In MongoDB, primary nodes are the node that can accept write. These are also known as master nodes. The replication in MongoDB is single master so, only one node can accept write operations at a time.

Secondary nodes are known as slave nodes. These are read only nodes that replicate from the primary.

In MongoDB, a replica set is a group of MongoDB servers that maintain the same data set, providing high availability and fault tolerance. Within a replica set, there are two types of members: primary and secondary.

Primary Node:
- The primary node is the main member of the replica set.
- It receives all write operations (inserts, updates, deletes) from clients and propagates these changes to the secondary nodes.
- There can be only one primary node in a replica set at any given time.
Secondary Node:
- Secondary nodes replicate the data from the primary node.
- They can be used for read operations, providing scalability for read-intensive workloads.
- If the primary node fails, one of the secondary nodes can be automatically promoted to become the new primary node, ensuring high availability.

The primary node and secondary nodes communicate with each other using the replication process, ensuring that data stays consistent across the replica set. This setup helps distribute the workload and provides fault tolerance, as the system can continue to operate even if one or more nodes experience issues.

In summary, a MongoDB replica set consists of a primary node that handles write operations and one or more secondary nodes that replicate data for high availability and scalability.