Keboola - Notice history

Resolved – Snowflake Service Restored

Snowflake has completed mitigation efforts and has restored service for the affected AWS US West (Oregon) region. The vendor has reported that processing efficiency improvements have been implemented and service health has returned to normal.

Following a period of monitoring, we have resumed all flow scheduling and processing activities. Backlogged workloads have been successfully processed, and we are no longer observing the internal Snowflake errors that were causing storage job failures.

At this time, all systems are operating normally and this incident is considered resolved.

Root Cause

The incident was caused by a Snowflake service disruption affecting AWS US West (Oregon). According to Snowflake, elevated load in a critical internal coordination service impacted request processing for a subset of workloads, resulting in query failures and internal SQL execution errors.

We will continue to monitor platform health, but no further customer impact is expected.

We are seeing significant improvements in service health and are no longer observing the Snowflake internal errors that were previously causing storage job failures.

Based on these improvements, we have resumed flow scheduling and processing. Queued work is currently being processed, and the platform is working through any backlog that accumulated during the incident.

Snowflake has reported that they are continuing to investigate the underlying cause of the issue and are implementing mitigations to improve processing efficiency within their infrastructure. While service appears stable at this time, the vendor incident remains open and an official resolution has not yet been declared.

We will continue to closely monitor the environment for any recurrence of errors or performance degradation and will provide additional updates as more information becomes available.

Snowflake continues to work on resolving the underlying service incident that is causing storage job failures and internal errors.

To prevent further impact and avoid additional failed processing, we have temporarily suspended scheduling of new flows. Existing queued work will not be scheduled until the Snowflake incident is resolved.

Once Snowflake confirms service restoration and we have validated system stability, we will resume flow scheduling. Processing will automatically continue from the point at which scheduling was suspended, and queued work will be backfilled to catch up on any delayed executions.

We are actively monitoring the situation and coordinating with Snowflake. We will provide further updates as more information becomes available.

This incident is now resolved. The affected project owners have been contacted directly.

We've identified the affected projects and disabled the Branched Storage on Snowflake feature there. Owners of the affected projects are being contacted.

The issues affecting our Jira-powered support widget have been resolved. Atlassian has confirmed that their services are fully operational.

Tickets submitted during the outage were delivered with a delay and we have responded to all of them. If you have not received a response, please re-submit your request.

We apologize for the inconvenience. The issue was on the third-party vendor's side, not in the Keboola platform.

Atlassian pipeline pipeline was presumably fixed now and we're starting to gradually receive submitted tickets. We apologize for the delay in processing. We'll re-enable the support widget once we confirm it works reliably. Next update in 8 hours.