I’m using GridDB for a distributed database setup and recently encountered the following error while performing operations across nodes in the cluster:

from griddb_python import StoreFactory, GSException

def perform_cluster_operations():
    try:
        factory = StoreFactory.get_default()
        gridstore = factory.get_store({
            "host": "239.0.0.1",
            "port": 41999,
            "cluster_name": "defaultCluster",
            "username": "admin",
            "password": "admin"
        })

        # Retrieve container for performing transactions
        container = gridstore.get_container("containerName")
        if container is None:
            raise Exception("Container not found")

        # Attempt a read transaction on the container
        row = container.get(1)
        print(f"Row retrieved: {row}")

        # Perform a write transaction to update a row
        container.put(2, {"id": 2, "value": "newValue"})
        print("Row updated successfully.")

    except GSException as e:
        print(f"Transaction failed: {e.what()}")
        raise

if __name__ == "__main__":
    perform_cluster_operations()

I connect to a GridDB cluster and attempt to perform read and write operations on a container.
The operations involve simple transaction handling to retrieve and update rows.
However, I frequently encounter the TXN_REQUEST_IGNORED ERROR 10906. This error suggests that the event is unknown, possibly due to mismatched cluster versions across the nodes.

Here are some additional details about my setup:

Operating System: Ubuntu 20.04
Python Version: 3.8
GridDB Version: Mixture of versions on different nodes (some nodes may have version 4.5, others version 4.4)
Cluster Size: 5 nodes
I suspect this issue may be related to having different versions of GridDB running on different nodes. I’m looking for guidance on:

How to verify the version of GridDB running on each node.
Whether this version mismatch could be causing the TXN_REQUEST_IGNORED ERROR 10906.
Steps to upgrade or synchronize the versions across the cluster to avoid this error.
Has anyone encountered this issue with GridDB? Any insights on how to resolve it?