Mongo DB Backup

This Docker image helps create an on demand image on mongodb atlas and upload it to S3.

upload_to_s3.py

This script uploads a specified file to S3 with a specific name if specified, otherwise it will use the file name.

usage: upload_to_s3.py [-h] [-f FILE] [-o OBJECT_NAME] [-b BUCKET]

Upload local file to S3

options:
  -h, --help            show this help message and exit
  -f FILE, --file FILE  The local file to upload to S3
  -o OBJECT_NAME, --object-name OBJECT_NAME
                        The name of the object to upload to the bucket. If not specified will the file name
  -b BUCKET, --bucket BUCKET
                        The name of the bucket to push the file to

Note

Environment variables starting with S3_ (case insensitive) will be passed down to the boto3 s3 client constructor while emitting S3_.

Configuration

If running on Docker, I recommend making 2 .env files and mounting them as environment variable in the container. For example, using the env-file flag with docker run.

These are the files I recommend on creating:

• mongodb_atlas.env

# API Authentication Details
MONGODB_ATLAS_PUBLIC_API_KEY=********
MONGODB_ATLAS_PRIVATE_API_KEY=**********
# Project ID and ORG in which the API creds have been set
MONGODB_ATLAS_ORG_ID=**********************
MONGODB_ATLAS_PROJECT_ID=******************
# Optional
MONGODB_ATLAS_OUTPUT=json

• s3.env

# These are a must have unless a profile is configured
S3_AWS_ACCESS_KEY_ID=************
S3_AWS_SECRET_ACCESS_KEY=****************
S3_REGION_NAME=us-east-1
# When using a different provider other than AWS, set this to point to its servers
# For example with Linode
S3_ENDPOINT_URL=https://us-east-1.linodeobjects.com

Command Line Arguments

The script requires the following 3 command line argument:

• $1 - The cluster name to create snapshot from
• $2 - The description of the snapshot that will be created
• $3 - The S3 bucket name to push the snapshot to

Example

Let's put it all together.

In this example we are creating a snapshot for a database called mydb, and uploading it to a bucket called mongodb-backups.

docker run --rm \
   --env-file mongodb_atlas.env \
   --env-file s3.env \
   idogakamai/mongodb-backup \
   mydb "snapshot description" mongodb-backups