERVER command. Note that the command does not return; that is you will not get the DGMGRL prompt after issuing the command.

DGMGRL> CONNECT sysdg@North_Sales.example.com;
Password: password
Connected as SYSDG.
DGMGRL> START OBSERVER;
Observer started

When starting the observer interactively, Oracle recommends that connection credentials not be supplied as command line parameters to the DGMGRL connect command. This practice prevents other users on the system from using a utility (for example, the UNIX ps utility) to display connection credentials. It also prevents clear-text passwords from being visible on the user's terminal.

When starting the observer from a script, Oracle recommends that you use a method that supports 'connect /', so that database connection credentials do not have to be embedded within the script. If you choose to use a client-side Oracle Wallet as a secure external password store (see Oracle Database Enterprise User Security Administrator's Guide), be sure to add credentials for both the primary and fast-start failover target standby databases. The database connect string that you specify when adding the credentials for each database must match the ObserverConnectIdentifer or DGConnectIdentifier database property.

Step 7   Enable fast-start failover.

You can enable fast-start failover while connected to any database system in the broker configuration. For example:

DGMGRL> ENABLE FAST_START FAILOVER;
Enabled.

Step 8   Verify the fast-start failover configuration.

Use the SHOW FAST_START FAILOVER command to display the fast-start failover settings:

DGMGRL> SHOW FAST_START FAILOVER;
 
Fast-Start Failover: DISABLED
 
  Threshold: 30 seconds
  Target: South_Sales
  Observer: observer.example.com
  Lag Limit: 30 seconds (not in use)
  Shutdown Primary: TRUE
  Auto-reinstate: TRUE
  Observer Reconnect: (none)
  Observer Override: FALSE
 
Configurable Failover Conditions
 Health Conditions:
   Corrupted Controlfile YES
  Corrupted Dictionary YES
  Inaccessible Logfile NO
  Stuck Archiver NO
  Datafile Offline YES
 
Oracle Error Conditions:
(none)

The following commands show that the FastStartFailoverTarget property is set up reciprocally once fast-start failover is enabled. The first command, issued for the current primary database North_Sales, shows the value of the FastStartFailoverTarget property to be the current target standby, South_Sales. The second command, issued for the target standby South_Sales, shows the current primary, North_Sales, as the target standby's future target standby should it ever take over as a primary.

DGMGRL> SHOW DATABASE 'North_Sales' FastStartFailoverTarget;
 FastStartFailoverTarget='South_Sales';
 
DGMGRL> SHOW DATABASE 'South_Sales' FastStartFailoverTarget;
 FastStartFailoverTarget='North_Sales';

6.8 Scenario 7: Enabling Fast-Start Failover When a Far Sync Instance Is In Use

Fast-start failover can be enabled in maximum availability mode when the fast-start failover target is a logical or physical standby database that receives redo data from a far sync instance.

To enable fast-start failover when a far sync instance is used to ship redo data to the standby database, the FastStartFailoverTarget property must first be set on both the primary and target standby database, as follows:

DGMGRL> EDIT DATABASE 'North_Sales' SET PROPERTY FastStartFailoverTarget='South_Sales';
DGMGRL> EDIT DATABASE 'South_Sales' SET PROPERTY FastStartFailoverTarget='North_Sales';

Then, fast-start failover can be enabled, as follows:

DGMGRL> ENABLE FAST_START FAILOVER;

Note that the far sync instance database is not specified as the fast-start failover target for either North_Sales or South_Sales.

6.9 Scenario 8: Performing Routine Management Tasks

There may be situations in which you want to change the state or properties of the databases in a broker configuration to perform routine maintenance on one or more databases. You might also need to temporarily disable broker management of databases in a configuration.

DGMGRL> EDIT DATABASE 'North_Sales' SET PROPERTY 'LogXptMode'=ASYNC;

Property "LogXptMode" updated 

EDIT DATABASE 'North_Sales' RESET PROPERTY LogXptMode;

EDIT CONFIGURATION RESET PROPERTY TraceLevel;

DGMGRL> EDIT DATABASE 'South_Sales' SET STATE='APPLY-OFF';
Succeeded.

DGMGRL> EDIT DATABASE North_Sales SET STATE=TRANSPORT-OFF;
 
Succeeded.

DGMGRL> EDIT DATABASE North_Sales SET STATE=TRANSPORT-ON;
 
Succeeded.

DGMGRL> DISABLE CONFIGURATION;

DGMGRL> DISABLE DATABASE 'South_Sales';
Disabled.

DGMGRL> DISABLE FAR_SYNC 'FS';
Disabled.

6.10 Scenario 9: Performing a Switchover Operation

You can switch the role of the primary database and a standby database using the SWITCHOVER command. Before you issue the SWITCHOVER command, you must ensure:

• The state of the primary and standby databases are TRANSPORT-ON and APPLY-ON, respectively.

• All participating databases are in good health, without any errors or warnings present.

• The standby database properties were set on the primary database, so that the primary database can function correctly when transitioning to a standby database (shown in the following examples in boldface type).

• Standby redo log files are configured on the primary database.

• If the configuration is in maximum availability mode, then the current primary is configured to receive redo via SYNC or FASTSYNC mode if it will receive redo directly from the new primary. If it will receive redo via a far sync instance, then the far sync instance is configured to receive redo via SYNC or FASTSYNC mode and the current primary is configured to receive redo via ASYNC mode. If the configuration is in maximum protection mode, then the current primary is configured to receive redo via SYNC mode.

• If fast-start failover is enabled, you can perform a switchover only to the standby database that was specified as the target standby database.

Step 1   Check the primary database.

Use the SHOW DATABASE VERBOSE command to check the state, health, and properties of the primary database, as follows:

DGMGRL> SHOW DATABASE VERBOSE 'North_Sales';
 
Database - North_Sales
 
  Role:            PRIMARY
  Intended State:  TRANSPORT-ON
  Instance(s):
    north_sales1
 
  Properties:
    DGConnectIdentifier             = 'North_Sales.example.com'
    ObserverConnectIdentifier       = ''
    LogXptMode                      = 'ASYNC'
    RedoRoutes                      = ''
    DelayMins                       = '0'
    Binding                         = 'optional'
    MaxFailure                      = '0'
    MaxConnections                  = '1'
    ReopenSecs                      = '300'
    NetTimeout                      = '30'
    RedoCompression                 = 'DISABLE'
    LogShipping                     = 'ON'
    PreferredApplyInstance          = ''
    ApplyInstanceTimeout            = '0'
    ApplyLagThreshold               = '0'
    TransportLagThreshold           = '0'
    TransportDisconnectedThreshold  = '0'
    ApplyParallel                   = 'AUTO'
    StandbyFileManagement           = 'AUTO'
    ArchiveLagTarget                = '0'
    LogArchiveMaxProcesses          = '5'
    LogArchiveMinSucceedDest        = '1'
    DbFileNameConvert               = 'dbs/bt, dbs/t, dbs/ct, dbs/t'
    LogFileNameConvert              = 'dbs/bt, dbs/t, dbs/ct, dbs/t'
    FastStartFailoverTarget         = 'South_Sales'
    InconsistentProperties          = '(monitor)'
    InconsistentLogXptProps         = '(monitor)'
    SendQEntries                    = '(monitor)'
    LogXptStatus                    = '(monitor)'
    RecvQEntries                    = '(monitor)'
    StaticConnectIdentifier         = 'DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=north_sales1.example.com)(PORT=2840))
(CONNECT_DATA=(SERVICE_NAME=North_Sales_DGMGRL.example.com)(INSTANCE_NAME=north_sales1)(SERVER=DEDICATED)))'
    StandbyArchiveLocation          = 'USE_DB_RECOVERY_FILE_DEST'
    AlternateLocation               = ''
    LogArchiveTrace                 = '255'
    LogArchiveFormat                = 'db1r_%d_%t_%s_%R.arc'
    TopWaitEvents                   = '(monitor)'
 
Database Status:
SUCCESS

In particular, you should examine the boldface properties and the current status of the primary database. See Chapter 4 for information about managing databases.

Step 2   Check the standby database that is the target of the switchover.

Use the SHOW DATABASE command to check the status of the standby database that is the target of the switchover. For example:

DGMGRL> SHOW DATABASE 'South_Sales';
 
Database - South_Sales
 
  Role: PHYSICAL STANDBY
  Intended State: APPLY-ON
  Transport Lag: 0 seconds (computed 0 seconds ago)
  Apply Lag: 0 seconds (computed 0 seconds ago)
  Apply Rate: 1.44 MByte/s
  Real Time Query: OFF
  Instance(s):
    south_sales1
 
Database Status:
SUCCESS

Step 3   Confirm that the database is ready for a role change.

Prior to performing a role change, you can use the VALIDATE DATABASE command to perform an exhaustive set of checks on the database to confirm that it is ready for a role change.

The examples shown in this step use the VALIDATE DATABASE command for all three databases in the DRSolution configuration: a primary, logical standby, and physical standby database. The configuration looks as follows:

DGMGRL> SHOW CONFIGURATION;
 
Configuration - DRSolution
 
  Protection Mode: MaxAvailability
  Databases:
  North_Sales - Primary database
    West_Sales - Logical standby database
    South_Sales - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

Example 6-10 Validate the Primary Database

Validate the primary database, as follows:

DGMGRL> VALIDATE DATABASE 'North_Sales';
 
  Database Role:    Primary database
 
  Ready for Switchover:  Yes

Example 6-11 Validate the Logical Standby Database

Validate the logical standby database, as follows:

DGMGRL> VALIDATE DATABASE 'West_Sales';
 
  Database Role:     Logical standby database
  Primary Database:  North_Sales
 
  Ready for Switchover:  Yes
  Ready for Failover:    Yes (Primary Running)
    Warning: Physical and snapshot standby databases will
    be disabled if a role change is performed to this database

Example 6-12 Validate the Physical Standby Database

Validate the physical standby database, as follows:

DGMGRL> VALIDATE DATABASE 'South_Sales';
 
  Database Role:     Physical standby database
  Protection Mode:   MaxAvailability
    Error: Switchover to this standby is not possible since there
    are no other standbys that can support the protection mode
 
  Primary Database:  North_Sales
 
  Ready for Switchover:  No
  Ready for Failover:    Yes (Primary Running)
 
  Transport-Related Property Settings:
    Property                        North_Sales Value        South_Sales Value
    LogXptMode                      ASYNC                    SYNC

Because the configuration protection is set to maximum availability mode and database South_Sales is the only one that has its LogXptMode property set to SYNC, an error is displayed to indicate that a switchover is not possible. Note that when a database receives redo from a database or a far sync instance that has the RedoRoutes property configured with a transport mode, that mode overrides the transport mode specified by LogXptMode.

Step 4   Issue the switchover command.

Issue the SWITCHOVER command to swap the roles of the primary and standby databases. The following example shows how the broker automatically shuts down and restarts the old primary database as a part of the switchover. (See the usage notes in Section 7.1.3 for information about how to set up the broker environment so that DGMGRL can automatically restart the primary and standby databases for you.)

DGMGRL> SWITCHOVER TO 'South_Sales';
Performing switchover NOW, please wait...
Operation requires a connection to instance "south_sales1" on database "South_Sales"
Connecting to instance "south_sales1"...
Connected as SYSDBA.
New primary database "South_Sales" is opening...
Operation requires startup of instance "north_sales1" on database "North_Sales"
Starting instance "north_sales1"...
ORACLE instance started.
Database mounted.
Switchover succeeded, new primary is "South_Sales"

After the switchover completes, use the SHOW CONFIGURATION and SHOW DATABASE commands to verify that the switchover operation was successful.

Step 5   Show the configuration.

Issue the SHOW CONFIGURATION command to verify that the switchover was successful.

DGMGRL> SHOW CONFIGURATION;
 
Configuration - DRSolution
 
  Protection Mode: MaxAvailability
  Databases:
    South_Sales  - Primary database
      West_Sales  - Logical standby database
      North_Sales - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

6.11 Scenario 10: Performing a Manual Failover Operation

You invoke a failover operation in response to an emergency situation, usually when the primary database cannot be accessed or is unavailable. See Section 5.2 before you fail over to decide which standby database should be the target of the failover. The following scenario describes a failover to the remote database called South_Sales.

Step 1   (Optional) Check the readiness of the target standby.

To validate the target standby database to ensure that it's ready to become the new primary database, use the VALIDATE DATABASE command, as shown in the following example:

DGMGRL> VALIDATE DATABASE 'South_Sales';
 
Database Role: Physical standby database
Primary Database: South_Sales
 
Ready for Switchover: Yes
Ready for Failover: Yes

Step 2   Connect to the target standby database.

To perform the failover operation, you must connect to the standby database to which you want to fail over as a user that has the SYSDG or SYSDBA privilege. For example:

DGMGRL> CONNECT sysdg@South_Sales.example.com;
Password: password
Connected as SYSDG.

Step 3   Issue the failover command.

Now you can issue the failover command to make the target standby database the new primary database for the configuration.

DGMGRL> FAILOVER TO 'South_Sales';
Performing failover NOW, please wait...
Failover succeeded, new primary is "South_Sales"

Note that after the failover completes, the original primary database cannot be used as a standby database of the new primary database unless it is reinstated or re-created (as described in Section 5.4.3).

Step 4   Show the configuration.

Issue the SHOW CONFIGURATION command to verify the failover.

DGMGRL> SHOW CONFIGURATION;
 
Configuration - DRSolution
 
  Protection Mode: MaxAvailability
  Databases:
    South_Sales  - Primary database
      Warning: ORA-16629: database reports a different protection level from the protection mode
 
      North_Sales - Physical standby database (disabled)
        ORA-16661: the standby database needs to be reinstated
 
Fast-Start Failover: DISABLED
 
Configuration Status:
WARNING

Note that in this example, the configuration was operating in maximum availability mode. The protection mode was preserved after the failover. The configuration also has a warning status. The SHOW DATABASE command for the new primary shows that the warning is the result of not having an enabled physical standby database. As a result, the warning status indicates that the protection level of the configuration is not the same as the configured mode.

Step 5   Show the new primary database.

DGMGRL> SHOW DATABASE South_Sales;
Database - South_Sales
 
  Role:            PRIMARY
  Intended State:  TRANSPORT-ON
  Instance(s):
    south_sales1
 
  Database Warning(s):
    ORA-16629: database reports a different protection level from the protection mode
 
Database Status:
WARNING

Step 6   Show the old primary database.

Issue the SHOW DATABASE command to see that the former (failed) primary database was disabled by the broker as a consequence of the failover. It must be reinstated (as described in Section 5.4.3).

DGMGRL> SHOW DATABASE 'North_Sales';
Database - North_Sales
 
  Role: PHYSICAL STANDBY
  Intended State: APPLY-ON
  Transport Lag: (unknown)
  Apply Lag: (unknown)
  Apply Rate: (unknown)
  Real Time Query: OFF
  Instance(s):
    north_sales1
 
Database Status:
ORA-16661: the standby database needs to be reinstated

6.12 Scenario 11: Reinstating a Failed Primary Database

If your former primary database was configured with Flashback Database, you can easily reinstate the failed primary database as a standby database of the new primary database. The failed primary database will be reinstated as a standby type that matches the old standby database. For example, if you failed over to a physical standby database, the old primary will be reinstated as a physical standby database.

To reinstate the failed primary database, start it to the mounted state. Then run DGMGRL, connect to the new primary database and reinstate the old primary database.

Step 1   Restart the Old Primary Database.

% dgmgrl connect sysdg
Password: password
Connected as SYSDG.
DGMGRL> startup mount;
ORACLE instance started.
Database mounted.

Step 2   Reinstate the old primary database.

% dgmgrl connect sysdg
Password: password
Connected as SYSDG.
DGMGRL> REINSTATE DATABASE 'North_Sales';
Reinstating database "North_Sales", please wait...
Reinstatement of database "North_Sales" succeeded
 
Database dismounted.
ORACLE instance shut down.
Operation requires startup of instance "north_sales1" on database "North_Sales"
Starting instance "north_sales1"...
ORACLE instance started.
Database mounted.
Continuing to reinstate database "North_Sales" ...
Reinstatement of database "North_Sales" succeeded

After the primary has been reinstated, issue the SHOW CONFIGURATION and SHOW DATABASE commands to confirm that the old primary has been successfully reinstated.

Step 3   Show the Configuration and Databases.

DGMGRL> SHOW CONFIGURATION
 
Configuration - DRSolution
 
  Protection Mode: MaxAvailability
  Databases:
    South_Sales  - Primary database
      North_Sales  - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

DGMGRL> SHOW DATABASE 'South_Sales';
 
Database - South_Sales
 
  Role:            PRIMARY
  Intended State:  TRANSPORT-ON
  Instance(s):
    south_sales1
 
Database Status:
SUCCESS
 
DGMGRL> SHOW DATABASE 'North_Sales'
 
Database - 'North_Sales'
 
  Role: PHYSICAL STANDBY
  Intended State: APPLY-ON
  Transport Lag: 0 seconds (computed 1 second ago)
  Apply Lag: 0 seconds (computed 1 second ago)
  Apply Rate: 0 Byte/s
  Real Time Query: OFF
  Instance(s):
    north_sales1
 
Database Status:
SUCCESS

6.13 Scenario 12: Converting a Physical Standby to a Snapshot Standby

If you have a physical standby database that you would like to convert to a snapshot standby database, use the DGMGRL CONVERT DATABASE command. Redo data will continue to be received by the database while it is operating as a snapshot standby database, but it will not be applied until the snapshot standby is converted back into a physical standby database.

A physical standby database must be configured with a fast recovery area to convert it to a snapshot standby database. This is because a guaranteed restore point is created during the conversion process, and guaranteed restore points require a fast recovery area.

DGMGRL> convert database 'South_Sales' to snapshot standby;
Converting database "South_Sales" to a Snapshot Standby database, please wait...
Database "South_Sales" converted successfully

DGMGRL> SHOW CONFIGURATION;
 
Configuration - DRSolution
 
  Protection Mode: MaxPerformance
  Databases:
    North_Sales  - Primary database
      South_Sales  - Snapshot standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
SUCCESS

When you are ready to revert the database back to a physical standby database, use the DGMGRL CONVERT DATABASE command again as follows. Any updates made to the database while it was operating as a snapshot standby database will be discarded. All accumulated redo data will be applied by Redo Apply services after the database is converted back to a physical standby database.

DGMGRL> CONVERT DATABASE 'South_Sales' to PHYSICAL STANDBY;
 
Converting database "South_Sales" to a Physical Standby database, please wait...
Operation requires shutdown of instance "south_sales1" on database "South_Sales"
Shutting down instance "south_sales1"...
Database closed.
Database dismounted.
ORACLE instance shut down.
Operation requires startup of instance "south_sales1" on database "South_Sales"
Starting instance "south_sales1"...
ORACLE instance started.
Database mounted.
Continuing to convert database "South_Sales" ...
Database "South_Sales" converted successfully

6.14 Scenario 13: Monitoring a Data Guard Configuration

The scenario in this section demonstrates how to use the SHOW command and monitorable properties to identify and resolve a failure situation.

Step 1   Check the configuration status.

The status of the broker configuration is an aggregated status of all databases and instances in the broker configuration. You can check the configuration status first to determine whether or not any further action needs to be taken. If the configuration status is SUCCESS, everything in the broker configuration is working properly. However, if you see a status of WARNING or ERROR, then something is wrong in the configuration.

For example, in the following display, you can see that the primary database has multiple warnings:

DGMGRL> SHOW CONFIGURATION;
Configuration - DRSolution
 
  Protection Mode: MaxPerformance
  Databases:
    North_Sales  - Primary database
      Warning: ORA-16809: multiple warnings detected for the database
 
      South_Sales  - Physical standby database
 
Fast-Start Failover: DISABLED
 
Configuration Status:
WARNING

Step 2   Check the database status.

To identify the warnings on the primary database, show its status using the SHOW DATABASE command:

DGMGRL> SHOW DATABASE 'North_Sales';
Database - North_Sales
 
  Role:            PRIMARY
  Intended State:  TRANSPORT-ON
  Instance(s):
    north_sales1
      Warning: ORA-16737: the redo transport service for standby "South_Sales" has an error
      Warning: ORA-16714: the value of property LogArchiveTrace is inconsistent with the database setting
      Warning: ORA-16715: redo transport-related property ReopenSecs of standby
 database "South_Sales" is inconsistent

Database Status:
WARNING

Step 3   Check the LogXptStatus monitorable property.

The SHOW DATABASE output in step 2 shows a Warning for error ORA-16737. To identify the exact transport error, use the LogXptStatus monitorable property:

DGMGRL> SHOW DATABASE 'North_Sales' 'LogXptStatus';
LOG TRANSPORT STATUS
PRIMARY_INSTANCE_NAME  STANDBY_DATABASE_NAME  STATUS 
         north_sales1            South_Sales  ORA-12541: TNS:no listener

The output shows that the listener for the physical standby database is not running. To fix this error, start the listener for the physical standby database South_Sales.

Step 4   Check the InconsistentProperties monitorable property.

The SHOW DATABASE output in step 2 also shows a Warning for error ORA-16714. To identify the exact error, use the InconsistentProperties monitorable property:

DGMGRL> SHOW DATABASE 'North_Sales' 'InconsistentProperties';
INCONSISTENT PROPERTIES
 INSTANCE_NAME  PROPERTY_NAME   MEMORY_VALUE SPFILE_VALUE BROKER_VALUE
 north_sales1   LogArchiveTrace 511          255          255

The current database memory value (511) is different from both the server parameter file (SPFILE) value (255) and the Oracle Data Guard broker's property value (255). If you decide the database memory value is correct, you can update the Oracle Data Guard broker's property value using the following command:

DGMGRL> EDIT DATABASE 'North_Sales' SET PROPERTY 'LogArchiveTrace'=511;
Property "LogArchiveTrace" updated

This command will result in the broker updating the SPFILE value so that the value of LogArchiveTrace is kept consistent.

Step 5   Check the InconsistentLogXptProps monitorable property.

Another warning shown in the SHOW DATABASE display in Step 2 is ORA-16715. To identify the inconsistent values for the redo transport database property, ReopenSecs, you can use the InconsistentLogXptProps monitorable property:

DGMGRL> SHOW DATABASE 'North_Sales' 'InconsistentLogXptProps';
INCONSISTENT LOG TRANSPORT PROPERTIES
 INSTANCE_NAME  STANDBY_NAME  PROPERTY_NAME  MEMORY_VALUE  BROKER_VALUE 
   south_sales1   South_Sales     ReopenSecs           600           300

The current database memory value (600) is different from the Oracle Data Guard broker's property value (300). If you think the broker's property value is correct, you can fix the inconsistency by re-editing the property of the standby database with the same value, as shown in the following example:

DGMGRL> EDIT DATABASE 'South_Sales' SET PROPERTY 'ReopenSecs'=300;
Property "ReopenSecs" updated

You can also reenable the standby database or reset the state of the primary database to TRANSPORT-ON to fix this inconsistency.