2011 · Brazil · O - Public administration and defence · Distributed denial-of-service attack causing short-term web outage
Analysed 2026-11-07 by Hermes
Open this case in the tool →Distributed denial-of-service attack causing short-term web outage
Expertise to Execution
Evidence coverage: 1/6 stages, weight 8/60
Decisive stage: recovery (clearest evidence) · Alternative: Strategy to Execution
(1) The Presidency's official portal and the Brasil.gov portal are the public online face of the Brazilian federal government, run by the state's own federal IT agency Serpro; the strategic expectation for such sites is availability under foreseeable load - and flood attacks had already hit Brazilian public sites before this incident (wire reporting on this case notes the Workers' Party site knocked out for 24 hours in April 2010, and Serpro itself described investigating attempts to block the Presidency's security as standard practice). (2) In practice, on 2 January 2011 an intentional access flood by the two-person group Fatal Error Crew overwhelmed the Presidency portal's stated capacity of 4,500 simultaneous accesses and left it and Brasil.gov out of service or unstable for several hours; Serpro restored stability the same evening, and only after the attack created a blocking strategy against the denial-of-service method used and said it would take preventive measures for the future; no system or database was invaded and no data was taken. (3) The classification rests on the recovery/availability stage: the remedy (flood filtering and adequate capacity) was technical knowledge the victim's own specialists plainly had - they built the blocking strategy within days - yet it had not been realised in the pre-attack operations and architecture of the country's most prominent public websites; no source documents a policy that existed and was not followed, nor a leadership decision that deferred or underfunded the protection, so this is a material-origin, execution-break gap (Expertise to Execution) rather than a doctrinal one; the entry vector is known but not load-bearing, because no intrusion occurred. (4) RM32 (Expertise to Execution lens): experts could not keep the Presidency portal available under attack because the pre-incident architecture capped it at 4,500 simultaneous accesses with no flood protection, so a self-described pair's flood took both federal portals down for hours (causal role: Enabling). RM21 (Expertise to Execution lens): Serpro's own statement that investigating blocking attempts against the Presidency site is standard practice shows the flood threat had been faced before, yet the known mitigation was not implemented or validated after those earlier events - it was built only after a flood finally succeeded (causal role: Perpetuating).
RM32 — Capacity / availability-protection gapRM21 — Recurring weakness / learning failureEuRepoC row: victim is only a category in the case file (receiver: State institutions / political system, Government / ministries); the Presidency's own portals were the targets, run by Serpro. Disclosure was attacker-initiated (the group claimed the attack on Twitter; the dataset records 'Incident disclosed by attacker'), but Serpro confirmed the incident after technical analysis the same day and gave an accurate account, so RM39 was not coded - its exclusion on non-unreasonable, non-concealed disclosure applies. The group's claimed motive (fun, showing Brazilian server security is fragile, protest against Dilma Rousseff's election) and its claim of 56,000 defaced sites are unverified attacker claims reported by press and are not used as evidence. Follow-up reporting (June 2011, other incidents, not this row) shows the Presidency site was repeatedly flooded again by other groups the same year. No charges against the attackers were reported. Sector O follows the method-spec rule that government bodies fall under Public administration and defence.