sidbelbase / backup-my-postgres

Periodically backups postgres databases from EC2 to S3 Bucket using python

Backup My Postgres

This project is to backup the database from EC2 Instance to s3 Bucket using python. It automatically dumps PostgreSQL backups to Amazon S3.

Prerequisites

• Rename .sample.env to .env and add your credentials.
• Add an AWS account from AWS Console that has the access to the S3 bucket with the following permissions: s3:PutObject

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": ["s3:PutObject"],
      "Resource": "arn:aws:s3:::<bucket>/<object>/*"
    }
  ]
}

• Run the following command to create the bucket:

aws s3 mb s3://example-backup-bucket --profile your-profile --region "ap-south-1"

NOTE: run.py is using Asia/Kathmandu Timezone & the database names are hard coded as DATABASES, please modify before running this script.

Usage

• Install dependencies:

  pip install -r requirements.txt

• Use python3 to run the script:

  python3 run.py

Cron Job

Setup in EC2 Instance

• Run the following command to setup the cron job:

  sudo service crond start (if not running)
  crontab -e

• Use cron to run the script every day at 02:00AM NEPALI TIME (UTC+5:45)

  0 22 * * * python3 /home/ubuntu/backup-my-postgres/run.py

Made with ❤️ in Nepal.

backup-my-postgres's Issues

Use .env for database names as well

Instead of this line, it will be better to read databases from .env files
https://github.com/sidbelbase/backup-my-postgres/blob/faf5a617c39918f49b08835d06f979b6aa647c66/run.py#L17

Example:

DATABASES = environ.get("DATABASES").split(",")