CriticalVulnerabilityActive

N-able takes N-central offline after 33 alleged zero-days are published (Able2Shell)

A researcher published exploit code for 33 unpatched N-central flaws; N-able took hosted N-central offline and told on-prem users to shut down pending a fix

Published
Updated
Fixed in
No fix yet. N-able says it is building and validating a hotfix and has given no release time
Exploitation
Not yet. N-able reports no active exploitation or unauthorized data access as of its 10 November 00:00 UTC update, but working exploit code is public

Live scores

CVE-2026-18577
EPSS
0.146
likely to be attacked soon
VEDAS
0.985
strong exploit evidence
On CISA KEV
Yes, since 2026-08-03
Detection rule
Nuclei · Suricata
EPSS (chance of attack)
100.1CVE-2026-18577 EPSS over time2026-082026-11-11
VEDAS (exploit evidence)
100.5CVE-2026-18577 VEDAS over time2026-082026-11-11

Thin horizontal line: the “likely exploited” threshold (EPSS 0.10, VEDAS 0.5). Shaded “no data” areas are periods when cve-scores did not publish.

Live scores

CVE-2026-86218
EPSS
0.154
likely to be attacked soon
VEDAS
0.982
strong exploit evidence
On CISA KEV
Yes, since 2026-09-08
Detection rule
Nuclei · Suricata
EPSS (chance of attack)
100.1CVE-2026-86218 EPSS over time2026-092026-11-11
VEDAS (exploit evidence)
100.5CVE-2026-86218 VEDAS over time2026-092026-11-11

Thin horizontal line: the “likely exploited” threshold (EPSS 0.10, VEDAS 0.5). Shaded “no data” areas are periods when cve-scores did not publish.

Live scores

CVE-2026-86206
EPSS
0.011
some chance of attack soon
VEDAS
0.986
strong exploit evidence
On CISA KEV
Not listed
Detection rule
Nuclei · Suricata

Live scores

CVE-2026-86207
EPSS
0.013
some chance of attack soon
VEDAS
0.981
strong exploit evidence
On CISA KEV
Not listed
Detection rule
Nuclei · Suricata

Live scores

CVE-2026-125878
EPSS
0.004
low predicted chance of attack
VEDAS
0.053
little exploit evidence
On CISA KEV
Not listed
Detection rule
No public rule yet

Live scores

CVE-2026-125879
EPSS
0.004
low predicted chance of attack
VEDAS
0.048
little exploit evidence
On CISA KEV
Not listed
Detection rule
No public rule yet

Live scores

CVE-2026-125880
EPSS
0.005
low predicted chance of attack
VEDAS
0.049
little exploit evidence
On CISA KEV
Not listed
Detection rule
No public rule yet

Live scores

CVE-2026-125881
EPSS
0.005
low predicted chance of attack
VEDAS
0.049
little exploit evidence
On CISA KEV
Not listed
Detection rule
No public rule yet

Live scores

CVE-2026-125882
EPSS
0.005
low predicted chance of attack
VEDAS
0.052
little exploit evidence
On CISA KEV
Not listed
Detection rule
No public rule yet

Live scores

CVE-2026-125883
EPSS
0.009
low predicted chance of attack
VEDAS
0.053
little exploit evidence
On CISA KEV
Not listed
Detection rule
No public rule yet

Live scores

CVE-2026-91000
EPSS
0.002
low predicted chance of attack
VEDAS
0.055
little exploit evidence
On CISA KEV
Not listed
Detection rule
No public rule yet

What happened

N-central is N-able’s remote monitoring and management (RMM) platform, used by managed service providers (MSPs) and IT teams to monitor, patch and run scripts on customer devices from one console.

Late on Sunday 8 November 2026, a GitHub account named Oracle-Security published a release called Able2Shell. It documents 33 alleged N-central vulnerabilities, with write-ups, screenshots and proof-of-concept code. BleepingComputer names the researcher as Sean Bales. TechTarget says the repository’s owner operates under the handle BinaryNinja. The repository README says the author is not affiliated with Vector 35 or its BinaryNinja product. GitHub README

On Monday 9 November, N-able opened a critical security incident. It took its hosted N-central environments offline in the Americas, Europe and Asia-Pacific, and advised on-premises customers to temporarily take their N-central servers offline “until further notice”. N-able status page

N-able describes the disclosure as “non-responsible communication of critical vulnerabilities” found by an independent researcher. It has not said which of the 33 claims are valid, but it confirmed it is building a hotfix, which BleepingComputer reads as support for at least some of the claims. BleepingComputer

What the researcher published

The repository has a README that indexes 33 entries, plus Python scripts for them. GitHub README SecurityOnline counts 26 scripts. SecurityOnline

  • Test target: all testing was done on N-central Appliance 2026.4.0.31, a pre-release build. The author says “most if not all” of the flaws should work on earlier versions, including the latest release, and that every flaw worked on the newest pre-release at the time of posting. GitHub README
  • Bug classes: authentication bypass, credential disclosure, arbitrary file write, directory traversal, cross-site scripting (XSS), XML external entity (XXE) injection and remote code execution (RCE). BleepingComputer SecurityOnline adds OGNL expression injection and several routes to root. SecurityOnline
  • Other impacts the README lists: account enumeration, audit-log tampering, remote-control session termination, device impersonation, and disclosure of database and Ecosystem Agent credentials. Field Effect also lists administrator account creation and configuration changes. Field Effect
  • No login needed for many: the index titles 12 of the 33 entries as anonymous or pre-authentication. That is a count of our own from the README titles, not a figure from the author. GitHub README

The chains reporters highlight

  • Unauthenticated XXE to server takeover. An unauthenticated XXE in the service-import function is used to steal a trusted local session, reset an account password and run commands on the server as the nable user. The same exploit can create an enabled Default Administrator account with attacker-chosen credentials. The researcher told BleepingComputer this is the most critical of the set. BleepingComputer
  • Stored XSS to root. A stored XSS can run commands as root when an administrator views malicious patch metadata. BleepingComputer The README describes the entry point as anonymous patch-metadata injection, with two follow-on routes (OGNL as nable, and an Audit Export injection to root). GitHub README
  • nable to root. A privilege escalation lets code running as the nable account run commands as root. BleepingComputer Huntress Labs also noted the flaws can be chained for privilege escalation to root. TechTarget

BleepingComputer has not independently verified every reported flaw. BleepingComputer Previdian says the working proof of concept demonstrates a full RCE chain from XXE to command execution. TechTarget No CVE IDs have been assigned to the 33 flaws so far. SecurityOnline

Why it was published without notice

Bales told BleepingComputer he did not report the N-central flaws to N-able first. He said he had reported two unrelated RCEs and a privilege escalation in a different N-able product five or six months earlier, that they remain unpatched, that his follow-up emails over the previous four to six weeks went unanswered, and that he received no bug-bounty payment. He said this led him to conclude “they don’t take disclosure seriously”. BleepingComputer

BleepingComputer asked N-able about those claims and the company did not address them. Bales also said N-able’s CEO and CTO contacted him after the release and asked him to take down the repository and work with them, without threats. He also confirmed that finding the flaws was AI-assisted. BleepingComputer These are the researcher’s statements, and N-able has not confirmed them.

What N-able has said

N-able’s status page, last updated 10 November 00:52 UTC, shows a major outage for hosted N-central in all three regions and for on-premises N-central in all three regions. N-able status page

  • 9 November, 15:15 UTC: incident opened. N-able says it has not seen active exploitation, is working on a hotfix and expects it “soon”.
  • 9 November, 19:59 UTC: N-able reports no evidence of active exploitation or unauthorized data access, and says it will share an estimated restoration time once it has one.
  • 10 November, 00:00 UTC: N-able says engineering is “on track with no setbacks”, that validation is continuing, and that it has seen no indication any customer environment was compromised.
  • Cadence: updates every four hours, or sooner if significant new information appears.

While N-central classic is down, services under N-able Login stay up, including Asset View, patching, vulnerability management, scripting, policies and tagging. They can be reached at n-able.app with an N-able Login, whether the partner is hosted or on premises. N-able also added preview features there, including Take Control from Asset View and on-demand Windows patching. N-able status page

Previdian’s founder told TechTarget on Monday that his honeypots had seen no exploitation attempts yet, but that he expected that to change quickly. He said attackers would need at least a few hours to test and weaponize the code. TechTarget

Am I affected?

  • Hosted N-central (Americas, Europe, Asia-Pacific): yes, per N-able’s status page. N-able took these environments offline itself. N-able status page
  • On-premises N-central, any region: treat every build as exposed. N-able has not published a list of affected builds, and the author says the flaws work on the latest release too. SecurityOnline GitHub README
  • Internet-exposed management interface: highest risk. Several published proofs of concept describe unauthenticated paths, so network reachability is the main condition for exploitation. Field Effect
  • What is reachable from N-central: the real exposure is the access N-central holds. Field Effect says a compromised RMM server can reach downstream systems through its existing trust, including software deployment, script execution, credential access and remote administration. Field Effect

What to do now

  1. Keep on-premises N-central servers offline until N-able gives further guidance saying it is safe to bring them back. N-able status page
  2. Use n-able.app for day-to-day work in the meantime, with your N-able Login, for patching, scripting, vulnerability management and Take Control. N-able status page
  3. Check whether N-central was reachable from the internet before you took it offline, and when it comes back, limit it to a dedicated administration network. Field Effect
  4. Review activity from before the shutdown. Look at authentication events, new administrator accounts, privilege changes, configuration edits, software deployments, script runs, remote-control sessions and automation-policy changes. Pay particular attention to the period just before servers went offline. Field Effect
  5. Map what N-central can reach. List the domain controllers, virtualization platforms, backup systems, identity providers and security tools it has privileged access to, so you know the scope if it was compromised. Field Effect
  6. Rotate credentials if you find anything unexpected. That means N-central administrators, service accounts and integrations. Then watch managed systems for software deployments, script execution and remote-control sessions that originate from N-central. Field Effect
  7. Use the indicators from earlier this year too. During the September intrusion, Huntress saw an unknown user create an account named to blend in, and the attacker installed Cloudflared, a tunneling tool. Cybersecurity Dive Huntress also advised auditing for unexpected users, such as ones with .invalid email addresses. Huntress
  8. When the hotfix ships, install it before reconnecting. N-able has not said what it will be called or which builds it covers.

Why this matters

N-central has had a bad year. This is the latest in a run of N-central security events in 2026:

  • In July and August, attackers used CVE-2026-18577, an authentication bypass, to gain administrative access and pivot to managed systems through built-in remote management. Field Effect The first fix had been incomplete, which is why that bug exists. Cybersecurity Dive See our earlier advisory.
  • In September, N-able fixed CVE-2026-86206 and CVE-2026-86207, which let attackers bypass access controls and create unauthorized administrative accounts. Rapid7’s Stephen Fewer discovered them. Help Net Security Huntress then reported an intrusion in a fully patched N-central instance. Cybersecurity Dive
  • On 5 September, N-able shipped Hotfix 4 (build 2026.3.1.14) for CVE-2026-86218, a pre-authentication RCE rated CVSS 10. Its public advisory said it had no confirmation of exploitation, while a separate notice sent to customers said the flaw “has been observed being exploited in the wild”. Help Net Security Field Effect says CISA later added CVE-2026-86218 to its Known Exploited Vulnerabilities catalog. Field Effect SecurityOnline says CISA now lists five N-central flaws as exploited. SecurityOnline

The shutdown is the signal. Taking the whole hosted platform offline and asking self-hosted customers to do the same is a large step. Field Effect reads the scope of N-able’s response as a sign the company considers the flaws capable of meaningful operational impact if exploited. Field Effect

Public exploit code shortens the clock. Previdian’s view is that attackers need only hours to weaponize published code. TechTarget

Open questions. N-able has not said which of the 33 claims are real, how many flaws the hotfix will cover, when service returns, or whether the researcher’s earlier, unrelated reports on another N-able product were handled. The researcher’s claim that AI helped find the flaws is his own statement to BleepingComputer, and the other sources we read do not address it. We will update this page when N-able publishes the hotfix and CVE identifiers.

Update, 11 November 2026

  • N-able released N-central 2026.4 HF2 (build 2026.4.1.26) on 10 November, and Huntress says all hosted servers have been upgraded and brought back online. (huntress.com)
  • The Hacker News says N-able expects a follow-up hotfix to provide further security updates, and still urges customers to keep N-central servers offline until the updates have been successfully applied. (thehackernews.com)
  • watchTowr says it observed exploitation attempts against N-able N-central honeypots overnight after the Able2Shell release, and Defused Cyber says it saw probing that included opening anonymous agent sessions, pulling appliance config and hitting the remote control endpoint without authentication. (thehackernews.com)
  • N-able has confirmed it is seeing scanning activity associated with the published proof-of-concept code, while saying it has no evidence of successful exploitation in any customer environment. Sophos X-Ops says it has not observed exploitation. (sophos.com)

Update, 11 November 2026

  • BleepingComputer reports N-able has released N-central 2026.4 HF2, which patches some of the leaked flaws, including the ones N-able calls “the most critical”. CVE IDs now include CVE-2026-125878 (OGNL injection), CVE-2026-125879 (local privilege escalation to root), CVE-2026-125880 (stored XSS), CVE-2026-125881 (SSRF), CVE-2026-125882 (RCE via upload path traversal), CVE-2026-125883 (XXE) and CVE-2026-91000 (insecure IPC, local privilege escalation). Whether HF2 covers every disclosed flaw is not confirmed.
  • In a Tuesday update, N-able said all hosted N-central servers in all regions have been upgraded. It “strongly” recommends on-premises customers upgrade as soon as possible and follow its upgrade guidance before restoring normal access.
  • Threat intelligence firm Defused said it saw probing of N-central for the disclosed flaws. N-able has not flagged exploitation in the wild, and no confirmed compromise is reported.
  • Shadowserver counts just over 148 internet-exposed N-central servers, down from 1,338 on Monday.
  • The researcher says the bugs affect pre-release N-central 2026.4.0.31 and likely earlier versions, which N-able has not confirmed. He told BleepingComputer he skipped notifying N-able over unpatched issues in another N-able product; N-able disputes this and calls claims that it ignored responsible disclosure inaccurate.

Sources