Cisco Unified MeetingPlace Release 8.0 -- Configuring Two New Application Servers in a Failover Deployment After Failure of Both Existing Servers

From DocWiki

(Difference between revisions)
Jump to: navigation, search
(Before You Begin)
m (1 revision)

Latest revision as of 17:36, 29 October 2010

Main page: Cisco Unified MeetingPlace, Release 8.0

Up one level: Configuration

Note: Only complete this procedure after complete server failure, when both Cisco Unified MeetingPlace Application Servers are damaged, with unrecoverable data from these servers.

This procedure describes how to configure two new Application Servers, in a failover deployment, by using data from the most recent archive and backup files.

  1. Create two new Application Servers in a failover deployment.
  2. Restore the production database on the first Application Server (Node 1) from the most recent backup.
  3. Force replication of all data from the first Application Server (Node 1) to the second Application Server (Node 2).

You may complete these commands at the Application Server command-line interface or by using a SSH client over the network.

Before You Begin

You must have a L0 (Level 0) backup file. The restoration of the Level 0 backup file restores all of your Application Server data as of the date of the Level 0 backup. Additional L1 (Level 1) and L2 (Level 2) incremental backup files, created after the Level 0 backup file, will provide a more recent image of the database.

Note: Unless you have a backup that was made immediately prior to the server failure, the restored database may not contain the more recent activities on the Cisco Unified MeetingPlace Application Server.

Tip: You must be the root user to run the failover and replication commands in this procedure.

  • Open an SSH client window and sign in as the mpxadmin user.
  • Change to the root user by entering su - and enter the root password.

  1. Create two new Application Servers in a failover deployment. See Setting Up Failover for Two Newly Installed Application Servers.
    You should now have a fully configured Application Server failover deployment, with Node 1 in active mode and Node 2 in standby mode. Replication between the two servers is active.
  2. Enter the following commands on Node 1:
    1. As the root user (see the Tip in Before You Begin), turn off replication:
      mp_replication switchOFF -r <node_2> -f
    2. Set Node 1 to standby mode.
      failoverUtil setServer standby
    3. Copy the most recent Level 0 backup, plus all other backups created after the Level 0 backup, from the archive server to the /db/backup/compressed_backup directory on the first Application Server (Node 1).
    Note: Remember the name of the most recent backup file. You will select this file during the restore operation.
  3. Restore the Application Server data. See Restoring Data By Using the CLI on the Application Server.
    1. Select L as you copied the backup and archive files to your local server.
    2. Follow the steps in this procedure to restore (copy and transfer) the various archived files into their respective directories on the new Application Server (Node 1).
    3. When prompted for the backup file, enter the number corresponding to the most recent backup.
    4. Select R to restore the files.
  4. After the restore operation is complete, reboot both Node 1 and Node 2.
  5. When Node 1 comes back online, then as the root user (see the Tip in Before You Begin), turn off replication on Node 1.
    mp_replication switchOFF -r <node_2> -f
  6. Force a full replication of data from Node 1 to Node 2.
    1. Node 1- Turn on replication and force a full replication from Node 1 to Node 2:
      mp_replication switchON -r <node_2> -S -F <node_1> -f
      Depending on the size of your Application Server database, this replication may take a half hour or longer to complete. Wait for the replication to complete, and for the Linux prompt to return.
    2. Node 1-Enter failoverUtil copyConfigFiles.
      When prompted, enter the hostname of Node 2, its IP address, a username (must be mpxadmin) and password (use the mpxadmin password).
    3. Node 2-Enter failoverUtil restoreConfigFiles.
    4. Node 1-Switch Node 1 into active mode by entering failoverUtil setServer active.
    5. After the setServer active operation is complete on Node 1, enter reboot on both Node 1 and Node 2.
    6. As the root user (see the Tip in Before You Begin), confirm that Node 1 is the active Application Server and that Node 2 is the standby Application Server, with replication turned on.
      failoverUtil status
      mp_replication status
  7. If the status is not set properly, then use the mp_replication switchON and failoverUtil setServer active commands to correct the problem.
    Both Application Servers are now set in a failover deployment and are ready for production use.
Related Topics

Rating: 0.0/5 (0 votes cast)

Personal tools