You do not need to be a critical infrastructure operator to get pulled into a ransomware mess. A growing business looks like a payday too, especially if you have remote access, shared drives, and a busy team.

CISA’s recent #StopRansomware advisory on Gunra is a good reminder of the basics: attackers get in through exposed systems, steal data, then encrypt what you need to run the business. The difference between a bad week and a business-stopping event is usually preparation.

What the Gunra advisory is really telling you

If you strip away the technical detail, the Gunra advisory is describing a familiar pattern.

  • Initial access through internet-facing vulnerabilities. Gunra actors have been observed exploiting CVEs in internet-facing devices to get a foothold. If something is exposed to the internet and unpatched, it is a target.
  • Double extortion. The play is not just encryption. Data theft comes first, then the pressure to pay increases because the threat is operational downtime plus potential data exposure.
  • A short clock. The advisory notes pressure to pay within a matter of days. That timeline matters because it pushes businesses into rushed decisions, which is exactly when mistakes happen.

The practical takeaway for a small business is not “learn Gunra”. It is “assume a ransomware incident includes data theft, and build recovery that does not depend on negotiating”.

The minimum backup design that survives ransomware

A backup that sits on the same network, uses the same credentials, and can be deleted like any other file is not a ransomware backup. CISA’s joint ransomware guidance is blunt about this: keep backups offline, encrypted, and test them regularly.

Here is the minimum design we like for small and mid-sized businesses.

  • An off-site copy that is isolated. Keep at least one backup copy that is not directly reachable from your day-to-day network. “Offline” can mean physically disconnected, or logically isolated so ransomware cannot see it and encrypt it.
  • Immutability or write-once protection. Where possible, use backup storage that cannot be altered or deleted for a set retention period. NIST calls out immutable storage as a ransomware-relevant control, because it stops attackers from quietly destroying your safety net.
  • Encryption you control. Backups should be encrypted, and you should know where keys live. This matters for two reasons: it reduces the impact of stolen backup data, and it forces you to prove you can actually restore.
  • Separate credentials from your main directory. If an attacker compromises your Windows domain or cloud admin accounts, they should not automatically get the keys to your backup platform.
  • Gold images for critical systems. CISA’s ransomware guide recommends maintaining “gold images” so you can rebuild servers or key workstations quickly, rather than trying to resurrect a broken machine under pressure.

If you are thinking, “We use Microsoft 365, so we are fine”, slow down. Microsoft keeps the service running, but recovery from ransomware often involves more than email. You still need a plan for endpoints, file shares, line-of-business apps, and identity.

What to test quarterly (so recovery is not a guess)

Quarterly testing is where most plans fall apart, because it forces you to answer uncomfortable questions with real timings.

A useful quarterly drill is not “can we restore a file”. It is “can we restore the business”.

  • Restore one server or one critical workload end to end. Pick something that would actually hurt if it was down (file server, accounting system, line-of-business database). Rebuild it the way you would after ransomware, not the easy way.
  • Prove your backups are isolated. Try the restore from an account that is similar to what an attacker might compromise. If that account can delete backups or change retention, fix that.
  • Validate your RTO and RPO in plain English. How long until your team can work again (RTO), and how much data you can afford to lose (RPO). Write the numbers down, then measure whether your restore hits them.
  • Test “gold image” rebuild speed. Time how long it takes to stand up a clean system from a standard image, patch it, install core apps, and hand it to a user.
  • Run a communications check. NIST’s ransomware profile recommends planning for out-of-band communications. In a real incident, email and Teams might be unavailable. Decide now what you will use instead, and make sure leadership can access it.

Keep the output simple: what you tested, what broke, how long it took, and what you will change before the next quarter.

The first 24 hours: decisions that keep you operating

The first day is about control and clarity. CISA’s ransomware response guidance is built around the same idea: stop the spread, assess impact, and restore from safe backups.

Here is a practical decision list you can use.

  • Decide who is in charge. One incident lead, one business lead, one technical lead. If everyone is in charge, nobody is.
  • Contain first, then investigate. If encryption is actively spreading, you may need to isolate systems quickly. The goal is to stop additional damage while preserving enough evidence to understand what happened.
  • Assume data was taken. With double extortion, treat it as a potential data exposure until proven otherwise. That changes how you think about notifications, legal support, and customer communications.
  • Pick a restore priority list. Start with what makes money and serves customers. For many businesses that is: identity and sign-in, phones, email, accounting, then file access.
  • Do not rebuild into the same problem. If attackers got in through an exposed system or weak remote access, closing that hole is part of recovery, not a later improvement.
  • Make a call on law enforcement and reporting. Even if you do not expect an immediate fix, reporting can help with guidance, coordination, and documentation.

One small but important point: keep a printed copy of your recovery contacts and key steps. NIST explicitly notes that plans should be available offline in case you lose access to internal documents during an incident.

A calm way to turn this into a real plan

If you want a Gunra ransomware prevention and recovery plan for small business that is more than a PDF in a folder, keep it tight.

  • One page of priorities. What must come back first, second, third.
  • One diagram of where backups live. Include what is immutable or offline, and who can access it.
  • One quarterly calendar reminder. Restore drill, patch review for internet-facing systems, and a quick check that backup retention has not drifted.

If you would like help designing ransomware-resistant backups and a quarterly recovery drill your team can actually run, the Flexnet Networks team can help you put the plan in place and prove it works.

Sources