Between approximately 12:50 AM and 02:50 AM UTC on August 26, 2025, users of our Japan site experienced a service disruption. During this time, customers may have encountered login failures, slow response times, or a temporary inability to access certain features such as signing or sending documents for signature.
A problem in a production deployment, led to resource exhaustion in application servers when attempting to connect to a backend resource.
To resolve the issue, a partial failover of frontend UI and API systems was performed, re-establishing connections to backend resources. Additionally, some systems required a restart to clear failed connections.
As a follow-up to this incident, additional validation steps have been introduced into the impacting production pipeline.