2020 · O - Public administration and defence · Ransomware deployment via exploited internet-facing servers and brute-forced remote access (multi-victim campaign)
Analysed 2026-11-07 by Hermes
Open this case in the tool →Ransomware deployment via exploited internet-facing servers and brute-forced remote access (multi-victim campaign)
Expertise to Execution
Evidence coverage: 5/6 stages, weight 38/60
Decisive stage: entry (clearest evidence) · Alternative: Strategy to Execution
(1) The expectation at these victims sat with their own technical staff: internet-facing and remote-access systems are expected to carry published critical updates (the ZeroLogon fix of 11-08-2020, with NCSC and CISA active-exploitation warnings from September 2020), RDP is expected to be hardened against brute forcing, EDR alerts and log retention are expected to be operationalised, and backups are expected to be out of an intruder's reach. (2) In practice the campaign selected victims precisely because those operational basics were missing: ESET states CosmicBeetle does not choose its targets but finds servers with critical security updates missing, and its ScPatcher component installed the ZeroLogon updates on machines the operators had already compromised - proof that the victims ran unpatched systems years into active exploitation. At the one victim documented forensically (the Polish company in the Cyber Arms / Zaufana Trzecia Strona case study that ESET built on), the attacker entered over a VPN access nobody could reconstruct because logs did not reach that far, moved laterally by brute forcing internal Windows services, took dominance of Active Directory, downloaded sensitive datasets unseen, and deleted data from backups before encrypting - while the victim's own EDR only sounded serious at the encryption stage. (3) Every load-bearing stage therefore shows the same pattern: the knowledge of what to do was material and specialist-level (a published patch, credential and privilege hygiene, alert triage, backup isolation) and the chain broke in day-to-day execution - no source documents a victim-side policy that existed or a leadership decision that deferred one, so the gap is Expertise to Execution, not a doctrinal failure. The classification rests on the entry stage, where origin and break are best documented (population-level forensic evidence from ESET plus the NCSC alert that the remedy was available and known), and is corroborated by the detection, containment, egress and recovery gaps of the documented victim; the specific initial-access route of each individual victim is not known per case and is not individually load-bearing. RM18 (Expertise to Execution lens): the ZeroLogon vulnerability was known and remediable from 11-08-2020, yet remained exploitable on this campaign's victims until the attackers patched the machines themselves - the remedy was available but never executed, delayed by victim-side capacity and legacy (causal role: Enabling). RM34 (Expertise to Execution lens): the documented victim had a good EDR in place, but it was insufficiently tuned and handled for this intrusion - custom Delphi tools passed as harmless and alerts only escalated at disk encryption, so detection came at the impact moment (causal role: Obscuring). RM41 (Expertise to Execution lens): after entry the attacker could brute force and escalate to dominance of the Active Directory environment, showing privilege tiering and directory hygiene that specialists know but that was not realised in the estate (causal role: Amplifying). RM31 (Expertise to Execution lens): backup arrangements were within the attacker's reach - data was deleted from the backups and shadow copies killed before encryption - so the recovery capability that resilience depends on was not isolated (causal role: Constraining).
RM18 — Vulnerability / patch latencyRM34 — Monitoring / detection / alert-handling gapRM41 — Privilege-escalation / directory-hygiene pathRM31 — Resilience / backup preparedness gapMulti-victim EuRepoC row: the victim is a population of ESET-telemetry victims in 20+ countries (Thailand, Israel, Poland, Brazil, Turkey, Mexico named), so Country is left blank per the method spec (diffuse/unspecified-population campaigns, never guessed). Sector: the output check requires one exact CBS SBI entry, so the sector is taken from the most concretely documented victim in the row's victim set, the local government institution in Poland named by ESET (O Public administration and defence); the victim population spans many sectors (hospital, insurer, school, entertainment provider) and the single code understates that - researchers may prefer to blank it in the workbook. Year 2020 from eurepoc_extra.start_date (01.05.2020), the start of the tracked campaign. The row's description matches its stated victim (a campaign description for a campaign victim set) - no data-quality mismatch. The Zaufana Trzecia Strona / Cyber Arms article is a Polish-language incident-response case study of one victim of the same toolset (early 2023, within the case window); ESET names it the first publication on Spacecolon and builds its report on it, so it is used here as the incident-level lifecycle evidence for detection, containment, egress and recovery. Actor unattributed by ESET (Turkish-speaking developer suspected); per the method an unattributed actor is not a claim and does not lower confidence. BleepingComputer (opened 07-11-26, June 2024 reporting) states ESET now tracks the group as the NoName gang, successor ransomware ScRansom, exploitation of further SMB vulnerabilities (EternalBlue, FortiOS SSL-VPN, noPac) and probable RansomHub affiliation - continuity of the same pattern after the 2020-2023 window, not part of the classification. eurepoc_extra.intelligence_impact is coded ambiguously (corruption without leakage OR leakage without corruption); the ESET report and the Polish case study document both data theft and destruction, and that ambiguity did not affect any stage judgment. FortiOS initial access is only low-confidence in the ESET report and was not coded.