I am using ElasticSearch-6.2.2 on three nodes in the same Cluster. Each of those nodes has a dedicated Cassandra Server integrated along with it. There are times when my application will need all three and there are times when one will suffice.
So when I tried to test my application, I configured, in the application only the first node's IP and ran only the first node's instance and the other two were stopped. However, there seems to be an issue when I do this.
But, when the application was configured with all three nodes of the cluster, it worked fine, even though the first instance was the only one running and the other two nodes of the cluster were still off. I'm trying to reproduce the incident for logs. In the meantime, does anyone have any idea why this is happening?
Thanks in advance,