> ## Documentation Index
> Fetch the complete documentation index at: https://help.nops.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Delete GCP Disk Snapshots Older Than 90 Days

> Delete Compute Engine disk snapshots older than 90 days, or shorten the retention of the schedule that made them.

Disk snapshots are billed for the data they hold for as long as they exist. Many are kept long after the disk they protect has changed or been deleted. nOps finds snapshots that are at least 90 days old and shows their storage cost as the most you could save by deleting them.

| | |
| - | - |
| **Cloud** | Google Cloud |
| **Category** | Storage |
| **Service** | Compute Engine |
| **Action** | Delete |
| **Effort** | Low |
| **Restart needed** | No |
| **Can be rolled back** | No |
| **Analysis window** | Last 90 days |
| **Source** | nOps |

## Why it matters

Snapshot schedules and manual backups add up over time. Old snapshots are rarely restored but are billed every month.

## How nOps finds it

nOps reads up to 160 days of Compute Engine snapshot storage charges. A snapshot's age is counted from the earlier of the day it was first billed and the creation date in its name (for snapshots made by a schedule). A snapshot is flagged when it is at least 90 days old and was billed in the last 2 days. Snapshots whose names include words such as keep, retain, permanent, golden or do-not-delete, and snapshots named by backup tools (such as Veeam, Velero, Commvault, Rubrik, Cohesity, Druva or Clumio), are left out. The saving is the snapshot's storage cost over the last 7 days, scaled to a month, and must be at least \$5 a month. Projects whose spend on the service is at least 90% paid by credits or promotions are left out.

## Before you start

Know your backup retention rules, and which snapshots are used to create images or restore disks.

## How to implement

1. In the console, open **Compute Engine → Snapshots** and sort by creation date.
2. Check whether each old snapshot is needed for recovery, compliance or to build images.
3. Delete the ones you do not need: `gcloud compute snapshots delete SNAPSHOT_NAME`.
4. If the snapshot was made by a snapshot schedule, give the disk a schedule with a shorter retention (for example 30 or 60 days) so old snapshots are deleted automatically.

## Risks and considerations

* Deleting a snapshot is permanent.
* Snapshots are incremental. Data that newer snapshots of the same disk still need moves to the next snapshot, so the real saving can be lower than the estimate.
* Snapshots managed by a backup tool should be changed through that tool's retention settings.

## FAQs

> [accordion] Why is the saving shown as an upper limit?
>
> Because snapshots share data. Deleting an old snapshot only frees the data no newer snapshot of the same disk still uses.

## Related

* [Delete Disks of Long-Stopped GCP VMs](https://help.nops.io/inform/supported-recommendations/delete-disks-of-long-stopped-gcp-vms)
* [Delete EC2 Snapshots Older Than 90 Days](https://help.nops.io/inform/supported-recommendations/delete-ec2-snapshots-older-than-90-days)
* [Delete Azure Snapshots Older than 90 Days](https://help.nops.io/inform/supported-recommendations/delete-azure-snapshots-older-than-90-days)
* [Supported recommendations](https://help.nops.io/inform/supported-recommendations.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.