中文
L4 · Septic Tank BlewFortinet · GlobalProtect · credentialsk0t-edge-credential-campaign

k0t edge-device campaign exposed over 70,000 credentials and established persistent network access

Group-IB and NyxLab report a distinct credential dataset and observed intrusions involving exposed FortiGate, FortiWeb and GlobalProtect devices. Over 70,000 credentials are not a count of victims; ransomware affiliations remain unconfirmed.

Disclosed
Oct 21, 2026
Public disclosure date
Severity
Septic Tank Blew
2 CVEs tracked here
Splash zone
FortiGate weak passwords; vulnerable FortiWeb; GlobalProtect override cookies with a reused portal/gateway certificate
Fortinet · GlobalProtect · credentials
In the wild
Yes
Group-IB and NyxLab observed persistent access and data exfiltration

What happened

The researchers found negligible overlap with the earlier FortiBleed credential dataset. Links to FortiBleed and ransomware groups remain assessments, not proven common operators. Exact victim attribution is incomplete. This report documents newly disclosed campaign evidence rather than a new vulnerability.

For GlobalProtect, check Network → GlobalProtect → Portals/Gateways → Agent authentication settings. Disable authentication override or generate a new dedicated cookie certificate; do not reuse a portal/gateway certificate. Upgrade every portal and gateway using the vendor’s version-specific table.

What’s in the bowl

CVE-2025-64446

FortiWeb initial-access path

Campaign tooling exploits FortiWeb, not FortiGate, through this vulnerability.

CVE-2026-0257

GlobalProtect authentication-bypass path

Requires authentication override cookies and vulnerable certificate configuration; the campaign forged cookies using the portal certificate.

Crime-scene timeline

  1. Oct 21, 2026

    Group-IB and NyxLab publish their joint investigation

  2. Feb–Jun 2026 (approximate)

    Many observed FortiGate compromises date to this period

Who said what

We list each claim and who made it. Primary sources (the vendor, the researcher, the registry) are marked. Where reports disagree, we grade by the worst credible case.