2019 · Netherlands · P - Education · Ransomware attack
Analysed 2026-11-07 by Hermes
Open this case in the tool →Ransomware attack
Strategy to Execution
Evidence coverage: 4/6 stages, weight 35/60
Decisive stage: containment (tie-break: largest share) · Alternative: Strategy to Leadership
(1) Maastricht University was a deliberately open, decentralised university with formal security artefacts in place: a CERT since 1994, a CISO since 2003, major-incident plans, awareness campaigns, an existing policy separating the domain-administrator account from maintenance of regular servers, and a decided 24/7 SOC/SIEM from January 2020. The Inspectorate of Education nonetheless found that cyber threat was not at the top of the risk list, ransomware was not included in the major-incident plans and there was no total overview of the ICT organisation before the attack. (2) In practice, two phishing e-mails opened on 15 and 16 October 2019 gave the attackers a foothold; because detection, monitoring and follow-up were inadequate, the intrusion stayed unnoticed for some ten weeks; on 21 November 2019 a server with missing security updates yielded full domain rights, since the domain-administrator account was used on regular servers contrary to the policy itself; limited segmentation then let Clop encrypt 267 Windows servers on 23 December 2019, including the online backups of a number of critical systems, and with no offline backups and a rebuild estimated at two to three months, UM paid a 30-BTC ransom on 30 December 2019. The entry vector (phishing) is documented but not load-bearing; the classification rests on the containment stage, with detection and recovery as load-bearing supporting stages and the response stage documented as adequate. (3) This is a knowing-doing gap in the doctrinal column's diagonal to operations: the formal artefacts existed but did not translate into operational reality - the domain-admin separation policy was violated in day-to-day maintenance, segmentation intentions were not implemented in the largely open VLAN architecture, and the continuity requirement for recoverable backups was realised only as online, encryptable snapshots. The detection gap reads differently: the need for 24/7 monitoring was strategically known yet the Executive Board did not prioritise or resource cyber risk in time, which is a leadership break (Strategy to Leadership); because no source documents either gap as causing the other, that pre-breach conflict was settled by the largest-share tie-break in favour of Strategy to Execution. (4) RM41 (Strategy to Execution lens): least-privilege and tiering requirements were not implemented - the domain-administrator account ran on regular ordinary servers contrary to the policy, giving the attacker domain-wide control from one workstation (causal role: Amplifying). RM30 (Strategy to Execution lens): segmentation requirements were not implemented in architecture and configuration, the VLANs being relatively open to one another, so the blast radius grew to 267 servers (causal role: Amplifying). RM34 (Strategy to Execution lens): the monitoring requirements (SIEM/SOC) that strategy had decided on were not yet operated as intended during the intrusion, alerts from the virus scanner went unanswered and were disabled, leaving a ten-week undetected dwell time (causal role: Obscuring). RM31 (Strategy to Execution lens): the continuity requirement for recoverable backups was not translated into isolated or offline copies, so online backups of critical systems were encrypted and recovery depended on paying the ransom (causal role: Constraining).
RM30 — Segmentation / containment weaknessRM31 — Resilience / backup preparedness gapRM34 — Monitoring / detection / alert-handling gapRM41 — Privilege-escalation / directory-hygiene pathCase-file source page https://www.maastrichtuniversity.nl/about-um/cyber-attack-um-2019 returns HTTP 404 and no Wayback snapshot exists (CDX empty); the university's published Fox-IT management summary and response PDF is cited in its place. The Inspectorate of Education report (June 2020) is cited via the English translation hosted at securitydelta.nl; the publisher URL at onderwijsinspectie.nl now 404s. Attribution to TA505/GraceRAT is Fox-IT's modus-operandi assessment as reported by BleepingComputer/SecurityWeek, not an individual identification. The entry stage is recorded as contributing only: the SURF interview describes pre-attack awareness campaigns, while the Inspectorate found those campaigns centred on GDPR/privacy rather than malware. No data-theft disclosure beyond credentials and network topology was found in any source, so the egress stage is recorded without a gap; the ransom decision is not coded RM42 because the Inspectorate documents it as a governed, costed decision.