tblong / aws-snapshot-tool

A Python/Boto script that automates snapshot making of EBS volumes and rotating.

aws-snapshot-tool

aws-snapshot-tool is a python script for AWS Lambda to make it easy to roll snapshot of your EBS volumes.

Simply assign a tag for each volume that is to have snapshots, configure the aws-snapshot-tool config.sample and the AWS Lambda script is ready. Based of how many snapshots that is needed to keep and scheduling that is desired for AWS Lambda to execute the AWS Lambda code, it will handle rolling snapshots on any schedule so that setting up the retention policy is simple.

Features:

• Python based: Leverages boto and is easy to configure and upload to AWS Lambda server
• Simple tag system: Just add a tag to every EBS volume that need snapshots
• Configure retention policy: Configure how many snapshots you want to retain
• Optimized for AWS Lambda scheduling: Allows the scheduling to be set within AWS Lambda not the config.py file.
• SNS Notifications: aws-snapshot-tool works with Amazon SNS out of the box, so notification of snapshots is available

Usage

1. Create a project folder for all of the projects resources.
2. Install the Boto package, by using pip, inside of the project folder(not Boto3) (see: Install:Pip Install:Boto)
3. Place both the snapshots.py and config.sample inside of the project folder
4. Rename config.sample file to config.py
5. Create a SNS topic in AWS and copy the ARN into the config.py file
6. Subscribe with a email address to the SNS topic to receive updates
7. Create a snapshot user in IAM and put the key and secret in the config.py file
8. Create a security policy for this user (see: iam.policy.sample)
9. Decide how many versions of the snapshots you want and change this in the config.py file
10. Change the Region and Endpoint for AWS in the config.py file
11. Optionally specify a proxy if needed, otherwise set it to '' in the config.py file
12. In the config.pyfile, specify the tags that are used to search Volumes and take snapshots
13. For each Volume that is to have snapshots, give a Tag with a Key and a Value, and put this data in the config.py file. Default: "autosnap" and the value "true" and "Customer" and the value "Customer Name"
14. Upon completion of configuring config.py, within the project folder, zip up all of the contents(makesnapshots.py, config.py, boto folder, and boto-2.38.0.dist-info folder) as one zip file
15. Upload the zip file within the project folder to AWS Lambda
16. Under the settings within AWS Lambda, configure the scheduling for the snapshots to be taken(see:Scheduled Events)

Additional Notes

The user that executes the script needs the following policies: see iam.policy.sample
This script is compatible with Boto, not Boto3
During step 14, the folder "boto-2.38.0.dist-info" may be a different version

aws-snapshot-tool's Issues

Put time zone on the string that returns the date/time

change finished time format

The format should match the start time format %Y-%m-%d %H:%M:%S

add support for multiple tags for snapshot criteria

Currently the only tag being searched for is tag_name. This poses a problem if a backup script will be running on two separate EC2 instances across two separate customers. Once the same script is installed on another customer's set of instances, the second backup script will invoke a snapshot of the first customer's instances since the same key-value pair criteria is used. Add flexibility to the config.sample and makesnapshots.py files where both a customer=name and autosnap=true key-value pairs must match before a snapshot is both taken and removed.

Non checking for snapshot name

snapshot name is not checked

resource tags not migrating to snap from volume

Need to fix set_resource_tags(resource, tags) to properly check args

config file namespace improvement

https://docs.python.org/2/howto/doanddont.html#from-module-import-name1-name2

The config file's config dictionary is currently being imported directly into makesnapshots.py. Improve the import per the above link.