I am currently trying to connect to a preexisting Cosmos DB using Gremlin and Mavin for Java. My database requires a username and password to connect. I currently pass in the code with endpoint, username, password, and port to create a cluster. I then use the cluster to start a client and use the client to create a root traversal which I can use to query the DB. My queries return network errors, but I suspect the problem might be in the creation of a cluster since when searching for hosts on the cluster, I get a result of 0 hosts.

I have already tried using a connection string instead of the separated username, password, etc. but that hasn’t worked either. I also tried a Python script which failed.

Here is the connection code:

private static GraphTraversalSource connect(String endpoint, String username, String password, int port) {
        System.out.println("STARTING");
        System.out.println("creating cluster");
        Cluster cluster = Cluster.build()
            .addContactPoint(endpoint)
            .port(port)
            .credentials(username, password)
            .enableSsl(true)
            .maxWaitForConnection(100000)  // Increase connection timeout
            .maxContentLength(65536)
            .create();
        int numHosts = cluster.availableHosts().size();
        System.out.println("Cluster created successfully with the following hosts:");
        System.out.println(numHosts);
        System.out.println("Connecting");
        Client client = cluster.connect();
        System.out.println("Determining traversal source");
        // Define the source and target graph traversal sources
        final GraphTraversalSource G = AnonymousTraversalSource.traversal().withRemote(DriverRemoteConnection.using(client));
        System.out.println("Finding count");
        try {
            long vertexCount = G.V().count().next();
            System.out.println("Source count: " + vertexCount);
        }
        catch(Exception e) {
            System.out.println("FAILED TO QUERY");
            System.exit(1);
        }
        return G;
    }

The output is as follows (I replaced the actual address and hostUri with endpoint for the address):

STARTING
creating cluster
Cluster created successfully with the following hosts:
0
Connecting
Determining traversal source
Finding count
09:28:44.187 [gremlin-driver-loop-1] ERROR o.a.t.g.d.Connection - Server closed the Connection on channel 2017caf4 - scheduling removal from ConnectionPool (Host{address=endpoint/internalIP, hostUri=wss://endpoint:port/gremlin})
Connection Pool Status (size=2 max=8 min=2 toCreate=0 bin=0)
==> Connection{channel=2017caf4 isDead=true borrowed=1 pending=1 markedReplaced=false closing=false created=2024-07-23T16:28:43.549866Z thread=gremlin-driver-conn-scheduler-1}
> Connection{channel=83a2625b isDead=false borrowed=0 pending=0 markedReplaced=false closing=false created=2024-07-23T16:28:43.549858Z thread=gremlin-driver-conn-scheduler-2}
-- bin --
FAILED TO QUERY

Process finished with exit code 1