Neo4j Protocol Connection Issue

[jp]

Looking for help with connecting to neo4j instance.

We are able to connect to our neo4j server via the bolt protocol with "bolt://remote.hostname:7687", however when using the neo4j protocol "neo4j://remote.hostname:7687" we receive the below error:

2022-09-16 22:53:03.386+0000 INFO Query started: id:72473 - 0 ms: 0 B - bolt-session bolt neo4j-cypher-shell/v4.4.9 client/our-ip-address server/ip-address> <none> - ****- CALL dbms.routing.getRoutingTable($context, $database) - {context: {address: 'remote.hostname:7687'}, database: <null>} - runtime=null - {}

cypher-shell -a neo4j://remote.hostname:7687 -u **** -p ****
Failed to connect to neo4j://remote.hostname:7687, fallback to bolt://remote.hostname:7687
Connected to Neo4j using Bolt protocol version 4.1 at bolt://remote.hostname:7687 as user ****.

Any help on how we can go about resolving this issue would be great!

[Franbr31]

Hi, we're facing the same issue. Any update on this?

Thanks!

[andy_hegedus]

Hi,

Similar issues with the latest Python driver.  My fix at the moment is to set the port to 11003 instead of 7687.

I noticed the Neo4J browser was using this port so I tried it.

Andy

[jp]

Issue related to Kubernetes infrastructure and neo4j not routing correctly during ingress. We determined that our applications can communicate and connect with neo4j protocol fine when they are on the same hosted cluster, however we could not use neo4j protocol locally.

We determined this issue could be resolved two different ways.

1. A more permanent solution by externally exposing the neo4j address to allow outside traffic in. This involves creating a load balancer to handle this traffic.

2. Using telepresence to connect directly to the Kuberenetes cluster and navigate through neo4j protocol

Option 2 has been an easy and successful work around for us

[jp]

@Franbr31