On August 11, 2026, an electrical failure in the central electrical system of CNT (National Telecommunications Corporation) caused the temporary drop of services hosted in one of sus centros de datos. The list of those affected was long: Civil Registry (national suspension of issuance of ID cards and passports), Presidency, Chancellery, the ministries of Education, Sports, Culture and Interior, Celec, Cosede, Dinarp, the Social Registry Unit, INEC, the Judicial Police (Dinased), the municipalities of Quito and Rumiñahui, Immigration services at airports, and even the AMT and the traffic lights in Quito.
CNT reacted quickly: 90% of the services were restored the same day, the rest in the following hours. That deserves recognition.
It wasn’t the first time
On June 21, 2026, a similar electrical failure had already occurred, affecting mobile telephony and state platforms such as the Ministry of the Interior, the Comptroller’s Office, AMT, Sercop and Jóvenes en Acción. Two serious power failures in less than two months is no coincidence — it’s a pattern.
Neither of the two crashes was reported as a cyberattack; both were attributed to technical/electrical failures. But context matters: the Ecuadorian digital ecosystem was already compromised this year (DIGERCIC, Banco de Machala, Ministry of Health — 100 GB of exposed medical data, as we covered in Ecuador Under Attack). And CNT is not new to this: it was the victim of a cyber attack in July 2021.
The detail that doesn’t add up
What makes this fall particularly uncomfortable is the technical detail: CNT’s data center in Guayaquil has Tier III (Uptime Institute) and LEED certification, with an investment of ~USD 21 million on 12,000 m². A Tier III center has, by definition, N+1 energy redundancy. That a “fortuitous” electrical failure knocks down critical State services in a center with that level of certification — and that it is repeated in 60 days — suggests something more than bad luck: maintenance deficiencies, or a dependence on a single point of failure that should not exist in infrastructure of this caliber.
What do Tier III certifications really mean?
To understand why this drop is so problematic, it is worth explaining what a Tier III certification entails. The Uptime Institute data center rating system defines four levels:
- Tier I: basic infrastructure, without redundancy. A single point of failure can take down the entire center.
- Tier II: partial redundancy in critical components (power, cooling), but still has single points of failure.
- Tier III: N+1 redundancy in all critical components. Does not require shutdown for maintenance. Expected uptime: 99.982% (maximum ~1.6 hours of downtime per year).
- Tier IV: 2N+1 total redundancy, fault tolerance. Expected uptime: 99.995%.
A certified Tier III should be able to handle an electrical failure in a component without it resulting in a loss of services. Power comes in through at least two independent paths, there are UPS (uninterruptible power supplies) and backup generators. For a power failure to knock down critical services in a Tier III means that either the redundancy is not working as it should, or the configuration has a single point of failure that the certification did not cover.
The cadence of falls: a temporal analysis
The timeline of the incidents at CNT is revealing when examined closely:
- July 2021: CNT is the victim of a cyber attack that affects its internal systems.
- June 21, 2026: electrical failure that affects mobile telephony and state platforms (Ministry of the Interior, Comptroller’s Office, AMT, Sercop, Jóvenes en Acción).
- August 11, 2026: electrical failure that knocked down the Civil Registry, Presidency, Chancellery, multiple ministries, Immigration, Quito traffic lights and more.
The 51-day distance between the two 2026 power failures is particularly concerning. That’s not enough time to implement and verify significant infrastructure improvements, but it’s enough time that public memory has cooled. This raises legitimate questions about whether the June post-incident audits identified and corrected root causes, or whether this was an administrative report that did not translate into operational changes.
Why does the State depend so much on a single data center?
The concentration of critical State services in a single CNT data center is, in itself, a design vulnerability. When Civil Registry, Presidency, Immigration, the Judicial Police and the traffic lights of an entire city depend on the same infrastructure, you are creating a “single point of failure” on a national scale.
Redundancy is not just a technical concept within a data center — it is an architectural principle that should be applied at the state infrastructure level. Ideally, critical services should be distributed across at least two geographically separated data centers, preferably operated by different providers. So when one falls, the other can take on the load.
Ecuador has other certified data centers that could serve as backup. The question is why such distribution has not been implemented, and the answer likely involves a combination of factors: costs, existing contracts, bureaucracy, and the lack of a regulatory framework requiring geographic redundancy for critical public services.
Real impact on citizens and companies
The August 11 crash was not a technical inconvenience — it had tangible consequences:
- Civil Registry suspended: people who were traveling the next day were unable to obtain a passport. Companies that needed updated IDs for legal procedures were left without options.
- Immigration at airports: international travelers faced delays or inability to process their departure from the country.
- Quito traffic lights: the AMT reported failures in the traffic light system, generating vehicle chaos in several areas of the city.
- Municipal government: the municipalities of Quito and Rumiñahui were unable to access internal systems, affecting citizen procedures.
- Education and culture: the affected ministries reported interruptions on digital platforms.
This is without counting the economic impact: companies that depend on these services to operate (consultants that need RUC Consultation, legal offices that need access to the Civil Registry, etc.) lost productivity for hours.
The question we leave open
A few days ago we published an analysis on estado real de los data centers en Ecuador: the country already has certified infrastructure, real investment, a growing industry. The question we left open was another — whether that capacity translates into services that people really feel, starting with whoever waits in line at the Civil Registry.
This fall unintentionally answers that question: it does translate — for better and for worse. When the data center works, the process is faster. When it falls, half of the public administration falls with it.
What should change?
Some measures that the public sector should seriously consider:
- Geographic redundancy: distribute critical services in at least two different data centers.
- Independent technical audits: that Tier III certifications are verified annually with real failure tests, not just documentary ones.
- Business continuity plan: have documented and tested procedures for each critical service when the primary infrastructure fails.
- Digital resilience regulations: a legal framework that requires minimum levels of redundancy for essential public services.
- Post-incident transparency: publish the technical reports of the failures so that citizens and the private sector can evaluate the real evolution.
Be careful — these things happen, and no infrastructure in the world is immune. But when the same single point of failure knocks down the Civil Registry of an entire country twice in two months, it is no longer an isolated incident. It’s a vulnerability worth keeping a close eye on.
Sources: Primicias, Ecuavisa, Expreso, El Universo, Teleamazonas, Radio Pichincha, Radio Govea (Aug 11-12 and June 22-24, 2026); CNT Guayaquil DC technical sheet (Datacentermap).