Knock Outage History

Knock experienced an issue with their Agent feature in the dashboard that was classified as a minor service incident. The problem affected the Agent functionality for approximately 1 hour. A fix was implemented and the service was restored to normal operation with monitoring in place.

Knock experienced slow workflow processing and message delivery due to extremely high load on their services, causing delays in notifications and broadcasts. The team worked to redistribute workload and isolate the source of the issue, gradually restoring normal processing across different service areas. The incident was resolved after 3 hours with all workflow and broadcast processing delays cleared.

Knock experienced a major incident where tracked links failed to resolve and redirect properly. The service disruption lasted 11 minutes before engineers deployed a fix that restored normal link resolution functionality. The team continued monitoring to ensure the resolution was stable.

Knock experienced delayed ingestion of API logs that put pressure on their ingestion pipeline and degraded the ability to browse API logs. The team shipped a fix and saw initial signs of recovery, followed by full recovery of the API log delays. The incident lasted 1.3 hours and was resolved through monitoring and implementing a targeted fix.

Knock experienced delays in creating and processing message events, causing customers to see delayed message events in the Dashboard and delayed outbound webhook events. The service implemented a fix that resolved webhook delivery lag first, followed by Dashboard availability recovery. The incident lasted 1.2 hours with full resolution achieved.

Knock experienced a 25-minute service incident where users were unable to promote workflows within the platform. The issue was identified and resolved through a deployed fix, while notification delivery remained unaffected throughout the incident.

Knock experienced a log processing issue that caused delays in log availability for 31 minutes. Workflows and message processing continued to function normally during the incident. The issue was identified and resolved with a fix implementation, followed by service monitoring.