EFS Cleanup Idle Filesystems
AWS Resource Type:
Amazon Elastic File System (EFS)
This opportunity identifies EFS instances that have been idle for greater than 30 days, creates a backup, and deletes the file system.
Criteria for identifying the opportunity:
Opportunities for EFS instances should meet one of the following criteria:
- no active mounts or mount removals from EFS clients in the last 30 days
- has active mounts with no active client connections in the last 30 days
Potential savings (range in % on annual basis):
Potential savings vary depending on the storage class of the file system, how the backup is stored, and how long the backup is retained for. Our initial findings suggest users could save as much as 11% on their annualized EFS costs.
What happens when the Fixer is executed?
The Fixer first removes any active mounts associated with the EFS instance. Then, the Fixer uses AWS Backup to create a backup of the EFS instance. This backup is retained indefinitely in warm storage using a vault called “cloudfix-vault”. Once the backup process is complete, the Fixer deletes the EFS instance.
Users are free to remove or configure their own lifecycle policies for the backup that is created - see this AWS article for further information.
Is it possible to rollback once CloudFix implements the fixer?
Yes. A user can do a manual rollback by executing the `Cloudfix-Runbook-EfsCleanupIdleFileSystems-Rollback-prod` runbook. The user must provide the ARN of the recovery point in AWS Backup created by the Fixer runbook.
Can CloudFix implement the fix automatically once I accept the recommendation?
Does this fix require downtime?
No, given the target file systems have been identified as idle.
- CloudFix Knowledge Base
- Automating backups and optimizing backup costs for Amazon EFS using AWS Backup
- Restoring an Amazon EFS file system
Article is closed for comments.