Trying to connect to postgresql on droplet programmatically via python and pg8000/psycopg2

Hey!

If SSH tunneling works but direct connection times out, it’s most likely a network block, not a Postgres config issue.

First, run this on your Droplet:

sudo netstat -plant | grep 5432

You should see something like:

tcp        0      0 0.0.0.0:5432      0.0.0.0:*       LISTEN      1234/postgres

That means Postgres is listening on all interfaces. If you only see 127.0.0.1:5432, then it’s only listening on localhost. So you need to update your postgresql.conf to set listen_addresses = '*' and restart the service to allow external connections.

Also, just in case check if a Cloud Firewall is attached to your Droplet. If so, make sure there’s a rule allowing inbound traffic on port 5432 from your dev machine’s IP.

If you’re using UFW on the Droplet, allow the connection with:

sudo ufw allow from <your-ip> to any port 5432 proto tcp

As a workaround, you can keep using the SSH tunnel from Python like this:

import psycopg2
conn = psycopg2.connect(
    host='localhost',
    port=5432,
    database='your_db',
    user='your_user',
    password='your_pass'
)

Just make sure the tunnel is open when running the code.

Alternatively, if you want to avoid dealing with firewall rules, SSH tunnels, and manual setup, you can consider using DigitalOcean’s Managed PostgreSQL.

Let me know how it goes!

- Bobby

Heya, @36d56192d3114513857618b22686be

You can make sure your PostgreSQL server isn’t binding only to localhost. In postgresql.conf, check that:

listen_addresses = '*'

And in pg_hba.conf, try changing your line to:

host    all    all    0.0.0.0/0    md5

If none of this helps, you can use a reverse SSH tunnel or deploy a lightweight VPN (like Tailscale) to connect securely from your dev machine to the DB without opening any ports publicly.

Regards