Recover (RC): Restoring Operations After an Incident
Recovery plan execution and recovery communication
TL;DR (for busy owners)
- What it is: Recover (RC) is restoring assets and operations affected by an incident, and communicating clearly while you do it.
- Why it matters: Attackers increasingly delete or encrypt backups first. If the identity that can delete production data can also delete its backups, recovery may not be possible.
- Quick action: Confirm that backups of critical systems live where a compromised production administrator cannot delete them — and that a restore has actually been tested.
The two Recover categories
| Code | Category | In plain English |
|---|---|---|
| RC.RP | Incident Recovery Plan Execution | Restoration activities are performed to bring systems and services back |
| RC.CO | Incident Recovery Communication | Restoration is coordinated with internal and external parties |
How recovery fails
An attacker with administrator rights deletes snapshots and backup recovery points in the same account before deleting production data, or schedules the deletion of the KMS key that encrypts them. Backups that existed on paper are gone. See the Impact tactic guide for how this is done and prevented in AWS.
Fast checks (30 minutes)
Recover readiness
- Critical systems and their recovery priority order are documented (RC.RP)
- Backups are stored in a separate account or protected with vault lock / Object Lock (RC.RP)
- A restore of a critical system was tested in the last 12 months (RC.RP)
- KMS keys protecting backups cannot be scheduled for deletion by workload administrators (RC.RP)
- There are prepared templates for customer and staff updates during recovery (RC.CO)
Evidence to keep
Recovery evidence
Backup plan and vault-lock configuration
Restore test record with time taken and issues found
Recovery communication templates and approvers
Map to frameworks
NIST CSF 2.0 Recover categories
Where Securitain fits
Securitain does not perform backups or restores. Its IAM analysis can show which identities hold destructive permissions — such as deleting recovery points or scheduling key deletion — so you can make sure backups are out of their reach.
Next steps
A restore you have never tested is a hope, not a plan
Schedule one restore test of a critical system this quarter and record how long it took. That single exercise usually reveals the most important recovery gaps.
Sources & References
- 1.The NIST Cybersecurity Framework (CSF) 2.0 — NIST CSWP 29
Official framework text
- 2.AWS Backup Vault Lock
Immutable backups on AWS
- 3.NIST SP 800-61 Rev. 3
Incident response, including recovery
Discuss backup isolation and recovery testing for your AWS environment.
Plan your recovery with a security architect