I have a DSE 6.8 cluster with two DC - Cassandra nodes and Analytics Solo nodes.
Also I have a separate node where OpsCenter is installed and it's Lifecycle manager was used for cluster installation.
I am planning to use this OpsCenter node as Spark driver node - for submitting Spark applications. I want to have here installed Livy/job-server and also Zeppelin.
So, I need to use "dse exec" command to run all this third-party software. But "dse" package is not installed at this node yet.
My question is: will I do this right if I perform next steps:
1. Will install "dse: package at OpsCenter node.
2. Will disable "dse" service at OpsCenter node (systemctl stop dse && systemctl disable dse)
3. Will export dse configuration from existing node and import to OpsCenter node.
4. Finally use "dse exec" at OpsCenter node to launch software I need.
I am in doubt if using separate node is a right way for using Livy/Job-server and Zeppelin. With implementing step 2 at OpsCenter node I will have a separate dse node that will be not used. Can this impact somehow my cluster? Are there any guidelines or things I should keep in mind?