Marqeta Outage History

Marqeta experienced a delay in receiving the final daily clearing file from Visa for May 14th, causing customers to receive fewer clearing webhooks than normal and delays in VSS reports. The incident affected Reporting & Analytics and Webhooks components for 18.6 hours while Marqeta worked with Visa to resolve the file delivery issue. The delayed clearing files were eventually received and processed, with all expected clearing data successfully reconciled and validated against the Visa network summary.

Marqeta performed scheduled maintenance on their private Sandbox environment to update network traffic handling and routing as part of infrastructure improvements. The maintenance affected users' ability to interact with the private sandbox environment. The maintenance was completed successfully after approximately 8 days.

Marqeta experienced elevated STIPs (transaction processing issues) on the Pulse network for 19.4 hours starting May 11 at 10:20 PM PDT, impacting some cardholders using the Pulse network for payment processing. The issue was caused by Marqeta systems failing to process a new data pattern from the Pulse network, which was traced to a single acquiring partner. The problem was resolved on May 12 at 8:40 AM PT after Pulse worked with the responsible acquirer to implement a fix, followed by a monitoring period to ensure stability.

An AWS us-east-1 outage caused widespread disruption to Marqeta's services for 94.4 hours, affecting payment processing, reporting & analytics, digital wallets & tokenization, risk & disputes, webhooks, sandbox, and 3DS systems. Marqeta engineers mitigated the impact by redirecting traffic away from the affected AWS region and availability zone while monitoring AWS's recovery efforts from a thermal issue in their data center. All services were fully restored once AWS resolved the underlying infrastructure problems and Marqeta remediated associated data discrepancies.

An external vendor's unexpected cluster maintenance starting at 4:00 PM PST on May 4th caused severe delays in data replication, affecting Marqeta's Diva reports and MQD endpoints for over 60 hours. The incident impacted the Reporting & Analytics components, preventing timely access to transactional data. The issue was resolved on May 7th at 5:40 PM PST when all replicated assets in the system became fully synchronized.

Marqeta's 3DS service experienced degraded performance for 24.6 hours, causing challenge timeouts and STIPs that impacted customers. The engineering team identified the root cause and worked on implementing a resolution over the course of the incident. Service was fully restored and customers should no longer experience any impact.

Marqeta experienced an increased rate of declines for card-present transactions using magnetic stripe swipe, causing physical cards in good standing to be incorrectly declined at point-of-sale locations. The issue affected the Payment Processing component and persisted for 8.5 hours before engineers successfully resolved all instances. Cardholders would have experienced unexpected transaction failures despite having valid cards and sufficient account status.

Marqeta experienced intermittent database connectivity issues that caused elevated STIPs (service timeouts) and partial disruptions to Core API, Payment Processing, and Digital Wallets & Tokenization services. The incident was traced to database connection failures with a third-party vendor, affecting customers with service interruptions and API traffic issues. The problem was resolved after 94 hours by applying a tactical fix to the database environment, which restored all services to normal performance levels.

Marqeta experienced delays in bank file delivery affecting ACH Processing for some customers over a 3.3-hour period. Engineers identified the root cause and restarted the delivery process to resolve the issue. All bank files were successfully delivered by 12:17 PM PT with no further delays expected.

Marqeta experienced network disturbances causing elevated STIPS (System Timeout in Progress) on the Mastercard network during two distinct periods at 11:14 AM and 12:45 PM Pacific Time. The incident affected payment processing services for 4.4 hours. The issue was fully resolved after Marqeta's engineering team worked with Mastercard to address the network disturbance and confirmed normal service operations.