← Breach analysis by the Dual Vulnerabilities framework

Taiwanese Government Websites DDoS (Filipino hacktivists, 2013)

2013 · Taiwan · O - Public administration and defence · DDoS disruption and defacement of government websites

Analysed 2026-11-07 by Hermes

Open this case in the tool →

What happened

DDoS disruption and defacement of government websites

Dual vulnerability type

No DV evidenced

Alternative: Expertise to Execution

Why this classification

(1) Strategic expectation: the victim is the shared web estate of Taiwan's central government, served over the Government Service Network (GSN) with a dedicated Executive Yuan Office of Information and Communication Security (OICS), an ICT Service Center that issues alerts to agencies, and TWNCERT for incident coordination; availability of official websites and their protection against disruption is the standing doctrinal expectation this apparatus exists to realise. (2) In practice: in the keyboard war that followed the fatal shooting of a Taiwanese fisherman by a Philippine official vessel on 9 May 2013, Philippines-based units of an anonymous group launched DDoS attacks and some website defacement against a number of central and local government websites (Office of the President, Ministry of Foreign Affairs, Ministry of National Defense, Ministry of Economic Affairs, Coast Guard Administration, plus Taipei and Tainan local-government and some private sites); the GSN investigated as soon as agencies reported weakening response and bandwidth spikes, asked Chunghwa Telecom's HiNet to block the source IPs, the OICS traced most attacks to the Philippines, the Presidential Office website that went down at 9:17 a.m. was back after 10 a.m., the Defense Ministry saw only slow website access with its internal network unaffected, TWNCERT had hacker-controlled Facebook pages removed and asked PHCERT for help, all attacked sites returned to normal, and the Executive Yuan disclosed the campaign publicly and in detail on 12 May. (3) Why there is no dual vulnerability: the documented lifecycle is one of alignment, not gap - detection was prompt (signs detected the previous Friday evening, investigation started immediately on agency reports), containment was executed (source-IP blocking through the upstream provider, IP lists distributed for detection and blocking), recovery was fast (Presidential Office site restored within about 45 minutes, all sites back to normal, disruption under 24 hours per the dataset), response coordination ran across agencies, vendor, platform and foreign CERT without documented friction, and disclosure was self-initiated, same-period and corroborated by independent reporting rather than contradicted by it. No source documents an expectation that was not acted on, a warning that was not escalated, or a decision or resourcing failure behind the one condition the evidence does show: public government websites temporarily overwhelmed by a politically-motivated crowd DDoS. An Expertise to Execution reading (crude flood tools repeatedly succeeding suggests known anti-DDoS measures were not implemented on government web infrastructure) is the runner-up but is rejected, because no source documents the origin or break of that candidate gap and the framework requires positive evidence, not coder inference, for a DV claim; the case is coded as a negative control on the strength of the victim's own primary disclosure corroborated by secondary reporting. RM32 Capacity / availability-protection gap and RM04 Control dependency concentration are recorded as the documented conditions of the case (brief availability loss under flood load; the affected services and their blocking capability concentrated on the shared GSN platform with mitigation routed through one upstream supplier), without a dual-vulnerability claim.

Risk markers

Sources

Notes

eurepoc row; the case file carries no receiver_name, so the victim is named from the agencies listed in the Executive Yuan press release; the row description (Philippino hackers launch DDoS attacks against Taiwanese government websites in response to Taiwanese hacking attacks) matches the stated victim, with the dataset's misspelling of Filipino. Atypical context: this was the Philippine-side counter-attack in a mutual hacktivist keyboard war after the 9 May 2013 shooting of a Taiwanese fisherman; the row dates the incident 13.05.2013 while the campaign ran in waves from 10 to 13 May - it is still a case with victim organisations, so the attacker-side Unknown rule does not apply. The victim-side lifecycle record rests on the victim government's own press release (12 May 2013) - primary but self-reported; no regulator or independent investigation of the incident occurred, which is the main caveat on the No DV evidenced reading. Attribution rests on receiver attribution (OICS IP tracing) plus attacker self-attribution via social media, unverified. The dataset source_url (softpedia.com) no longer opens; the analysis cites the Internet Archive snapshot of that exact article that was opened and read on 07-11-26. Website defacement against Taiwanese sites is reported by the Executive Yuan without mechanism detail, so no integrity or exposure marker is coded for it. Egress not in play: data_theft none in the case file. Sector coded O for the victim bodies as public administration; Country and Year from eurepoc_extra.receiver_country and start_date. Unattended run: no action requiring consent was needed or taken.