← Breach analysis by the Dual Vulnerabilities framework

Hotels & government organisations worldwide espionage (FamousSparrow, 2019)

2019 · I - Accommodation and food service activities · State-aligned cyber-espionage via exploitation of internet-facing servers (data theft, system misuse)

Analysed 2026-11-06 by Hermes

Open this case in the tool →

What happened

State-aligned cyber-espionage via exploitation of internet-facing servers (data theft, system misuse)

Dual vulnerability type

Expertise to Execution

Evidence coverage: 2/6 stages, weight 13/60

Decisive stage: entry (clearest evidence) · Alternative: Strategy to Execution

Why this classification

(1) The victims were a diffuse set of organisations - mainly hotels, plus governments, international organisations, engineering companies and law firms - and the documented expectation for them is an operational baseline: patch disclosed vulnerabilities on internet-facing servers promptly and do not expose back-office business and management web applications (Exchange mail servers, SharePoint, the Oracle Opera hotel-management system) to the internet at all; this is specialist-level knowledge and ESET's report itself states the remedy. (2) In practice FamousSparrow entered since at least August 2019 through known vulnerabilities in exactly those internet-facing applications - including ProxyLogon, exploited from 3 March 2021, one day after the patch, amid a publicly reported exploitation frenzy that had run since January 2021 - then deployed the SparrowDoor backdoor and Mimikatz and exfiltrated data; the victims themselves never noticed, and the campaign surfaced only via ESET telemetry in September 2021. (3) The classification rests on the entry stage (clearest evidence: a primary forensic report documents the vector for these victims): the unmet expectation ORIGINATES with technical specialists - an available, known remedy needing no strategic decision - and the chain BROKE in operations and architecture, the applications remaining vulnerable and internet-exposed. That is a gap within the material column: Expertise to Execution. The long dwell time (detection stage, assessed at inference level) points to the same type and reinforces it without changing the choice. No source documents a doctrinal artefact or leadership decision at these victims, so a Strategy-type reading is not evidenced and is kept only as the alternative. RM18 (Expertise to Execution lens): the exploited vulnerabilities were known and remediable - ProxyLogon patched on 2 March 2021 with exploitation publicly reported since January 2021, SharePoint and Oracle Opera flaws disclosed and patched earlier - and the victims' remediation lagged behind the available fix (causal role: Enabling). RM27 (Expertise to Execution lens): back-office business and management web applications, including hotel-management software, were left unnecessarily reachable from the internet, so operations did not maintain a reduced exposure surface (causal role: Enabling). RM34 (Expertise to Execution lens): activity dating back to 2019 was never picked up by the victims' own monitoring and surfaced only through a third party's telemetry years into the campaign, a dwell time showing detection capability was not realised in practice (causal role: Obscuring).

Evidence by stage

Risk markers

Sources

Notes

EuRepoC multi-victim row: no individual victim organisation is publicly named, so the victim is named descriptively per EuRepoC convention; receiver_country spans 10 countries, so Country is left blank per the method-spec convention for multi-organisation campaigns (Year taken from eurepoc_extra.start_date, 01.08.2019). The victim set spans several sectors (hospitality, government, law/engineering); Sector records the main documented target group, hotels, and no single victim organisation's sector otherwise applies. Attribution caveat: the case file's initiator_name lists Salt Typhoon/FamousSparrow/GhostEmperor/UNC2286 as one actor; Microsoft conflates these clusters while ESET maintains FamousSparrow is a distinct group with loose links (ESET report, 26-03-2025) - the DV assessment is victim-side and unaffected. Two dataset source links describe different incidents that only mention FamousSparrow in passing and were not used for the analysis: The Record's MGM Resorts article (September 2023 attack) and The Hacker News article on the Barracuda zero-day campaign (UNC4841, August 2023). ProxyLogon timing: ESET telemetry puts FamousSparrow's exploitation one day after the patch, so the marker RM18 is coded on the campaign's documented pattern of exploited known vulnerabilities (Exchange, SharePoint, Oracle Opera) and the ~2-month public exploitation window since early January 2021 (DIVD: 46,000 of ~250,000 scanned Exchange servers still unpatched in March 2021), not on a same-day patch turnaround alone. No victim statement, regulator finding or notification record was found in the sources opened; response and recovery are not assessable (no disruption, third-party disclosure).