Bringing up the recovery or primary TrueSight Infrastructure Management Server in application high-availability and disaster recovery mode


Bringing up the recovery site TrueSight Infrastructure Management servers

Stop the primary site Infrastructure Management servers if they are running.

Perform the following steps to bring up the recovery site TrueSight Infrastructure Management servers.

Step 1: Set the DNS server

If you have registered the alias on the DNS server, make the appropriate changes so that the DNS alias redirects to the recovery site TrueSight Infrastructure Management servers.

Step 2: Transfer the latest \pw files

If no storage replication system is used, transfer the latest installedDirectory\pw files from the primary site servers or backup location to the corresponding recovery site servers (from computer A to A_DR and from computer B to B_DR.) 

Note

BMC recommends that you use any third-party file synchronization and backup software to copy the installedDirectory\pw directory because traditional copying can result in any unexpected failure.

Step 3: Update the mcell.dir entry

At the recovery site, on both primary and secondary Infrastructure Management servers, update the $BPPM_SERVER_HOME/pw/server/etc/mcell.dir entry with the recovery site, primary Presentation Server host name. 

For example:

gateway.pn_serverts_event_gatewaymc<Primary Presentation Server host name>:<port number>

Step 4: Transfer the latest database backup files

If using Oracle Data Guard:

To make the standby Oracle database in active state, an Oracle DBA must switchover from the primary to the standby Oracle database.

If not using Oracle Data Guard:

  1. If no database replication system is used, transfer the latest database backup files from the backed-up source to the recovery database server. 
  2. On the recovery database server, restore the database files using either the database replication tool or the database replication solution.
  3. Start the recovery Oracle database instance.

Step 5: Run the pw dbconfig command

At the recovery site, run the pw dbconfig set command on both primary and secondary Infrastructure Management servers to point to the active Oracle database. For example:

pw dbconfig set -database oracle -host Recovery_Oracle_hostname:1521 -sid ORA12C -username db_user_pronet_tsim_a -password <password> -repusername db_user_report_tsim_a -reppassword <password>

For more information about the command options, see pw-dbconfig.

Step 6: Start the recovery site TrueSight Infrastructure Management servers

  1. Run the pw sys start command to start the primary Infrastructure Management server processes.
    Ensure that the primary server has successfully started by running the pw license list command.
  1. Run the pw sys start command to start the secondary Infrastructure Management server processes.
    Ensure that the secondary server has successfully started by running the pw license list command. 

Step 7: Configure the remote event cells

This step enables the remote event cells to retrieve the IP address of the recovery TrueSight Infrastructure Management Server. This is only required if you have registered the alias on the DNS server.

  1. To flush and reset the cache, open a command prompt on a remote event cell and run the following command:
    ipconfig /flushdns
  2. Restart the remote cell.
  3. Configure all the other remote event cells using the same procedure.

Step 8: Run the pw alarmsync command

At the recovery site, run the pw alarmsync command on the primary Infrastructure Management server to synchronize the database and the impact management cell database.

Bringing back the primary site TrueSight Infrastructure Management servers

After you fix the primary site TrueSight Infrastructure Management servers, you can switch the operations from the recovery site back to the primary site.

Stop the recovery site Infrastructure Management servers if they are running.

Step 1: Set the DNS server

If you have registered the alias on the DNS server, make the appropriate changes so that the DNS alias redirects to the primary site TrueSight Infrastructure Management servers.

Step 2: Transfer the latest \pw files

If no storage replication system is used, transfer the latest installedDirectory\pw files from the recovery site servers to the corresponding primary site servers (from computer A_DR to A and from computer B_DR to B.) 

Note

BMC recommends that you use any third-party file synchronization and backup software to copy the installedDirectory\pw directory because traditional copying can result in any unexpected failure.

Step 3: Update the mcell.dir entry

At the primary site, on both primary and secondary Infrastructure Management servers, update the $BPPM_SERVER_HOME/pw/server/etc/mcell.dir entry with the primary site, primary Presentation Server host name. For example:

gateway.pn_serverts_event_gatewaymc<Primary Presentation Server host name>:<port number>


Step 4: Stop the recovery site Infrastructure Management servers

  1. Run the pw sys stop command to stop the secondary Infrastructure Management server processes at the recovery site.
    Ensure that all internal processes are stopped by using the pw process list command.
  1. Run the pw sys stop command to stop the primary Infrastructure Management server processes at the recovery site.
    Ensure that all internal processes are stopped by using the pw process list command.

Step 5: Transfer the latest database backup files

If using Oracle Data Guard:

To make the primary Oracle database in active state, an Oracle DBA must switchover from the standby to the primary Oracle database.

If not using Oracle Data Guard:

  1. If no database replication system is used, transfer the latest database backup files from the backed-up source to the primary database server. 
  2. On the primary database server, restore the database files using either the database replication tool or the database replication solution.
  3. Start the primary Oracle database instance.

Step 6: Run the pw dbconfig command

At the primary site, run the pw dbconfig set command on both primary and secondary Infrastructure Management servers to point to the active Oracle database. For example:

pw dbconfig set -database oracle -host Primary_Oracle_hostname:1521 -sid ORA12C -username db_user_pronet_tsim_a -password <password> -repusername db_user_report_tsim_a -reppassword <password>

For more information about the command options, see pw-dbconfig.

Step 7: Start the primary site TrueSight Infrastructure Management servers

  1. Run the pw sys start command to start the primary Infrastructure Management server processes.
    Ensure that the primary server has successfully started by running the pw license list command.
  1. Run the pw sys start command to start the secondary Infrastructure Management server processes.
    Ensure that the secondary server has successfully started by running the pw license list command.

Step 8: Configure the remote event cells

This step enables the remote event cells to retrieve the IP address of the recovery TrueSight Infrastructure Management Server. This is only required if you have registered the alias on the DNS server.

  1. To flush and reset the cache, open a command prompt on a remote event cell and run the following command:
    ipconfig /flushdns
  2. Restart the remote cell.
  3. Configure all the other remote event cells using the same procedure.

Step 9: Run the pw alarmsync command

At the primary site, run the pw alarmsync command on the primary Infrastructure Management server to synchronize the database and the impact management cell database.