Astra Serverless: Sporadic read failures in GCP

This incident has been resolved.

All of the databases have been recovered. We are continuing to closely monitor the situation.

Restoration efforts in the us-central-1 region continue to make steady progress. Most affected databases have been fully operational. A few remaining instances are undergoing final recovery steps. Our teams are closely monitoring all systems to ensure stability, and we will continue to share updates as recovery progresses.

We are continuing restoration efforts in the us-central1 region. Most of the impacted databases are now fully operational, while a few are still undergoing recovery steps. We are actively monitoring all remaining instances and will share updates.

Restoration efforts remain ongoing as we continue targeted cleanup activities. While we’ve made significant progress and many databases have been successfully recovered, a few in the us-central1 region are still in the process of restoring. We will continue to provide updates as we reach key milestones.

Restoration efforts are ongoing as we continue targeted cleanup activities to fully recover all databases across GCP regions. Although we have made progress, the work is still in progress. We will provide further updates as we reach key milestones.

We are continuing to work through cleanup efforts to restore all databases across all GCP regions.

We are bringing GCP us-central1 back online now to allow traffic. Multiple efforts are underway to assess the impact on all GCP databases and ensure everything is working as expected. Some databases are still experiencing timeouts or unavailable exceptions at this time.

We are seeing a recovery in all regions but us-central1. We are continuing to work with GCP for this region and will bring the region back online once the GCP issue is resolved.

We are pausing all traffic to the GCP us-central1 region to protect against potential data loss due to failure to interact with the remote storage. Once GCP has recovered in this region, we will be bringing the region back online.

Here is the GCP incident: https://status.cloud.google.com/incidents/ow5i3PPK96RduMcb1SsW

Please be aware that a global GCP-impacting event is impacting Astra databases hosted in GCP. We are working with GCP and are closely monitoring the situation.

We are currently investigating an issue with sporadic read failures in Astra Serverless databases in GCP. This issue is isolated to GCP only.