question

I believe the configuration parameter you're looking for is this:

https://docs.datastax.com/en/developer/java-driver/4.9/manual/core/pooling/#heartbeat

The error being referred to is a driver configuration timeout and not something from the scc.

The quick answer is that you do not change the heartbeat timeout. Doing so will just hide the underlying issue and you will still eventually run into other problems.

The Spark connector uses the Java driver under the hood. Heartbeats are a client-side mechanism similar to the Linux TCP keepalive. It's the driver's way of sending fake traffic on a connection to make sure network devices like firewalls don't close idle sockets.

If a driver fails to get a response from a heartbeat, it either indicates:

• it lost network connectivity to the nodes in the cluster, even if it was just a temporary network glitch, or
• the nodes are unresponsive usually because they are overloaded.

You need to review the logs on the Cassandra nodes for clues as to why a heartbeat request timed out and address that issue. The heartbeat itself is not the problem -- it's a symptom of another problem which you need to identify and resolve. Cheers!