← Breach analysis by the Dual Vulnerabilities framework

Maersk NotPetya (Sandworm GRU, 2017)

2017 · Denmark · H - Transport and storage · Destructive malware (wiper) attack via a compromised software update supply chain

Analysed 2026-11-07 by Hermes

Open this case in the tool →

What happened

Destructive malware (wiper) attack via a compromised software update supply chain

Dual vulnerability type

Strategy to Leadership

Evidence coverage: 3/6 stages, weight 20/60

Decisive stage: containment (tie-break: pre-breach condition) · Alternative: Strategy to Execution

Why this classification

(1) Strategic expectation: Maersk was the world's largest shipping company with formal IT strategy under execution (centralisation of the group IT estate in Maidenhead was under way, per the company's own Maersk Post), and in 2016 its seniority formally accepted a security redesign - green-lit and budgeted - that explicitly addressed patching, outdated operating systems and, above all, insufficient network segmentation, after IT executives warned that malware could spread wildly beyond its initial foothold. (2) Practice: on 27 June 2017 a compromised M.E.Doc update on a single PC in an Odessa finance office gave NotPetya a foothold; the worm used the leaked EternalBlue exploit (patched since March 2017) and Mimikatz credential theft to cross the flat global network, shutting down headquarters, 76 ports and the booking platform within hours; the ~150 syncing domain controllers were wiped simultaneously with no isolated backup, leaving one accidentally-offline machine in Ghana as the sole directory copy, and the company spent ten days rebuilding 4,000 servers, 45,000 PCs and 2,500 applications at a cost of $250-300 million, running ports on paper in the meantime. (3) Why a Strategy to Leadership gap: the classification rests on the containment stage. The expectation was doctrinal - not a vague good intention but an approved, budgeted programme - and the chain broke at the decision level, not in a resourced implementation: the redesign's success was never made a KPI for the most senior IT overseers, so it carried no bonus or ownership, and they never carried it forward. That is espoused strategic intent diverging from enacted leadership priorities - the S2L pattern - rather than a resourced plan that operations failed to realise. The entry vector (a state-hijacked update of tax software legally required for Ukrainian operations) is documented but carries no evidenced victim-side assurance failure and is not load-bearing; the technical entry gaps (patch latency, Windows 2000) are contributing. The equally documented recovery gap - a continuity architecture that was never isolated - is a secondary Strategy to Execution pattern; under the tie-break the pre-breach containment condition prevails over the post-breach recovery stage (both secondary-sourced, both load-bearing, equal weight). (4) RM30 (Strategy to Leadership lens): leadership never made containment and blast-radius reduction an architectural priority - segmentation was the chief flaw its own executives named, the redesign was approved, yet the single flat global domain let one Odessa PC infect 76 ports worldwide (causal role: Amplifying). RM02 (Strategy to Leadership lens): leadership privileged competing delivery objectives over the stated security strategy without accountable trade-off governance - the redesign was left untied to any senior IT overseer's KPI or bonus and was silently abandoned, an unmanaged trade-off against the group's own security programme (causal role: Predisposing). RM31 (Strategy to Leadership lens): continuity objectives were not backed by realistic design and investment - directory resilience rested on 150 synchronised, production-connected domain controllers with no isolated backup, so the same worm that broke the network also broke the recovery capability (causal role: Constraining). RM18 (Strategy to Leadership lens): leadership priorities allowed known exposure to persist despite remediation expectations - patching was flagged as imperfect and some servers still ran unsupported Windows 2000, and the programme meant to fix it was left undone while the EternalBlue remedy had been available since March 2017 (causal role: Enabling).

Evidence by stage

Risk markers

Sources

Notes

Case description matches the stated victim (Cyberaanval op Maersk / The Untold Story of NotPetya). The first source_url in the case file (archief.nipv.nl IFV PDF Cyberaanval op Maersk) could not be opened: it returns an HTML error page and has no Wayback snapshot, so the analysis rests on the second source_url plus Maersk's own disclosures; the replacement Maersk Post reportage PDF (victim's own publication, September 2017) was used as the victim-side source. The 2016 security-redesign account (green-lit, budgeted, never carried forward, no KPI) rests on Wired's reporting from anonymous Maersk sources - one-sided on that point, no primary document confirms it. Maersk's 28 June announcement labelled the malware Petya and stressed containment; the $250-300m figure is the chair's Davos estimate as reported by Wired. Attribution: UK government (15 February 2018) and the US DOJ indictment of six GRU Unit 74455 / Sandworm officers (October 2020, via the FBI wanted page); no separate earlier comparable Maersk incident was asserted. The M.E.Doc supply-chain trust boundary (a vendor legally required for Ukrainian operations) is described in the rationale but not coded as RM13: no source documents a victim-side assurance failure there. Link check: investor.maersk.com blocks automated access (HTTP 403); the page was opened and read during research.