Black Box ET10000A manual Preparation for DR listening, Actions on DR activation failover occurs

Appendices

UDP 45599

UDP 46699

Ordering of actions is important.

You should install in the following steps:

1Power on both servers

2Assign IP to server #1

3Assign IP to server #2

4Make sure that server #1 can see server #2 on the network

5Run /etc/init.d/policyserver-install on server #1 ( same order of IP addresses on both )

6IMPORTANT: WAIT for server #1 to fully complete the install and startup

7Run /etc/init.d/policyserver-install on server #2 ( same order of IP addresses on both )

Once installation is complete you can view the web interface from either of the cluster nodes IP addresses.

To verify that the cluster is in place check the Platform -> Utilities page DB Nodes and Appserver Nodes.

Preparation for DR listening

Until EncrypTight supports a fully replicated data layer at the DR cluster site, you must shut-down the database server on the second node. Login as root and issue the following command:

% /etc/init.d/postgresql-9.0 stop

This will cause that DB node to go inactive. You can verify this in the Platform -> Utilities page on the DB Nodes Tabd.

Actions on DR activation (failover occurs)

When failover occurs, in order to ensure the DR cluster is fully redundant, including at the data-layer, you must restart the database server on the second node, and activate it via the UI. Login to the second server as root and issue the following command:

% /etc/init.d/postgresql-9.0 start

Once the database has started, login to EncrypTight Manager as a Platform Admin, navigate to the Platform-> Utilities page, locate the inactive database on the DB tab, select it, right-click and select "Activate". This will synchronize the database and the DR site will be fully HA.

Failback

When the DR site fails back to the main site, you should once again stop the database on the second DR appserver.

EncrypTight Manager Installation Guide 33