N-able shuts down N-central, its remote management tool, after flaws were published without warning
New attack chains published on November 12 reportedly target even servers patched on the 10th. N-able is restricting some features while it waits for a new patch. Here is what to do and check.
Translated from French by AI. The original French article is the reference version.

On the night of November 8 to 9, a repository called “Able2Shell” appeared on GitHub. It describes 33 alleged vulnerabilities in N-central, with code to exploit them. On November 9, the US vendor N-able took its hosted service offline in the Americas, Europe and Asia-Pacific. It also asked customers who run the product on their own servers to shut them down until further notice.
N-central is a remote management tool, what the industry calls an RMM (remote monitoring and management) platform. Managed service providers use it to monitor, update and remotely control their customers' computers and servers. According to TechTarget, N-able claims in its marketing materials to serve “more than 500,000 organizations.” On November 10, the vendor released a patch, version 2026.4 Hotfix 2 (build 2026.4.1.26). It fixes seven vulnerabilities with CVE identifiers, including two rated 9.3 out of 10: CVE-2026-125883 and CVE-2026-125881.
Why the vendor chose to cut everything off
One of the published attack chains requires no credentials. It uses a flaw in XML file processing (known as XXE) to run commands on the server under the “nable” service account. According to the researcher, it could also create an administrator account. Tenable's analysis says other chains lead to full control (root), but they require an account or an administrator who views a booby-trapped page. And a remote management server can, by design, act on every device it manages.
N-able denounces a “public disclosure and irresponsible communication” of critical vulnerabilities. The researcher, whom BleepingComputer identifies as Sean Bales, says they published without warning the vendor because other flaws had allegedly not been fixed. They had reported those flaws five to six months earlier in another N-able product. The vendor has not responded to these accusations.
Attempts detected, no confirmed intrusion
N-able says it is seeing scans linked to the published code, with no evidence that any customer has been breached. According to Beazley Security, the company watchTowr detected exploitation attempts against its honeypot servers, and Defused observed probing aimed at these flaws. Huntress, for its part, had seen no exploitation among its partners as of November 10. The number of N-central servers visible from the internet fell from 1,338 on November 9 to just over 148, according to the Shadowserver Foundation, cited by BleepingComputer.
The patch may not fix everything. N-able has not said which of the researcher's findings correspond to its seven CVEs, and it has announced a second patch. This is not the tool's first warning: according to Tenable, three other N-central vulnerabilities were added in August and September to the US CISA agency's catalog of exploited vulnerabilities. Managed service provider tools attract attackers, as the hack of ConnectWise's backup service showed.
What to do if you use N-central
No intrusion has been confirmed, so there is still a window to update. Here is N-able's guidance, as relayed by Tenable and Rapid7.
- Server installed on your premises: urgently upgrade to version 2026.4.1.26, and keep the server offline until the update is done.
- Agents and probes: update the agents first, then the probes, because an updated probe no longer communicates with an old agent. This also applies to customers of the hosted service, even though N-able says it has upgraded all of its servers.
- Affected features: the patch removes the legacy mobile device management (MDM) and temporarily disables the built-in remote control. The Take Control tool remains available.
- Traces to look for: Huntress advises reviewing logins since November 8, unusual remote sessions, recently deployed scripts and new administrator accounts.
- If you are a customer of a managed service provider: ask which remote management tool it uses and whether it is up to date.
It remains to be seen when the second patch will arrive. And whether attackers will manage to exploit the published code before the last servers are updated.
Update of November 13, 2026
The patch N-able released on November 10 for its N-central remote management tool does not close the matter. On November 12, the researcher behind the “Able2Shell” flaws, who now signs as “MSPNightmare,” published two new sets of exploits. They say these work against that patch (2026.4 Hotfix 2), according to Tenable's FAQ. The first, 4EverAble2Shell, would allow full control (root) of the server without credentials. However, it requires N-central's old programming interface, the “Legacy API,” to be enabled. The second, Agent2Shell, reportedly targets the agents and probes installed on Windows machines.
N-able responds that the server flaw depends on a non-default setting: the Legacy API enabled and exposed to the internet. While waiting for a new patch, the vendor is restricting some features as a precaution. It has cut the integration with Intune, Microsoft's device management tool, in its hosted environments. It is temporarily disabling it for customers who use it on their own servers, and asks them to check that the Legacy API is indeed disabled. N-able says it has seen no sign of compromise.
A first exploitation has nonetheless been reported. The company Huntback says an attacker used 4EverAble2Shell against production servers to install XMRig, software that mines the Monero cryptocurrency. The attacker's mining account listed 21 machines, all N-central servers according to Huntback, 19 of them already on Hotfix 2. These figures have not been independently verified. Huntress, which spoke before this report, had seen no exploitation among its partners. As of November 12, none of the seven flaws fixed by Hotfix 2 was in the US CISA agency's catalog of actively exploited vulnerabilities, according to Tenable.
Method: AI-assisted monitoring, verified sources, article reviewed and approved by Frédéric LENQUETTE. Our editorial policy.
Share this article