You can configure an up-to-date backup recovery server in a remote location for resuming operation in the event of a disaster.

You can add a disaster recovery node to a clustered environment or to a stand-alone Data Center Operation (DCO) server. However, disaster recovery requires a Cluster node for high availability and disaster recovery license.
PostgresSQL streaming replication is used to move data to the disaster recovery node. Communication is necessary through ports 22, 80 and 5432. The DR node also needs to able to 'ping' the IP address that it gets attached to.

The connection between the DCO server or cluster and the disaster recovery node must be stable and quick enough to transfer a full backup.
If you are running a cluster setup and a problem occurs during transfer to the disaster recovery node, the cluster will clean up and detach the disaster recovery node. If your system integrates with StruxureWare Data Center Expert (DCE), it will also trigger an alarm.

Configuring disaster recovery

  1. Allow the Disaster Recovery node to connect to the DCO server or cluster.
    1. Open the Webmin server configuration interface on the DCO server (or cluster).
      1. Type the server's IP address in a Web browser.
      2. On the download page, click Server Configuration.
        This opens the Webmin server configuration interface on the DCO server (or cluster).
      3. Log into Webmin and navigate to StruxureWare DC Operation>Disaster Recovery.
    2. Accept the fingerprint and allow the Disaster Recovery node (.53 in the example) to connect to this node (.52 in the example). Figure 1.
  2. Configure the Disaster Recovery node.
    1. Install DCO on a dedicated remote server with stand-alone Disaster Recovery configuration (.53 in the example). See how.
    2. Open the Webmin interface on the Disaster Recovery server.
      1. Type the Disaster Recovery server's IP address in a Web browser.
      2. On the download page, click Server Configuration.
        This opens the Webmin server configuration interface on the Disaster Recovery server.
      3. Log into Webmin using the user credentials created during installation (step 2a) and navigate to StruxureWare DC Operation>Disaster Recovery.
    3. Accept the fingerprint and attach this node as Disaster Recovery node. Figure 2.

You have now configured Disaster Recovery. In case of a disaster and shutdown of the DCO server (or cluster), you can promote the disaster recovery to a stand-alone DCO server.

Note

If the Disaster Recovery node is monitored, for example by DCE via SNMP, it is important to set the device scan timeout in DCE high enough for the DR to respond. Otherwise, it will return "communication lost". This, of course, all depends on the network speed between the Disaster Recovery node and the DCO server.

Figure 1


Figure 2

Promoting disaster recovery

  1. Open the Webmin interface on the Disaster Recovery server (.53 in the example).
    1. Type the Disaster Recovery server's IP address in a Web browser.
    2. On the download page, click Server Configuration.
      This opens the Webmin server configuration interface on the Disaster Recovery server.
    3. Log into Webmin and navigate to StruxureWare DC Operation>Disaster Recovery.
  2. Verify the data shown is correct and select to Promote disaster recovery node. You have now restored a copy of your DCO server.

Re-enabling after disaster recovery

Reconfigure your main server as it was before the disaster occurred and then follow the steps below to reenable the disaster recovery node:

  1. Create a backup of the disaster recovery server.
  2. Restore the backup on your main DCO server(s).
  3. Repeat the steps in Configuring disaster recovery to reconfigure your setup as it was before promoting disaster recovery.

 

See also

Cluster Setup
Backing up and Restoring Backup

Skip to end of metadata
Go to start of metadata
RELATED COMMUNITY QUESTIONS
WAS THIS ARTICLE HELPFUL?