AWS Backup for VMware Incremental VM backups

Question

We're currently experiencing issues with AWS Backup for VMware. The service is not performing incremental backups as expected, which is causing problems:

Full Backups Instead of Incremental: Each backup operation is creating a full copy of our VMs rather than just capturing changes since the last backup.

Vault Bloat: As a result, our backup vault is becoming severely bloated. We're seeing exponential growth in storage usage, far beyond what we anticipated.

Increased Costs: The bloated vault is leading to higher storage costs, as we're essentially storing multiple full copies of each VM.

We've verified our configuration and ensured we're using supported VMware versions. Despite this, the issue persists. We're reaching out to the community to see if anyone has encountered similar problems or can offer advice on resolving this critical issue.

Has anyone else experienced this with AWS Backup for VMware? Any suggestions for troubleshooting or workarounds would be greatly appreciated.

I have already confirmed the following:
- The option **ctkEnabled** is equal **True**.
- ESXi v7.0
- VM Hardware v 12.0

Jimbo Surlo · Answer

The notion is that AWS backups should be sending data writen since the last backup since you have Change Block Tracking enabled. As VMware presents storage to your VMs, the OS on the VMs write to that storage. If the application writes staging data to disk, even if it deletes it, then the blocks have changed. As an example, windows would let you undelete a file - so, new blocks are written when the file is created, then when the file is deleted, the file table is changed, but the blocks are still there and noted as changed. The worst case would be if one would defrag the hard drive - blocks are moved around the virtual disk - and thus many if not all changed blocks.

Don't know if the above is relevant to your environment.

Further data: If you haven't seen this, all looks good - except you didn't note if you rebooted your VMs. https://docs.aws.amazon.com/aws-backup/latest/devguide/vm-troubleshooting.html

Don't know if the above is relevant to your environment.

Further data: If you haven't seen this, all looks good - except you didn't note if you rebooted your VMs.
https://docs.aws.amazon.com/aws-backup/latest/devguide/vm-troubleshooting.html

dsonger-AWS · Answer

Check your backup plans and more specifically retention settings.  Lower retention settings can cause an incremental to look like full backup since the backup data would be refreshed more frequently.  
"Understand the cost impact of incremental forever backups. Legacy backup solutions used schemes that needed periodic full backups, along with incremental backups. Backups in AWS are based on incremental forever snapshots that only need a single full snapshot to be taken. For example, a 100GB Amazon Elastic Block Store (Amazon EBS) volume with a 2% daily change rate would accumulate ~160GB of snapshot data over a single month of daily backups, with a 30-day retention. Using a legacy approach with weekly full backups, the same volume would accumulate ~600GB of snapshot data."

https://aws.amazon.com/blogs/storage/optimizing-aws-backup-costs/

Giovanni Lauria · Answer

You’re facing issues with AWS Backup for VMware creating full backups instead of incremental ones, causing storage bloat and higher costs.

Despite verifying configurations and VMware versions, the problem persists. To address this, double-check your backup settings, review logs for errors.

AWS Backup for VMware Incremental VM backups

Relevant content