Fivetran Outage History

Multiple Twitter Organic and Twitter Ads connections in Fivetran failed for 23.5 hours due to "HTTP 503 Service Unavailable" errors caused by the X API not returning rate limit information in response headers, which led to repeated 429 rate limit errors and sync failures. The issue was resolved when X fixed their API problem, and Fivetran deployed a hotfix to automatically reschedule connections that encounter account-level rate limits without proper response headers to prevent future sync failures.

Fivetran experienced OAuth authentication issues that affected multiple data connections for 1.1 hours. The problem impacted General Services, preventing users from properly authenticating their data source connections. The issue was identified, a fix was implemented, and the incident was resolved after monitoring confirmed the solution was working.

Fivetran experienced an issue where PostgreSQL connections using RDS Proxy were failing due to an unsupported lock_timeout parameter error. The incident affected General Services, Google Cloud SQL for PostgreSQL, and PostgreSQL components. The issue was identified and resolved after 1.9 hours of service disruption.

The Fivetran REST API experienced an issue where certain endpoint calls failed with 404 errors for 3.9 hours on March 26, 2026. The problem was caused by a recent change made by Fivetran, which was subsequently reverted to restore normal API functionality. The incident was fully resolved with error rates returning to normal levels.

Fivetran's Twitter and Twitter Ads connections experienced failures with "HTTP 503 Service Unavailable" errors due to a third-party API issue on X (Twitter)'s side. The incident affected Twitter Ads components and lasted 10.2 hours from March 26, 2026 01:00 UTC to 14:30 UTC. The issue was resolved when X fixed their API, with connection recovery observed gradually and error rates returning to normal levels.

Multiple Fivetran connectors syncing to BigQuery destinations failed with "GCS_BG_STORAGE" errors due to an internal configuration issue. The problem affected the Destinations component and prevented connectors from syncing data successfully. Fivetran resolved the internal configuration error after 1.7 hours, restoring normal instance rates and allowing affected connectors to resume syncing.

Fivetran experienced a 3.5-hour incident where some Box connections failed due to HTTP "429 Too Many Requests" errors. The issue affected Box integrations and prevented data synchronization for impacted connections. The problem was identified, a fix was implemented and monitored, and the incident was fully resolved.

Fivetran's iTunes Connect connections experienced sync failures with "Unknown failure" errors due to intermittent unavailability of Apple's App Store Connect Reporter API. The issue lasted approximately 1 hour from 06:17 to 07:14 UTC on March 22, 2026, and was caused by a temporary API failure on Apple's side. The problem resolved automatically when Apple's App Store Connect Reporter became available again, allowing syncs to resume normal operation.

Fivetran experienced sync failures across multiple connections using Snowflake as a destination due to a third-party Snowflake downtime issue. The incident affected Destinations and General Services components, preventing data synchronization for customers using Snowflake destinations. The issue was resolved after 3.4 hours once Snowflake recovered their US regions and Fivetran confirmed all syncs returned to normal operation.

Fivetran experienced sync delays affecting services in the GCP Australia Southeast1 region for 3.2 hours. The issue impacted General Services, causing disruptions to data synchronization processes. The problem was identified, a fix was implemented and monitored, and the incident was fully resolved.