Rackspace Breached via ScienceLogic SL1 Zero-Day Vulnerability CVE-2026-9537
Managed hosting giant Rackspace confirmed a zero-day remote code execution vulnerability (CVE-2026-9537) affecting the ScienceLogic SL1 (formerly EM7) IT operations platform was actively weaponized against its internal infrastructure. The underlying defect resides within an undocumented, non-ScienceLogic third-party utility packaged directly inside core SL1 appliance distributions, including Data Engine (DE), Configuration Database (CDB), and Administration Portal (AP) modules. Unauthenticated remote adversaries leveraged the flaw to execute arbitrary system code, achieving initial access into three internal Rackspace monitoring web servers that support the multi-tenant MyRack customer management environment.
During the intrusion, threat actors harvested sensitive telemetry and infrastructure metadata, exfiltrating customer account names, internal device identifiers, IP addresses, and AES-256 encrypted internal device agent credentials. While customer virtual machines and core compute clusters remained uncompromised, Rackspace immediately took customer-facing monitoring dashboards offline and rotated all internal service credentials. ScienceLogic rushed emergency patches across SL1 versions 12.1.3+, 12.2.3+, 12.3+, and backported hotfixes to legacy 10.x and 11.x deployments, while withholding the name of the vulnerable third-party library to limit secondary exploitation.
Why it matters:
- Technical & Architecture Risk: Bundled third-party utilities in enterprise appliances bypass software composition analysis and standard web application firewalls, enabling pre-auth code execution in deeply trusted management planes with access to sensitive internal network telemetry.
- Strategic & Defender Imperative: Infrastructure teams must isolate monitoring engines behind strict zero-trust boundary controls, audit all bundled dependencies on virtual appliances, immediately apply ScienceLogic hotfixes, and proactively rotate any agent credentials stored in exposed monitoring databases.