Ransomware Hit the Backups First. Would Your Restore Work?

On Friday, November 13th, 2026, Kaspersky's Global Emergency Response Team (GERT) published a case study of a ransomware intrusion that destroyed an organization's whole virtualization estate over 10 days: four VMware ESXi hosts, every virtual machine on them, both vCenter appliances, and all connected backup infrastructure. The ransom notes named the BlackNevas group. No virtualization flaw was exploited. The attackers used credentials taken from other systems and normal VMware management tools. (Kaspersky GERT on Securelist)
The order of events is the lesson. Kaspersky says the attacker disabled antivirus on the backup environment days before the main attack, then stopped backup services, encrypted the recovery infrastructure, and disrupted administrative access, all before the servers were wiped. In Kaspersky's words, "backup infrastructure was not collateral damage."
This is a Monday brief for founders, CTOs, and security and compliance leads at SaaS and B2B tech vendors that hold, or are working toward, a SOC 2 report or ISO 27001 certification. You do not need to know VMware to act on it. You need three answers from your IT team, each with a date.
Watch the 60-Second Briefing: Watch our narrated YouTube Short on ransomware that hits backups first, on YouTube (@emrynetworks).
What should SaaS and tech leaders take away from this case?
Having backups is not the same as being able to restore. Attackers now plan for your recovery path, and they remove it first.
- What happened: a 10-day intrusion that ended with servers, virtual machines, and backups destroyed. Kaspersky did not name the victim or its industry.
- How they got in: stolen credentials, then legitimate admin features. No virtualization flaw was exploited.
- What failed first: the backup environment. Antivirus off, backup jobs stopped, recovery systems encrypted.
- What to verify now: backup admin logins kept apart from everyday admin logins, an alert that reaches a person when backups stop, at least one offline or immutable copy, and a dated restore test.
How often do backups fail after ransomware?
More often than most teams expect. Cyber insurer At-Bay studied 186 ransomware claims that involved backups. Only 63% of those businesses restored their data from backup, and 31% could not, which At-Bay describes as more than 1 in 4. That was true even though 92% of its policyholders reported having functioning backups in place. Businesses that did restore had claim costs 41% lower than those that could not. (At-Bay, Backup Breakdown, October 2023)
That study is from October 2023, so treat it as a pattern, not this year's number. The pattern matches what the Kaspersky case shows: a backup can exist and still be useless if the systems, logins, and steps needed to restore it are gone.
Why do backups fail?
Four common reasons:
- Shared credentials. The same admin account runs production and the backup console, so one stolen login reaches both.
- No one watching the backups. A stopped job or a disabled antivirus agent on the backup server raises no alert, or the alert goes to an inbox no one reads at night.
- Every copy is reachable. All copies sit on systems an attacker with admin rights can change or delete.
- No real restore test. Jobs report "success," but no one has rebuilt a full system from them, timed it, and written down the result.
What is double extortion, and does a backup stop it?
CISA's #StopRansomware Guide describes double extortion as ransomware crews both encrypting systems and stealing data, then pressuring victims by threatening to release the stolen data. (CISA #StopRansomware Guide) A clean restore fixes the outage. It does not take back data that was copied out.
Kaspersky did not report data theft in this case. The point for SaaS vendors is broader: your recovery plan has to cover both threats. Backups and restore drills handle the first. Access control, logging, and fast detection limit the second, and your incident response plan should say how you would tell customers.
What do SOC 2 and ISO 27001 expect for backups and restores?
Both frameworks expect backups to be tested, not assumed.
- SOC 2: Trust Services Criterion A1.3 (Availability) reads: "The entity tests recovery plan procedures supporting system recovery to meet its objectives." A1.2 covers backup processes and recovery infrastructure.
- ISO 27001:2022: Annex A control 8.13 (Information backup) says backup copies of information, software, and systems should be maintained and regularly tested in line with your backup policy.
- CISA guidance: maintain offline, encrypted backups of critical data, and regularly test their availability and integrity in a disaster recovery scenario.
Neither framework names a 'restore drill' or sets one test schedule for every company. A1.3 applies when the Availability category is in the SOC 2 report's scope; it requires testing recovery plan procedures, and a dated restore test is the evidence auditors commonly ask for. ISO 27001:2022 8.13, when included in your Statement of Applicability, says backups should be regularly tested under your backup policy. Your policy sets the schedule, and your evidence has to show you kept to it. An auditor will usually ask for the date of the last restore test, what was restored, how long it took, and who signed off.
What should your team check this week?
Send these three questions to your IT team or provider in writing and keep the replies with your SOC 2 or ISO 27001 evidence. Ask for a date, a number, and a screenshot, not a reassurance.
- Are backup admin logins separate from everyday admin logins? Ask who can sign in to the backup console, whether those accounts use phishing-resistant MFA, and whether any of them are also domain or cloud admin accounts.
- Would a person get an alert tonight if backups stopped or backup antivirus was turned off? Ask who receives that alert after hours and when it was last tested.
- When did we last test a full restore, how long did it take, and where is the log? Ask for the restore record: the date, the system restored, the time to recover, and any problems found.
What if nobody can answer?
That is the finding. If no one knows when the last full restore happened, the backups have not been proven. Write it down, set a date to run a restore test, and bring in help if you need it.

How does Standard IT differ from Defensible IT on backups?
Standard IT sees green backup jobs and calls it done. Defensible IT proves the restore works and keeps the backups out of an attacker's reach.
Standard IT says "the backups ran last night." Defensible IT can show the separate backup logins, the backup alert history, and the dated restore log.
Standard IT looks like this:
- Backup console reachable with the same admin logins used for everything else.
- Backup failures land in an email folder that no one checks after hours.
- Every copy can be changed or deleted by an admin account.
- "Our backups are fine," with no restore record to show it.
Defensible IT looks like this:
- Separate backup admin accounts with MFA, managed through compliance-first managed IT services.
- 24x7 managed detection and response, so a stopped backup job or a disabled antivirus agent on a backup server gets a person that night.
- At least one offline or immutable copy that a stolen admin login cannot erase.
- A restore drill on the schedule your policy sets, timed and logged as evidence for SOC 2 A1.3 and ISO 27001 8.13.
How does Emry Networks help regulated businesses in Jacksonville FL, Manchester NH, and nationwide?
Emry Networks provides managed IT, cybersecurity, and IT compliance program support for SaaS and B2B tech vendors, medical and dental groups, law firms, financial advisors, defense contractors, logistics firms, and other regulated industries. We serve regulated SMBs nationwide from hubs in the Jacksonville, FL and Manchester, NH areas. Pricing is all-inclusive and flat monthly, so overnight incident response does not bring a surprise invoice. Every engagement follows the 3-phase Emry Assurance Roadmap (often delivered as a 90-day blueprint):
- Phase 1 - Red Zone (Regulatory Discovery): Map who can reach your backups, how copies are stored, and what restore evidence exists against your SOC 2 or ISO 27001 controls.
- Phase 2 - Transition (Security Hardening Sprint): Separate backup logins, add MFA, set up an offline or immutable copy, and route backup alerts to 24x7 coverage.
- Phase 3 - Green Zone (Continuous Management & Drift Control): Run restore drills on your policy's schedule, check for drift, and hand you dated evidence for audits and customer questionnaires.
Not sure your backups would restore? Book a 30-Minute Risk Assessment. In thirty minutes we review your backup access, backup alerts, and restore evidence, then tell you plainly what we find. Prefer the phone? Call (603) 202-3200.
What are the next steps for your team?
Book a 30-Minute Risk Assessment to review your backup access, alerting, and restore evidence. Before that call, send your IT team the three questions above.
- Book a 30-Minute Risk Assessment with our team. Prefer the phone? Call (603) 202-3200.
- Take the free Compliance Readiness Assessment to see where you stand.
- Read is your IT provider leaving you exposed?
Frequently asked questions
What happened in the Kaspersky BlackNevas backup case?
Kaspersky's Global Emergency Response Team published a case study on November 13th, 2026 describing a 10-day intrusion that destroyed four ESXi hosts, their virtual machines, both vCenter appliances, and all connected backups. The attacker disabled antivirus on the backup environment days early, then stopped backup services and encrypted the recovery systems before wiping the servers.
How often do backups fail after a ransomware attack?
In At-Bay's October 2023 study of 186 ransomware claims that involved backups, 31% of businesses could not restore from their backups and 63% could. At-Bay describes that as more than 1 in 4 backups failing when they were needed.
Does a backup protect against double extortion?
Only partly. A backup helps you recover encrypted systems, but it does not stop a criminal from publishing data they already stole. CISA calls the mix of encryption and data theft double extortion. Access control, logging, and fast detection limit the theft side.
What do SOC 2 and ISO 27001 expect for backups and restore tests?
When the Availability category is in scope, SOC 2 criterion A1.3 says the entity tests recovery plan procedures supporting system recovery. It does not name a 'restore drill'; a dated restore test is the evidence auditors commonly ask for. ISO 27001:2022 Annex A control 8.13 says backups should be maintained and regularly tested under your backup policy. Your policy sets the schedule, and you need dated evidence that each test happened.
How does 24x7 MDR help protect backups?
In the Kaspersky case, the backup antivirus was turned off days before the main attack. With 24x7 managed detection and response, an alert like that gets a person the same night instead of the next business day, and the response leaves dated notes for your compliance file.
Next steps
See how we help businesses in your industry, or book a short call to talk through where you stand today.
Read more from our team
Explore insights on compliance and security.
Ready to strengthen your compliance?
Get hands-on assessment and guidance from our compliance experts.


