Connect to AWS RDS Postgres database with python

Question

I have an existing postgres table in RDS with a database name my-rds-table-name

I've connected to it using pgAdmin4 with the following configs of a read-only user:

host_name = "my-rds-table-name.123456.us-east-1.rds.amazonaws.com"
user_name = "my_user_name"
password = "abc123def345"

I have verified that I can query against the table.

However, I cannot connect to it using python:

SQLAlchemy==1.2.16
psycopg2-binary==2.7.6.1
mysqlclient==1.4.1

With:

import psycopg2
engine = psycopg2.connect(
    database="my-rds-table-name",
    user="my_user_name",
    password="abc123def345",
    host="my-rds-table-name.123456.us-east-1.rds.amazonaws.com",
    port='5432'
)

It fails with

psycopg2.OperationalError: FATAL:  database "my-rds-table-name" does not exist

Similarly, if I try to connect to it with sqlalchemy:

sqlalchemy.exc.OperationalError: (psycopg2.OperationalError) FATAL:  database "my-rds-table-name" does not exist

What am I missing?

Answer 1

Thank's John Rotenstein for your comment.

As he pointed out, my-rds-table-name is the database instance name, not the database name, the default database name is postgres .

import psycopg2
engine = psycopg2.connect(
    database="postgres",
    user="my_user_name",
    password="abc123def345",
    host="my-rds-table-name.123456.us-east-1.rds.amazonaws.com",
    port='5432'
)

Answer 2

Using sqlalchemy you can do the following:

engine = create_engine('postgresql://postgres:postgres@<AWS_RDS_end-point>:5432/postgres')

Then you can update your database. For example:

df = pd.DataFrame({'A': [1,2], 'B':[3,4]})
df.to_sql('tablename', engine, schema='public', if_exists='append', index=False)