RabbitMQ Cluster - Load Balancing Using Azure application gateway

[ela19p]

I have configured a 3 node rabbitmq cluster(v3.12). The nodes are testdkg001@rabbit ,testdkg001@rabbit and testdkg001@rabbit. The cluster and queue fail over is working fine. I have a mq connector property files where I need to define the rabbitmq hostname and port. Here I defined the hostname "testdkg001.org.com" from one of the cluster and it redirects the traffic to the node where the queue is active. if there is any issues with that node , the queue is transferred to the next available node and the connection is automatically rerouted to the working rabbitmq node. But if the "testdkg001.org.com" - entire server goes down , the mq connector will need to establish the connection to the next working node in the cluster which requires load balancing . I'm using azure application gateway and defined a default host name which will redirect the traffic to any one of the available hosts. But if I define the default host name in the mq connector properties files its throwing connection errors
`[org.springframework.jms.listener.DefaultMessageListenerContainer] Could not refresh JMS Connection for destination 'test.queue.001' - retrying using FixedBackOff{interval=5000, currentAttempts=2, maxAttempts=unlimited}. Cause: null.

If I add the default hostname into the /etc/host file and provide one of the nodes ip address ,it can connect without any issues. So any advice on how to define the application gateway rules in azure.