What it means
Ransomware incidents can affect more than encrypted files. Attackers may steal data, interfere with backups, compromise identities, and use threatened disclosure to apply pressure. Restoring an application therefore does not resolve every part of the incident. A response must consider safety, business continuity, the attacker's remaining access, the evidence available, and possible exposure of information. Preparation combines reducing likely entry points with a practiced response and recovery process. The goal is to preserve essential operations and restore a trustworthy service, not merely make an error message disappear.
See the idea / Resilience
Restored data needs a fresh working-task check.
Compare the same encrypted dispatch data with an intact or also-encrypted recovery copy.
In this synthetic incident, an unauthorized operator has encrypted selected DispatchDesk data D205 and left demand N205. Event metadata E205 is retained. Independent recovery copy B205 contains the same pre-event records at September 30 20:00 UTC, with no later legitimate changes. Compare a protected copy with a copy also encrypted by the operator. Nora may restore only this copy to an isolated test service; production return is not authorized.
01 / 04
Evidence
D205 is encrypted in the specified scenario. N205 demands payment; protected B205 remains readable.
More about this step
The controlled encryption event is known from the exercise setup. E205 preserves its selected event metadata and pre-event digest, not a usable recovery copy. N205 is an attacker demand, not proof of decryption ability or disclosure. B205’s state differs between isolated runs. The operator’s decryption key is not available to the recovery workflow. Identity and read grants are unchanged.
Step through at your own pace.
Keep in mindThis exercise starts after a specified encryption event; an error or ransom note alone would not prove its cause. Harmless local buffers model affected data and backup reachability, not malware, a real isolated network or real attacker access. Trusted recovery authority and suitable test configuration are assumed. A successful selected test does not establish a clean environment, complete recovery or absence of data theft. Payments, safety systems, other services and production return are outside the exercise.
Read the visual, step by step
Protected copy
- Evidence. D205 is encrypted in the specified scenario. N205 demands payment; protected B205 remains readable. The controlled encryption event is known from the exercise setup. E205 preserves its selected event metadata and pre-event digest, not a usable recovery copy. N205 is an attacker demand, not proof of decryption ability or disclosure. B205’s state differs between isolated runs. The operator’s decryption key is not available to the recovery workflow. Identity and read grants are unchanged.
- Read. A fresh authorized dispatch request returns 503 without records. Valid identity and permission do not make encrypted application data usable. Maya’s established identity has the ordinary dispatch-read grant. DispatchDesk tries to parse its selected active data and cannot serve the dispatch rows, returning 503 and no records. This observes the failed business task rather than inferring it solely from a note or changed file extension. Evidence E205 and demand N205 remain; other systems and possible disclosure are unassessed.
- Restore. Nora submits the exact approved restoration to the isolated test service. B205’s rows are restored, but no fresh task result exists yet. The restore endpoint checks Nora’s trusted identity and exact B205/test-service scope. It independently parses the selected copy before replacing test data. Intact B205 passes and its bytes become the test service’s data. Successful restoration is an application change, not a working-task test or permission to reconnect production.
- Retest. An independent authorized read of the test service returns the exact pre-event dispatch rows. Production return and investigation remain open. A new GET checks the resulting test-service state using Maya’s unchanged identity and read grant. It returns 200 with the exact B205 rows, including order D205, quantity 10 and dispatch bay A. Only this selected isolated task is demonstrated. E205, N205 and the selected backup are retained. Possible theft, continuing attacker access, other dependencies and production return require separate work.
Copy also encrypted
- Evidence. D205 is encrypted in the specified scenario. N205 demands payment; reachable B205 was also encrypted. The controlled encryption event is known from the exercise setup. E205 preserves its selected event metadata and pre-event digest, not a usable recovery copy. N205 is an attacker demand, not proof of decryption ability or disclosure. B205’s state differs between isolated runs. The operator’s decryption key is not available to the recovery workflow. Identity and read grants are unchanged.
- Read. A fresh authorized dispatch request returns 503 without records. Valid identity and permission do not make encrypted application data usable. Maya’s established identity has the ordinary dispatch-read grant. DispatchDesk tries to parse its selected active data and cannot serve the dispatch rows, returning 503 and no records. This observes the failed business task rather than inferring it solely from a note or changed file extension. Evidence E205 and demand N205 remain; other systems and possible disclosure are unassessed.
- Restore. Nora submits the exact approved restoration to the isolated test service. The service rejects unreadable B205 and does not activate a replacement. The restore endpoint checks Nora’s trusted identity and exact B205/test-service scope. It independently parses the selected copy before replacing test data. Also-encrypted B205 fails that validation. The endpoint returns 422 with no replacement data applied. The attempted restore does not obtain the operator’s key or recover the rows.
- Retest. An independent authorized read of the test service still returns 503 and no rows. Production return and investigation remain open. A new GET checks the resulting test-service state using Maya’s unchanged identity and read grant. It returns 503 without rows because no readable replacement was applied. A failed restore request cannot be promoted to verified recovery. E205, N205 and the selected backup are retained. Possible theft, continuing attacker access, other dependencies and production return require separate work.
This exercise starts after a specified encryption event; an error or ransom note alone would not prove its cause. Harmless local buffers model affected data and backup reachability, not malware, a real isolated network or real attacker access. Trusted recovery authority and suitable test configuration are assumed. A successful selected test does not establish a clean environment, complete recovery or absence of data theft. Payments, safety systems, other services and production return are outside the exercise.
Flow references: CISA: ransomware, backups and controlled recovery
AN ILLUSTRATIVE SCENARIO
A selected telemetry read is tested after restoration
In the synthetic incident, selected orbital pass data is encrypted and a demand appears. Vera’s authorized read fails. A protected B205 copy can be restored to an isolated test service, after which a fresh read returns its exact rows. If the reachable backup was also encrypted, restoration is refused and the task still fails. Neither a working test nor restored data establishes a clean environment, reverses disclosure or authorizes production return.
Put it to work
- Reduce exposure through timely remediation, strong identity controls, restricted administration, and segmentation. Identify the services whose loss would create the greatest safety or business impact and prepare fallback operations.
- Maintain protected recovery copies and an incident plan with technical, operational, legal, and communication contacts. Practice isolation and restoration decisions using a scenario appropriate to the organization.
- During an incident, coordinate containment, evidence preservation, access revocation, and recovery. Reconnect systems only after the responsible team verifies the recovery conditions and continuing monitoring is in place.
How to check your work
Run a tabletop exercise and an isolated restoration test. Check that staff can locate contacts and backups during an identity outage, maintain the essential workflow, and demonstrate the conditions required before returning a recovered service to use.
Connect the ideas
- Backup
A copy of data kept so integrity and availability can be restored after loss, preferably beyond the production identity’s reach.
- Incident
An event or set of events that actually or potentially causes a security loss requiring coordinated handling.
- Blast radius
How much additional loss a failure or a response action can cause beyond the original asset.