Fivetran Outage History

We have resolved this incident.

This incident has been resolved. We have observed that instance rates are returning to normal levels and affected connectors are syncing successfully. Incident Summary Description: We identified an issue for Reddit Ads which resulted in syncs failing with a "HTTP 500 (Internal Server Error)" error. Timeline: This issue began on April 1st at 16:00 UTC and was resolved on the same day at 21:30 UTC. Cause: The Reddit Ads API started returning 500 failures to connector requests. Resolution: The 500 errors were resolved source-side and a hotfix has been implemented to catch similar failures in the future.

This incident has been resolved. We observed error rates returning to normal levels, and all affected connections are now running successfully. Incident Summary Description: We identified an issue where GitHub connection syncs were intermittently failing with the error: "No server is currently available to service your request." Timeline: The issue began on April 1st at 10:00 UTC and was resolved by 13:00 UTC. Cause: The failures were caused by intermittent unavailability of the GitHub API, which led to sync disruptions. Resolution: The issue was resolved automatically once the GitHub API resumed normal operation.

Description: We identified an issue affecting the Pipedrive connection, which resulted in syncs failing with Null exception: Cannot invoke "java.time.temporal.Temporal.until(java.time.temporal.Temporal, java.time.temporal.TemporalUnit)" because "startInclusive" is null Timeline: This issue began on March 31, 2026 at 11:06 UTC and was resolved on March 31, 2026 at 16:20 UTC. Cause: A configuration issue impacted existing connections. Resolution: Applied a fix to improve handling and prevent the issue.

This incident has been resolved. We have observed that error rates have returned to normal levels, and the Twitter Organic and Twitter Ads services are operating as expected. Incident Summary Description: We identified an issue affecting multiple Twitter Organic and Twitter Ads connections, which were failing with the error: “HTTP 503 Service Unavailable.” Timeline: The issue began on March 30th at 00:00 UTC and was resolved on March 31st at 16:30 UTC. Cause: The X API stopped returning rate limit information in the response headers. This caused Twitter Organic and Twitter Ads connections to encounter 429 errors, leading to sync failures. Resolution: X has resolved their API issue. Additionally, we deployed a hotfix to ensure that if the source hits account-level rate limits and does not return response headers, it will no longer cause sync failures. Instead, the connections will be automatically rescheduled.

Description: We identified an issue affecting the Criteo connector, which resulted in syncs failing with 503 Service Unavailable and 504 Gateway Time-out errors. Timeline: This issue began on March 30, 2026 at 08:17 UTC and was resolved on March 30, 2026 at 09:17 UTC. Cause: The failures were caused by an outage on the third-party side, as confirmed on their official status page. https://status.criteo.com/incidents/dj24tx2vf4w3 Resolution: The issue was resolved after third-party restored their services. Syncs resumed normal operation once the upstream outage was mitigated.

This incident has been resolved. We have observed that error rates have returned to normal levels, and the service is operating as expected. Incident Summary Description: We identified an issue with the OAuth Authentication that is affecting multiple connections, resulting in a 404 error. Timeline: The issue began on 26/03/2026 at 09:09 UTC and was resolved on 26/03/2026 at 16:42 UTC. Cause: During an ingress upgrade, an error caused OAuth callback requests to return HTTP 404 responses. This led to authentication failures for connections relying on OAuth. Resolution: The change was reverted, which restored normal OAuth functionality for Fivetran connections.

Incident Summary Description: We identified an issue where some Fivetran REST API requests were timing out. Timeline: The issue began on 26/03/2026 at 10:17 UTC and was resolved on 26/03/2026 at 12:05 UTC. Cause: The issue was caused by resource constraints in our infrastructure, causing timeouts when resources were unavailable. Resolution: We have increased resource limits to mitigate the issue, and we are not seeing connection failures after that change.

This incident has been resolved. We have observed that error rates have returned to normal levels, and the service is operating as expected. Incident Summary Description: We identified an issue with the PostgreSQL Connections Using RDS Proxy Failing with the Error: The source database does not support a provided lock_timeout parameter Timeline: The issue began on 26/03/2026 at 05:40 UTC and was resolved on 26/03/2026 at 07:45 UTC. Cause: The issue was caused by a recent code change introducing a bug from Fivetran. Resolution: The bug was fixed and the connection services were restored.

This incident has been resolved. We have observed that error rates have returned to normal levels, and the service is operating as expected. Incident Summary Description: We identified an issue with the Fivetran REST API that caused certain endpoint calls to fail with a 404 error. Timeline: The issue began on 25/03/2026 at 14:04 UTC and was resolved on 26/03/2026 at 06:55 UTC. Cause: The issue was caused by a recent change on our end. Resolution: The change was reverted, restoring normal API functionality.

This incident has been resolved. We have observed that error rates have returned to normal levels, and the service is operating as expected. Incident Summary Description: We identified an issue with the multiple Twitter and Twitter Ads connections, which are failing with the error "HTTP 503 Service Unavailable". Timeline: The issue began on 26/03/2026 at 01:00 UTC and was resolved on 26/03/2026 at 14:30 UTC. Cause: A third-party issue caused Twitter and Twitter_Ads connections to fail. Resolution: X (Twitter) has fixed their API.

This incident has been resolved. We have observed that instance rates are returning to normal levels, and affected connectors are syncing successfully. Incident Summary Description: We identified an issue causing sync failures, with multiple connectors in the BigQuery destination failing due to a "GCS_BG_STORAGE" error. Timeline: This issue began on March 25, 2026, at 00:18 UTC and was resolved on March 25, 2026, at 01:49 UTC. Cause: Connections syncing to BigQuery destinations are impacted due to an internal configuration error. Resolution: The necessary steps were taken to resolve the internal configuration issue, after which instance rates returned to normal and the affected connectors resumed syncing successfully.

We have resolved this incident. Description: We identified an issue where multiple Mixpanel connections are failing their sync with a socket timeout error due to an issue on the third-party side. Timeline: Incident Start: 2026-03-23 at 18:50 UTC Incident End: 2026-03-23 at 19:15 UTC Cause: The issue was caused by an intermittent issue on the third-party side. Resolution: The issue was automatically resolved from the third-party side without any intervention from the Fivetran end.

Description: Some Box connections unexpectedly failed their sync runs, but recovered on subsequent runs. Timeline: Incident Start: 2026-02-23 at 16:30 UTC Incident End: 2026-02-23 at 18:00 UTC Cause: The issue appears to be caused by a temporary error in our credential handling service, but we are continuing to investigate this post incident resolution. Resolution: The App Store Connect Reporter is now available, and syncs are succeeding normally.

Description: We identified an issue with Apple App Store Connect that caused sync failures due to intermittent unavailability of the App Store Connect Reporter. Timeline: This issue began on 22 March 2026 at 06:17 UTC and was resolved on 22 March 2026 at 07:14 UTC. Cause: The issue was caused by a temporary API failure on the App Store Connect side. Resolution: The App Store Connect Reporter is now available, and syncs are succeeding normally.

This incident has been resolved. We have observed that instance rates are returning to normal levels, and affected connectors are syncing successfully. Incident Summary Description: We identified an issue where multiple connection syncs under the Snowflake destination are failing due to a Snowflake Downtime. Timeline: This issue began on 21-03-2026 at 05:26:30 UTC and was resolved on 21-03-2026 at 06:17:30 UTC. Cause: External Snowflake incident affecting the impacted AWS region(s), leading to elevated Snowflake exceptions and downstream sync failures. Resolution: The issue was resolved on the Snowflake side, and connections are now syncing successfully

Description: We identified an issue for Hubspot which resulted in syncs failing due to increased API rate limiting. Timeline: This issue began on 20th March, 2026 at 20:45 UTC and was resolved on 20th March, 2026 at 23:00 UTC. Cause: A recent update intended to improve how our connection manages HubSpot authentication led to an unintended increase in how frequently access tokens were refreshed. This resulted in a higher number of authentication requests than expected, which triggered HubSpot’s rate limits and caused some syncs to fail. Resolution: We rolled back the change, which immediately resolved the issue. Syncs are now succeeding normally.

We have resolved this incident.

Incident Summary Description: A few Amazon S3 connections experienced intermittent failures due to an AWS-side issue, with the error: "Unable to execute HTTP request." Timeline: The issue began on March 18, 2026, at 10:30 UTC and was resolved by 13:40 UTC. Root Cause: The intermittent sync failures were caused by AWS-side timeout issues. Resolution: The issue was automatically resolved once connectivity from the AWS side was restored.

This incident has been resolved. We observed that Snapchat Ads connection sync success rates have returned to normal levels, and syncs are now running as expected. Description: We identified an issue where Snapchat Ads connections started failing with error "500 Internal Server Error" Timeline: The issue began on March 18, 2026, at 04:03 UTC and was resolved on March 18th, 2026, at 05:01 UTC. Cause: The issue was caused by an intermittent issue with a third-party API service that resulted in 500 errors. Resolution: No changes were required on our side. The issue was automatically resolved on the third-party side, and syncs are back to normal.

This incident has been resolved. We observed that HubSpot connector sync success rates have returned to normal levels, and syncs are now running as expected. Description: We identified an issue where some HubSpot connector syncs were failing with the error: "Unknown failure." Timeline: The issue began on March 17, 2026, at 17:10 UTC and was resolved at 17:26 UTC. Cause: The issue was caused by timeouts while attempting to reach HubSpot's OAuth endpoint. Resolution: No changes were required on our side. The issue was automatically resolved once connectivity to HubSpot's OAuth endpoint was restored.

Description We identified an issue with our public REST API that caused API calls to fail with a "column api.disabled_at does not exist" exception. Timeline: This issue began on 22:15 UTC on March 16, 2026 and was resolved on 22:50 UTC on March 16, 2026. Cause: A migration change caused the issue. Resolution: A fix was implemented to address the migration issue, after which instance rates and all implementations of the Fivetran public REST API returned to normal.

This incident has been resolved. We have observed that instance rates are returning to normal levels, and affected connectors are syncing successfully. Incident Summary Description: We identified an issue that caused delays in connection syncs and transformations. During this time, new syncs in AWS and Azure were delayed, while ongoing syncs continued to run normally. In the GCP regions, ongoing syncs failed, and new syncs were delayed. Timeline: This issue began on March 15, 2026, at 6:20 UTC and was resolved on March 15, 2026, at 11 UTC. Cause: A recent infrastructure change caused the issue. Resolution: A fix was implemented to address the infrastructure issue, after which instance rates returned to normal, and affected connectors resumed syncing successfully.

This incident has been resolved. We have observed that instance rates are returning to normal levels and affected connectors are syncing successfully. Incident Summary Description: We identified an issue for Qualtrics which resulted in syncs failing with HTTP Error Code 429. Timeline: This issue began on 10th March at around 10PM UTC and was resolved on 14th March at 10:30PM UTC. Cause: This issues was caused by Third party 429s for distribution History endpoint which started spiking after 10th March at around 10PM UTC, we suspect if qualtrics had enforced additional rate limits at an app level in addition to the brand level api limit of 300 per minute Resolution: We increased the retries and also added exponential sleep between api calls to reduce the load on qualtrics server

This incident has been resolved, and we have observed that the sync is now completing successfully. Incident Summary Description: We have identified an issue where some HubSpot connector syncs are failing with the error Failed to sync 1 endpoint(s) with error: {CAPTURE_DELETES_USING_WEBHOOKS=com.fivetran.platform.interfaces.security.credentials.RefreshCredentialsException: HTTP 477 } Timeline: The issue began on March 14, 2026, at 09:01 UTC and was resolved on March 14, 2026, at 10:43 UTC. Cause: The recent migration of the Fivetran public app by HubSpot has caused the issue. Resolution: A fix has been implemented to resolve the issue.