Access to RDS in AWS - HTTPConnectionPool: Max retries exceeded with url - python-3.x

Question

I have a Docker container running on an AWS ECS task. Basically this container needs to get a dataframe that contains customer information in each row. Each row is then accessed and data stored in AWS S3 is read depending on the customer information in that row. Multiple metrics and pdf reports are generated for each row. In initial tests that I have performed, the customer information was provided in a csv within the container. This csv was read and then the container run successfully.
The next step in the testing was to read the customer information from a RDS instance (MySQL) hosted in AWS. So basically, I query the RDS instance and get a dataframe (analogous to the csv described above with customer information). Here, is where I am having problems. I am getting the following error after running the ECS task in Fargate (port number not shown):
HTTPConnectionPool(host='localhost', port=[port_number]): Max retries exceeded with url: /session/03b59c11-8e28-4e6a-9b66-77959c774858/window/maximize (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f515cf97580>: Failed to establish a new connection: [Errno 111] Connection refused'))
When I look at the logs in CloudWatch, I can see that the query is executed successfully and the container starts reading each row of the dataframe. However, it breaks after N amount of rows are read (or probably there is a time limit?).
This are the functions in the container that I am using to execute a given query:
def create_connection_mysql():
"""
Opens a connection to a mySQL server and returns a MySQLConnection object.
Returns
-------
connection : MySQLConnection
MySQLConnection object.
"""
# read credentials from environment file
host = os.getenv("HOST")
user = os.getenv("USER_MYSQL")
passwd = os.getenv("PASSWORD")
database = os.getenv("DATABASE")
# create connection
connection = connect(
host=host,
user=user,
passwd=passwd,
database=database,
use_unicode=True,
charset="utf8",
port=3306
)
print("Connection to MySQL DB successful")
return connection
def execute_query(connection, query):
"""
Executes a query on mySQL and returns the result as a dataframe.
Parameters
----------
connection : MySQLConnection object
A MySQLConnection object to access the database.
query : str
A valid query to perform on the provided connection.
Returns
-------
result : DataFrame
Returns a DataFrame with the results of the query.
"""
# execute query
cursor = connection.cursor(buffered=True)
cursor.execute(query)
# output as a pandas DataFrame
colnames = cursor.column_names
result = pd.DataFrame(cursor.fetchall(), columns=colnames)
# close connection and return result
cursor.close()
connection.close()
return result
Because these functions are working correctly in the container, I wonder what could be causing this problem with the RDS instance.