Blend Outage History

An Encompass outage caused degraded performance affecting exports, disclosure, and document sync functionality within Blend's Lending Platform and LOS Integrations for 21.5 hours. The issue originated from Encompass (ICE) systems and impacted mortgage technology services. Encompass resolved the underlying problem on their end, with all services returning to a stable state after monitoring showed high success rates across production connections.

Blend Beta environments experienced a major outage lasting 3.7 hours where users encountered 502 errors when attempting to access the Lending Platform. The engineering team identified and validated the root cause, confirming that only Beta environments were affected while production systems remained stable. The issue was fully resolved and normal functionality was restored to all Beta environments.

Blend's ticketing system experienced a 21.3-hour outage caused by infrastructure issues with Freshdesk's mail servers, preventing email-based support tickets from being delivered or processed. The incident was compounded by a broader Microsoft service outage that may have contributed to the email delivery problems. The issue was resolved after tickets began flowing into the system again, though users were advised to resend any communications from the affected period to ensure they were received.

Integration connection errors in Blend's document generation service prevented successful disclosure issuance and disrupted synchronization of disclosure status updates between Blend and client loan origination systems. The engineering team identified and implemented a short-term fix to resume workflows, then worked to replay events for impacted loans. The incident was fully resolved after 6.4 hours with all affected loan events addressed.

Blend's API Gateway experienced issues for 3.2 hours, causing some clients to encounter 502 Bad Gateway errors and disruptions with document generation, disclosures, and exports. The team resolved the incident by increasing the number of API Gateway pods and adjusting auto-scaling settings while proactively tracking any failed exports. The service was stabilized and the incident was fully resolved.