← Breach analysis by the Dual Vulnerabilities framework

DigiNotar Certificate Authority breach (unattributed, 2011)

2011 · Netherlands · J - Information and communication · Network intrusion leading to issuance of fraudulent digital certificates (CA compromise)

Analysed 2026-11-07 by Hermes

Open this case in the tool →

What happened

Network intrusion leading to issuance of fraudulent digital certificates (CA compromise)

Dual vulnerability type

Expertise to Execution

Evidence coverage: 5/6 stages, weight 43/60

Decisive stage: entry (tie-break: largest share) · Alternative: Strategy to Execution

Why this classification

(1) DigiNotar was the primary certificate authority for Dutch government services under the PKIoverheid framework - a 'trust provider' whose product is assurance itself, contractually audited and vetted, with formal expectations of secure PKI operations, incident reporting and revocation (Dutch Safety Board; Kroes). (2) In practice the Fox-IT final report found an organisation in which known, routine controls were simply absent: two public web servers ran a CMS version with publicly known vulnerabilities and were the point of entry on 17 June 2011; the 24-segment/DMZ segmentation design was never defined or enforced in the firewall; CA and netHSM logs were not centrally or separately stored, so neither the intrusion (undetected for over a month) nor its full scope could ever be reconstructed; and after a routine check found rogue certificates on 19 July, the company stayed silent for five weeks and kept making reassuring statements that its own later disclosures contradicted. (3) The decisive gaps sit in the material column: for the entry and detection stages the expectation was specialist knowledge of available remedies - patches/upgrades for known DotNetNuke vulnerabilities, the vendor-recommended separate log server - and the chain broke at execution, the remedies never being realized in operations: a textbook cybersecurity knowing-doing gap (Expertise to Execution). The containment design that existed but was not enforced is a parallel pre-breach Strategy to Execution condition, and the five-week silence with public claims later contradicted is a clear secondary Strategy to Leadership pattern at the response stage; under the tie-break rules the post-breach response stage yields to the pre-breach conditions, and between the two parallel pre-breach conditions (neither documented as causing the other) the tie-break goes to the largest share, so the type follows the material-column gaps. The entry vector is not what classifies the case: it is load-bearing here only because the unexecuted remedy behind it (patching, logging, egress control) recurs across entry, detection and egress. (4) RM18 (Expertise to Execution lens): the exploited vulnerability was known and remediable - the outdated DotNetNuke versions had publicly known security issues and the remedy was a routine patch/upgrade, yet patching of the Internet-facing servers was never executed, leaving them exploitable for the 17 June intrusion (causal role: Enabling). RM30 (Expertise to Execution lens): the containment architecture the experts had designed - 24 segments plus dual DMZ - could not be implemented in operations: zones were not strictly described or enforced and firewall exception rules plus RDP tunneling let one DMZ foothold reach all eight CA servers, so the blast radius became the entire certificate estate (causal role: Amplifying). RM33 (Expertise to Execution lens): the investigation was crippled by design - CA management logs ran on the compromised CA servers themselves and netHSM logs were deleted at every shutdown instead of stored on a separate log server as the vendor recommends, so Fox-IT could only make inconclusive observations and the final list of affected certificates could not be determined (causal role: Obscuring). RM39 (Expertise to Execution lens): because the forensic constraints left scope unestablishable, the company's disclosure from late July to early September stayed incomplete and repeatedly inaccurate - it insisted all rogue certificates were revoked while the active Google certificate was missed and was revoked only after Google's public disclosure (causal role: Constraining).

Evidence by stage

Risk markers

Sources

Notes

cyberdefensemagazine row, blind (researcher analysis removed); the row description matches the stated victim (DigiNotar crisis aftermath / bankruptcy filing), so no data-quality mismatch. Of the two source URLs in the case file, the Leiden student-thesis viewer (studenttheses.universiteitleiden.nl item 2665708) could not be opened during analysis (HTTP 422, content not extractable); the Wired bankruptcy article was opened and used. Fox-IT citations go to the roselabs.nl mirror of the official Fox-IT 'Black Tulip' final report (published August 2012 by the Ministry of the Interior); the rijksoverheid.nl publication page did not respond to fetches in this environment, so the mirror that was actually opened is cited. Actor attribution is unverified: responsibility was claimed in the press by a hacker identifying himself as a 21-year-old Iranian student (Wired); the claim rests on attacker statements and is not a claim the DV rests on. The Dutch Safety Board's formal investigation centred on government oversight of PKIoverheid (Logius/OPTA relied on audits and had insufficient risk awareness); it is cited here for the doctrinal trust-framework expectation, while the victim-side lifecycle evidence comes from the Fox-IT final report. Personal data: no separate confirmed figure for data theft at DigiNotar itself was established in the sources opened; the Fox-IT report documents file exfiltration via the stepping-stone servers without a victim-side personal-data count, so none is recorded.