Cannot connect to DataStax Enterprise cluster from Python app

Question

I am having some difficulty in connecting to a Centos 7.x server hosted DataStax Cassandra 6.8.

I am able to successfully connect locally inside the Centos Shell and the nodetool status shows the cluster Up and Normal.

Things I tried in cassandra.yaml file -

changed the listen_address parameter from localhost to the IP address of the server. Result -> DSE is not starting.
Commented the listen_address line. Result -> DSE is not starting
Left the parameter of listen_address blank. Result -> DSE in not starting.

as mentioned above - OS - CentOS 7 DSE Version - 6.8 Install method RPM

Python program -

#cluster = Cluster()
cluster = Cluster(['192.168.1.223'])

# To establish connection and begin executing queries, need a session
session = cluster.connect()

row = session.execute("select release_version from system.local;").one()
if row:
    print(row[0])
else:
    print("An error occurred.")

Exception thrown from python ->

NoHostAvailable: ('Unable to connect to any servers', {'192.168.1.223:9042': ConnectionRefusedError(10061, "Tried connecting to [('192.168.1.223', 9042)]. Last error: No connection could be made because the target machine actively refused it")})

Both my PC and my server are on the same network and I am able to ping from each other.

Any help is highly appreciated.

Thanks

Erick Ramirez Erick Ramirez · Accepted Answer · 2021-08-25T11:35:25

The same question was asked on https://community.datastax.com/questions/12174/ so I'm re-posting my answer here.

This error indicates that you are connecting to a node which is not listening for CQL connections on IP 192.168.1.223 and CQL port 9042:

No connection could be made because the target machine actively refused it

The 2 most likely causes are:

DSE is not running
DSE isn't listening for client connections on the right IP

You indicated already that you are not able to start DSE. You 'll need to review the logs located in /var/log/cassandra by default for clues as to why it's not running.

The other possible issue is that you haven't configured native_transport_address (rpc_address in open-source Cassandra). You need to set this to an IP address that is accessible to clients (your app) otherwise, it will default to localhost (127.0.0.1).

In cassandra.yaml, configure the node with:

listen_address: private_ip
native_transport_address: public_ip

If you are just testing it on a local network, set both properties to the server's IP address. Cheers!

[EDIT] I just saw your conversation with @Alex Ott. I'm posting my response here because it won't fit in a comment.

This startup error means that the node couldn't talk to any seed nodes so it won't be able to join the cluster:

ERROR [DSE main thread] 2021-08-25 06:40:11,413 CassandraDaemon.java:932 - \
  Exception encountered during startup
java.lang.RuntimeException: Unable to gossip with any peers

If you only have 1 node in the cluster, configure the seeds list in cassandra.yaml with the server's own IP address:

seed_provider:
    - class_name: org.apache.cassandra.locator.SimpleSeedProvider
      parameters:
          - seeds: "192.168.1.223"

Cannot connect to DataStax Enterprise cluster from Python app

1 Answers