Global AWS Outage Impacts Costa Rican Institutions
A widespread global outage of Amazon Web Services (AWS) on Monday, October 20, 2025, led to significant disruptions in digital services across Costa Rica, including those of the Costa Rican Social Security Fund (CCSS) and the Ministry of Finance. The incident, which began in the early morning hours, underscored the increasing dependency of critical national infrastructure on major cloud providers.
The CCSS's Directorate of Information and Communication Technologies (DTIC) confirmed that internet service interruptions experienced by the institution were a direct consequence of this global AWS failure. Several of the CCSS's institutional equipment and systems operate on the AWS platform, leading to partial repercussions on specific services.
Affected Services and Technical Details
Beyond the CCSS, the Ministry of Finance also reported system issues. Its Virtual Office (OVi), which serves as the gateway to the crucial Tribu-CR tax system, experienced interruptions in sending messages and processing electronic invoicing. Additionally, the Compañía Nacional de Fuerza y Luz (CNFL), a public utility, saw its customer service channels, including WhatsApp, Messenger, email, and its call center, suffer intermitencies due to the AWS disruption.
The global outage was traced to a DNS resolution failure in DynamoDB within AWS's US-EAST-1 region, located in Northern Virginia. This region is a critical hub for numerous services worldwide. The problem started around 9:11 AM (7:11 GMT), causing widespread inaccessibility to various online platforms.
Recovery Efforts and Broader Implications
AWS acknowledged the issue, reporting 'significant error rates in the orders sent' and confirming a deep technical failure within its infrastructure. Recovery efforts began swiftly, with AWS reporting that the underlying DNS issue was largely mitigated by 6:35 AM ET (approximately 10:35 UTC). However, some services, particularly Elastic Compute Cloud (EC2), continued to experience instability during the recovery phase. Full restoration of all cloud operations was announced by 6 PM ET on October 20, concluding a 15-hour global outage.
This incident affected over 1,000 companies and millions of users globally, including major platforms in banking, gaming, and social media. The widespread nature of the outage has reignited discussions about the risks associated with the centralization of digital services and the critical importance of robust business continuity planning and diversified cloud strategies for organizations worldwide.
6 Comments
Eric Cartman
This incident is being overblown. Services were restored.
Kyle Broflovski
For how many services AWS runs, a 15-hour outage is actually rare.
Eric Cartman
While the disruption to social security and tax systems is concerning, it's also a reminder that even the most advanced infrastructure can fail. The focus should be on building hybrid solutions and local redundancies, rather than just pointing fingers at the cloud provider.
Kyle Broflovski
While the outage was disruptive and highlights cloud dependency risks, the article also shows how quickly AWS addressed a complex global issue. Organizations need better fallback plans.
Eric Cartman
The incident reveals a genuine vulnerability in relying heavily on a single cloud region. However, it also emphasizes the global reach and rapid response capabilities of a provider like AWS, assuming local entities have resilient designs.
anubis
Why didn't Costa Rica have proper redundancy or disaster recovery plans?