AWS Lambda to generate CSV file from RDS PostgreSQL

One of the requirement was to generate csv file for set of queries from RDS PostgreSQL and upload the csv file to s3 bucket for power bi reporting. Powerbi connects to s3 url and generate report. There is no gateway to connect to PostgreSQL instance from power-bi, hence we need to have a mechanism to upload the data to s3 so that powerbi can import it and generate reports. How can you generate csv file and upload to s3 bucket ?

There are multiple ways we can achieve this, one is to use ssm command send over as shell script and use copy command for postgreSQL to generate csv file and push it to s3. Another approach is use pandas module and dataframe to convert the data to csv and push it to s3.

Both the examples are as below. I retreived username and password from parameter store. For more information about parameter store you may refer here.

import boto3
import time
import sys
import os
ec2 = boto3.client('ec2')
ssm_client = boto3.client('ssm')
def lambda_handler(event, context):
def execute_ssm_command(client, commands, instance_ids):
resp = client.send_command(
Parameters={'commands': commands},
nodename = os.environ['region']+'.'+os.environ['env']+'.'+os.environ['app']+'.'+os.environ['company']+'.'+os.environ['role']
print os.environ['date']
print os.environ['code']
uname = ssm_client.get_parameter(Name='dbusername', WithDecryption=True)
username = uname['Parameter']['Value']
pwd = ssm_client.get_parameter(Name='dbpassword', WithDecryption=True)
password = pwd['Parameter']['Value']
for reservation in response["Reservations"]:
for instance in reservation["Instances"]:
for tag in instance['Tags']:
if (tag['Key'] in 'Name'):
if (tag['Value'] in nodename):
print tag['Value']
commands = ['if [ -z "${HOME:-}" ]; then export HOME="$(cd ~ && pwd)"; fi','sudo yum install mail -y', 'sudo yum install postgresql96 -y', '#!/bin/bash ', 'PGPASSWORD='+str(password)+' psql -h postgres -U '+str(username)+' -d dbname -c "\copy (select * from report where date= '+os.environ['date']+' and code= '+os.environ['code']+') to stdout csv header">/$HOME/s3Reportqa.csv', 'aws s3 cp /$HOME/s3Report.csv s3://'+os.environ['bucketpath']+'/', 'printf "Hi All, csv file has been generated successfully. " | mailx -vvv -s "report" -r "" -S smtp="smtp" ""']
temp = instance["InstanceId"]
instance_ids = [temp]
print os.environ['region']+'.'+os.environ['env']+'.'+os.environ['app']+'.'+os.environ['company']+'.'+os.environ['role']
print instance_ids
execute_ssm_command(ssm_client, commands, instance_ids)



Second Method (Using Pandas module)

import psycopg2
import request
import boto3
import pandas as pd
from pandas import Series, DataFrame
import csv
conn_string = "host='dbinstancename' dbname='databasename' user='username' password='password'"
conn = psycopg2.connect(conn_string)
cursor = conn.cursor()
print ("Connected")
tablename = 'report'
date = '2019-07-11'
code = 'RAMA'
#cursor.execute("SELECT * FROM " + tablename +" Where date ="+ date + " and code = "+code+";")
cursor.execute("SELECT * FROM " + tablename +" limit 100;")
myresult = cursor.fetchall()
item_list = []
for i in myresult:
item = {'col1':i[0],
'col3' :i[2],
'col4' :i[3],
'col5' :i[4],
'col6' :i[5],
'col7' :i[6],
'col8' :i[7],
'col9' :i[8]}
concat = str(i[0]) + str(',') + str(i[1]) + str(',') + str(i[2]) + str(',') + str(i[3]) + str(',') + str(i[4]) + str(',') + str(i[5]) + str(',') + str(i[6]) + str(',') + str(i[7]) + str(',') + str(i[8])
# print (concat)
df = pd.DataFrame(data=item_list,columns=['col1','col2','col3','col4','col5','col6','col7','col8','col9'])
print (df.head(40))
# importing the result to csv begins
print('csv generated')
# to push the datafram results to s3, we can use boto3 s3 resource as below
s3_resource = boto3.resource('s3')
s3_resource.Object(bucket, 'df.csv').put(Body=csv_buffer.getvalue())

You can schedule the lambda using cloudwatch events every 5 minutes to update the data in s3.

Hope you enjoyed the post.


Ramasankar Molleti


Published by Ramasankar

Hi. I’m Ramasankar Molleti. I’m a passionate IT professional with over 14 years of experience on providing solutions for customers who are looking on cloud computing, Database Migration, Development, and Big Data. I love learning new technologies and share my knowledge to community. I am currently working as Sr Cloud Architect with focus on Cloud Infrastructure, Big Data. I work with developers to architect, build, and manage cloud infrastructure, and services. I have deeep knowledge and experience on working with various database platforms such as MS SQL Server, PostgeSQL, Oracle, MongoDB, Redshift, Dyanamodb, Amazon Aurora. I worked as Database Engineer, Database Administrator, BI Developer and successfully transit myself into Cloud Architect with focus on Cloud infranstructure and Big Data. I live in USA and put my thoughts down on this blog. If you want to get in touch with me, contact me on my Linkedin here: My Certifications: Amazon: AWS Certified Solutions Architect – Professional AWS Certified DevOps Engineer – Professional certificate AWS Certified Big Data – Specialty AWS Certified Security – Specialty certificate AWS Certified Advanced Networking – Specialty certificate AWS Certified Solutions Architect – Associate Microsoft: Microsoft® Certified Solutions Associate: SQL Server 2012/2014 Microsoft Certified Professional Microsoft® Certified IT Professional: Database Administrator 2008 Microsoft® Certified Technology Specialist: SQL Server 2008, Implementation and Maintenance

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: