Disconnections in Neo4j Cluster Deployment via LoadBalancer Exposure

Hello all,

I have deployed Neo4j cluster following those instuctions.

I have also deployed the proposed LoadBalancer helm charts.
The issue is that I have several disconnections in my browser and then via refreshing the page the issue is getting resolved. I have also changed the LoadBalacer's Service parameter externalTrafficPolicy to Cluster. With this option, it goes better, but still there are issues. Also I get messages in my browser like "Server is taking a long time to respond..." I am connecting as user neo4j
to neo4j://<LoadBalancers-externalIP>

What is more, sometimes when trying to delete nodes and relationships from database GUI in my browser I get the error:

But after some browser's refreshes, the neo4j GUI hits the cluster leader and the deleting is taking place. How can I figure out in order the neo4j GUI in my browser always hitting the LEADER pod? Moreover should I miss any setting in neo4j.conf for the disconnections issue? I have not set up neo4j certificates. Could this help in disconnections issues?