News and insights

N-able takes N-central offline: when your IT provider's main tool is switched off

A remote management tool lets an IT provider patch, monitor and fix its customers’ computers from one console. That reach makes the tool valuable to attackers. It also means one problem with the tool reaches every customer at once, even those who never log in to it. On 9 November, the maker of one such tool switched it off.

Updated 13 November 2026: N-able released a fix on 10 November, and new exploits published on 12 November reportedly work against it where an older feature is switched on and reachable from the internet.

What happened

N-central is N-able’s remote management tool. IT providers and in-house IT teams use it to look after many devices. N-able can host it, or customers can run it themselves.

On 8 November 2026, a security researcher published details of 33 critical flaws in N-central. They were posted on GitHub with working code to exploit them. The researcher did not tell N-able first, so there was no fix ready. N-able called it a “public disclosure and non-responsible communication of critical vulnerabilities”.

Huntress, a cyber security company that works with IT providers, has looked at the flaws. It says they can be combined into four attacks, named “Able2Shell”. Three let an attacker run their own programs on the N-central server. The fourth gives full control of it. The researcher tested N-central Appliance 2026.4.0.31, a pre-release version, and expects earlier ones to be affected. The flaws had no CVE numbers, the standard reference numbers for flaws, at the time of writing.

On 9 November, N-able took the N-central service it hosts offline in the Americas, Europe and Asia-Pacific. It asked customers who run N-central on their own servers to switch them off until it gives further guidance.

N-able said on 9 November it had seen no sign of attackers using the flaws, and no unauthorised access to data.

Huntress notes that attackers have used earlier N-central flaws to push their own tools out to customers’ networks.

The fix, and new exploits against it

On 10 November, N-able released N-central 2026.4 Hotfix 2 (build 2026.4.1.26) as a critical security update. N-able upgrades hosted servers itself. It told on-premises customers to upgrade as soon as possible. Everyone also needs to upgrade agents and then probes, the parts installed on managed computers and networks. N-able reported scanning but no evidence that any customer environment had been successfully exploited.

On 12 November, the researcher behind Able2Shell published two more exploits. They reportedly work against the new build. “4EverAble2Shell” would let an attacker with no account run commands on the server with full control. “Agent2Shell” would run commands on managed Windows computers, if the attacker knows the server’s appliance ID.

N-able says it is investigating. It says the server issue needs a non-default setup: the Legacy API, an older way for other software to connect, switched on and reachable from the internet. The Legacy API is off by default, and N-able doesn’t recommend using it. Hosted N-central is not affected.

Previdian, which runs decoy servers to watch for attacks, saw attempts within hours. watchTowr, a security research company, reported probes against servers with the Legacy API on. At the time of writing, neither had seen a successful attack.

What to do now

The first three steps come from N-able. The last two are from N-able’s earlier guidance, as reported by Huntress and by GBHackers, a security news site.

  • If you run N-central yourself, upgrade the server to 2026.4.1.26 as soon as possible.
  • Switch off the Legacy API now if it is on, as N-able advises, and make sure the server isn’t reachable from the internet in ways you didn’t intend.
  • Upgrade agents, then probes, whether you use hosted N-central or run it yourself.
  • Look back from 8 November. Review sign-ins to N-central for access you can’t explain.
  • Check for changes nobody asked for. Look for unusual remote sessions, newly deployed scripts and new administrator accounts.

If an IT provider runs your systems, ask whether they use N-central and how they are managing your devices meanwhile.

These Blueprint requirements cover the same ground:

  • RW-SUP-02 Supplier access limited to the job (Foundation). Know what your IT provider’s tools can reach, and what they do.
  • RW-DET-06 Urgent security warnings acted on (Foundation). Act quickly on warnings from your main suppliers.
  • RW-DEV-10 Critical weaknesses fixed within 48 hours (Established). Install updates within 48 hours for critical flaws in systems that can be reached from the internet.
  • RW-NET-18 Everything exposed to the internet is known (Established). Check what your management server exposes, and close anything that shouldn’t be reachable.
  • RW-IDA-19 Admin activity reviewed (Established). Check what admin accounts have done, and follow up anything unexplained.
  • RW-SUP-09 An exit plan for each critical supplier (Established). Know how you’d carry on if a critical supplier’s service stopped.
  • RW-SUP-15 Critical suppliers in your incident plans (Advanced). Agree beforehand how you and your IT provider will work together in an outage.

How RiskWitness helps

An outage like this raises a few questions quickly. Here is how RiskWitness helps you answer them.

“Are we affected?” Suppliers and Assets. If your IT provider is recorded as a supplier, their record lists the assets they provide or hold. Add their management tool as one of them. You can then see in one place what it reaches, and which risks and controls depend on it.

“What happens while it’s switched off?” Risks and risk acceptance. With the tool down, your devices may not be patched or watched. That is a risk of its own. If you decide to live with it for now, accept it with a reason and an end date. If it’s outside your tolerance, a named approver signs it off. Either way, it can’t quietly become permanent.

“Is our server reachable from the internet?” Assets and Controls. Record on the tool’s asset where it’s hosted and who owns it. Add a control for keeping the Legacy API off and the server off the internet, and record a test of it. The test result shows whether the control is working.

“Has anyone got in?” Actions and Incidents. Turn each step above into an action with an owner and a due date. Overdue actions show on the dashboard, so nothing stalls unnoticed. If the review finds something, record an incident linked to the tool and your provider. RiskWitness shows which regulators may need telling, and by when.

“Can we trust our provider’s answer?” Supplier questionnaires and assurance evidence. Send your provider a questionnaire on what they found and did. Keep their reply as assurance evidence, alongside their contract clauses and reviews, ready for when you renew.

“What will we change?” Controls, Frameworks and Reports. Map the controls this tested to the RiskWitness Blueprint requirements above, and record how each test went. If you had no plan for a long outage, add an action to write one. Then produce a PDF report or slide deck for leadership.

Every step is kept in a history that shows who did what, when, and the old and new values. Nobody can edit or delete it. When leadership, an auditor, a customer or a regulator asks how you responded, the answer is already written down.

Sources