Ransomware Attack in San Jose? Here’s What to Do in the First Hour

Photo of author

By Markus Winkelhock

The first hour decides more than the next week

Isolate the infected machines. Don't shut them down yet.

Pull the network cable or disable Wi-Fi. If you can, cut the affected VLAN or switch ports at the firewall. Powering a machine off can wipe the memory where investigators sometimes find encryption keys and the attacker's tools. Disconnecting keeps that evidence and still stops the spread.

Then call people in this order:

  1. Your IT lead or managed service provider.
  2. Your cyber insurance carrier. Most policies require notice quickly, and many only cover work done by firms on their approved panel.
  3. A lawyer with breach experience.
  4. The FBI, through ic3.gov or the San Francisco field office, which covers San Jose.

Don't email the attackers from a work account. Don't wipe anything to "start fresh." Wiping destroys the evidence you need to learn how they got in.

Figure out what you actually lost

Ransomware attacks are rarely just encryption anymore. Many crews copy your data out first, then lock the systems. So you're answering two questions, not one: what can't I open, and what did they take?

The first question is about recovery. The second is about legal exposure. Keep them separate, because the answers often differ. You may restore everything from backup in three days and still owe notification letters.

Ask your responders to check for:

  • The initial entry point (usually a stolen password, an unpatched VPN or firewall, or a phished employee)
  • Any new admin accounts the attacker created
  • Large outbound transfers in firewall or cloud logs
  • Whether your backup server was touched

That last one matters most. Attackers know to hunt for backups first.

Should you pay?

I wouldn't, and I'd tell a client the same.

Payment doesn't guarantee a working decryptor. Even when the tool works, it's often slow and leaves files corrupted, so you end up restoring from backup anyway. Paying also marks you as a business that pays. And if the group is on a U.S. Treasury sanctions list, paying can create its own legal problem. The Treasury's OFAC has published an advisory on this, and your lawyer should check it before anyone touches a wallet.

There's one honest exception. If you have no usable backups and the data is the business, such as a clinic's patient records or a design firm's only copy of client files, payment may be the least bad option. Make that call with counsel and your insurer, not alone at 2 a.m.

Paying is a last resort, not a plan.

Restore in the right order

Don't bring everything back at once. You'll reinfect it.

Rebuild from the center outward. A sensible order:

  1. Identity. Domain controllers, Active Directory, or your cloud identity provider. Reset every privileged password and the Kerberos KRBTGT account twice if you run Active Directory.
  2. Core network and security tools. Firewalls, DNS, endpoint protection.
  3. The two or three systems that make money. Billing, ERP, your main database, whatever stops revenue when it's down.
  4. Everything else. File shares, printers, the conference room display.

Restore into a clean, isolated network first. Scan the restored data before reconnecting. And don't restore the backup taken the night before the encryption just because it's the newest. The attacker may have been inside for two to four weeks. That's a rule of thumb from how these intrusions commonly play out, not a fixed number. Go back further, test, and pick a restore point from before the first sign of activity.

What downtime costs

A rough way to see why speed matters. Say you have 40 employees at a loaded cost of $60 an hour, and they can't work for a full eight-hour day. That's about $19,200 in idle labor per day, before lost sales, overtime, or missed deadlines.

A week offline for that company is north of $100,000 in payroll alone. Compare that with the price of the protections below.

Silicon Valley rents and salaries make these numbers climb fast. Your own figures will differ, so plug in your actual headcount and rates.

Backups: what I'd choose

Skip the argument over whether to use cloud or tape. What matters is whether an attacker with your admin password can delete the copy.

My pick for most small and mid-sized San Jose businesses: immutable cloud backup plus one offline or logically separate copy. Immutable means the storage refuses deletion or changes for a set retention period, even from an admin account. Major cloud storage providers offer this feature, usually called object lock.

What I wouldn't bother with is a USB drive plugged permanently into the server, or a NAS joined to your domain. Those get encrypted along with everything else. I've seen it often enough that I'd call it the default outcome.

Rough numbers, and treat them as ballpark:

  • Immutable cloud storage runs roughly $5 to $25 per terabyte per month, depending on provider and retrieval fees.
  • A managed backup service for a 30-to-100-person office commonly lands between $300 and $1,500 a month, depending on data volume and how many servers you protect.
  • Keep 30 to 90 days of restore points. Thirty is the minimum I'd accept.

Test a restore every quarter. Not "check the green light." Actually restore a server to a spare environment and time it. If it takes 36 hours, you've just learned your real recovery time.

Reporting and legal duties in California

If attackers took personal information, California's breach notification law (Civil Code section 1798.82) generally requires you to notify affected California residents without unreasonable delay. If more than 500 residents are affected, you also send a sample notice to the state Attorney General. Healthcare providers have separate HIPAA duties, and public companies face SEC disclosure rules.

Don't guess at which rules apply. A one-hour call with a privacy attorney, often $400 to $900 an hour in the Bay Area (a rule of thumb, not a quote), is cheap next to a missed deadline.

Also keep a written timeline from the first minute: who noticed what, when, and what you did. Regulators, insurers, and your own team will all ask.

Who to hire in San Jose

You have three realistic options:

  • An incident response firm on your insurer's panel. This is what I'd pick. Coverage usually flows through them, and they've done it dozens of times.
  • Your managed service provider. Fine for rebuilding, but ask whether they've handled an active ransomware case. Many haven't.
  • Doing it in-house. Not for anything beyond a single infected laptop.

Better still, sign a retainer before you need one. Many IR firms offer prepaid hours, and having a name and phone number ready saves a day of searching while your systems are down. Expect hourly rates in the $300 to $600 range for responders, again a rule of thumb that varies by firm and urgency.

The five controls that pay for themselves

You can't prevent everything. These five cover most of what actually gets businesses hit:

  1. Multi-factor authentication on email, VPN, and every admin account. Use app-based or hardware keys, not SMS where you can avoid it.
  2. Patching internet-facing devices (firewalls, VPNs) within days, not months.
  3. Endpoint detection and response on every laptop and server, watched by someone, at roughly $5 to $15 per device per month.
  4. Separate admin accounts so your daily-use login can't wipe the backup server.
  5. Immutable backups, as described above.

Total cost for a 50-person company usually runs a few thousand dollars a month, all in. One bad week costs more than that.

A plan on one page

Write down, today, and print it:

  • Phone numbers for your IT contact, insurer, lawyer, and IR firm
  • Where your backups live and who can delete them
  • Which three systems come back first
  • Who talks to employees, customers, and the press

Your email may be down when you need it. Paper works.

Recovery goes better for the companies that decided things before the attack. Decide yours this week.

Leave a Comment