Setting up gateway nodes to ensure cluster communication during failover

Both the primary and the DR clusters require designating gateway nodes for access to whichever side is acting as the cache. By designating gateway nodes on both clusters, you can ensure that even during failover, cluster communication continues properly.

To handle a possible node failure, you need to specify at least two nodes on each cluster to be gateway nodes. To specify two nodes on the primary cluster as gateway nodes, use the command similar to the following:

mmchnode -N Node1,Node2 --gateway

Using the example setup mentioned in Example setup for protocols disaster recovery, the command to specify gateway nodes on the primary cluster is as follows:

# mmchnode -N clusternode-vm1,clusternode-vm2 --gateway
Tue Apr 28 20:59:01 MST 2015: mmchnode: Processing node clusternode-vm2
Tue Apr 28 20:59:01 MST 2015: mmchnode: Processing node clusternode-vm1
mmchnode: Propagating the cluster configuration data to all
affected nodes. This is an asynchronous process.
# Tue Apr 28 20:59:04 MST 2015: mmcommon pushSdr_async: 
mmsdrfs propagation started

Tue Apr 28 20:59:08 MST 2015: mmcommon pushSdr_async: 
mmsdrfs propagation completed; mmdsh rc=0

Similarly, you need to specify at least two nodes on the DR cluster as gateway nodes. Using the example setup, the command to specify gateway nodes on the DR cluster is as follows:

# mmchnode -N clusternode-vm1,clusternode-vm2 --gateway
Tue Apr 28 20:59:49 MST 2015: mmchnode: Processing node clusternode-vm2
Tue Apr 28 20:59:49 MST 2015: mmchnode: Processing node clusternode-vm1
mmchnode: Propagating the cluster configuration data to all
affected nodes. This is an asynchronous process.
# Tue Apr 28 20:59:51 MST 2015: mmcommon pushSdr_async: 
mmsdrfs propagation started

Tue Apr 28 20:59:54 MST 2015: mmcommon pushSdr_async: 
mmsdrfs propagation completed; mmdsh rc=0