Cassandra节点无法互通：AWS EC2集群端口监听与连通问题求助

Hey there, let's work through this Cassandra connectivity issue step by step—those timeout errors usually boil down to a few common culprits, so let's check each one:

1. First, Confirm Cassandra is Actually Running #

It sounds obvious, but if Cassandra isn't started, it won't listen on port 7000 at all. Run these commands to verify:

• Check service status:

  sudo systemctl status cassandra
  

• If it's inactive, start it and enable auto-start on boot:

  sudo systemctl start cassandra
  sudo systemctl enable cassandra
  

• You can also check for running Cassandra processes:

  ps aux | grep cassandra
  

2. Fix Cassandra's Configuration File (cassandra.yaml) #

The most common misconfig here is the listen_address setting, which controls how nodes communicate with each other. Edit the file (usually at /etc/cassandra/cassandra.yaml):

• Find listen_address and set it to your EC2 instance's private IP (like 172.31.30.37—don't use localhost or 0.0.0.0 here, since nodes need to reach each other via private IPs).
• Double-check the seeds setting: This should list the private IP(s) of at least one seed node in your cluster (e.g., if your first node is 172.31.30.37, all three nodes should have this IP in their seeds list).
• Optional: If you need remote client access, set rpc_address to 0.0.0.0, but this isn't required for node-to-node communication.

After making changes, restart Cassandra to apply them:

sudo systemctl restart cassandra

3. Unblock Port 7000 on Ubuntu's Local Firewall (ufw) #

If Ubuntu's ufw firewall is enabled, it might be blocking incoming/outgoing traffic on port 7000:

• Check firewall status:

  sudo ufw status
  

• If it's active, allow TCP traffic on port 7000:

  sudo ufw allow 7000/tcp
  

• If you want to allow ping (ICMP) between nodes too, run:

  sudo ufw allow from <your-cluster-private-subnet> to any proto icmp
  

  (Replace <your-cluster-private-subnet> with your EC2 subnet, like 172.31.0.0/16)

4. Critical: Update AWS EC2 Security Group Rules #

This is the #1 gotcha for EC2-based clusters—even if your local firewall is open, AWS security groups will block traffic unless explicitly allowed:

• Go to your EC2 console, select one of your Cassandra instances, and navigate to its Security Groups tab.
• Edit the inbound rules to add:
  • Type: Custom TCP
  • Port Range: 7000
  • Source: Select the ID of your cluster's security group (this allows all instances in the same security group to communicate on port 7000)
• If you want ping to work, add another inbound rule:
  • Type: All ICMP - IPv4
  • Source: Your cluster's security group ID

5. Verify Port Listening & Test Connectivity #

Once you've made the above changes, check if Cassandra is now listening on port 7000:

netstat -na | grep 7000
# Or use ss for more detailed info:
ss -tulpn | grep cassandra

You should see a line like tcp 0 0 172.31.30.37:7000 0.0.0.0:* LISTEN (matching your instance's private IP).

Then test telnet again from another node:

telnet <target-private-ip> 7000

If it connects, you're good to go! If not, check Cassandra's system logs at /var/log/cassandra/system.log for errors—look for messages about failed binding to port 7000 or configuration issues.

内容的提问来源于stack exchange，提问作者Payal